· isfp iowa strengthening family program jag joint advisory group kav knowledge, attitude and...

��

��

��

� � � � � � � � �

��

��

��

� �� !�"#�

�� !� �� "��#$��%��&��"��'�� (��)�!�� '� ��*��

�� + �� ' ��,� ��-�� "�� "�� "��

�� !��!�.

"�� # ��)��"� �� ,�� *� ��"$�� "/�

��$ #�$��!�� '-� ��/� � ��

��

��

��

© Commonwealth of Australia 2004

ISBN: 0 642 82482 7

This work is copyright. Apart from any use as permitted under the Copyright Act 1968, no partmay be reproduced by any process without prior written permission from the Commonwealthavailable from the Australian Government Department of Communications, InformationTechnology and the Arts. Requests and inquiries concerning reproduction and rights should beaddressed to the Commonwealth Copyright Administration, Intellectual Property Branch,Australian Government Department of Communications, Information Technology and the Arts,GPO Box 2154, Canberra ACT 2601 or posted at <http://www.dcita.gov.au/cca>.

The National Drug Research Centre and the Centre for AdolescentHealth

The Prevention of Substance Use, Risk and Harm in Australia: a reviewof the evidence

Wendy Loxley, John W Toumbourou, Tim Stockwell, Ben Haines, KatieScott, Celia Godfrey, Elizabeth Waters, George Patton, Richard Fordham,Dennis Gray, Jann Marshall, David Ryder, Sherry Saggers, Lena Sanci andJo Williams

with contributions by

Susan Carruthers, Tanya Chikritzhs, Simon Lenton, Richard Midford,Pamela Snow and Catherine Spooner

Publication approval number: 3475/JN8607

Design: Themeda

Printing: Union Offset

This research paper does not necessarily reflect the policy position of the AustralianGovernment.

��

��

��

��

��

� ��

��

�� !�"# $�%&'� !% !�'(�)� * ��

��+��,-".��"/�!��"#"'% .� �� 0

�� !!�%#,�"/�1%&'�&,�� #1�1%&'2%�$ !�1�� %3 �� 0

��0��%,.�-!�4�,�"#�.%�4�#!�"# �� 5

��6��-"#"3�-��4 $& !�"#,�"/� $-"�"$� #1�1%&'��#!�%4�#!�"#, �� 7

��5��$�' $�,! !&,�"/�1%&', �� 8

��+��

+�� &33 %( ��

+�+��#!%"1&-!�"# ��

+�� !!�%#,�"/�&,� ��

+�0�� %3, ��

+�6� "-� $�1�!�%3�# #!, ��+

+�5��,9� #1�.%"!�-!�4��/ -!"%, ��+

+�7��#!�%4�#!�"#,�$�!�% !&%� ��

� ��

�� :�� ;�� <

�� &33 %( �� <

��+��#/$&�#-��"/�. %�#!,=�1%&'�&,��1&%�#'�.%�'# #-(� #1�� %$(�-��$1�""1 ��<

�� !!�%#,�"/�1%&'�&,�� #1�� %3� 3"#'� 1"$�,-�#!, ��+�

��0��"#'�!&1�# $�. !!�%#,�"/�&,��/%"3� 1"$�,-�#-��!"� 1&$!�""1 ��+6

��6��"#'�!&1�# $�. !!�%#,�"/�� %3�/%"3� 1"$�,-�#!�1%&'�&,� ��+7

�- ��

��0�� :�� ;�� 2��

0�� &33 %( ��

0�+�� !!�%#,�"/�1%&'�&,�;�%�,9� #1�� %3��#�!��&,!% $� #�.".&$ !�"# ��+

0��1&$!�. !!�%#,�"/�1%&'�&,�;�%�,9� #1�� %3 ��6

0�0�� !!�%#,�"/�1%&'�&,�;�%�,9� #1�� %3� 3"#'�"$1�%��&,!% $� #, ��0�

0�6�� !!�%#,�"/�1%&'�&,��%�,9� #1�� %3� 3"#'��#1�'�#"&,��&,!% $� #, ��06

0�5�� !!�%#,�"/�1%&'�&,��%�,9� #1�� %3� 3"#'�-&$!&% $$(� #1�$�#'&�,!�- $$(1�4�%,��)��*��&,!% $� #, �� 0<

0�7�� ,,"-� !�"#�"/�1%&'�&,�� #1�3�#! $�� $!��.%">$�3, ��6?

0�8�� !!�%#,�"/�1%&'�&,�;�%�,9� #1�� %3� 3"#'�."$�-��1�! �#��,� #1�.%�,"#�%, ��6�

0�<��3. -!,�"#�!��>%" 1�%�-"33&#�!( ��6+

� �� !�� "� �� #�

��6�� :�� 5�

6�� &33 %( �� 5�

6�+��#!%"1&-!�"# �� 5�

6��,"-� $�1�!�%3�# #!,�"/�� $!� ��5+

6�0��"#-$&,�"# ��5<

��5�� :�� :�� 7�

5�� &33 %( �� 7�

5�+��#!%"1&-!�"# �� 7�

5��%�4�#!�"#�,-��#-�� 1�4�$".3�#! $�. !�@ (,� ..%" -��!"�.%�4�#!�"# ��7+

5�0�A& $�!(�"/��4�1�#-� ��7�

5�6��%�"%�!"�>�%!� �� 70

5�5��#/ #-(� #1�.%�2,-�""$ ��77

5�7��%�3 %(�,-�""$�) '�,�6�!"��* �� 78

5�8� �-"#1 %(�,-�""$�) '�,��+�!"��7* ��8?

5�<��1&$!�""1�) '�,��8�!"�50* �� 85

5��?��!�%�3�#!� #1�"$1� '��) '�,�56B* �� 88

5��-&3&$ !�4��#/$&�#-��"/��$�4 !�1�%�,9�/ -!"%,� #1�1�.%�,,�1.%"!�-!�4��/ -!"%, ��88

5��+��%��,�"&$1�@��! %'�!�.%�4�#!�"#��//"%!,C��,9;�.%"!�-!�"#� #1�!��.%�4�#!�"#�. % 1"� ��<?

-�� !"��#�$%�

� ��$�� $�� !%��!�� !��

��7�� <6

7�� &33 %( �� <6

7�+��%�"%�!"�>�%!� �� <5

7��#/ #-(� #1�� %$(�-��$1�""1�) '�,�?�!"�0* ��<7

7�0��%�3 %(�,-�""$�) '�,�6�!"��* �� ?+

7�6��"#-$&,�"#,D-��$1�""1��#!�%4�#!�"# �� ?

��8�� :��

8�� &33 %( ��

8�+��1"$�,-�#-��) '�,��+�!"�+0* ��

8�� -�""$2> ,�1�1%&'��1&- !�"# �� 8

8�0� -�""$�"%' #�, !�"#� #1�>�� 4�"&%�3 # '�3�#! �� +5

8�6��%��#!�%4�#!�"#� #1�.��%��1&- !�"# �� +7

8�5��"&!��,."%!� #1�%�-%� !�"#�.%"'% 3, �� +<

8�7��#!"%,��. ��

8�8��"33&#�!(2> ,�1�1%&'��1&- !�"# ��

8�<��%�4�#!�4��- ,��3 # '�3�#! �� +

8��?��"33&#�!(�3">�$�, !�"# ��

8�� $!��,�%4�-��%�"%��#! !�"# �� 5

8��+��3.$"(3�#!� #1�!% �#�#' �� 0?

8�� "-� $�3 %9�!�#'��#!�%4�#!�"#, �� 0?

8��0�� @;�%�'&$ !�"#� #1�."$�-�#' �� 06

8��6��"#-$&,�"#,D�#!�%4�#!�"#�!�%"&'�� 1"$�,-�#-� �� 05

��<��2�� 07

<�� &33 %( �� 07

<�+��#!%"1&-!�"# �� 08

<��%" 12> ,�1�.%�4�#!�"#��#�!� !�4�,�/"-&,�#'�"#�-��$1%�#� #1�("&!� �� 08

<�0��%" 12> ,�1�.%�4�#!�"#��#�!� !�4�,�/"-&,�#'�"#� 1&$!, �� 6?

<�6��#! $�� $!��.%"3"!�"#� #1�.%�4�#!�"# �� 6�

<�5��"33&#�!(��3.%"4�3�#! �� 65

<�7��"#-$&,�"#,D>%" 12> ,�1�.%�4�#!�"# �� 65

-� ��

� ��&��$�� $��!%�� !!� '��'�� &(

��?�� 6<

�?�� &33 %( �� 6<

�?�+��%� !3�#!�%�,."#,�,�!"� $-"�"$� #1�"!��%�1%&'�.%">$�3, �� 5�

�?�� $!��,�%4�-��%�"%��#! !�"#D>%��/��#!�%4�#!�"#, �� 56

�?�0��"33&#�!(2> ,�1�.%�4�#!�"#�.%"'% 3, �� 55

�?�6��%"'% 3,��#��#1�'�#"&,�-"33&#�!��, �� 5<

�?�5��%"'% 3,�!"�%�1&-��1�3 #1� 3"#'�!��$1�%$( �� 7�

�?�7��"%9.$ -��.%"'% 3, �� 7+

�?�8� "-� $�3 %9�!�#'D ! !�E# !�"# $�3�1� �- 3. �'#,;� #1�>""9$�!�- 3. �'#, �� 76

��:��:�� 8�

�� &33 %( �� 8�

��+��"> --" �� 85

��$-"�"$ �� 87

��0�� %3 -�&!�- $, �� <+

��6��"$ !�$��,&>,! #-�, �� <�

��5��%/"%3 #-�� #1��3 '��#� #-�#'�1%&',�)��,* �� <0

��+��:��:�� <6

�+�� &33 %( �� <6

�+�+��%"$��"/�$ @��#/"%-�3�#!��#�%�1&-�#'�1�3 #1��#�!��-"33&#�!( �� <5

�+��%"$��"/�$ @��#/"%-�3�#!��#�%�1&-�#'�1�3 #1� 3"#'�&,�%, �� <8

�+�0� &..$(2,�1��1%&'�$ @��#/"%-�3�#! �� +?�

�+�6��,� %-�� #1��4 $& !�"#�#��1, �� +?<

�+�5��"@�'""1��,�!��4�1�#-�C �� +�?

��F�� +�+

�� &33 %( �� +�+

��+��%"-�1&%�,�/"%�1�4�%!�#'�("&#'�"//�#1�%,��#!"�� %$(��#!�%4�#!�"#�,�%4�-�, �� +��

��4�%,�"#�.%"'% 3,��#�!��'�#�% $�-"33&#�!( �� +�0

��0�%&'�-"&%!, �� +�5

��6��%"'% 3,��#�.%�,"# �� +�8

��0�� +++

�0�� &33 %( �� +++

�0�+�� %3�%�1&-!�"#��#��&,!% $� #�1%&'�."$�-( �� ++0

�0��#�!,�"/�� %3�%�1&-!�"# �� ++6

�0�0��"> --" �� ++6

�0�6��$-"�"$� #1�� %3�%�1&-!�"# �� ++5

-�� !"��#�$%�

�0�5��1&-�#'�".� !��"4�%1",� �� +�?

�0�7��1&-�#'�� %3,� ,,"-� !�1�@�!��#G�-!�#' �� +��

�0�8�H #-��1%&',=�).��#�!�($ 3�#�,;�� ;�9�! 3�#�* �� +�0

�0�<��#I"1� I�.�#�, �� +�6

�0��?��!%"$ �� +�5

�0�� %3�3�#�3�, !�"#�1%&'��1&- !�"# �� +�5

�0��+��"#-$&,�"#, �� +�5

� ��#��%�� !�� (

��6�� :�� +�<

�6�� &33 %( �� +�<

�6�+��#!%"1&-!�"# �� +0?

�6�� !!�%#,�"/�1%&'2%�$ !�1�� %3��#��&,!% $� �)�� .!�%��* �� +0�

�6�0�� !!�%#,�"/�%�,9(�1%&'�&,��#��&,!% $� �)�� .!�%�0* �� +0�

�6�6� "-� $�1�!�%3�# #!,�"/�1%&'�&,�� #1�%�$ !�1�.%">$�3,�)�� .!�%�6* �� +0+

�6�5��,9� #1�.%"!�-!�4��/ -!"%,�)�� .!�%�5* �� +0�

�6�7��%��,�"&$1�@��! %'�!�.%�4�#!�"#��#4�,!3�#!C��,9;�.%"!�-!�"#� #1�!��.%�4�#!�"#. % 1"��)�� .!�%�5* �� +0�

�6�8��4�1�#-��> ,��/"%�1%&'�.%�4�#!�"#�)��0* �� +00

�6�<��#1�'�#"&,�.�".$�� #1�1%&'�.%�4�#!�"# �� +08

�6��?��"#-$&,�"#, �� +0<

�� &

�..�#1��"> --"D 1"$�,-�#!, �� +60

�..�#1��$-"�"$D 1"$�,-�#!, �� +65

�..�#1�� ## >�,D 1"$�,-�#!, �� +6<

�� #

� �!��

� >$��-$ ,,�/�- !�"#�"/�. !!�%#,�"/�%�,9(�1%&'�&,�� #1�� 3.$�,�"/�.",,�>$�� ,,"-� !�1� %3,�!"�� $!�;�, /�!(� #1�@�$$>��#' �� 6

� >$��+��$ ,,�/�- !�"#,�"/�!��$�' $�,! !&,�"/�1%&',�"/�-"#-�%#��#��&,!% $� �� 8

� >$��%�4 $�#-��"/�1�//�%�#!�1%&'2&,�#'�. !!�%#,��#��&,!% $� #,� '�1�>�!@��#�0� #1��<�(� %,�� +??� ��+0

� >$��+��%�4 $�#-��"/��$$�' $�1%&',��4�%�&,�1�>(��+�!"��7�(� %�"$1, ��+0

� >$��%�4 $�#-��"/�1%&'�&,��#�$ ,!�3"#!��/"%��+E�� #1��5E�7�(� %�"$1, ��+6

� >$��0��,!�3 !�1�1%&'2- &,�1�1� !�,�)�<<8*;��%,"#�� %,�"/��/��",!�)�<<8*;�",.�! $�,�. % !�"#,�)�<<7E<8*� #1��-"#"3�-�-",!,�)�<<8E<<*�/"%!"> --";� $-"�"$� #1��$$�-�!�1%&', ��

� >$��0�+��%�4 $�#-��"/�1%&'�&,��#�$�/�!�3�� #1�$ ,!��+�3"#!�,�/"%� $$�.�%,"#,�� +??� ��0

-��

� >$��0��%�4 $�#-��"/�1�//�%�#!�1%&'�&,�#'�. !!�%#,��#��&,!% $� #,�"4�%�!�� '�"/�5?�� +??� ��00

� >$��0�0�%&'�&, '��#�#"#2�#'$�,��,.� 9�#'�> -9'%"&#1,�/%"3��<<8�� 0<

� >$��0�6��"2"--&%%�#-��"/�,&>,! #-��&,�� #1�3�#! $�� $!��1�,"%1�%, ��6?

� >$��0�5��%�4 $�#-��"/�,&>,! #-��&,��#�.�".$��@�!�� .,(-�"!�-�1�,"%1�% ��6?

� >$��0�7�� !�,�"/�.",�!�4��!�,!,��#��+??� ��6�

� >$��0�8�� !�,�"/�1%&'�&,��#�,�#!�#-�1�3 $��"//�#1�%,��#�� 6�

� >$��5��.%�4 $�#-�� #1�%�$ !�4��%�,9,� ,,"-� !�1�@�!��#1�4�1& $�%�,9� #1.%"!�-!�4��/ -!"%,� #1�!��&,��"/�!"> --"� #1� $-"�"$��#�!��$ ,!��?�1 (, ��8<

� >$��+�� "&%-�,�"/�,&..$(�"/�3 G"%��&,!% $� #��$$�-�!�1%&', �� +?�

� >$��6��>G�-!�4�,�/"%�1�//�%�#!�."$�-(�G&%�,1�-!�"#,� #1�".�% !�"# $�,�!!�#',�@�!��#� �%"!�-!�"#� #1��,9��1&-!�"#��..%" -��!"��%�4�#!�"#� �� +6?

�..�#1��"#,�J&�#-�,�"/�("&!��!"> --"�&,��/"%�$ !�%�>�� 4�"&%� #1�� $!� #1�,"-� $�/&#-!�"#�#' �� +60

�..�#1��:"$$"@2&.�,!&1��,�� 3�#�#'�1�4�$".3�#! $�-"#,�J&�#-�,�"/ 1"$�,-�#!� $-"�"$�&,��>�� 4�"&%, �� +65

�..�#1��:"$$"@2&.�,!&1��,�� 3�#�#'�1�4�$".3�#! $�-"#,�J&�#-�,�"/ 1"$�,-�#!�- ## >�,�&,��>�� 4�"&%, �� +6<

��'��

:�'&%�� (,!�3,�3"1�$�/"%�!��.%�4�#!�"#�"/� $-"�"$� #1�"!��%�1%&'�.%">$�3, �� <

:�'&%��+��,9� #1��%"!�-!�"#�3"1�$�/"%�!��.%�4�#!�"#�"/�1%&'2%�$ !�1�� %3 ��?

:�'&%��5��%"."%!�"#�"/�@��9$(�,3"9�%,��#�(� %,�7;�<� #1��>(�$�4�$,�"/�%�,9 �� <�

:�'&%��5�+��%"."%!�"#�"/�>�#'��1%�#9�%,��#�(� %,�7;�<� #1��>(�$�4�$,�"/�%�,9 ��<�

:�'&%��5��%"."%!�"#�"/�@��9$(�- ## >�,�&,�%,��#�(� %,�7;�<� #1��>(�$�4�$,�"/�%�,9 �� <+

:�'&%��5�0��%"."%!�"#�"/�@��9$(�&,�%,�"/�"!��%��$$�-�!�1%&',��#�(� %,�7;�<� #1��>(�$�4�$,�"/�%�,9 ��<+

:�'&%��6��,&33 %(�"/�3 �#�- !�'"%��,�"/��#!�%4�#!�"#�%�-"33�#1�1�/"%-"#!�#&�1� #1��#� #-�1��#4�,!3�#! �� +6�

�� !"��#�$%�

�� This work could not have been accomplished without the input, expertise and support of colleagues. Weparticularly acknowledge the following for their input into specific areas.

Monica Barratt Information management

Susan Carruthers Injecting drug use; blood-borne viruses

Tanya Chikritzhs Alcohol epidemiology and policy

Fran Davis Administrative support

Richard Fordham Economics of illicit drug use

Dennis Gray Indigenous issues; social determinants

John Hall Editing and proof reading

Simon Lenton Harm reduction; legal status of cannabis

Jann Marshall Early childhood interventions

Richard Midford Community interventions; workplace; media

David Ryder Multicultural substance use and interventions

Sherry Saggers Indigenous issues; social determinants

Lena Sanci Drug education and health service intervention

Pamela Snow Youth recreation and community programs

Catherine Spooner Early childhood epidemiology and interventions

Janney Wale Editing and proof reading

Jo Williams Adolescent tobacco use and interventions

Susan Wilson Formatting and secretarial support

Early versions of some of the material presented in this report were presented at the Preventing SubstanceUse, Risky Use and Harm Conference in Perth (February, 2003) and through the Victorian Drug InfoClearinghouse.

(�/� ��-�� .

We have also been advised and assisted by the following experts to whom we owe considerable thanks. Itshould be emphasised that any omissions or errors of interpretation are the responsibility of the authors.

Edward Helmes; David Hill; Toni Makkai; Richard Mattick; Vicki White; Alex Wodak

Finally, we would like to thank the Australian Government Department of Health and Ageing and theIntergovernmental Committee on Drugs for entrusting us with this important project. We hope that thefinal product will assist with the development of more effective prevention strategies and policies inAustralia.

� ��

AA Alcoholics Anonymous

ABCI Australian Bureau of CriminalIntelligence

ABS Australian Bureau of Statistics

ACS Australian Customs Service

ADCA Alcohol and Other Drugs Council ofAustralia

ADHD Attention Deficit Hyperactivity Disorder

ADIA Australian Drug Intelligence Assessment

AFP Australian Federal Police

AIDR Australian Illicit Drug Report

AIHW Australian Institute of Health andWelfare

ALRC Australian Law Reform Commission

AMPS Alcohol Misuse Prevention Study

ANAO Australian National Audit Office

ANCD Australian National Council on Drugs

ARISIT Action Research Intervention andSystem Improvement Team

ATP Australian Temperament Project

ATS Amphetamine-type stimulants

AUSTRAC Australian Transaction Reports andAnalysis Centre

BBV Blood-borne virus

BCR Benefit-cost ratio

BEST Behavioural Exchange Systems Training

CALD Culturally and linguistically diverse

CAP Community Action Project

CAR Children at Risk

CBA Cost-benefit analysis

CEA Cost-effectiveness analysis

COAG Council of Australian Governments

COMPARI Community Mobilisation for thePrevention of Alcohol-Related Injury

CPI Community Partnerships Initiative

CTP Community Trials Project

DAMEC Drug and Alcohol MulticulturalEducation Centre

DARE Drug Abuse Resistance Education

DOFA Department of Finance andAdministration

DPP Director of Public Prosecutions

DUCO Drug Use Careers of Offenders

DUMA Drug Use Monitoring in Australia

EAP Employee assistance program

ETS Environmental tobacco smoke

FAS Foetal alcohol syndrome

FAST Families and Schools Together

FEI Family Empowerment Intervention

FFT Functional Family Therapy

GBL Gamma butyrolactone

GHB Gamma hydroxybutyrate

GSAP Good Sports Accreditation Program

HIC Health Insurance Council

HSPP Hutchinson Smoking Prevention Project

IDEP Illawarra Drug Education Program

IDRS Illicit Drug Reporting System

IDU Injecting drug user

IGCD Intergovernmental Committee on Drugs

ISFP Iowa Strengthening Family Program

JAG Joint Advisory Group

KAV Knowledge, attitude and values

LECP Law Enforcement Cooperation Program

LIFT Linking the Interests of Families andTeachers

LSD Lysergic acid diethylamide

LST Life Skills Training

��

�� !"��#�$%�

MCDS Ministerial Council on Drug Strategy

MDMA Methylenedioxymethamphetamine

MMT Methadone maintenance therapy

MSIC Medically supervised injecting centre

MSO Most serious offence

MST Multisystemic Treatment

NAIP National Alcohol Indicators Project

NDRI National Drug Research Institute

NDS National Drug Strategy

NDSHS National Drug Strategy HouseholdSurvey

NESB Non-English speaking background

NHMRC National Health and Medical ResearchCouncil

NHPA National Health Priority Area

NHSP National Heroin Signature Program

NIDC National Illicit Drugs Campaign

NIDS National Illicit Drugs Strategy

NRT Nicotine replacement therapy

NSSDS National Secondary School Drug Survey

NSMHWB National Survey of Mental Health andWell-being

NSP Needle and syringe programs

NTC National Tobacco Campaign

OMCG Outlaw motorcycle gangs

OTC Over-the-counter

PACE Parenting Adolescents a CreativeExperience

PATHS Promoting Alternative ThinkingStrategies

PDFY Preparing for the Drug Free Years

PED Performance and Image EnhancingDrugs

PMA Paramethoxyamphetamine

POC Proceeds of Crime

PYLL Person Years of Life Lost

RBT Random breath testing

RCT Randomised controlled trial

ROI Return on investment

RSAP Residential Student Assistance Program

SES Socioeconomic status

SHAHRP School Health and Alcohol HarmReduction Project

SIDS Sudden Infant Death Syndrome

SNAP Smoking Nutrition Alcohol and PhysicalActivity

SSDP Seattle Social Development Project

TAC Transport Accident Commission

TAFMI Targeted Adolescent/FamilyMultisystems Intervention

TC Therapeutic Community

TOPS Treatment Outcome Prospective Study

UATSIPS Urban Aboriginal and Torres StraitIslander Peoples Supplement

UN United Nations

UNDCP United Nations Drug ControlProgramme

VAHC Victorian Adolescent Health Cohort

WHO World Health Organization

ZTP Zero Tolerance Policy

��

�� !"��#�$%�

This Monograph has trawled a vast sea ofinformation relevant to drug use, risk and harm inthe Australian community, to capture the essentialwisdom that can guide the national preventionagenda. The document has been prepared toprovide an evidence base for a comprehensivenational prevention agenda to be implementedalong with synergistic actions across multiplegovernment departments and sectors of society.Prevention refers to measures that prevent or delaythe onset of drug use as well as measures thatprotect against risk and reduce harm associated withdrug supply and use. The Monograph encompassesthe full spectrum of prevention interventionmeasures; evaluated Australian approaches to theprevention of drug supply, use, and harm;approaches to prevent or delay the uptake of licitand illicit drugs by children and young people;current application of prevention policy and strategyin Australia; and gaps in prevention knowledge andeffort.

To undertake this task, we have identified the majorpatterns of use, risk and harm accruing to drug usein Australia using the most contemporary data;overviewed the social and structural determinants ofhealth and drug use, internationally and inAustralia, with a particular concern for IndigenousAustralians; reviewed the literature on risk andprotective factors relating to drug use and otherpsychosocial problems; reviewed the literaturerelating to the evidence for a range of preventionstrategies, both in Australia and internationally;included every drug type identified by the NationalDrug Strategy (NDS); where possible, distinguishedbetween different age groups, from conception toold age; and considered the implications of thesefindings.

Our major conclusion is that an integrative policyframework, which we refer to as a Protection andRisk Reduction Approach to Prevention, should beadopted. This approach integrates knowledge ofdevelopmental processes throughout the life-course,with knowledge of broader macro-social influenceson behaviour and health outcomes. The Protectionand Risk Reduction Approach emphasises theimportance of reducing the known developmentalrisk factors that lead children and young people tobecome involved with risky drug use and harm,while also enhancing protective factors. While thereis evidence that interventions in the early yearsmake a difference later in life, it is also the case that

risk factors for harmful drug use arise in childhood,adolescence and later in life and reductions to drug-related harm may be limited if preventioninvestment were to exclude risks emerging at otherlife stages. The framework also acknowledges thattargeted, early intervention strategies focused onstrengthening protective factors will be useful forchildren and youth with a high number ofdevelopmental risk factors. The approach alsoemphasises brief interventions, treatment and harmreduction strategies, acknowledging that suchstrategies can reduce drug-related harm for drugusers who have a high number of risk factors, whilealso improving developmental opportunities forchildren. Law enforcement is an essential element ofthis approach to prevention not just in controllingthe supply of drugs, but also in influencingcommunity values about drug use, diverting earlyoffenders and acting to protect the community fromcrime and social disorder.

Central to the thesis of this review is that acomprehensive Prevention Agenda needs to beguided by a shared understanding of the nature ofthe harms to be prevented and, correspondingly, ofthe underlying patterns of risky drug use. Theconcept of ‘risk’ has been expanded to include moredistal risk factors, including social and demographicfactors as well as early developmental risk (andprotection) factors. The first substantive part of thereview overviews current information about thenature and prevalence of the most serious harmsassociated with different patterns of use of differentsubstances.

Other models that have informed the work includethe Public Health Systems model, which illustratesthe levels of increasing breadth and complexity atwhich prevention activities can be focused, fromwork with individuals to national and internationalapproaches. This model also conceptualisesdeterminants of health and drug use on acontinuum from macro to micro: social andstructural determinants are distal influences; riskand protection factors are more proximal. Themodel is both top-down and bottom-up: the macroclearly influences the micro, but equally clearly themicro influences the macro. Activity at any one ofthe levels can influence not only that level but,indirectly, all other levels. This approach allows themapping of systems, pathways and strategies thatconnect among and between risk factors, protectivefactors and drug use outcomes.

��

��- ��

To better organise the systems approach toprevention, we have emphasised the localcommunity as one of the primary levels forintegrating and coordinating planning within aProtection and Risk Reduction Approach toPrevention. An emphasis on the local communityoffers prospects for addressing some of the broadsocial determinants related to both socialdisadvantage and disconnection that underlie aspectsof drug-related harm. A structured approach to localcommunity organisation also offers a promisingmethod for the coordinated application of evidence-based prevention strategies aimed at the reductionof developmental risk factors and enhancement ofprotective factors. By emphasising the communitylevel, the implications of the systems model for acoordinated approach to prevention across differentjurisdictions become clearer.

The review found that tobacco is the leading causeof premature death and hospitalisation among allAustralians. However, alcohol causes the deaths andhospitalisation of slightly more children and youngpeople than do all the illicit drugs combined, andmany more than does tobacco. There are likelyfuture health costs associated with current drug use,including the costs of hepatitis C among injectingdrug users in Australia. There are also known to bea range of social harms impacting on individualusers of illicit drugs who receive criminalconvictions.

The main features of risky drug use patterns inAustralia follow.

� There has been a dramatic reduction in levels ofsmoking in Australia in recent decades, butsmoking rates by young people and youngwomen in particular have been less resistant tochange and are a concern for future levels oftobacco-caused mortality and morbidity.

� Alcohol consumption in Australia has recentlyincreased slightly overall and more markedlyamong young people. Two-thirds of PersonYears of Life Lost through risky alcohol use aredue, at least partly, to the short-term or acuteeffects from alcohol intoxication.

� Cannabis is the most widely used illicit drug inAustralia, though its use may have declined veryrecently. Around 10% of people become regularheavy users of cannabis and risk long-termhealth consequences and dependence. Cannabisuse during adolescence is associated with latermental health and conduct problems, thoughthe causal processes remain unclear.

� Injection is the main risk behaviour in relationto health-related harms from other illicit druguse. The 2001 National Drug StrategyHousehold Survey (NDSHS) indicated that lessthan 2% of the adult population reportedinjecting illicit drugs at some time in the last 12months. Injection is associated with the risk ofdependence, opiate overdose and thetransmission of blood-borne viruses (BBV).

� Heavy ‘binges’ on amphetamine-type drugs areassociated with reckless and aggressivebehaviour and, when sustained over days, mayprecipitate a psychosis.

� There are marked temporal and developmentalsequences concerning the ages of first use andthe order of onset of drugs. The mechanisms bywhich legal drugs serve as ‘gateways’ in somesense for illegal drugs are not clear. Earlyadolescent use of cannabis significantlyincreases the risk of later use of other illicitdrugs, but nonetheless, only around 10% ofcannabis users progress to use other illicit drugs.

It is clear that patterns of drug use and related harmsare not distributed randomly across the populationbut that there are defined groups in contemporaryAustralia that are over-represented in the statistics.These groups are usually also those that are over-represented in statistics on general ill-health. Firstly,across all drug types, being male and being youngare each independently highly predictive ofinvolvement in risky drug use and harm. Secondly,almost any measure of disadvantage will besimilarly associated with increased risk and harmfrom drugs, regardless of gender and age. Theassociation of drug use and measures of socialdisadvantage is strongest for the illicit drugs versusthe licit and also for more intensely problematicpatterns of drug use, including dependence. Relatedto findings of social disadvantage, there areindications that social disconnection is increasinglya modern driver underlying drug-related harm.

Comprehensive reviews of longitudinal and otherstudies examining significant influences on the druguse of young people have identified factors such asfamily functioning, school performance, peerinfluences, temperament and local drug availabilityas predictive of who will use drugs. These variableshave been combined to form overall surveymeasures of risk for, and protection against, avariety of problem behaviours including drug use.Children with high scores on the risk scale and lowscores on the protection scale are more likely todrink in a risky fashion, smoke, use illicit drugs,

�-�� !"��#�$%�

experience mental health problems and exhibitconduct disorder. Evidence suggests that earlyinitiation and frequent youth drug use is mostclearly predicted by the cumulative number ofelevated risk factors, rather than by any specific riskfactor.

Further analysis of a major Australian data set onrisk, protection and adolescent problem behaviourswas conducted to inform the present project. Thisanalysis suggests that ‘whole of population’ oruniversal strategies are of particular importance inrelation to reducing the more prevalent harmsassociated with tobacco and alcohol use. However,strategies targeted to high-risk children andadolescents may be necessary to prevent the harmsassociated with illicit drug use. As high-risk youthgenerally have high levels of drug use, the moretargeted strategies will also benefit the prevention ofharms associated with legal drugs and cannabis. Tomaximise their effectiveness, targeted strategiesshould be initiated early in the developmentalpathway and aimed to reduce risk factors, enhanceprotective factors and to prevent or delay drug use.

Evidence is summarised for the relative effectivenessof interventions and policies from pre-conceptionthrough to prenatal care, antenatal care, infancy,pre-primary, primary school, adolescence, youngadulthood, adulthood and old age. The quality ofthe research was highly variable; the types ofoutcomes examined ranged from known risk factorsfor later drug use, to age of onset of use of differentdrugs, to intensity of drug use, dependence andexperience of problems relating to drug use. Insome areas, it was possible to state whether or notwide implementation could be confidentlyrecommended; in others there were theoreticalreasons to recommend that interventions should betrialled; and in others there was no relevantliterature upon which to draw.

The evidence for investment in early life-stageinterventions suggests a range of opportunities forencouraging healthy child development and therebypreventing children’s drug use and progression toheavy and harmful use. Prior to birth and also inchildhood, the healthy development of children canbe impaired through parental tobacco, alcohol andillicit drug use. Further innovation investment willbe required to develop and evaluate health servicereorientation programs that can be more effectivelyapplied to address these problems.

There is an emerging evidence base for universalinterventions focusing on adolescent use of alcoholand tobacco. A combination of well-designed and

executed regulatory approaches, supported by othercomponents such as school-based interventions,holds the most promise. There is very strongevidence and a sound rationale for the enforcementof laws prohibiting sales of both tobacco andalcohol to persons under legal purchasing age forthese legal drugs. Similarly, there is evidence for theeffectiveness of measures that control the price ofalcoholic drinks favoured by young people.

Common benefits can be obtained through broad-based preventive interventions addressing a widerange of health, social and criminal problembehaviours. Firstly, there are benefits associatedwith universal programs to reduce or eliminate thesocial and developmental risk factors that predict thedevelopment of problem behaviours. Secondly,benefits can be obtained through programs thattarget individuals and groups with a high numberof developmental risk factors in settings such asdisadvantaged areas, family crisis, police and courtcontacts and mental health. Thirdly, benefits areavailable through programs for adolescents andadults who have high rates of drug-related harm.Examples include broad-based health promotioninterventions delivered by primary care healthprofessionals such as general practitioners (GPs),occupational health workers and also community-wide health screening and brief interventionprograms. The potential public health benefits forthe broad application of screening and briefintervention programs targeting a range of healthrisk behaviours has not yet been realised inAustralia.

Universal regulatory interventions for legal drugsare essential. Regulation of the supply of bothtobacco and alcohol products, supported by a rangeof public education measures, is strongly supportedin the research literature. Young people, as well asheavy drinkers and smokers, are most affected byprice increases. At present there are consistenttaxation policies in place to maintain the high priceof cigarettes in Australia and these should besustained. For alcohol there are sound policies inplace, from a public health point of view, in relationto beer and spirits. The main weakness in currentpolicy is the absence of an alcohol content-based taxon wines, resulting in the availability of very cheapbulk wines favoured by vulnerable groups andproblem drinkers. Taxation policy also encouragesthe consumption of wine-based fruit drinks(‘alcopops’) and premixed spirits that areparticularly marketed to young people.

Restrictions of sales of both alcohol and tobacco tominors can be effectively enforced. There is

�-� ��

community support for strict enforcement of theselaws but also evidence, especially in relation toalcohol, that underage youth access is relativelyeasy. Physical availability of alcohol in terms ofnumbers of outlets and hours of sale has increasedin Australia over the last decade and Australian andoverseas evidence now clearly identifies late nighttrading for hotels and nightclubs as a source ofalcohol-related violence and road trauma. Thedevelopment and enforcement of laws to punishand deter drink-driving in Australia have beenmajor successes for public health and safety withuniform laws in place across Australia.

There is strong evidence that public educationcampaigns can contribute to reductions in smokingand risky alcohol use, but usually only if theysupport other policy measures such as tax increasesand law enforcement. The community is aneffective way of organising and deliveringprevention targeted at legal drugs, especiallyalcohol, but community-based interventions thattarget structural policy change at the local level aremore effective than approaches with the lessfocused aim of community mobilisation. Thuscommunity action to restrict trading hours in highrisk communities, to increase enforcement of drink-driving and liquor laws and to restrict local alcoholavailability are reported to have achieved the mostpositive results.

In relation to the prevention of illicit drug use, therole of law enforcement is central. Laws shapecommunity values and opinions about drug use. Onthe one hand, they express social disapproval thatreinforces social norms against illicit drug use andon the other hand, they act as a deterrent againstuse. The current National Diversion Initiativedemonstrates the importance of the lawenforcement role in the apprehension of early usersand referral to education, treatment or support.Drug courts are another important way in whichdrug users are given the opportunity to deal withtheir drug use in an effective and structured way.Many programs in prison are directed towards thehigh rates of illicit drug use among prisoners.

The impact of laws prohibiting the sale, supply anduse of certain drugs is very hard to ascertain fromcurrent scientific evidence. Acknowledging thedifficulties in working in this area, a majorinvestment in research is recommended to improvethe future evidence base for illicit drugs policy.Literature evaluating the impact of changing theprecise legal status of cannabis, including someimportant Australian studies and reviews, indicatethat moving from criminal to civil penalties for use

and possession of small quantities is not associatedwith significant increases in the prevalence ofcannabis use and may reduce the social costs relatedto conviction.

There are important intersections between the aimof population-level prevention of drug-related harmand what has traditionally been considered to betreatment. There is emerging evidence thatinvestment in various forms of treatment will havebenefits in terms of community level reductions incrime, road trauma, hospital admissions and otherserious drug-related harms.

These savings can be enhanced by:

� expanding brief intervention programs thattarget smoking and risky drinking, to a widerange of primary health care, workplace andother community-based settings;

� ensuring that treatment programs offeredinclude approaches with the strongest evidencebase and that these are made widely accessible,perhaps through a greater emphasis on deliveryat the community rather than institutional level;and

� incorporating interventions to support childrenin families with drug-using parents in order toattempt to break inter-generational patterns oftransmission of problem drug use.

There have been few formal evaluations ofIndigenous intervention projects. In part, but notwholly, this is related to Indigenous concerns aboutwhat constitutes culturally appropriate indicatorsand methods for evaluation. The evaluations thathave been undertaken have reached similarconclusions to those reports and submissions thataddress substance misuse among IndigenousAustralians. The first recommendation is the need toaddress the underlying social determinants ofIndigenous inequality. This includes the call forreal, but appropriate, economic development forIndigenous people.

The second recommendation is the need forIndigenous people to be involved as equal partnersat all stages in the development and implementationof strategies to address substance misuse. There isevidence from both Australia and overseas of theefficacy of Indigenous ownership and control ofinterventions to more generally address ill-health.As important as Indigenous involvement is, it isinsufficient without adequate resourcing—the thirdtheme to emerge from recommendations to addresssubstance misuse. There has clearly been an increasein funds identified by the Commonwealth for

�-�� !"��#�$%�

expenditure on Indigenous affairs over the pastthree decades. Whatever the levels of funding,however, they have failed to meet the welldocumented needs or to remedy the social andeconomic inequalities that underlie and perpetuatethe high levels of substance misuse amongIndigenous Australians. An important component ofadequate resourcing is the building, withincommunities and community organisations, ofcapacity to continue to provide adequate andappropriate services. This includes infrastructuraldevelopment, research capabilities, and staffdevelopment and support.

Over the past three decades, government policy onIndigenous health and substance misuse hasacknowledged a link between substance misuse andunderlying social issues. However, despite thisacknowledgement, substance misuse policy andservice planning has largely been developed inisolation from policies in other portfolio areas suchas land, employment, education and housing.Furthermore, substance misuse services have beenimplemented inconsistently across different regionsof Australia as attested by the mismatch betweenregional funding allocations and population levels.Concerns over such problems underlie the fourththeme—that is, the need for a holistic andcoordinated approach that includes Indigenouscommunity-controlled organisations, all levels ofgovernment, and all sectors.

The Protection and Risk Reduction Approach toPrevention holds advantages not simply forreducing the harm to Australian society from druguse, but also for broader social improvement goals.Investments in prevention should aim to maximisethe potential for early childhood development,while also acknowledging that development andsocialisation have ongoing threads in later years.The benefits that could flow from such investmentsrange from the maximising of human potential,through to increasing productivity andachievement, with ultimate outcomes for improvingboth the wealth and wellbeing of the nation.

�-��

��

��

� �� !��"�#��$��%��

��

Efforts to monitor drug use trends, to documentdrug-related harms, and to develop and testinterventions to prevent drug use and harms havegenerated an expanding stream of informationflowing into a broader ocean of literature. For thoseseeking to extract wisdom from this broad sea oftext, the contrary currents must be navigated andcarefully designed nets applied.

This Monograph trawls the existing evidence base tocapture the essentials regarding ‘what works’ in theprevention of substance use problems. It bringstogether what is known about the prevention ofdrug use, risk and harm in the Australiancommunity with some particular foci: socialdeterminants of health; risk and protective factorsthrough the life span; developmental milestones,transitions and trajectories; and systems approachesto drug prevention. The document has beenprepared so that a comprehensive nationalprevention agenda can be implemented withsynergistic actions across multiple governmentdepartments and sectors of society.

Over the last two decades, the Australiancommunity has become increasingly concernedabout drug-related harm and the cost of drug use,both licit and illicit, which was estimated in 1998–9to be $34.4 billion.1 Laws and regulations,community education, school drug education, harmreduction programs and treatment (particularly,lately, as it relates to diversion from criminaljustice) have all addressed these concerns, withvarying degrees of effectiveness.

��

Australia has a balanced national drug policy that isbased, in large part, on the recommendations of the1977 report of the Senate Standing Committee onSocial Welfare (The ‘Baume Report’) that argued,inter alia, that total elimination of drug abuse wasunlikely.2 In a submission to the House ofRepresentatives Standing Committee on Family andCommunity Affairs, the Department of Health andAgeing stated that the overall aim was ‘to minimisethe harmful effects of drug use in Australiansociety’, or harm minimisation. This was defined as‘encompassing supply reduction strategies to disruptproduction and supply of illicit drugs, demandreduction strategies to prevent the uptake of harmfuldrug use, and harm reduction strategies to reduce

drug-related harm for individuals and communities’(p71).2 The current NDS and its forerunners werecreated with strong bipartisan political support andinvolve cooperation between the Commonwealthand State/Territory Governments as well as the non-government sector. The current Strategy runs until2003.

In 1997, the Commonwealth Government allocated$516 million over four years to the National IllicitDrug Strategy (NIDS) which is a major componentof the current NDS. NIDS provides a balanced andintegrated approach to reducing the supply of anddemand for illicit drugs. Initiatives funded includediversion programs, treatment, prevention, training,monitoring and evaluation, research and measuresto intercept more illicit drugs at the border andwithin Australia. Prevention initiatives include theCommunity Partnerships Initiative and the NationalIllicit Drugs Campaign which are reviewed in thisdocument.

Within the NDS, it has been recognised that there isa need for a systematic approach to prevention thatspans all elements and is underpinned by a practicalapproach that supports implementation ofprevention strategies. This Monograph provides theevidence base to underpin the Prevention Agendaproposed for Australia. The goal in setting theevidence base for the Prevention Agenda is toestablish an integrated map of the systems,pathways and strategies that act as interconnectionsamong and between risk factors, protective factorsand outcomes related to the prevention of drug-related harm. The aim has been to produce acomprehensive review of international scientificliterature and experience relating to prevention inthe context of drug supply, use and harm.

Prevention in this context refers to measures thatprevent or delay the onset of drug use, as well asmeasures that protect against risk and reduce harmassociated with drug supply and use. It is intendedthat the Prevention Agenda will adopt strategiesinformed by the evidence base on the uptake ofdrugs in the first instance, and reduce drug use ingeneral, as well as seek innovative approaches toharm minimisation. To ensure social investment iswell targeted, prevention priorities should bedirected particularly at drug use that is associatedwith a measurable risk of harm.

��

�

� �� !��"�#��$��%��

��

The Monograph encompasses:

� the full spectrum of prevention interventionmeasures,

� evaluated Australian approaches to theprevention of drug supply, use and harm,

� approaches to prevent or delay the uptake oflicit and illicit drugs by children and youngpeople,

� current application of prevention policy andstrategy in Australia,

� gaps in prevention knowledge and effort.

To undertake this task, we have:

� identified the major patterns of use, risk andharm accruing to drug use in Australia using themost contemporary data,

� overviewed the social and structuraldeterminants of health and drug use,internationally and in Australia, with a particularconcern for Indigenous Australians,

� reviewed the literature on risk and protectivefactors relating to drug use and otherpsychosocial problems,

� reviewed the literature relating to the evidencefor a range of prevention strategies, both inAustralia and internationally,

� included every drug type identified by the NDS,

� where possible, distinguished between differentage groups, from conception to old age,

� made recommendations for policy and research.

This is an ambitious work. We are not aware of anyother single volume that has brought together sucha diverse range of literature relating to prevention ofnot only use, but also risk and harm, from theperspective of demand, supply and harm reduction.Inevitably, there are some areas that are coveredmore thoroughly than others, that is the nature ofthe literature.

Our approach to this task has been inclusive. Boththe National Drug Research Institute and the Centrefor Adolescent Health are fortunate in employingsome of the best expertise in Australia in the areasof licit and illicit drug use, Indigenous substanceuse, and childhood and adolescent health. We havemade extensive use of the thoughts and writings ofour colleagues and experts from other institutionswho, in many cases, have drafted new material for

this Monograph. We have also sought advice fromother experts who have helped us to select materialand review drafts. We have listed these colleaguesand experts and their areas of contribution in theacknowledgements.

The remainder of this chapter introduces some ofthe concepts and terminology that were consideredimportant in the development of this Monograph.We close with an outline of the legal status of thesubstances of concern to this review.

�� !

Central to the thesis of this review is that acomprehensive Prevention Agenda needs to beguided by a shared understanding of the nature ofthe harms to be prevented and, correspondingly, ofthe underlying patterns of risky drug use. As will bediscussed further below, the concept of ‘risk’ needsto be expanded to include more distal risk factorsfor harmful patterns of drug use, including socialand demographic factors as well as earlydevelopmental risk (and protection) factors. Thefirst substantive part of the review overviewscurrent information about the nature and prevalenceof the most serious harms associated with differentpatterns of use of various substances.

There are four main categories of drug use patternsthat pose a risk of adverse health, safety, social,developmental and economic consequences, anyone or all of which may be present. These aresummarised and illustrated in Table 1.1 and areadapted from a classification system first developedfor the World Health Organization.3 The firstcategory is the mode of administration of the drug: oralingestion (eating or drinking) tends to be leastassociated with harmful consequences and alsoresults in slow absorption into the blood stream andcentral nervous system; smoking and inhalationtend to produce immediate effects, as doesinjection. Injection of drugs allows large quantitiesof drug to be absorbed within the bloodstreamalmost instantly and poses particular risks foroverdose. The second category of risky drug useinvolves intoxication, which is associated with anincreased risk of both intentional and unintentionalinjuries, but also some acute medical and sometimespsychiatric conditions. There are many seriousillnesses associated with regular and prolonged use ofsome drugs. These have been particularly welldocumented for tobacco and alcohol.

Finally, there are problems of dependence. Dependenceis currently defined in the Diagnostic and Statistical

"��

Manual of Mental Disorders, produced by theAmerican Psychiatric Association (DSM-IVR)4, as a‘maladaptive pattern of substance use, leading toclinically significant impairment or distress’. It isfurther stipulated that at least three of the followingsigns must have been evident in the previous 12months: increased drug tolerance, repeatedwithdrawal symptoms, drug use to relievewithdrawal symptoms, high salience of drug-seeking behaviour over other activities, a narrowrepertoire of drug use, subjective awareness of acompulsion to use drugs and rapid reinstatement ofsymptoms after a period of abstinence. It should benoted that this definition is an attempt tooperationalise a clinical concept developed mainlyfrom the work of Edwards and colleagues in the1970s and 1980s.5 The original formulationstressed the importance of: (i) considering drugdependence as existing on a continuum of severity,and (ii) comprising both physiological processes(neuro-adaptation to prolonged exposure of thecentral nervous system to drug effects) and learningprocesses (e.g. learning to use drugs to relievewithdrawal distress). Problems of dependence aredirect in terms of a propensity to experiencephysical and psychological withdrawal symptoms,of mild to severe intensity, if drug supply runs outor runs low. Indirect problems of a social, legal andeconomic nature may accrue as a result of impairedcontrol over drug intake leading to patterns of usethat are in conflict with personal, work and socialcommitments.

Each of the patterns of risky drug use shown inTable 1.1 can be associated with differing degrees ofdisruption of functioning in all main areas of lifedepending on duration, frequency and intensity ofdrug use. Equally, almost none of the potentialharms identified from risky patterns of drug use areinevitable; rather their risk of occurrence isincreased by differing degrees, although there isconsensus that all tobacco use is harmful.Furthermore, the degree of risk associated with aparticular pattern or episode of drug use will alsodepend on many other contextual factors, whichmay include broad social circumstances, diet,context of use, concurrent activities, the nature ofthe drug concerned and any concurrent use of otherdrugs.

The prevention of drug-related harm needs to beable to identify and reduce major patterns of riskydrug use. At present, Australian laws identify manydrugs as so hazardous to health, safety and/orwellbeing that any use is deemed to be illegal.Tobacco use is legal for persons over 18 years of agebut there is no medically recognised safe level ofsmoking tobacco. Alcohol consumption on theother hand, while a small risk for some cancerswhen consumed in moderation is also, at low dailydoses, associated with reduced risk of heart diseaseand some strokes. Different prevention strategies arerequired that reflect these different patterns of riskand harm.

��

��

� � ��

�� !�" �� #�� $ �� %�%�� &��

� ��

�� '�� !� �� (��$�� "��

��

) �� )��

'�� '�� * �� " ��

�

# �� !��"�#��$��%��

Some sub-populations experience particular risks ofharm. Relevant harms for children and youngpeople include adverse developmental influences.Ensuring healthy child development introducesprevention goals that include reducing pre-birthexposure to toxins (tobacco and alcohol),preventing child exposure to tobacco smoke anddelaying or preventing initiation of alcohol andother drug use.

Indigenous communities and families experiencehigh levels of drug-related harm and preventionapproaches that have relevance to these populationshave been given particular emphasis throughout thedocument.

�� $�� $��

A number of perspectives on prevention haveinformed the present work.

&'('&��)*+��#��+��,-��-��+

This model (Figure 1.1, p. 9), which was adaptedby Lenton6 from an earlier model by Holder,7

illustrates the levels of increasing breadth andcomplexity at which prevention activities can befocused, from work with individuals to national andinternational approaches. The model conceptualisesdeterminants of health and drug use on acontinuum from macro to micro: social andstructural determinants are distal influences whilerisk and protection factors are more proximal. Themodel is both top-down and bottom-up: the macroclearly influences the micro, but equally clearly themicro influences the macro. Activity at any one ofthe levels can influence not only that level but,indirectly, all other levels. This approach allows themapping of systems, pathways and strategies thatconnect among and between risk factors, protectivefactors and drug use outcomes.

&'('.�� -�+/�-�+�� -

For several decades it has been common to classifyhealth interventions into primary, secondary andtertiary prevention where primary prevention aimsto reduce risks and prevent new cases, secondaryprevention seeks to limit harm in the early stages ofa disorder, and tertiary prevention treats the long-term sequelae and consequences of the disorder. Analternative conceptualisation of prevention that hasgained much currency is that provided by the USInstitute of Medicine in 1994.8 This categorisation isbased on the level of risk of disorder in variousgroups targeted. Universal interventions are directed

at whole populations at average risk; selectiveinterventions target groups at increased average riskand indicated interventions target those individualswith early emerging problems. The latter model hasbeen used alongside a broader risk and protectionmodel outlined below, and the term targeted-intervention has been used in this document to referto a combination of selective and indicatedinterventions.

&'('0��-1�� +

A major focus of this report is on common earlypathways to a variety of psychosocial problemsincluding crime, mental illness and suicide, andevidence that such pathways may also be significantfor problematic involvement with both legal andillegal drugs. The risk and protective factorsassociated with antisocial and criminal behaviourhave been described and discussed in the work ondevelopmental and early intervention approaches tothe prevention of crime in Australia,9 and also inrelation to adolescent substance use.10 The specificapplication of these approaches to the primaryprevention of drug use among children and youngpeople has formed a major part of our work. Riskand protective factors originate within a variety ofenvironments, such as the family and educationsystems, and are influenced by community andcultural factors9 and these have been taken intoconsideration.

The developmental pathways model holdspotentially important implications for theprevention of drug-related harm to children andyoung people. Strategies emerging from thedevelopmental pathways model include objectivesrelevant to positive youth development, such associal participation and wellbeing. Inter-sectoralAustralian approaches to the prevention of socialproblems, which include programs informed by theemerging research in developmental health andwellbeing, may be among the more promisinginitiatives for the prevention of drug use and drug-related risk and harm. The developmentalperspective also emphasises the importance ofindividual and contextual factors earlier in thepathway that contribute to the development ofskills, resources and strategies that individuals bringto current situations, and frequently mentions theimportance of situational factors, particularly inadolescence (e.g. peer interactions). In that sense itincorporates situational and harm reductionstrategies.

The present report attempts a synthesis between theevidence base emerging from the developmental

%��

pathways approach and the literature on efforts toreduce drug-related harm at the population level bytargeting more proximal risk factors, that is, riskypatterns of drug use, high risk drugs and high riskdrug use settings. Thus risk factors in this modelrange from the distal (e.g. early developmental andsocial) to more proximal (patterns and places ofdrug use) factors that can be shown to predisposetowards harmful drug use. Similarly, protectivefactors have been identified in the earlydevelopmental and social literature that areassociated with a lowered likelihood of later drug-related harms for those with a high number of riskfactors. This concept of protection can be expandedto incorporate more proximal factors such asreduced drug availability and low risk patterns ofuse (including abstinence). Many harmminimisation strategies can also be defined asprotective factors.

The idea that there are common risk and protectivefactors for substance use and other problembehaviours is also captured, together with the ideaof reciprocal interactions between other problembehaviours and substance use. Evidence that mentalhealth and substance use problems often co-occurand that each can be risk factors for the other isconsidered in this report.

This model is summarised in Figure 1.2 (p.10).

&'('(�� 2�� -��+�� )-�� )�-

The ‘Prevention Paradox’11 demonstrates that toprevent the most amount of harm, it may benecessary to focus (through universal interventions)on the majority who are not as seriously involved inharmful drug use as are the smaller proportion ofhigh risk users. To date this issue has beenconsidered in relation to alcohol but less so inrelation to other drugs, although arguably a casecould be made for tobacco, given that preventioninitiatives discourage non-smokers fromcontemplating smoking as well as motivating othersto quit. A new analysis of a large data set ispresented in this Monograph that addresses theextent to which, in general, universal or targetedinterventions are more appropriate for the riskypatterns of use for each of the major drug categoriesconsidered. This analysis also has relevance to asocial determinants view of problems relating todrug use since it addresses whether the bulk ofharmful drug use can be located withindisadvantaged and otherwise high risk populations,or more generally within the wider community.

��"�&� � !��$�� $��

An economic evaluation framework is invaluablewhen considering the many different strategies ofdrug prevention available. In particular, it isimportant that policy makers clearly understandwhat a program costs in terms of the resources itconsumes and what type of outcomes can beexpected. Surprisingly, these two pieces ofinformation are more often than not missing anddrug programs are poorly evaluated. Too littleeconomic evidence has been forthcoming to ensurethat an efficient use of society’s resources is beingmade across the drug policy arena. Much evidencestill rests on the individual impact or effectivenessof programs without specifying the resourcesneeded to bring this about.

In the methodology literature there is some debateconcerning the type of economic evaluation to usein drug interventions, in particular the techniquesof cost-effectiveness analysis (CEA) and cost-benefitanalysis (CBA). Both are useful supplements ifstudies are well designed in the first place. WhereasCEA measures a given amount of pre-specifiedoutcome obtained by a program for a given amountof expenditure, CBA converts the outcome(s)achieved into a monetary amount and compares thiswith the cost. This allows benefit-cost ratio (BCR)or a return on investment (ROI) to be calculated. Inparticular, benefit-cost analysis forces explicitcomparisons between costs and benefits bymeasuring both in the same unit. However, onlythose programs with the highest positive BCR (atnet present value) can usually be taken up becauseof budget constraints. Hence many clinicians preferCEA believing these analyses to be more clinicallymeaningful and less political in nature than cost-benefit studies.

Benefit-cost ratios may prove useful when based onresults that capture the full benefits in monetaryterms. However, cost-benefit studies do have somelimitations in the way intangible benefits are treatedbecause of the reliance on monetary valuation.Intangible benefits such as a peace of mind, securityfrom crime, freedom from dependency etc are hardto put into monetary terms and so are thus oftenoverlooked without employing more sophisticatedmethodologies such as ‘willingness to pay’ orcontingent valuation.

If a range of outcomes additional to the costs ofdrug use are monetised for prevention programs,stronger benefit to cost ratios may result. Reductionof risk factors such as academic failure is likely to

�

' �� !��"�#��$��%��

lead to a host of cost reductions in addition toreductions in drug use problems. These should bemonetised to fully appreciate the full impact ofinterventions on cost reductions. For example,reduction of academic failure is likely to lead togreater completion of high school, increasedattendance at college and greater job opportunities,which can be monetised as benefits of early school-based prevention efforts.

��#��

The legal status of the substances of interest in theMonograph has been summarised in the followingtable, using categories suggested by McDonald etal.12 Only the major drugs of concern are included.

A wide range of laws and regulations control orrestrict the availability of licit drugs, notably alcohol

and tobacco. These include licensing restrictions,advertising controls, the regulation of sale andsupply of alcohol and tobacco to minors andtaxation initiatives that impact on the cost of legaldrugs. Other control strategies include initiatives tolimit the free availability of inhalants, particularly tominors.

The distinction between ‘licit’ and ‘illicit’ drugs isnot absolute. Some drugs, such as alcohol andtobacco, can be legally sold to adults but not tominors; some prescription drugs can be obtainedlegally but are also used for non-medical purposes,or obtained with stolen or bought prescriptions, by‘doctor shopping’ or on the black market. Somecommunities have chosen to limit or proscribe thesupply of alcohol and the use of alcohol, as well asillicit drugs, is prohibited in prisons.

��

�� + �� *'��,+-.,-*�� "��

� �� "��!��

�� !�� /'�� '�� 0

� �� /��0�� '�� 1��2��3��4��

� �� '�� 5

�� '��

� ��

� ��

� �� &��& ��

� ��6��

� ��

'�� " ��

� ��

� � �� /�� 0

� �� /� 70

" ��!�� 78��$�� '� ��

(��

�"�� '�� /��* ��0

��

��

��

��

��

��

��

��

��

�

��

��

��

��

��

��

� � ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� ��

�!�

��

��

��

��

� ��

��

��

��

"��

��

��

��

��

��

��

��

��

��

��"�

��#�

��

��"��

��

��

��

$��

��

��

��

%��

��

��"

��

�� &

� �'

��

��

��

��

� ��

��

��

��

"��

��

��

��

��

��

��

��

��

��

��"�

��

!��

��

��

��

(��

��"

��

��

��

��

$��

��

��

��

��

��

��

��

)��

��

��

*��

��

� ��

��

��

��

��

��

��

��

� �*�

� ��

��

��

+��

��

��

��

��

$��

��

��

��

��

��

��

,��

��

�,��

��

��

��

��

&��

��-

��

�,��

-��

��

�.��

��

�

/��

��

��

��

��

��

��

�.

%"��

��

��

0��

��

�

��

� ��

��#��

� ��

��

��

��

%�"�

��

)��

��

�.��

� ��

�"��

��

��

��

��

��

� ��

��

� ��

!��

��

��

��

&��

��

��

��

��

��

��

(��

��

��

� ��,

��"

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�"��

��

��

� �"��

��

��

�

&��

��

��

��

��

��

��

"��

�

&��

��

��

��

��

��

��

��"

��

�

��

��

��

��

��

� ��

��

��

��

��

��

��

��

(��

��

��

��

��

��

��

��

�

��

��

��

��

�

��

�

��

��

��

��

�

��

��

�

��

��

��

��

�

��

��

��

�

��

��

��

�

�

��

��

��

��

��

��

��

�

��

��

��

��

��

�

��

��

��

��

�

��

��

��

��

��

�

��

��

��

��

��

��

��

��

��

��

��

�

��

��

��

��

��"

��

��

��

��

!�

��

��

��

��

1��

��

��

��

�

!��

��,��

��,��

��

��

)��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

�.�

��

��.

��

$��

��

� ��

��

��

��

��

��

1��

��

��

��

�

!��

��,��

��

��

��

��)�

��

��

��

��

��

��

��

��

�.�

��

��.

��

)��

��

�,��

��

��

��

��

� ��

��

��

��

��

��

)��

��

��

1��

��,��

��

*��

��

� ��

��

�&�

��

)��

��

��

��

*��

��

� ��

��

+��

��

��

��

��

�%!�

��

��

��

��

��

��

��

��

"��

��

2��

��

"��

��

��

!��

��

��

�+�

��

��

��

��

3��

��

��

�$�

��#

4&�

3��

��

��

��

��

�

�) �� !��"�#��$��%��

"�� (�� $� ��

'��

��

* � ��

� ��

* � ��$� ��

��

'��

��

*�� /�� 0��

� ��

(�� 9��

+��

( ��

��

�� !!��

This chapter outlines the methods used in the review, including criteria for literature inclusion,and procedures for locating and evaluating literature. Both longitudinal and cross-sectionalstudies were included in the consideration of the contribution of specific patterns of drug use toharm, and in examining determinants of patterns of drug use. Environmental influences thatoperate across aggregates of individuals at international and national levels were defined as socialand structural determinants, while risk factors were more narrowly defined according to theirinfluence in increasing the probability that an individual or group would subsequently becomeinvolved in early or heavy drug use or experience drug-related harm. Protective factors weredefined according to their influence in moderating and mediating the influence of risk factors,while not influencing drug use directly.

We have reviewed evidence relevant to the effectiveness of intervention programs aiming toreduce risk factors, enhance protective factors and reduce harmful drug use; to individuals, theirfamilies and the community. The effectiveness of prevention programs was categorised using aclassification system of six mutually exclusive criteria that briefly summarise the status of researchevidence for each strategy. At every stage we have attempted to review materials relevant to thefull life-course and to differentiate age-specific findings and strategies.

��

To reduce harmful drug use requires distinctions ofnot just the different types of drugs being used, butalso variations in the frequency and amount of druguse over time (defined here as patterns of use). Themost recent and accurate statistical evidenceavailable was sought for estimates of the proportionof the Australian population using drugs at variouslevels of intensity and by various means. Nationallevel data sources were given priority over State andregional data, though national estimates ofgeographic variation in patterns of use are reported.To this end, reports from national governmentdepartments, statutory authorities and nationalresearch centres around Australia were sought.Where gaps were found, supplementation wassought with other evidence identified by in-houseexperts, and through searches of Institute* libraries.This strategy provided the most accurate picturecurrently available of the epidemiology of drug usein Australia.

��*�!�

To maximise benefits, prevention activities shouldbe based on a careful appraisal of the harmsresulting from patterns of drug use. Evidence forharms arising through drug use relies at one levelon associations between patterns of drug use andharms. A higher level of evidence requires,however, findings from follow-up andexperimental research to establish that an associated

��+��

To yield the wisdom within the vast sea of text, it isimportant to trawl the literature on drug use andharms systematically and carefully evaluate thecatch. In order to achieve the intended destinationof reducing drug-related harm, the preventionjourney should begin by carefully mapping theharms that result from drug use and the range offactors that influence involvement in suchbehaviour. In the section that follows, the methodsused to determine search, retrieval and appraisalprocesses are described.

The term ‘drug’ refers in this report to all categoriesdescribed within the National Drug StrategicFramework. In considering patterns of use andharms, separate consideration has been given to thelegally available drugs: alcohol, tobacco,pharmaceuticals, inhalants, and performance andimage enhancing drugs. For the illicit drugs,separate consideration has been given to cannabis,heroin, amphetamine-type stimulants,hallucinogens, phenethylamines (such as MDMA),cocaine, ketamine and poly-drug use.

In this report the term intervention refers to the coreprocesses through which interventions aretheoretically conceived to achieve behaviourchange. Programs are vehicles for delivering andsequencing interventions over time. Strategy is thebroad umbrella term for describing the coherentorganisation of programs within settings.

�

* NDRI and CAH

�� !��"�#��$��%��

harm can be reduced by preventing a specificpattern of drug use. To identify literature relevant toharms, in-house searches were conducted formaterial relating to frequency or relative risk ofharm, which included searching the Institutelibraries, consultation with in-house and otherexperts, and examination of all the citations andabstracts in peer reviewed journals, reports or otherdocuments arriving via mailing lists. Gaps werefilled with specific searches (PsycINFO, Medline,Internet) on harms for particular drugs and forparticular types of harms, settings and outcomes(e.g. overdose, dependence, crime, and workplaceissues). In addition, the ‘pearling’ technique (thatis, examining the reference lists of reports and thenobtaining the original report) was used to sourceinformation relating to this section of the report.The data were critically assessed and cross-checkedfor accuracy before being reported.

A particular focus in examining drug-related harmsfor young people has been evidence for the impactof drug use on development. To identifydevelopmental impacts, information from follow-up studies was sought and reviewed, whereavailable. Locating these studies utilised thetechniques described above.

The inclusion criteria for follow-up studiesrequired:

� measurement of drug use behaviour prior to theend of adolescence (age 25),

� that follow-up measurement included one ormore harms,

� an explicit sample selection framework (either acommunity or high risk population) withsampling adequate to permit unbiasedparameter estimation,

� that problems with attrition were addressed(either 80% or more of the initially recruitedsample were retained to follow-up or analysesfor attrition were adequate to ensure resultswere representative).

��"�� !��

Having examined evidence for the harms resultingthrough particular patterns of drug use, attentionwas turned to identifying factors that influenceinvolvement in harmful drug use. Although it isself-evident that drug use is influenced byimmediate decisions and circumstances, what isoften neglected is the important role that is playedby a wider set of influences that determine the moreimmediate drug use context. In this report,

environmental influences that operate acrossaggregates of individuals at the international andnational level were categorised as social andstructural determinants. Under social determinantswas considered the influence of class, gender,ethnicity and culture. The interaction of theseinfluences was considered on structuraldeterminants defined to include war and conflict,poverty and employment status as well as otherfactors relating to political organisation and macro-economic factors at the global, national and Statelevels. Literature on social determinants wasidentified via Institute staff, through noting andsourcing references to this area from the peer-reviewed journals and reports received via mailinglists, from bulletins posted on Update (an emailinformation service for drug researchers and serviceproviders), and from pearling reference lists. Thelimitations of this method of research are discussedwithin the text. Both Australian and internationalevidence was incorporated in this section.

��#�,��-�� $��

Knowledge has been accrued over the past threedecades about factors at a community and individuallevel that increase the probability of drug use.Predictors have been identified, through follow-upand intervention research, as factors that increasethe probability that an individual or group willsubsequently become involved in harmful drug use.In this report we define risk factors as a special class ofpredictors that continue to prospectively predictdrug use after controlling (or adjusting) for theinfluence of other known predictors. Risk factorswere sought that have potential to be modifiedthrough interventions. Protective factors were identifiedaccording to evidence that they moderate andmediate the influence of risk factors, while notbeing themselves predictors of drug use. Inaccepting this definition it became apparent thatmany harm minimisation strategies can be definedas protective factors. To identify risk and protectivefactors, behavioural follow-up studies werecarefully selected and reviewed. Follow-up studiesthat met the inclusion criteria described above weresought and reviewed. Follow-up studies examiningfactors in childhood and adolescence that predictedthe emergence of harmful drug use patterns werespecifically sought. In addition to drug usepredictors, influences such as social determinantsthat impact on the emergence of risk factors werealso considered.

For risk and protective factors relating to adults, anumber of exploratory literature searches were

��

conducted on PsycINFO and Medline. Thesesearches indicated that, with respect to adults, theliterature does not address ‘risk and protectivefactors’ for use, but rather, reports risk andprotective factors relative to particular harms. Thecurrent report is, therefore, one of the first attemptsinternationally to extend the study of risk andprotective factors to influences leading to harmfuldrug use throughout life, from prior to birththrough to late adulthood.

��%�+��$��

Having investigated factors influencing involvementin harmful patterns of drug use, our attentionshifted to the evidence for the effectiveness ofinterventions, programs and strategies aiming toreduce risk factors, enhance protective factors andreduce harmful drug use. The intervention literaturewas categorised to examine demand reductionstrategies, beginning with broad based preventionprograms and then examining programs morespecifically targeting life stages and vulnerablepopulation groups. The report then examinesstrategies aimed at supply reduction and harmreduction, as defined within the National DrugStrategic Framework.13 Demand reduction programswere further categorised by the type of strategyutilised and the impact on different patterns of druguse. Strategy definitions were amended fromToumbourou and colleagues,14 who classifiedprograms according to similarities in theintervention delivery setting, or jurisdictionsinvolved in service development and delivery.Where possible, evidence for specific programs wasalso summarised.

.'3'&�� +)-��

An outline was developed that provided generalguidelines for the inclusion of material in theproject. The earliest publication date for includedmaterial was set at 1990. Any review pre-dating1990 was only included if the material covered wasnot available in a more recent article of equalquality, or the material was of particular historicalsignificance. Inclusion criteria aimed to obtainevaluation reports of the best available evidence. Forstrategies where randomised trials were reported,evaluations focused on effectiveness evidence fromthese studies.

In considering evidence within a social systemsapproach to prevention we become aware of aparadox: strategies that have the potential for thelargest population impact often have the lowestpotential for controlled evaluation. For example,

programs attempting to modify internationaltreaties, or to intervene within international ornational drug supply processes, have profoundpotential to impact drug use but limited potentialfor controlled evaluation. In these cases, the bestavailable evidence is typically limited to the analysisof temporal changes in drug use behaviour underdiffering exposure conditions. Reports meeting theinclusion criteria were retrieved and systematicallyexamined to rate the standard of evidence and thenthis evidence was synthesised to arrive at anevaluation rating for each strategy by drugbehaviour outcome and, where possible, for specificprograms.

For the adult prevention literature, material fromexperts was given primacy due to the likelihood thatthis material would represent the best evidencefrom each field, and that experts often have accessto material that is not widely disseminated. Reviewsfrom The Cochrane and Campbell Collaborations,organisations that apply rigorous scientific standardsto their systematic reviewing process of theinternational literature in order to provide anauthoritative and up-to-date evidence base in healthcare and criminal justice, were incorporated at thesecond inclusion stage.15, 16 This information wasthen supplemented with published systematicreviews other than those from The Cochrane andCampbell Collaborations. Although such reviews areoften of a higher standard than normal literaturereviews, they differ from Cochrane and Campbellreviews in that they are usually less stringent intheir application of systematic principles, and arenot regularly updated with new material. Followingthis process, reviews of the Australian literaturewere then incorporated due to their applicability tothe Australian context. The final inclusion stageincorporated reviews of international literature thatwere most likely to provide evidence that may beapplicable to the Australian setting.

Due to the limited time frame for the project,review articles were selected over primary studies;however, where reviews proved insufficient,primary studies were also included. Primary studieswere selected for review from the resourcesavailable at Institute libraries or because of theirhistorical importance in demonstrating theeffectiveness of a specific strategy. Key informants atthe Institute were also asked to identify relevantreview articles relating to the prevention of drug-related harm, with respect to the main areas ofinvestigation.

The review articles identified by key informantswere then obtained, as were further relevant

�

�� !��"�#��$��%��

reviews cited in the reference lists. Searches ofInstitute libraries’ databases, the NDRI IndigenousAustralian Alcohol and Other Drugs BibliographicDatabase and the NDRI Indigenous AustralianAlcohol and Other Drugs Intervention ProjectsDatabase were also conducted. Information postingson UPDATE were inspected daily as were all peer-reviewed journals and reports received via mailinglists.

In the second phase of the search strategy, scholarlyelectronic databases and online libraries forpublished and unpublished literature wereexplored. The databases searched included:PsycINFO, Medline, Embase, Current Contents,Dissertation Abstracts, SIGLE, Social Work Abstracts,National Clearinghouse on Alcohol and DrugInformation (IDA), DRUG database, Alcohol andAlcohol Problems Science Database (ETOH), CINCH(the Australian Institute of Criminology library),The Cochrane Library, Campbell Collaboration(criminal justice version of Cochrane), theLindesmith Centre, and general internet searches forspecific reports in pdf and html format using theGOOGLE search engine.

The initial searches produced over 9000 studies. Alltitles and abstracts of articles produced in thesesearches were then examined in order to selectpotentially relevant reviews as per the areas ofinterest designated for investigation. The finalsearch strategy supplemented material with relevantliterature identified by experts contracted to assistwith the project (see acknowledgements).

.'3'.��+)�� 4)�+�,�� -�+��)��

As well as general information from the CochraneReviewers Handbook,17 three articles were used toinform the process of evaluating review quality.18–20

Reviews in the substance abuse area, like those inepidemiology and the medical sciences, have notbeen particularly systematic and tend to ignorequality guidelines.18 Common failures appear to bethe lack of clarity regarding scope, search criteriaand inclusion criteria, making it difficult to ascertainthe validity of conclusions and recommendations.18

Despite this, review articles may still be of areasonable quality even if some of these criteria arenot explicitly stated, although this requires a goodknowledge of the research area under investigation.Both Oxman18 and Rehm19 suggest that goodreviews demonstrate the following characteristics:

� provide a good summary and critical evaluationof the state of knowledge;

� have the capacity to influence prevention,treatment and care;

� can be used to guide decisions for healthpolicy/public health; and

� provide an assessment of gaps in the researchand stimulate future research.

Furthermore, guidelines have been developed toascertain the quality of review articles.18, 19 Suchreviews are classified as ‘systematic reviews’, andincorporate the following criteria.

� Problem formulation: a focused and clearlyformulated problem.

� Study identification: a clear description of searchmethods.

� Study selection: criteria for inclusion/exclusion arereported; bias in selecting studies is avoided.

� Study appraisal: validity of studies included isassessed with stated criteria.

� Data synthesis: an appropriate combination ofstudy findings based on reported criteria.

� Interpretation: conclusions that flow from theevidence and that are linked to the strength ofthe evidence.

� Recommendations: can in all parts be supported bythe evidence.

These criteria were considered for both assessingthe calibre of the evidence under review, and forreporting conclusions regarding the effectiveness ofspecific interventions. In particular, considerableattention has been given to ensuring that anyconclusions drawn are linked to the strength of theevidence and do not exceed the evidence reviewed,and that values attached to outcomes (such as costsand risks of harm) are also considered. To ensurethat key reviews were included, a number of expertswere consulted to identify important material and toreview drafts of the interventions sections.

Given the likelihood that few reviews would meetall of the criteria outlined above, the following corethemes were adapted from Rehm18 and applied tothis project:

� the role of experts and the limitations of relyingheavily on expert advice;

� the value and availability of systematic reviews;and

� assessing review quality with respect to bias,comprehensiveness, interpretation of results andvalidity.

�"��

In general, reviews were excluded where they reliedon expert views and provided no attempt tosystematically utilise levels of evidence criteria toappraise empirical research. Reviews were appraisedas of high, moderate or weak quality according totheir fit with the criteria articulated above. Astandardised evaluation checklist guided reviewers’assessments. Given the size and timeframe of theproject, individual reviews were not evaluatedseparately. Rather, the checklist was used byreviewers to derive a qualitative assessment of thestrength of the evidence for each type of strategy,and this is reported in the text.

Based on evidence from existing reviews and theappraisal of primary evidence, conclusions werereached regarding the effectiveness of strategies forreducing harmful drug use. To communicateconclusions regarding the ‘best buys’ for investment

in prevention, the categorisation developed byToumbourou et al. was used.14 Six mutuallyexclusive categories were developed to brieflysummarise the status of research evidence for eachstrategy in addressing different categories ofharmful drug use at different life stages. Thefollowing categories were used to conveyconclusions regarding the implication for futureinvestment in research and dissemination.

O Limited investigation

� Evidence is contra-indicative

� Warrants further research

�p Evidence for implementation, p proportionof studies with positive effects

�� Evidence for outcome effectiveness

�� Evidence for effective dissemination

��

O Limited investigation: no relevant effectiveness studies were located and there were no empirical ortheoretical grounds suggesting the intervention might potentially impact the outcome; may alsoindicate that the evidence is inconsistent or contradictory.

� Evidence was contra-indicative for the use of this strategy to prevent the targeted outcome:consistent null or negative findings in well-controlled evaluation studies.

� Warrants further research: applied to strategies that appeared theoretically sound or had somepromising evidence for their implementation or outcome, but the operational specifics of thedelivery format were not clearly resolved or had been investigated only in small scale orinadequately controlled studies. Policies and programs utilising these strategies might be consideredpriority targets for future research funding focusing on innovations to better define service delivery.

� p Evidence for implementation: published studies reported a sound theoretical rationale, a clearlyspecified service delivery format, acceptance within service delivery organisations, target populationrecruitment on a scale sufficient to usefully contribute to population health impacts, and adequateconsumer approval measured using indicators such as program retention. p The proportion ofpositive demonstrations of impacts on risk factors, protective factors or outcome behaviours isreported. Although this rating required a clear service delivery format, in some cases not all othercriteria were satisfied and in such cases this was indicated in the summaries. Policies and programsutilising these strategies might be supported for implementation where there are few costs andobvious benefits. In other cases wider implementation may await rigorously controlled outcomeevaluation to better establish benefits.

�� Evidence for outcomes: applied where positive outcomes were consistently published in well-controlled interventions. Interventions were required to be of sufficient scale to ensure outcomeswithin the constraints imposed by large-scale population health frameworks. Policies and programsutilising these strategies might be carefully monitored for their impacts while being supported forwide-scale dissemination.

�� Evidence for dissemination: published reports of impacts where programs were delivered on a largescale, not by research teams, but rather by government auspice bodies or other service deliveryagents. Evidence for dissemination was only sought for strategies demonstrating evidence foroutcomes. Policies and programs utilising these strategies might be accorded some priority fordissemination in the Australian context. Initial Australian dissemination trials should monitor forimpacts. Where possible, cost-effectiveness has been considered for programs using these strategies.

�

�# �� !��"�#��$��%��

It should be noted that these criteria for evidencefor outcome are congruent with definitionscommonly used by the National Health and MedicalResearch Council (NHMRC), The CochraneCollaboration and other groups (e.g. systematicreview of randomised trials). The category‘evidence for dissemination’ is an innovationdeveloped by Professor George Patton to address thechallenge of system change in the health promotionfield.

Each chapter that reviews interventions commenceswith a ratings table using these ratings. Wherepossible, interventions for individual drug types arereviewed and rated but the literature does notalways allow this specificity. Other areas(e.g. broad-based prevention strategies, reviewed inChapter 9) do not lend themselves at all to theseratings and strategies are summarised in terms oftheir applicability to drug use and harm.

In some cases, it might be wondered whether theauthors have been sufficiently sceptical of theevidence base. Where reviews of studies withsophisticated statistical controls were available, theyhave been given priority, but it was also feltimportant not to overlook promising Australianapproaches that may not yet have that level ofevidence attached to them. Where there has beenany doubt about a rating, the authors have erred onthe side of inclusivity.

�%��

��

� � � � � � � �

�

�' �� !��"�#��$��%��

�(��

�� !!��

This chapter overviews patterns of drug use and harm in the Australian population fromconception through to adolescence.

National survey data reveals increasing youth alcohol use through the 1990s despite an overallreduction of drinking in the general population for most of that decade. Most pregnant womencontinue to drink alcohol but, based on self-report data, less than 5% drink at levels likely toharm the foetus. There is a need for better monitoring of alcohol use in pregnancy. Acute harmssuch as accidents are disproportionately experienced in the younger end of the population, whilechronic harms are largely experienced in the older population. Exposure to alcohol prior to birthcan have negative developmental implications. Earlier initiation to alcohol use in childhood orearly adolescence leads to more frequent and higher amounts of alcohol consumption in mid-adolescence and this pattern of use has been associated with the subsequent development ofalcohol-related harms in late adolescence and adulthood. Adolescent alcohol use has also beenlinked to subsequent involvement in tobacco use and the onset of some patterns of crime anddelinquency.

Youth tobacco use also resisted the overall declines in the general population during the 1990s,but is now beginning to decline. In the mid-1990s parents were smokers in around one third offamilies, however, information for monitoring maternal smoking is inadequate. Parental smokingduring pregnancy and exposure to environmental tobacco smoke in childhood has beenassociated with a range of child health problems. Adolescent tobacco use increases the risk oftobacco dependence and is also a consistent predictor of subsequent mental health problems.

The non-medical use of pharmaceuticals appears prevalent in school surveys, with 22% of seniorhigh school students reporting previous use of tranquillisers for non-medical purposes.

Finally, there are marked temporal and developmental sequences concerning the ages of first useand the order of onset of drugs. It is apparent that early use of tobacco and alcohol is predictiveof later problems with both alcohol and illicit drugs. These associations appear to be maintainedafter controlling for a range of risk and protective factors. There is also evidence suggesting thatharm reduction strategies can alleviate the development of alcohol-related harm in adolescents.Intervention research may be warranted to establish the relative merits of use reduction versusharm reduction approaches for the prevention of alcohol-related harm in adolescents. Adolescentuse of cannabis significantly increases the risk of later use of other illicit drugs but, nonetheless,only around 10% of cannabis users progress to use other illicit drugs. Early age and frequentcannabis use appear to have negative developmental consequences, though there is only limitedevidence of morbidity and of premature death associated with its use.

��+�� .��

There are a number of ways in which parents’ druguse can impact on their children from the momentof conception, through pregnancy and childhood.

The use of a variety of drugs during pregnancy mayhave an adverse effect, depending on dose andfrequency, on physical development in utero and thismay also impact on later emotional development.

� Breast-fed babies of drug-using mothers havebeen shown to ingest in the milk smallquantities of the drug used by the mother.

� Children of parents with impaired control and/or other signs of drug dependence are morelikely to suffer from neglect and various formsof abuse.

� Parental patterns of drug use have a powerfulrole of influencing children’s drug use as youngadults, through modelling.

�

�) �� !��"�#��$��%��

� Parental smoking presents risks to infants andchildren as a consequence of passive smoking.

0'.'&��2��-)��)��)-��

The use of drugs by pregnant women and alsowomen who are planning to become pregnant is animportant issue for prevention policy. Adverseeffects of maternal drug use have been documentedon the unborn child and can affect physical andpsychological development in both the short- andlong-term. Specific effects have been bestdocumented in relation to the use of legal drugs, inparticular tobacco and alcohol. Evidence exists tosuggest that the use of a number of illegal drugs canalso be damaging though it is harder to separate outthe effects of drug use from a constellation of otherrisk factors such as nutrition and poor physical andmental health of the mother. The nature and extentof these risks will be briefly summarised.

There is some information about the use of drugsby pregnant women in the 1998 and 2001 NationalDrug Strategy Household Surveys (NDSHS) butprecise quantification of this risk behaviour iselusive. The 1998 NDSHS found that 75% ofwomen who were pregnant or breastfeedingreported consuming alcohol, tobacco or at least oneillicit drug in the previous twelve months. Amongthese women, 75% had drunk alcohol, 24% hadsmoked cigarettes, 18% cannabis and 8% used otherillicit drugs. These rates were lower than thosereported for women who were neither pregnant orbreastfeeding.21

The 2001 NDSHS included further questions aboutalcohol and drug use during pregnancy.22 Thisshowed that 41.5% of pregnant women and 45.8%of breast-feeding women had drunk alcohol. Only asmall minority, however, stated that they had noteither reduced their drinking or completelyabstained while pregnant (4.4%) or breast-feeding(5.8%). There is a need for better monitoring ofalcohol use amongst women who are pregnant.

Tobacco

Maternal smoking during pregnancy and earlychildhood is associated with impaired lung growthand diminished lung function.23 Animal model andhuman epidemiological data also clearly point to acausal relationship between prenatal tobaccoexposure and adverse behavioural and neuro-cognitive effects on children.24 Potential pathwaysfor effects include low birth weight and impairedin-utero brain growth.

Cigarette smoking is the single most importantfactor affecting birth weight in developedcountries.25 Several studies have confirmed thisfinding with many of these,25–28 including aCochrane review29, confirming the direct dose-response relationship.26, 30

The effect of prenatal exposure on birth weight ismore attributable to intra-uterine growthretardation than pre-term delivery.26, 29 Poor intra-uterine growth has a lasting effect on thesubsequent growth31 and development of children.32

Low birth weight infants are at increased risk ofemotional and behavioural problems33–35 and thesequelae of low birth weight also include loweredcognitive abilities and hyperactivity.34 An increase inneurological damage has been found among lowbirth weight children,36 which has in turn beenassociated with increased risk for subnormal IQ andlearning disorders. Low birth weight is alsoassociated with increased risk for reading and mathdisabilities.37 However, it remains unclear whethermodest decrements in birth weight associated withmaternal smoking have neuro-behaviouralconsequences among those who are not bornprematurely or of substantially low birth weight.

Children whose mothers smoked during pregnancyhave consistently demonstrated higher rates ofbehaviour problems than those not exposed. Oldsnotes in his 1997 paper38 that 10 out of 11 humanstudies reviewed found increased rates of childbehaviour problems and attention deficit/hyperactivity disorder-like behaviours, even aftercontrolling for many potential confounders. Thesestudies included samples from the newborn periodup through adolescence.

An increased chance of perinatal mortality,including Sudden Infant Death Syndrome (SIDS),has been noted with prenatal exposure to tobaccosmoke.39, 40 Although the effect of maternal smokingduring pregnancy on offspring behaviour isrelatively well studied, less is known about thelong-term effects such smoking has on child neuro-cognitive functioning. Several studies have foundreductions in IQ scores in children born to womenwho smoked in pregnancy, although these resultsare not uniform.41

Alcohol

There is some evidence that even ‘moderate’ alcoholconsumption by mothers during pregnancy canhave adverse effects on the unborn child. The 2001Australian Alcohol Guidelines42 were based on asystematic review of the international literature thatconcluded that maternal consumption of fewer than

��

seven drinks per week, and fewer than three drinkson any one day, was not associated with significantrisk.

Maternal alcohol dependence and frequent high-riskalcohol consumption can cause foetal alcoholsyndrome (FAS).

In infancy, FAS is elucidated by:

� intrauterine growth retardation, with persistentpostnatal poor growth in weight or height;

� a pattern of specific minor physical anomaliesthat include a characteristic facial appearance;and

� central nervous system deficits includingmicrocephaly, delayed development,hyperactivity, attention deficits, intellectualdelays, learning disabilities and, in some cases,seizures.43

FAS infants also have a 3.5 times elevated mortalityrate.44

Even in the absence of FAS, infants born to alcoholdependent mothers show an increased incidence ofintellectual impairments, congenital anomalies anddecreased birth weight.45

Many studies have identified delayed developmentin the first two to three years of life for childrenexposed to significant prenatal alcohol use46–48 andsome studies have followed up such children forlonger periods. By age 7, lower IQ scores, reading,spelling, arithmetic, higher rates of retardation,differences in height, weight, and headcircumference,49 and behaviour problems have beenobserved.45 Similar behavioural problems andimpairments in concentration and attention arebeing described for adolescents and young adults.49

Research in this area has been hampered by poormeasurement of levels of alcohol and other druguse. It is also likely that pregnant women do notalways provide reliable self-reports of drug usewhen participating in these studies.

Cannabis

There are theoretical reasons for expecting adverseeffects of heavy maternal cannabis use on foetalgrowth, but there is limited and inconclusiveevidence that this occurs.50 During pregnancy,maternal cannabis effects in animals and humanshave been documented on pituitary ovarianfunction, prolactin secretion and uterinecontractility.51 However, no relation has beendocumented between cannabis use and length of

gestation or birth weight.52 Birth weight reductionshave been associated with cannabis use indescriptions of those studying higher risk, lowerincome families but the results are conflicting.53

A few studies have suggested a link betweenprenatal cannabis exposure and features similar tofoetal alcohol syndrome,54, 55 but the separate effectsof cannabis use and heavy alcohol consumptionhave not been determined.52 Several neuro-behavioural findings in the newborn period point todecreased responsiveness, on visual responsivenessto animate and inanimate stimulation, and a higherpitched cry.52, 56, 57 Other characteristics of newbornsexposed to heavy maternal cannabis use are tremorsand increased startle in the first seven to 14 days oflife.58 Changes in sleep patterns have also beenreported including a decrease in quiet sleep, andlower sleep efficiency and maintenance as measuredby sleep EEG, as late as three years of age.59, 60

Frequent maternal cannabis use may be a weak riskfactor for Sudden Infant Death Syndrome but thisfinding requires further research.61 One early reviewconcluded that the smoking of cannabis duringpregnancy was sometimes associated with asignificantly increased risk of placental abruptionand a decrease in birth weight.62

Cocaine

There is a considerable US literature relevant to theimpact of maternal cocaine use on childdevelopment. Infants exposed prenatally to cocaine,however, are also exposed to a number of other riskfactors63 such as maternal use of other drugs, healthproblems including a higher incidence of HIV withor without AIDS-related illnesses, complicateddeliveries and intrauterine growth retardation.Postnatally, infants exposed to cocaine continue tobe exposed to ongoing parental substance problems,they are more often neglected and abused, and theyhave parents with more frequent depression andhigher overall stress and anxiety.64 Any one of thesefactors may influence the development of earlyattentional and arousal regulatory functions, laterlanguage, and potentially overall developmentalcompetency.65 There are suggestive findings whichpoint to impairments in more basic neuro-developmental domains of attention and arousalregulation, functions that underlie learning andinformation processing.65

Animal studies show that cocaine administrationduring pregnancy results in major maternalcardiovascular effects and that some of these effectsare enhanced in pregnancy.66

�

�� !��"�#��$��%��

Singer et al.67 in their review conclude that currentstudies are inconclusive but suggest that prenatalexposure to cocaine can have significant effects onthe growth and neurological development of theinfant, with the potential for later learning andbehavioral disabilities. Social-environmentalcorrelates of maternal cocaine use are confoundingfactors with known negative effects on childoutcome. Neuspiel68 and Mayes63 caution that overestimating the risks of intrauterine cocaine exposurecan have negative effects, including unnecessarytermination of pregnancy and labelling of cocaine-exposed children to create a self-fulfilling prophecyof later developmental problems.

Heroin

Infants exposed significantly in utero to opiates(heroin or methadone) may exhibit withdrawalsymptoms in the first days to weeks after delivery.69

Numerous studies have now replicated the findingsthat such exposure reduces birth weight and headcircumference.70–72 Similar findings in animalmodels that control for exposure to other drugssuch as alcohol or tobacco and for poor maternalhealth support the finding of an opiate effect onfoetal growth.73 Prenatal exposure to opiates alsocontributes to as much as an eight-fold incidence ofsudden infant death syndrome (SIDS).71, 74

The dramatic neuro-behavioural abnormalities seenin the newborn period generally diminish over thefirst month of life75 for the majority of infants andare thus assumed to reflect transitory opiatewithdrawal rather than evidence of permanentneurological dysfunction.71 Past the neonatal period,a number of studies have documented small and notusually significant delays in acquisition ofdevelopmental skills.65, 76 However, much moreconsistent and significant across studies have beenthe findings of persistent problems in poor motorcoordination, high activity level and limitedattention span among opiate-exposed infants in thefirst year of life.77, 78 These state and motorregulatory difficulties make it especially hard for aparent to provide appropriate care for the infant,especially for parents with their own state andattentional problems.71

Follow-up studies through early childhood ofopiate-exposed compared with non-opiate-exposedchildren have continued to report few to nodifferences in cognitive performance.65 However,opiate-exposed school aged children show higheractivity levels, are often impulsive with poor self-control, show poor motor coordination, and havemore difficulty with tasks requiring focusedattention. There is also an increased incidence of

attention deficit disorder among opiate-exposedschool aged children.71

Past the years of early childhood, there are fewstudies of the long-term effects of prenatal opiateexposure, and those available usually lack a non-exposed control group or are not based onlongitudinal designs.71 The data from these studiessuggest that by adolescence, opiate-exposedchildren exhibit an increased incidence of behaviourand conduct problems including impulsivity,involvement in criminal activities, heavy drug use,more antisocial behaviour, and increased schooldropout.79–81 It is altogether not clear how muchthese problems in conduct and impulse regulationare attributable to persistent effects of prenatalopiate exposure, and how much they are theconsequence of cumulative exposure to the discordand dysfunction often characterising householdswith substance use problems.

0'.'.��-5�� *�-�)-�

Postnatally, cannabis has been identified in the urineof breast-fed infants whose mothers continue to useafter delivery.82 However, no acute toxic effectshave been identified with this level of passiveexposure although a few studies suggest possibledevelopmental effects related to heavy postpartumexposure via breast milk.50 In one study, cannabisexposure via breast milk in the first postpartummonth was related to decreased motor developmentat one year, and there appeared to be a dose-relatedpattern to the level of association between exposureand motor delay.83 However, longer term studies ofthe outcomes of prenatal cannabis exposure are fewin number and there is a paucity of long-termfollow-up studies. The few findings make it difficultto conclude whether or not prenatal cannabisexposure has a direct effect on later developmentalfunctions such as memory.65

0'.'0� ��+�� -�� -��)��*+��-

Families, in particular those with drug-usingparents, have received considerable attention fromhealth providers and researchers in recent years.Children living with a substance-abusing orsubstance-dependent parent suffer physical,psychological and emotional abuse to a greaterextent and more often than children whose parentsdo not abuse substances.84–86 Angus and Hallreported that in 1994/5, approximately 22% ofemotional abuse cases in NSW were the result of aparent’s substance misuse.87

��

It is widely assumed that parents affected bysubstance use provide their children with less thanoptimal care.88 Women’s use of alcohol or otherdrugs is strongly dissonant with cultural ideals ofmotherhood and is highly stigmatised by society.89,

90 There has been only a very limited body ofresearch that has focused on the study of substancemisusers as parents. These studies suggest that manysuch parents provide adequate care of theirchildren91 and that substance-abusing parents mayhave active strategies to protect their children fromthe risks of their lifestyles.92 However, they alsosuggest that substance-misusing parents providepoorer quality care than other parents. Severaloutcome measures have been used to measureparenting, ranging from gross indicators of childmaltreatment to more subtle aspects of parent-childinteraction. The mechanisms by which parenting ischallenged or compromised when human mothersare drug dependent are not fully understood.88 It isunclear whether parenting is directly impaired bysubstance use or undermined by other factors suchas poverty, lack of education or poor mental health.

Evidence of the moderating effects of environmenton the development of children prenatally exposedto alcohol and other drugs is scant. Hans76 foundthat differences between the development ofchildren prenatally exposed to methadone andcomparison children were only expressed inchildren being reared in the most impoverishedcircumstances and environments. In a related line ofresearch on the effects of lead, Bellinger93 has alsodemonstrated different patterns of teratologic effectsdepending on social class.

One view of moderating effects is that postnatalinteraction and environmental conditions mayenable recovery from a prenatal insult.94 A secondview is that prenatal exposure does not necessarilylead to behavioural changes but rather causesvulnerabilities to other risk factors. Hans76 chose tointerpret her data in this way: that prenatally-exposed children were more vulnerable toconditions of extreme poverty than other childrennot exposed in utero.

0'.'(��-��--��-��1� �� +��

Passive smoking, or the inhalation of environmentaltobacco smoke (ETS), is known to have harmfuleffects. Smoking by parents increases the risk of avariety of diseases in their children.95 Exposure toETS during early life is associated with many adversehealth outcomes including infant mortality,96

respiratory illnesses such as bronchitis, wheezy

bronchitis and pneumonia,97–100 middle ear effusion(glue ear),101 Sudden Infant Death Syndrome,102, 103

reduced ventilatory function98, 104 and shorterstature.105, 106 Importantly, a number of studiessuggest that the effects of parental smoking on achild’s health may have long-term consequences.Middle ear effusion during infancy, for example,may impact on later linguistic and cognitivedevelopment106, 107 and lower respiratory tract illnessin early childhood may predispose for chronicobstructive pulmonary disease in later life.98, 108, 109

Further, in addition to the short- and long-termphysiological damage caused to infants and childrenby ETS, the chances are increased that offspringexposed to parents who smoke may themselves takeup smoking in adolescence or adulthood.110–112 Dollhas argued that exposure to ETS increases the risk oflung cancer later in life, probably exacerbateschronic obstructive lung disease, and may increasethe risk of ischaemic heart disease.113

It has been estimated that each year, passivesmoking causes 46 500 cases of asthma inAustralian children, and causes lower respiratorytract illness in 16 300 children.114

The magnitude of the effects of children’s tobaccoexposure on adverse behavioural and neuro-cognitive effects is not entirely clear, but isestimated to involve a deficit of four to five IQpoints and an odds ratio of 1.5. These decrementsare of concern in that their negative impact tends toaccumulate over the course of development.

�� !! ��

Data are available from regular national surveys forsecondary school students aged 12 to 17 years up to1999 and from the NDSHS for 14 to 19 year oldsup to 2001. Table 3.1 summarises the most currentestimates of the prevalence of different patterns ofdrug use. The main features are that teenage girlsare more likely to be regular smokers and risky orhigh-risk drinkers than are their male counterparts,though levels of use defined as risky for women arelower than for men. Slightly more teenage malesreport having used cannabis in the past 12 months,which is by far the most commonly used illicitdrug. Other levels of drug use are similar betweenthe sexes. Apart from cannabis, the only illicit drugsused with any frequency by this age group areamphetamines and ecstasy. ‘Recent use’ for illicitdrugs in the NDSHS is defined as any use in the last12 months, so the figures do not distinguishbetween occasional, experimental use and moreregular, intense patterns.

�

�� !��"�#��$��%��

The large sample size and standard format over timeof the National Secondary School Drug Surveys(NSSDS) permits detailed breakdowns and sometrend analyses for use of drugs by this age group.Table 3.2 shows breakdowns for illegal drugs byage and sex, with a comparison of overall changesbetween 1996 and 1999.

Main features are that:

� with the exception of inhalants and steroids(perhaps reported unreliably by youngerchildren) there are dramatic increases in levelsof illicit drug use from 12 to 17 years of age;

� with the exception of tranquilisers, males aremore likely to use these drugs than females; and

� cannabis has been tried by about half of all 17year olds but its use has declined since 1996.

��#�� $�� :3��:5�� ;<�'+'�1==:��

$��%��!�� &�� "��'��( �'��() )

��( �� :3%: :>%17�� /� �� 0 >%: 3%3

��$�� 3%= 3%?

< � �� ?@%5 ?3%?

*�� ?1%4 ?3%?(��.��$�� $ �� 4%4 :3%>

(��.��$�� $ �� 15%3 22%=

�� /�� 0 14%4 1>%>

�� 1>%> 11%>�� @%? >%4

�� @%? 3%2

!�" �� =%> =%>

+ �� =%3 =%3

��#�� :1��:?�� /�� A� �� :55>0

$�� !�� )��!� �� )��!� �� )�� *�� *��+�� ((,��(((

) � ) � ) �

�� ::%: ?%: @2%: 3>%3 $?%5 $>%>

�� 2%> :%5 :2%= ::%2 B=%> B:%2

+�� 1%@ :%4 :3%4 5%3 $1%@ $:%3�� 2%@ :%= ?%1 3%5 B=%1 B=%@

�� .�� 53%3 5>%3 5@%4 54%> $:%> $=%?

�� 1%> :%? >%1 1%3 $=%@ B=%3

8��6�� C :@%? ::%3 11%: 11%2 $=%3 $1%:!�� 21%2 22%? :?%1 :3%3 B=%> B=%>

7�� 2%1 :%? >%> 2%5 $=%1 B=%@

' ��C 2%4 1%4 1%5 =%5 B:%= B=%5

C ��$� �� % %��

�"��

The restricted time window of ‘use in last month’ islikely to give a more realistic sense of more regularpatterns of drug use. These are summarised for the1999 NSSDS in Table 3.3. These data again confirmthe pattern of dramatic increases in levels of druguse during the teenage years for all drugs other thaninhalants and analgesics. Putting aside the risksassociated with commonly available painkillers, it isclear that alcohol, tobacco, cannabis and, to a lesserextent, amphetamines are the only drugs used withany significant frequency.

��#�#� � �� :1.:2��:>.:?��

$�� )�� )��!��

*��%�# *��,%�+

�� 2: ?=8�� :1 22

�� > 1=

�� 3 ::

+�� : 2�� 1 3

�� .�� >? ?@

�� : :

8��6�� C 2 @!�� :> 3

7�� : :

' ��C : :

C ��$� �� % %��

Trend data from the NSSDS, between 1983 and1996, show significant increases in alcohol useincluding frequency and quantity consumed by theAustralian youth population through the 1990s,after a small decline in the late 1980s. This trendcontinued between 1996 and 1999 with an increasein the proportion of students reporting drinkingalcohol in the week before the survey. Inspection oftrends demonstrated that the prevalence of studentsdrinking in the previous week at ages 12 to 15tended to increase through the 1990s, while boththis indicator and drinking large (potentiallyharmful) amounts of alcohol also increased forstudents aged 16 to 17 years.116 The Australian trendof increasing youth alcohol use through the late1990s has also been reported in Canada and theUnited States.116

��/ �� !��

Although there is little information, available datasuggest that inhalant use may peak in earlyadolescence and may be more common in youthpopulations at risk for subsequent drug abuse. Inone of the few longitudinal studies, inhalant use byage 16 was found to be a unique predictor of heroinuse by age 32.117

Adolescence is an important period in thedevelopment of drug use. In this section, follow-upresearch studies are examined to explore therelationship between earlier and later patterns ofdrug use in adolescence. Appendices A, B and Csummarise the follow-up studies that are examinedin this section.

Tobacco

Evidence from two cohorts demonstrates thattobacco use in early adolescence predicts latertobacco dependence and daily smoking,118, 119 afteradjusting for other influences.

The relationship between adolescent tobacco useand the development of alcohol use has beenexamined in four separate cohorts. The availablelongitudinal evidence is slim but does not supportthe view that tobacco use in adolescence leads toincreased alcohol use or alcohol use problems.Three cohorts have examined the possibility thatfrequent and dependent tobacco use at ages 17 to18 leads to subsequent alcohol use problems at ages21 to 25. In general, the evidence does not supportthis proposed relationship.120, 121 McGee et al.reported a relationship but this was not adjusted forother influences.122

Evidence from three of four cohorts suggests thatadolescent tobacco use leads to cannabis use. Afteradjusting for other factors, tobacco use at age 15was shown in two studies to predict cannabis use ataround ages 17 to 18.122, 123 In the Los Angelescohort after controlling for the influence on poly-drug use, frequent tobacco use at age 17 tended tolead to frequent cannabis use at age 21.124 However,in a New York cohort early tobacco use did not leadto later cannabis use.121

Alcohol

Work from eight cohorts has examined theinfluence of alcohol use in childhood andadolescence on the subsequent development ofharmful alcohol use. Evidence suggests that earlierinitiation to alcohol use is related to more frequentand higher quantity alcohol consumption in

�

�# �� !��"�#��$��%��

adolescence, and these patterns are in turn related tothe development of alcohol-related harms inadolescence and adulthood. In the New ZealandChristchurch cohort, after controlling for otherknown risk factors, early age alcohol initiation(prior to age 6) led, at age 15, to more frequent,higher quantity and more harmful alcohol use.125 Ina Black-American sample with a high number ofrisk factors, early age use was also associated withmore alcohol use at age 15, but these analyses werenot adjusted.126 Costello et al. did not find age ofalcohol initiation predicted alcohol dependence atage 16.118 In two cohorts, patterns of frequentalcohol use were examined with controls for poly-drug use. In the Boston cohort, Guy et al. report thatfrequent alcohol use at age 14 (without poly-druguse) led to problems with alcohol at age 26.127 InLos Angeles in the 1980s, Newcomb and Bentleralso found that a similar pattern of alcohol use atage 17 linked to alcohol and cannabis problems atage 25.124 Both the frequency and amount ofalcohol use in adolescence appears to predict alcoholproblems in early adulthood. In analysis of the NZChristchurch cohort the amount of alcohol used atage 18 predicted age 21 alcohol harms, aftercontrolling for gender, sociodemographicdifferences, exposure to drinking environments,and prior alcohol use.128 In the same cohort, use atage 18 predicted dependence at age 21, but adjustedanalyses were not reported.122 In the Seattle cohort,frequent alcohol use measured from age 14 or 16predicted alcohol abuse and dependence combined,and also alcohol dependence, at age 21.129 In thesame cohort, trajectories of binge drinking from 13to 18 also predicted alcohol abuse and dependenceat age 21.130

Four cohorts have examined whether alcohol useincreases tobacco use. In most cases involvement inalcohol use has been shown to increase subsequentinvolvement in tobacco use, but at this stage thethree cohorts where this effect has been observed120,

121, 131 were not controlled for known confounders.In the Los Angeles cohort, frequent alcohol use(without poly-drug use) at age 17 did not predicttobacco use, after controlling for other factors.124

Evidence is accumulating that early involvement inalcohol use may predict subsequent cannabis use.These findings have been demonstrated inunadjusted analyses in New York,121 in an AmericanIndian and Black American cohort131 and in the NewZealand Dunedin cohort.122, 132 In the Los Angelescohort, no significant influence for age 14 frequentalcohol use (adjusted for poly-drug use) wasreported for frequent cannabis use at age 21, afteradjustment for other risk factors.124 In the Victorian

Adolescent Health Cohort (VAHC) both thefrequency and quantity of alcohol consumed at age14 predicted frequent cannabis use at age 15, andfor females the amount consumed at age 14predicted daily use at 16 to 17. These effects appliedafter controlling for smoking and other riskfactors.123

Prediction of illicit drug use from prior adolescentalcohol use has been the focus in three cohorts. Theavailable evidence suggests that alcohol is not adirect predictor. Kandel observed that earlier use ofalcohol was associated with a higher likelihood ofsubsequent illicit drug use and use of psychoactivepharmaceuticals in early adulthood.121 A similarrelationship was observed in the Chicago,Woodlawn cohort.117 However, in the Chicagostudy this effect was no longer significant once theimpact of social disadvantage, other drug use andinadequate education were considered.117 In the LosAngeles cohort neither the effect of frequent alcoholuse at age 13133 on illicit drug use at 21/22 or theeffect of frequent alcohol use at 17124 on illicit druguse problems at age 25 were significant afterfactoring in poly-drug use and other risk factors.

Cannabis

Evidence from five cohorts suggests that cannabisuse in early adolescence leads to more frequentcannabis use123 and to problems with cannabis and/or alcohol.124 Evidence from four cohorts suggeststhat cannabis use does not predict harmful alcoholuse. Three cohorts have examined whether cannabisuse leads to tobacco use and again the evidence doesnot support a direct link. Evidence from five cohortssupports the view that cannabis use in earlyadolescence leads to illicit drug use. Work in the LosAngeles cohort suggests the effect may occurparticularly through poly-drug use.124, 133 But in theChicago cohort the effect was maintained afteradjusting for other drug use. Although adolescentcannabis use is predictive of subsequent illicit druguse, it is not all-determining. Findings from theAustralian Temperament Project revealed that ofthose using cannabis at age 13 approximately 10%reported illicit drug use at age 15.461

Poly-drug use

Findings from a number of follow-up studiessuggest that poly-drug use in adolescence is a majorfactor leading to subsequent drug-related harms. Inwork using structural equation modelling, thefrequency of different forms of adolescent drug usehas been typically found to be linked through anunderlying association with poly-drug use andtypically this pattern of drug use is stable124, 135, 136

�%��

and predictive of drug-related harm inadulthood.127, 133, 137 Early involvement in poly-druguse is a marker of risk for later drug use problems.At age 13 approximately 9% of the AustralianTemperament Project cohort were engaged in poly-drug use and this pattern was predictive of moreserious substance use at age 15.461 Williams et al.noted that ‘taking any particular drug (by 15)significantly increases the likelihood of takinganother type of drug’ (p26),461 while use of threeor more different drug types was less common.Structural equation approaches typically find a latentfactor measuring early adolescent poly-drug use(ages 13/14) to be an important and uniquepredictor of drug use problems and otheradjustment difficulties at age 21/22133, 136, 138 and25/26.127 Preventing poly-drug use appears to be animportant goal for reducing drug-related harm.

��"�/ �� !�� !��

0'6'&��*��

Tobacco has acute effects in children andadolescents that are often overlooked. A relationshipbetween respiratory symptoms and smoking hasbeen found in primary school students139 andsmoking-related respiratory problems may becomeapparent within weeks of beginning to smoke.140

Adolescent smokers cough more than non-smokersof the same age.141 They are more likely to developrespiratory tract infections than non-smokers,142

more likely to experience shortness of breath withexertion,140 and have more asthma and allergysymptoms.143

Appendix A summarises follow-up research studiesrelevant to the consequences of different forms ofadolescent tobacco behaviour for subsequentdevelopment. Across the 11 cohorts 41 analyseswere reported. The main consequences measuredwere effects on later drug use, and health andmental health. The developmental consequences ofadolescent tobacco use for subsequent patterns ofdrug use are summarised above, demonstrating thatadolescent tobacco use increases the risk of tobaccodependence. A broader review of risk and protectivefactors for later adolescent substance use is providedin Chapter 6. Among these risk and protectivefactors are included parental and peer alcohol,smoking and other drug use.

The available follow-up studies reveal thatadolescent tobacco use increases the risk ofsubsequent health problems in early adulthood.This issue has been addressed in two cohorts and

both cases refer to the effects of early tobacco useafter controlling for the contribution of poly-druguse. In a Boston cohort, frequent tobacco use at age14 predicted respiratory problems by age 26.127 In aLos Angeles cohort, frequent tobacco use at age 14was associated with health problems at age 22.133

Results from four out of five cohorts suggest thatyouth tobacco use may predict subsequent mentalhealth problems. In the Los Angeles cohort,frequent tobacco use at age 14 (that was notoccurring with poly-drug use) predicted emotionaldistress and psychosomatic problems at age 22.133 Inthe Dunedin cohort, those using tobacco on a dailybasis by age 15 were more likely to have mentalhealth problems at age 18, after controlling forother factors.122 In a follow-up of a small (N = 133)New York cohort, youth tobacco use from 13 to 16increased the risk of difficult temperamentcharacteristics (such as negative emotions)emerging at age 21.144 Analyses with a cohort inNorth Carolina have linked frequent tobacco use atage 16 with the emergence of anti-social personalitydisorders and major depression, but not anxietydisorders, at age 22.138 These analyses adjusted forprior psychiatric symptoms. The Boston cohort wasthe only analysis where age 14 frequent tobacco usedid not predict mental health problems at age 26,after adjusting for other relationships. This analysiswas based on structural equation modelling andfound that the pathway to age 26 ‘psychiatricdistress’ was better explained by peer and self-reported socialisation at age 14.

The possibility that youth involvement in tobaccouse leads to social problems arises from one cohort.In analyses with the Los Angeles cohort, Newcomband Bentler reported that frequent tobacco use(without poly-drug use) at age 14 led to socialproblems at age 22.133 A further analysis of the samecohort found that frequent tobacco use combinedwith alcohol use at age 18 led to increases ingeneral deviancy at age 22.145

Relevant prevention targets for adolescent tobaccouse include preventing youth initiation of tobaccouse and encouraging tobacco users to quit.146

0'6'.��+��+

The research examining the consequences ofadolescent alcohol use for the development of moreextreme alcohol use behaviours, and also fortobacco use, illicit drug use and cannabis use aresummarised above. In this section the consequencesof adolescent alcohol for mental health, health andsocial problems are explored (see Appendix B).

�

�' �� !��"�#��$��%��

The development of mental health problems hasbeen examined in eight cohorts and the evidencedoes not support a direct link with anxiety ordepressive illness. Both the Boston cohort127 and theLos Angeles study133 have examined the effect offrequent alcohol use at ages 13 to 14 and found nosignificant effect on mental health problems at ages22 or 26 respectively. The cohort of BlackAmericans and Indians131 showed an increased levelof psychiatric diagnoses for those using alcohol atearlier ages and the Dunedin cohort122 alsomanifested a greater level of mental health problemsat age 18 for those drinking at age 15, though inthe Dunedin cohort this effect no longer appliedafter considering the effect of family climate, earlierbehaviour problems and smoking. Neither theDunedin cohort122 nor a study of children ofalcoholics147 found that alcohol influenced anxietydisorders. After adjustment for other factors, neitherthe Seattle cohort130 nor the Upper New Yorkcounties study138 found any impact on depression.However, the New York counties study did find aneffect for those reporting ‘heavy’ alcohol use at age16 to have higher diagnoses of both anxietydisorders and antisocial personality disorders by age22 and this applied after considering prior mentalhealth and substance use status.138

The role of alcohol in the development of criminaland delinquent behaviour has been examined inthree cohorts and findings suggest that frequentalcohol use around the ages 15 to 18 may increasethe risk of entry to delinquency and crime. In theSeattle cohort, binge drinking from 13 to 18 did nottend to predict entry to crime by age 21, aftercontrolling for a range of risk factors.130 In theChristchurch cohort, the quantity and frequency ofalcohol consumed at age 15 did predict entry toproperty crime at age 16148 and the emergence ofviolent crime,148 after appropriate control for arange of risk factors. In the Los Angeles cohort, thefrequency of alcohol use at age 17 also tended todemonstrate unique pathways to general deviance atage 22.145

In two cohorts, outcomes relevant to education havebeen examined and findings are inconclusive. In theSeattle cohort, binge drinking from 13 to 18influenced high school retention, after controllingfor other risk factors.130 Frequent alcohol use from11 to 15 did not influence attitudes to school, aftercontrolling for other predictors.149

In the Los Angeles cohort,133 a pathway wasobserved from more frequent alcohol use withoutpoly-drug use at age 13 to more social relationshipsand romantic attachments at age 22. One study

conducted in Oslo observed that female alcohol useat age 13 increased the incidence of subsequentexperiences of sexual victimisation, but the extent towhich this finding was due to other underlying riskprocesses was not examined.150

Heale et al. used data from the 1998 NDSHS toestimate the proportion of alcohol users in Australiadrinking in patterns that are likely to risk harm.151

The analysis revealed for young adults (18 to 24years) 90% of all alcohol was consumed in high-risk patterns, primarily due to drinking in wayswhich placed the drinker at risk of acute harm. Theelevated rate of high-risk drinking in the youngadult age group is due to young people being morelikely to drink a large amount of alcohol in a shortspace of time, typically on weekends.152 This is areason why young people are more likely toexperience acute harm and less likely to experiencechronic harm. These drinking patterns are reflectedin the types of harm, which typically includes drinkdriving and violence.

Findings suggest that prevention targets mightinclude delaying the age of first alcohol use. Beforethis target receives widespread support, furtherevidence will be needed in the Australian context.Currently, many parents appear unconvinced thatearlier age of alcohol use does lead to harms asparents are the main source for early adolescents toobtain alcohol.153 Efforts in Australia to teach youngpeople strategies to minimise harms and avoid risksassociated with alcohol show some evidence forsuccess. Such strategies, in concert with other publichealth initiatives, have been associated withreductions in alcohol-related road deaths154 andreductions in levels of youth alcohol use.155

Maintaining prevention targets aimed at minimisingor reducing risky alcohol use would appearwarranted. Intervention research may be warrantedto establish the relative merits of use reductionversus harm reduction approaches to the preventionof alcohol-related harm in adolescents.

0'6'0� � �*�-

A number of follow-up research studies haveexamined the consequences of adolescent cannabisuse for the development of later drug usebehaviours, and also for mental health and healthand social consequences (see Appendix C).

Two US studies have found conflicting results inregard to later health problems as a result ofadolescent cannabis use. Within the Boston cohort,no effect was found between frequent (not poly-

�(��

drug) adolescent cannabis use and health problemsin the early twenties; however, within the LosAngeles cohort, a significant relationship was found.

Several recent studies have documented arelationship between adolescent cannabis use andlater mental health problems; however, theevidence is not consistent across studies. An earlyfinding of relevance was the small study by Lernerand Vicary where cannabis use from ages 13 to 19predicted increasing levels of difficulttemperament.144 In the Brook and colleagues NewYork State cohort, heavier cannabis use at age 16predicted personality disorder at age 22, aftercontrolling for initial mental health status.138

However, in the same cohort, no effect on anxietyor depression was noted. In the New ZealandDunedin cohort, cannabis use at age 18 increasedthe risk of mental disorder at 21 (although from age15 to 18 mental disorder increased the risk ofcannabis use).122 In the Christchurch cohort,frequent cannabis use at ages 15 to 16 did predictlater major depression (but not anxiety or suicideattempts) at ages 17 to 18, after adjusting for arange of risk factors.156 Adolescent cannabis use hasalso been linked with the subsequent developmentof psychotic symptoms in the Christchurchcohort.157, 158 No independent (adjusted) effect wasfound between frequent cannabis use in earlyadolescence (13 to 14 years) and mental healthsymptoms in early adulthood (21 to 25 years)within either the Los Angeles133 or Boston127

cohorts, when the influence of cannabis use withoutpoly-drug use was examined.

The available longitudinal research indicates thatadolescent cannabis use is related to various socialproblems in late adolescence/early adulthood. Forexample, a significant relationship was foundbetween cannabis use at age 15 to 16 years andearly school drop out, unemployment, violentoffending and property offending at ages 17 to 18years, within the Christchurch cohort.156, 159 Theseresults are consistent with a South East US cohortwhere earlier age cannabis use was associated withschool drop out by age 18 years.160 Within the LosAngeles cohort, family problems at age 22 were alsofound to be exacerbated by cannabis use in earlyadolescence.133 A separate analysis of theChristchurch cohort found a significant effectwithin a shorter time frame between cannabis use atage 15 and school dropout and police contact byage 16. However, once these findings were adjustedto control for other factors including peerrelationships, social disadvantage and childhoodbehaviour problems, effects tended to disappear.159

In their review, Lynskey and Hall noted that earlycannabis use consistently predicted poor schoolperformance and non-completion of high school.They argued that this effect was not due directly tothe influence of cannabis on motivation or cognitiveability but acted through prior risk factors andaffiliation with low achieving peer groups.161

Many of the harmful consequences have beenassociated with earlier age use and frequent use inadolescence. Hence targets for reducing harmassociated with adolescent cannabis use mightinclude preventing youth initiation to cannabis use,reducing the number who progress to regular useand encouraging regular users to use lessfrequently.

0'6'(� ��

Although the specific developmental effects ofcocaine use in adolescence have not been widelystudied, a number of US follow-up studies haveexamined the developmental implications ofadolescent illicit drug use and in many of thesestudies cocaine has been prominent. These studiesreveal that illicit drug use initiated in earlyadolescence tends to be very stable162–164 andpredictive of the subsequent emergence of drug-related harms127 and problems associated withmental health,133, 138 crime162 and socialrelationships.133

In small amounts cocaine causes euphoria andfeelings of energy. Used repeatedly, adverse effectscan include agitation, anxiety, paranoia,hallucinations, dizziness, nausea and vomiting,tremors and aggression.165 Prolonged use leads totolerance and bingeing which is followed byperiods of intense depression, lethargy, and hunger.Cocaine smokers may experience lung problems,while injectors risk BBVs (see injecting, page 30).Repeated inhaling damages the nasal lining and thestructure separating the nostrils.165

Cocaine can be fatal in large doses although it isnowhere near as toxic as opioids.166 Cocaineoverdose is associated with cardiovascular incidentssuch as cardiac arrest and cerebral vascularaccidents. It was estimated that it was responsiblefor four deaths in Australia in 1998, along with 78Person Years of Life Lost (PYLL) and 59 hospitalepisodes.167 In the US, the combination of cocaineand alcohol has been found to be a factor incocaine-related deaths but research is needed toascertain whether this is a factor in Australiancocaine-related deaths.168

�

�) �� !��"�#��$��%��

Both smokers and injectors are more likely todevelop dependence than those who sniff or snortthe drug. Dependence on cocaine is associated withsignificantly higher levels of physical and mentalharm, higher levels of criminality andunemployment, along with lower levels of generalsocial functioning.169

In the United States, where cocaine use is far morecommon than in Australia, it has been found thatcocaine use is associated with high rates of anxietydisorders and affective disorders.166 In a Sydneystudy, 31% of surveyed non-injecting cocaine usersreported experiencing psychological problemsincluding depression, anxiety and paranoia as aresult of their cocaine use.170 As with amphetamine-type stimulants (ATS), prolonged bingeing oncocaine can result in a transient but severe paranoidpsychosis.166

��

�� !!��

Alcohol and other drug use substantially contribute to death, injury, illness, crime, mental healthand social problems both to drug users and the wider community. A broad range of preventableharm and patterns of drug use need to be considered in determining priorities for preventionpolicy. Understanding the underlying patterns of risky drug use is also a pre-requisite fordesigning effective interventions.

Drug use is associated with high legal and social costs to communities and families. There aremajor economic costs caused by the lost productivity resulting from drug-related deaths. Themost significant drug in regard to road trauma is alcohol. Crime is strongly associated withalcohol and drug use, particularly alcohol with violence and heroin with property crime.

Tobacco and alcohol use contribute the great bulk of preventable health problems associated withdrug use. Although regular tobacco use continues to decline, it is responsible for the highestlevels of mortality and morbidity of any drug, primarily due to cancers affecting older people.Alcohol represents the second largest contribution to drug-related harm. Unlike tobacco, adependent pattern of use does not contribute the most harms at a population level: the short-term effects generate the most years of life lost (e.g. from injury and poisoning) while the long-term effects of regular heavy use result in the more premature deaths among mostly older people.The main causes of alcohol-related deaths are cancer, alcoholic liver cirrhosis and road trauma.

While tobacco use continues to decline, risky alcohol use remains common (particularly amongyoung adults) and overall consumption of alcohol has increased in recent years. Most alcoholconsumed in Australia is not drunk within Australian Alcohol Guidelines for low-risk use.

Despite illicit drug use being under-estimated by population-based surveys, as many as 38% ofAustralians over the age of 15 admitted they had tried an illicit drug in 2001. Cannabis continuesto be the most commonly used illicit drug in Australia, followed by amphetamine type stimulants(ATS), the use of which has been increasing recently. Lifetime use of drugs believed to be‘ecstasy’ was reported by 6% of adults. Heroin has recently decreased in availability and wasreported to be used by only 1.6% of the adult population in 2001.

The major health problems caused by the use of illicit drugs are connected with injection, eitherthrough an opiate overdose or the transmission of blood-borne viruses. Injecting drugs is alsoassociated with the development of dependence. A marked drop in heroin overdose has occurredsince 2000, though this follows a 55-fold increase between 1964 and 1997. Major risks foropiate overdose are injecting after a period of abstinence and use in association with other CNSdepressant drugs, especially benzodiazepines and alcohol. While a rate of HIV infection of 2%among injecting drug users (IDUs) is among the lowest of any country, hepatitis C is a majorconcern and is present in two-thirds of Australians who have injected for sixyears or longer.Many sufferers go on to develop liver cirrhosis with often fatal consequences.

A handful of ecstasy-related deaths have been reported, most commonly from hyperthermia.There is little evidence that steroids are a significant public health concern.

Rates of substance use are strikingly higher in persons diagnosed with mental health disorders.There is growing evidence that heavy and dependent alcohol use can cause, as well as be inresponse to, mental problems. Most authoritative reviews conclude that cannabis can onlyexacerbate symptoms and precipitate psychotic episodes in vulnerable individuals. Approximately10% of people who experiment with cannabis develop dependence. Such a pattern of use

�

�� !��"�#��$��%��

increases the risk of lung diseases. The social and economic costs of a criminal conviction for usemay be the most significant harm for many cannabis users. The use of ATS, especially in a bingepattern, is associated with psychotic episodes as well as violent and risk-taking behaviours.

Illicit drug use, like alcohol and tobacco use, is more prevalent among younger people andamong males, Indigenous Australians and persons living in the north of Australia. Rates ofinjecting drug use and notifications for hepatitis C have increased significantly in recent years andare a growing cause for concern. Smoking and episodic risky alcohol use are still the main riskbehaviours for preventable drug-related harm. Inhalant use (including petrol sniffing) is alsocause for concern, especially among young Indigenous people in some remote locations. FewIndigenous people have escaped the effects of alcohol on family and community life. Higher ratesof mental health problems among Indigenous Australians are associated with the same underlyingsocial determinants as substance use but are also exacerbated by the misuse of alcohol.

Culturally and linguistically diverse (CALD) Australians generally have lower rates of drug use.

There are some special concerns with alcohol use among the elderly due to their lower toleranceand the likelihood of interactions with commonly prescribed drugs. However, rates of risky useare also very low in this age group. Light use of alcohol among the middle-aged and elderly isassociated with reduced rates of heart disease.

Prisoners and police detainees have higher rates of illicit drug use than are found in the generalcommunity. Injecting is common in prisons, as is needle sharing, which brings with itconsiderable risks for the transmission of blood-borne viruses.

Drug use—both licit and illicit—impacts on the community through its associations with crimeand violence, sexual assault, domestic violence, concerns about public safety and amenity;impacts on families, the workforce and road trauma; and also impacts on health throughpremature death, injury and illness.

�� 0��-��!��1��

Any discussion of priorities for prevention of drug-related harms must first specify the nature andprevalence of both the harms to be prevented andthe significant patterns of related risky drug use. Inthis chapter, the harms associated with drug use inAustralia are examined for a range of different drugtypes and for different population age groups.Specific attention is paid to settings where harmsmay be particularly prevalent and to populationsvulnerable to harm.

A classification of the many categories of drug-related harm has been presented in Table 1.1 of thefirst chapter.

It is easier to quantify the health, safety andeconomic costs of drug use than the social and legalcosts. Each year in Australia, alcohol, tobacco andother drugs contribute to almost one in five of alldeaths, to more than 185 000 hospital admissions;and cost billions of dollars in health care, lawenforcement and lost productivity.1, 167, 171 Less easyto quantify, but still important, are the social costsexperienced by some drug users, their partners,families and communities. While there has been asubstantial rise in the use of heroin and in heroin-

related deaths in recent years,172 the health costsassociated with alcohol and tobacco consumptionstill greatly exceed those from illicit drugs.

Table 4.1 provides an overview of key indices ofhealth and economic costs, by drug type and broadage groups, using most recent available data.

The key points revealed by this summary table arelisted.

� Tobacco is by far the leading cause of prematuredeath and hospitalisation among Australians.Most (77.8%) of tobacco-caused deaths involvepersons over 64 years of age. However,tobacco-caused deaths involving children andadults up to 64 years of age are still greater intotal than all deaths caused by alcohol and illicitdrugs combined for all age groups. Theeconomic costs of tobacco reflect this fact.

� Alcohol causes the deaths and hospitalisation ofslightly more children and young people thando all the illicit drugs combined and many morethan tobacco. These deaths are almost invariablycaused by either intentional or unintentionalinjuries. While alcohol is also responsible forthe deaths of many more adults and elderlypeople than are the illicit drugs, there are moredeaths believed to be saved among older people

��

as a result of, mainly, low-risk alcoholconsumption, principally among older women.

There will also be significant future health costsassociated with current drug use that are hard toestimate. The high prevalence of hepatitis C amonginjecting drug users in Australia means there will besubstantial mortality, morbidity and associatedeconomic costs as a result of higher incidence ofliver disease in this group, in future years. There isalso a range of social harms impacting on individualusers of illicit drugs who receive criminalconvictions.173

The major source of data on population prevalenceof drug use in Australia is the NDSHS. This is aperiodic self-report questionnaire, distributed to alarge stratified population sample surveyed in theirhomes. There have been seven household surveys,the most recent in 2001, which had a large samplesize (26 744) permitting reporting of state levelrates of use. There are limitations to the validity ofsuch drug consumption self-report data. In relation

to alcohol, self-reported levels of consumption inthe 1998 NDSHS only accounted for 46.5% of thetotal volume of alcohol actually consumed inAustralia as estimated from sales, production,import and export data.151 Users of illicit substancesmay under-report their use for a number of reasons,including the illegality of illicit drug use, and it isthose groups with the highest rates of usage that arethe most likely to under-report their levels of use.174

Additionally, the NDSHS does not include thehomeless, those in prison, or highly mobile peoplewho may have higher rates of drug use, and ittherefore under-samples the highest risk groups.175

Finally, the response rate to the 2001 NDSHS wasonly 51%, a drop of 5% compared with 1998. Thisis a very low rate and flags a concern about non-response bias—it is known that persons notcontacted readily in such surveys tend to be higherusers.176 It is necessary, therefore, to treat surveydata as but one indication of patterns of substanceuse and harm.

��'�� $�� /:5540�� D ��*�� *��/:5540�� /:55?.540�� /:554.550��

� �� -�� .�� /��/�� 01��2

��=E:3�� 55 1F554 :F@?=:@E23�� 4= 1F=5> ?F:4?

2@E>3�� 3F=31 >4F=5: @5F245

>@B�� :3F?55 :::F252 ?3F2?5

�� (34�( �5'36+( �'�36�6 ��4,

*��=E:3�� 1: >2? :F=23:@E23�� 4:3 1:F53: 1@F1=?

2@E>3�� :F12= 1@F2=5 12F@53

>@B�� :F=3> 5F>2> >F=11

�� #3�� 6+36�# 66356+ +�6,

�� =E:3�� 5 1>: 33:@E23�� >35 :?F15> :=F4?>

2@E>3�� 221 ?F@2> 2F=@?

>@B�� 22 141 353

�� 34�# �63#+6 �'3'+� ,�45

� �� '�� ;�(��' � ��/1==:0��

� '�� ;��*�� /1==10�

�

�� !��"�#��$��%��

Table 4.2 shows NDSHS rates of alcohol and druguse for the population 14 years and over.22 Thistable shows that a great majority of the populationhas used alcohol, and about half have used tobaccoat least once. Two-fifths have used an illicit drug atleast once, predominantly cannabis. Lifetimeprevalence rates of other illicit drugs are all less than10%. It is important to note, however, that anylifetime use of a drug is not inevitably risky druguse and more frequent use is not reported for illicitdrugs other than for injecting drugs, rates of whichremain at very low levels. Table 4.2 shows that in2001, a third of Australians aged 15 or overreported drinking regularly at risky levels for short-term harm and that slightly less than 20% of thepopulation smoked on a daily basis.

For most drug types other than alcohol, levels of usein the last 12 months were significantly lower in2001 than in 1998. This trend may in part be aconsequence of the lower response rate to thesurvey in 2001. Geographically, the 2001 NDSHSillustrates a marked north-south divide in terms oflevels of drug use. The NT has the highest rates ofsmoking, risky alcohol use and ‘recent’ illicit druguse, followed (by some distance) by WA andQueensland.

The association of mental health and drug useproblems is a substantial concern. However, it canbe difficult to isolate in every case to what degreedrug use causes mental health problems, and towhat degree mental health problems give rise todrug use, often in the context of self-medication.There is no doubt, however, that some patterns ofalcohol and other drug use can exacerbate pre-existing mental health problems or even precipitatethese for the first time. Examples of this includeusually temporary psychoses induced by a heavy‘binge’ of stimulant drugs, the association of suicideand para-suicide with both cannabis use and alcoholintoxication,177 the worsening of anxiety anddepression during withdrawal from alcohol andopiate drugs, and the increased prevalence andintensity of phobic anxiety states among heavy anddependent drinkers.178 Collectively, substance useand mental health problems represent the largestcontributors to the burden of disease amongyounger Australians. In those aged 15–24, 90% ofthe leading causes of disease and injury in men, and80% of the leading causes in women, are substanceuse disorders or mental health disorders.179 Co-morbidity is associated with a worse treatmentoutcome, a longer duration of mental illness, and ahigh rate of service seeking.179

��'�� :1�� ;�<�'+'�1==:��/��A�� :554�� 0

$�� !�� )�/�� )�� )�7�� )��0�((58�44�2 �� 0�((58�44�2

��/:B��0 5=%3 /B=%40 41%3 /$=%40

��;��$ �� 5%5��;��$ �� 25%3

8�� 35%3 /$=%>0 12%1 /$:%?0

�� 2?%? /$4%20 :>%5 /$@%=0

�� 22%: /$>%=0 :1%5 /$@%:0�� C 4%5 /B=%:0 �2%3 /$=%20

+�� ?%> /$1%20 :%:

��.� �� >%: /B:%20 �1%5 /B=%20

�� .�� C >%= /$@%@0 2%:�� 3%3 /B=%:0 :%2

8��6�� .�� C 2%1 /$2%=0 :%:

!�� 1%> /$:%20 =%3

!�" � �� :%4 /$=%20 �=%> /$=%:0+ �� :%> /$=%>0 �=%1 /$=%@0

-�� =%5 /$=%?0 =%1

' �� =%2 /$=%@0 =%1

) �� CC =%2 /$=%10 =%:

C ��$� ��

CC <�$��

�"��

Other forms of harm include harms to self andothers incurred when individuals are intoxicatedwith alcohol and/or other drugs: the NDSHS, forexample, reports that between 1998 and 2001 theproportion of the population undertaking a range ofactivities while under the influence of alcohol and/or other drugs generally decreased.22 In 2001,12.8% of the sample (18% males and 7.7% offemales) reported driving a car while under theinfluence of alcohol. Other activities performedwhile under the influence of alcohol includedverbal abuse (6.3%), going swimming (5.2%),going to work (4.3%) and creating a publicnuisance (2.9%). Activities performed whilst underthe influence of other drugs included driving(3.9%), going swimming (2.4%) and going towork (2.3%). In all cases, males were more likely toreport these activities than females.22

Approximately 6% of all Australians (6.6% of menand 4.7% of women) sustained a non-self inflictedinjury as a result of an alcohol- or other drug-related incident in the previous year. The mostprevalent, most serious injury was bruising orminor abrasions (40.7% of all those injured).Almost 3% suffered injuries sufficiently serious torequire hospital admissions. The youngest agegroups (14 to 39 years) were the most likely tosustain bruising or abrasions while the oldest group(60+) were the most likely to requirehospitalisation (6.1%).22

��1�� 0��-��!

The following section discusses particular patternsof risky use of different drugs, and issues regardingthe supply and availability of drugs.

('0'&��*��

Tobacco smoking is the single largest preventablecause of disease and death in Australia.167 It has beenestimated to cause 15% of all deaths, typicallythrough chronic health conditions resulting fromlong-term smoking across the life course.

The prevalence of regular tobacco use in Australia isbroadly similar to that found in the USA, UK,Canada and New Zealand, in which countries aboutone-fifth to one-quarter of adults smoke tobacco.180

Smoking is more common among people with loweducation, with low occupational status and wholive in rural Australia.181 Cigarettes are the dominantform of tobacco used and, in 2001, NDSHS smokersreported smoking approximately 16 per day onaverage.22

Trends: there has been a dramatic reduction insmoking over the past 50 years. The 2001 NDSHSand reports of Customs between 1995 and 2002confirm that this trend continues, notwithstandingevidence of an increasing black market incigarettes.182

Harms: the acute harms of smoking are largelyrestricted to novice users who generally experiencenausea, dizziness, and other symptoms. Tolerance tothese effects develops quickly.183 However, someacute harms, such as the reduced ability of theblood to carry oxygen, which results in diminishedexercise performance, are maintained.183

The chronic harms include 32 diseases andconditions related to active smoking.183 The maincauses of tobacco-related deaths in 1998 werecancer (40%), chronic obstructive pulmonarydisease (20%) and ischaemic heart disease (21%).184

Cessation of smoking, however, causes a rapiddecline in the risk of both mortality andmorbidity,183 which demonstrates the importance ofpolicies and interventions aimed at encouragingsmoking cessation. For those stopping at the ages of60, 50, 40, and 30, the cumulative risk of lungcancer by age 75 is 10%, 6%, 3%, and 2%,respectively.185 Collectively, these findings havesignificant implications for smoking policy. Over arelatively short period, the excess mortality could besubstantially reduced by current younger smokersgiving up the habit.185

In relation to impact on non-smokers, passivesmoking has been estimated to account for 12 newcases of lung cancer each year and 77 deaths fromcoronary heart disease.114 Australian Institute ofHealth and Welfare estimates were that 128 deathswere attributable to environmental smoking in1998.184

('0'.��+��+

The physical availability of alcohol has increasedmarkedly in Australia over the past two decades, asit has in most economically developed countries.186

Licensing laws have been continuously revised tosimplify and streamline procedures for acquiringliquor licences, trading hours in all jurisdictionshave been extended, and laws regarding service tointoxicated and underage drinkers are enforcedintermittently at best.187 There have been severalsignificant developments in the alcohol market overthe past 30 years.188 One has been the rise in thepopularity of Australian wine, both domesticallyand internationally. This has been encouraged by afavourable taxation regime for wine in comparisonwith beer and spirits. This situation encourages both

�

�# �� !��"�#��$��%��

the widespread distribution and consumption ofcheap packaged wine (cask wine is an Australianinvention) and the production of wine-based fruitdrinks (‘alcopops’). Another importantdevelopment has been the rise in popularity of low-and mid-strength beers, apparently encouragedboth by tax breaks for lower strength beers andaggressive enforcement of drink-driving laws acrossAustralia in the 1990s. Beer with a strength of lessthan 3.8% was estimated by the AustralianAssociated Brewers to comprise 40% of the totalAustralian beer market in 1999.189 Since 1 July2000, a new alcohol tax system has been inoperation that, on the one hand, reduced taxes onpremixed spirits to a rate slightly higher than fullstrength beer and, on the other, increased taxes onalcoholic sodas and lemonades with a wine base tothe same level as pre-mixed spirits. The resultappears to have been a dramatic switch in marketshare from full strength beer to pre-mixed spirits,though the net level of alcohol consumption hasremained static.190

Estimates from the 1998 NDSHS of how muchalcohol was drunk in Australia at risky levels forhealth by persons aged 15 and over, found that 39%of all alcohol was consumed at levels that posedhealth risks in the long-term; 51% was consumed atlevels posing short-term health risks. Two-thirds ofthe alcohol consumed in Australia in 1998 posed arisk of chronic and/or acute health consequences.For young adults, this figure was 90% of all thealcohols consumed by that age group—primarily bydrinking in ways that place the drinker at risk ofacute harm. These estimates are conservative sinceaverage consumption reported in that survey wasless than half that estimated from official statistics of1998 alcohol sales.191 These data should beconsidered in the context of the clear national andinternational trend for increased risky alcohol useby adolescents and young adults.

Trends: in the late 1980s and early 1990s, per capitaalcohol consumption fell in Australia, as in mosteconomically developed countries, then levelled offuntil the late 1990s when it has increased slightly.192

Harms: alcohol is responsible for the majority ofdrug-related deaths and substance-related hospitalepisodes in young people.193 Acute harm (as a resultof road trauma, violence, accidents, acuteintoxication) is disproportionately experienced inthe younger end of the population. Chronic harm,on the other hand, is largely experienced in an olderpopulation.194 While there are slightly more deaths

caused by the long-term toxic effects of alcohol thanfrom heavy episodic drinking, two-thirds ofalcohol-caused person years of life lost are due toeither the acute effects or conditions caused by acombination of acute and chronic effects ofalcohol.195 Combining estimates of deaths fromethanol toxicity, alcoholic beverage poisoning,other ethanol/methanol poisoning and aspirationyields a total of 96 deaths for Australia in 1997 thatcould be broadly described as alcoholic overdose.195

It should be noted that a significant number ofopiate overdose deaths involve concurrent heavyalcohol use and so this figure is likely to be anunderestimate.166

Data from the National Survey of Mental Health andWell-being (NSMHWB) indicate that 8.2% of thepopulation aged 18–50 have an alcohol use disorderas defined by DSM-IV criteria.196 The estimates ofeconomic costs of alcohol misuse in Table 4.1 donot include those associated with mental healthproblems and there is mounting evidence thatalcohol is implicated in the highly prevalentdepressive and anxiety-related disorders.Longitudinal research on the relationship betweenalcohol dependence and major depression showsthat alcohol dependence increases the risk of havingmajor depression one year later, and equally, thepresence of major depression elevates the risk ofhaving an alcohol dependence disorder one yearlater.197 A variety of studies report that the lifetimeprevalence of major depressive disorder in peopleseeking treatment for alcohol dependence is around40%. The co-occurrence of major depression andalcohol use disorders elevates the risks of bothviolence and suicidal behaviour.197 Alcoholdependence is a major risk factor for suicide.198

Low risk alcohol consumption is believed to have apositive effect on health by way of reducing the riskof ischaemic heart disease among older people.184

Because heart disease is the most common cause ofdeath, any assumption of such benefits frequentlyresults in a net positive effect of alcohol onmortality.199 There is a growing consensus againstpublishing such net figures for the outcomes of alltypes of drinking patterns combined. It is importantto separately identify the deaths associated withrisky/high-risk drinking on the one hand (3 294),and the lives saved in association with low-riskdrinking on the other (6 605).199 A net figurecompletely misrepresents this situation but has beenthe basis of most epidemiological and economiccost estimates of alcohol-related harm in Australia.

�%��

('0'0��)��+-

The non-medical use of pharmaceuticals has beenidentified as one of the major drug problems inAustralia.200 The Australian Illicit Drug Report(AIDR) states that prescribed drugs, includingopiates such as morphine and methadone;benzodiazepines, particularly diazepam andtemazepam; and amphetamines prescribed forAttention Deficit Disorder, such as dexamphetamineand methylphenidate (Ritalin), are obtained illegallyby feigning symptoms, ‘doctor shopping’, usingstolen and forged prescriptions and theft frompharmaceutical and surgical establishments.168

Another level of pharmaceutical use relates tomisuse or overuse of over-the-counter (OTC)drugs. This includes the purchase of preparationscontaining pseudoephedrine for the illicitmanufacture of amphetamines, which is discussedfurther in chapter 12.

The 2001 Illicit Drug Reporting System found thatthe non-medical use of methadone, morphine,benzodiazepines and anti-depressants was commonin a national sample of 951 injecting drug users.Diazepam was most commonly used, by up to 86%of respondents in the previous sixmonths.201 Highrates of benzodiazepine use are also found amongpolice detainees. In 2001, the Drug Use Monitoringin Australia study (DUMA) found that 21% of menand 33% of women in police lockups tested positiveto benzodiazepines.202

Trends: the 2001 NDSHS survey indicates a reductionin the use of benzodiazepines without prescriptionsince 1998.

Harms: benzodiazepines (e.g. Serepax [oxazepam],Valium [diazepam], Euhypnos [temazepam] andRohypnol[flunitrazepam]) are frequently prescribedin the treatment of a range of anxiety conditionsand for insomnia and are often associated withdependence and related problems. Benzodiazepinedependence produces a pronounced and distressingwithdrawal syndrome.203 It has been estimated that44% of long-term benzodiazepine users becomedependent.204

Flunitrazepam has recently been placed onSchedule 8 of the National Drugs and PoisonsSchedule because it has a higher abuse potentialthan other benzodiazepines, particularly in relationto the risk of death in heroin users.205 Other harmsfound to be associated with flunitrazepam includeviolent behaviour, prolonged sedation and short-term memory loss.205 This drug is also associatedwith drug-assisted sexual assault.

Heroin users frequently use benzodiazepines, eitheras a supplement to opiates or as an alternative intimes of scarcity. The combination of heroin andbenzodiazepine use, particularly when alcohol isalso being consumed, is a key factor in heroinoverdose.

('0'(�� )�-

While there are many types of performanceenhancing drugs, the anabolic/androgenic steroidsare most commonly used. Steroids are taken for twoprimary reasons: improving physical appearancethrough increased muscle mass and decreased bodyfat, and for improved athletic performance in somesports. They exert both effects by increasing musclemass and strength.206 Steroids have no effect on theaerobic capacity of athletes and are considered oflittle use in endurance-based sports.207 The 2001NDSHS reported use in the last 12 months by only0.2% of the population and lifetime use by 0.3%,22

but steroids are readily available in some gyms200

and via mail order from overseas internet sites.165

Use is largely confined to those involved inbodybuilding, some athletes and those in thesecurity industry.* 165 The gay population alsoreports a comparatively high rate of steroid use, ataround 4% using in the past sixmonths.208

Trends: non-steroid performance enhancing drugshave become more common recently among eliteathletes, for example, human growth and thyroidhormones, human chorionic gonadotrophin andlevodopa.209

Harms: there is little recorded evidence that steroidsare a significant public health concern in Australia.No deaths and only two hospital episodes in 1998are attributed to anabolic steroids.184 However, thelong-term effects of the drugs are not well knownand their short-term effects on personal appearancecan for some be highly undesirable. Overallreported levels of side effects and harms areconsidered low, especially given the high doses ofsteroids typically consumed.207 In males, harmsinclude gynecomastia (development of breasttissue), testicular atrophy and temporaryinfertility.207 In females, harms include decreasedbreast size, menstrual irregularities and growth offacial hair207, and clitoral hypertrophy. There arecase reports of premature halting of growth inyounger female steroid users, although these effectshave not been systematically studied.209 Harms forboth genders include skin rashes, acne, deepeningof the voice, increased water retention, jaundice andchanges in libido. Chronic harms have been known

* A discussion of the role of steroids in elite sport is beyond the scope of this document

�

�' �� !��"�#��$��%��

to include liver disease, cardiovascular disease andsterility.207 However, in the majority of users, healthproblems are minimal and they are not likely to belong-term, although the patterns of harm in long-term chronic users on heavy dose regimes remainsunknown.207, 209 Nearly all steroid users inject210 butrarely share needles207, 211 so there is little blood-borne virus (BBV) risk.207 There is little evidence todemonstrate increased risk of mental healthdisorders.207, 210, 212

('0'6� � �*�-

Around 33% of Australians have consumed cannabisat some point in their lifetime and approximately13% of Australians have consumed cannabis in thepast year.22 Cannabis is readily available in all partsof Australia,165, 213 being increasingly producedhydroponically.165

Gender and age are strong correlates of levels ofusage. Men are more likely to have tried cannabisand more likely to be frequent users. In theNSMHWB: while 10.3% of men used cannabismore than five times in the past 12 months, only4.3% of women had used at this rate.214 Increasedlevels of education are associated with likelihood ofhaving ever used cannabis, whereas lower levels ofeducation are associated with more frequent use.166

Australian data indicates that only around 10% ofpeople who ever use cannabis become daily users,with another 20 to 30% becoming weekly users.166

Trends: lifetime usage rates have risen substantiallyover the past twenty years but have dropped inrecent years. The estimated prevalence of lifetimecannabis use for those over the age of 14 was 12%in 1973, compared to almost 40% in 1998.215 Areduction in rates of cannabis use is evident forsecondary school children between 1996 and 1999as shown in the previous chapter.

Harms: while cannabis is not a harmless drug, thehealth risks associated with cannabis appear to besmaller than for most other drugs, legal or illegal.Despite its wide use, there were only 652 hospitalseparations and no deaths attributed to cannabis in1998.167

The acute harms are: anxiety, dysphoria, panic, andparanoia although these rarely lead to help-seeking.166 Chronic harms include the damage doneto lungs through smoking, though the full extent ofthis remains to be determined.216, 217 A confoundingfactor is that most cannabis smokers also smoketobacco.166 There is considerable anecdotal evidencefrom both cannabis users and clinicians thatprolonged use can have a negative impact on

attention, memory, and concentration. Carefullycontrolled studies indicate significant thoughmodest impairments of cognitive functioning.218

Most authoritative reviews conclude that cannabisdoes not cause psychosis in its own right166 but usecan exacerbate symptoms of schizophrenia andpsychosis in those already suffering theseconditions, and may precipitate the onset ofschizophrenic episodes in those already vulnerableto such an episode due to personal or familyhistories of schizophrenia.219 New evidence thatadolescents using cannabis may be especiallyvulnerable to such effects (see Chapter 3) warrantscareful scrutiny.

('0'7�� ,��-��)+� -�8��9

ATS are the second most commonly used illicitdrugs after cannabis,184 with most of this use beingmethamphetamine.168 ATS are produced and readilyavailable in Australia.165,213 The ATS classificationused in this report includes amphetamine,dexamphetamine, and meth(yl)amphetamine. Itexcludes phenethylamines, such as MDMA, whichare described in a later section. The most currentillicit ATS used in Australia is methamphetamine,which has a longer duration of action thanamphetamine.220 In the 2001 DUMA data collection,385 of 412 positive amphetamine screens wereconfirmed with meth(yl)amphetamine only or incombination with amphetamines.202 Regular ATSusers are more likely to inject than occasionalusers.166

Trends: amphetamine use has been widespread inAustralia since the mid-1980s. In the NDSHS, use inthe last 12 months increased among men from2.8% in 1995 to 5% in 1998, but fell slightly to4.2% in 2001. Increases in amphetamine use from2000 onward have been notable in the DUMAsurvey.202, 221 The AIDR noted that the heroin‘drought’ had led to a nation-wide increase in theuse of other drugs including ATS.168 The annualmulti-centre Illicit Drug Reporting System (IDRS),in 2001, noted that the prevalence and frequency ofmethamphetamine use increased between 2000 and2001 with increased use of ‘base’ and ‘ice’, whichare potent forms of the drug.168

Harms: ATS users generally suffer a high level ofmorbidity and a low level of mortality. Harms fromATS use are mostly related to mental rather than tophysical health. In 1997/98 there were three tosixdeaths and 409 hospital admissions related toamphetamines dependence or abuse.167 Physicalhealth problems include poor appetite, fatigue,tremors, sleep disturbances, cardiac arrhythmias,

�(��

headaches, joint pains and weight loss.213 Overdoseon ATS can occur, associated with circulatorycollapse, cerebral haemorrhage and myocardialinfarction.220

Mental health effects include acute effects such asanxiety or aggression, with typical after-effects offatigue and depressed mood.220 Regular use caninduce personality changes, with users typicallybecoming irritable, suspicious, dysphoric, anxious,and at times aggressive.220

Heavy users typically use less regularly than opiateusers and ‘binge’ over a period of a few days andnights, often followed by the use ofbenzodiazepines or other sedatives to comedown.166 Such bingeing can induce a temporarypsychosis identical in symptoms to an episode ofparanoid schizophrenia and this psychosis can bereliably induced in people with no history of, orpredisposition towards, mental illness.166 ATS usecan also severely exacerbate psychotic symptoms inthose already experiencing a psychotic mentalillness of some form.220

The existence of an amphetamine dependencesyndrome, comparable to that of alcohol andtobacco, has recently been documented.213

('0'3�%�"/�:#�� 1�� 8��++)�� -9

The availability and use of LSD are stable and low inmost jurisdictions. It is usually used infrequentlyand alongside other recreational drugs. Typically,hallucinogen users are young, unemployed malesborn in Australia or the UK.166

GHB (gamma hydroxybutyrate) is a relatively scarcedrug that is not investigated separately in theNDSHS. There has been a significant increase in thenumber of Customs seizures of GHB at the border,from three seizures in 2000/01 to 18 in 2001/02,165 suggesting a trend of increased popularity ofthe drug.

Ketamine use is relatively rare and is notinvestigated in the NDSHS. One study of ketamineusers found that users tended to be older, male,poly-drug users; many of whom were gay. Thegroup was well educated and relatively well-offfinancially. Use was predominantly in nightclubsand dance parties.168

Trends: in the NDSHS, lifetime prevalence ofhallucinogens increased from 7% in 1995 to almost10% in 1998, and then decreased to 8% in 2001.

Harms: one Australian death was attributable tohallucinogens in 1998, along with 234 hospitalepisodes.167 Deaths are very rare and are typically aresult of bizarre behaviour rather than overdose.166

The harms associated with hallucinogen use appearto be mainly of an acute psychological nature, suchas depression or a psychotic episode.166 There islittle evidence of dependent use or even a pattern offrequent use.166 The oft-mentioned ‘flashback’syndrome, where users experience elements of thehallucinogen experience at a later date, sometimesyears afterwards, is typically associated with fairlyheavy LSD use.166

A small study of Australian GHB users found otherdrugs were used 95% of the time, which probablyincreased negative effects.222 Users reporteddizziness, blurred vision, hot and cold flushes andvomiting. Half of the sample reported losingconsciousness after GHB, and almost half hadexperienced blackouts or memory lapses. One intwelve users reported having a fit or seizure afterusing GHB.222 The Australian Bureau of CriminalIntelligence (ABCI) reports that while there havebeen media reports of GHB deaths, the real rate ofsuch deaths is unknown.165

('0';�� ,+�� -�8$"$�/��-�-,9

MDMA (Methylenedioxymethamphetamine) andrelated drugs are defined in different surveys asMDMA, ecstasy, ‘designer’ or ‘party’ drugs. Thebroader term phenethylamines is used by theAustralian Standard Classification of Drugs ofConcern168 and is used in this report except wherespecific research is reported, when the terminologyis that used by the authors.

Drugs described as ‘ecstasy’ are easy to obtain inAustralia223 but most do not contain any MDMA andare more likely to be methamphetamine combinedwith a synthetic hallucinogen.165 Those containingecstasy or a related substance (e.g. MDA or MDEA)had an average purity of 33%, with a range of 3% to93%.223

Phenethylamines users have recently been profiledas being ‘young, relatively well educated, likely tobe employed or engaged in studies, and to havelittle contact with social authorities such as thepolice or treatment agencies’(p1).223

Phenethylamines users are less likely to beunemployed than users of other illicit drugs.166

Trends: the use of phenethylamines has been steadilyincreasing in popularity since the late 1980s, withusers becoming younger and using more frequentlyand heavily.166

�

�) �� !��"�#��$��%��

Harms: acute symptoms of use include increasedheart rate and body temperature, increased reflexes,muscular cramps, tremor, dry mouth, loss ofappetite, vomiting and sleeplessness.220 Manystudies have documented reduced mood andfeelings of anxiety in the few days following MDMAuse.224 Other common side effects over the next fewdays include insomnia, drowsiness, depression anddifficulty concentrating.225

Between 1995 and 1997, there were at least 12deaths in Australia associated with ecstasy use, ofwhich six involved PMA(paramethoxyamphetamine, an amphetamineanalogue with a high degree of toxicity).226 Deathshave variously involved persons with pre-existingcardiac conditions,220 hyperthermia, and ingestionof excessive amounts of water.166 Such deaths havetypically occurred at dance venues. Hyperthermicsyndrome can be successfully treated in the majorityof cases, if medical intervention is available.220

Concern exists that MDMA may cause later cognitivedeficits though there is no definitive evidence.227

Investigating psychiatric problems associated withecstasy has proven difficult due to high rates ofpoly-drug use among those presenting withpsychiatric symptoms.220 There are reports that thatecstasy use has been associated with paranoidpsychosis, flash-backs, suicidal ideation,depersonalisation, anxiety states, panic attacks, anddepression. However, reviews of these particularcase histories also indicates previous psychiatrichistories and high rates of psychiatric morbidityamong first degree relatives.220 Heavy use may beassociated with heightened anxiety andaggressiveness.225

('0'<� ��

Cocaine appears to be far less common in Australiathan it is in many other western nations, beingrestricted mainly to injecting drug users in Sydney.The use of crack cocaine is extremely rare inAustralia.165

Cocaine use may be associated with a broadersocioeconomic range of users than most illicitdrugs.200 There is some traditional associationbetween its use and young, well-paidprofessionals,200 but there are no representativeAustralian data to support this supposition. NDSHSdata indicates a higher rate of cocaine use amongunemployed people.166

Trends: injection use of cocaine has risen over time inSydney, from the first IDRS survey in 1996 whereno users nominated cocaine as their drug of choice,

to 39% in 2001. Cocaine may be supplanting ATS asthe stimulant of choice among many stimulantusers.166 Cocaine availability and purity in NSW hasbeen increasing steadily since 1996.228 Slightincreases in cocaine availability were noted in 2001in Victoria, ACT, SA, Queensland and WA.201 Aswith ATS, the strong increase in cocaine use bySydney’s injectors has been associated with thecorresponding decrease in availability of heroinover this time as a result of the well-publicised‘heroin drought’.228

There were no sizeable detections of cocaine amongpolice detainees in the 1999 and 2000 DUMA data.Cocaine began to be detected in Sydney in early2001 and increases were sustained throughout theyear. No such trends were observed in the non-Sydney sites. These trends reflect the localised andrapidly changing nature of illicit drug markets,which has relevance for prevention initiatives.202

Harms: there are no Australian data on the prevalenceof harms associated with cocaine use. Harmsassociated with injection are discussed genericallybelow.

('0'&=�#��

Estimates based on the 2001 NDSHS were that37 700 people had used heroin in the previousyear, while 252 600 had ever used heroin.22

However, estimates of the number of dependentheroin users from diverse indicators such asoverdose prevalence, arrests, methadoneprescriptions, ambulance call outs and needleexchange usage suggest there were 100 000 in1998.229 These discrepant results reflect the inabilityof surveys to accurately estimate prevalence of illicitdrug use.

Heroin users are typically older than other illicitdrug users, male and unemployed.166 They are alsotypically from a disadvantaged background, have ahistory of problems at school and at home, and areoften impulsive.229 Heroin is usually injected inAustralia. Non-injecting use of heroin is morecommon in Australia’s Indochinese population thanin other sociocultural groups.166

Trends: heroin use increased substantially throughthe 1990s; during this period there was evidencethat new cohorts of users were both younger oninitiation to heroin and more likely to be femalethan older cohorts.166 Use has fallen since December2000, however, when the current ‘drought’commenced.168 The latest AIDR reports thatalthough heroin remained scarce in the second halfof 2001, there is evidence that heroin was

��

becoming more available in major Australian capitalcities in the first half of 2002, though not to pre-drought levels.165

Harms: the main physical and mental healthproblems associated with heroin use are fatal andnon-fatal overdose, dependence and the high risk ofcontracting BBVs, because almost all heroin isinjected.229 It has been estimated that between 0.5and 1% of the heroin-using population dies eachyear from a drug overdose and that dependentheroin users have approximately a one-third chanceof dying from their heroin use.229

The total number of non-fatal overdoses per annumhas been estimated at 12 000 to 21 000,230 withsome studies indicating that the lifetime prevalenceof non-fatal overdose in heroin users is 67%.166

Heroin deaths increased from 10 per year in the late1960s to more than 500 per year from 1995onward but have declined since. The number ofopioid-related deaths among 15 to 44 year olds inAustralia decreased from 958 in 1999 to 725 in2000.231

Death from heroin overdose usually results fromrespiratory failure, which is a consequence of itscentral nervous system depressant effects.230 Riskfactors include concomitant use of benzodiazepines,alcohol, and other central nervous systemdepressants;166 loss of tolerance from periods ofreduced use, such as time in prison or in treatment;and fluctuations in heroin purity. There is noevidence of toxicity resulting from contaminants ofheroin in Australia.230

Studies, worldwide, have shown repeatedly thatheroin users have high rates of psychologicalproblems with the most common being depression,anxiety and antisocial personality disorder.166 Astudy of Australian methadone patients found veryhigh rates of psychiatric disorder but in over 70% ofcases with psychiatric dysfunction, the onset ofpsychiatric symptoms was reported to predate theuse of heroin.232

('0'&&�� +� -

The broad category of inhalants includes drugs ofdiffering harm potentials. Volatile solvents include awide range of different products, including types ofpaints, glues and petrol. These drugs have seriousharm potential in both the short- and long- term.233

Other inhalants (that are not volatile solvents)include nitrous oxide (once known as laughinggas), various types of asthma medication, householdgases such as butane and bottled domestic gas, andnitrites. Levels of use appear to be significant only

in some populations of adolescents and are thenrapidly replaced by other substances, legal andillegal, in later teenage years.

Lifetime rates of inhalant use in the adult populationdecreased in the NDSHS from 3.9% in 1998 to 2.6%in 2001 with a concomitant decrease in recent use.

('0'&.��+,5��)��)-�

Poly-drug use is a common pattern, not just amongusers of illicit drugs, but also among users of legaldrugs and pharmaceuticals. In this respect, cannabisuse more closely resembles licit than illicit drug use.NDSHS 1998 data show that there is a strongassociation between tobacco and alcohol use: 90%of recent smokers had recently consumed alcoholand one in three recent drinkers reported recenttobacco use.184 There is also an association betweentobacco, alcohol and cannabis use. In 1998, one infive recent drinkers reported recent cannabis use,which is a higher rate of consumption than thegeneral population, and 40% of recent smokers hadrecently used cannabis. Almost all recent cannabisusers had recently used alcohol, and 57% ofcannabis users had recently smoked.184 Amongpeople who had recently used pain-killers/analgesics for non medical purposes, 87% hadrecently used alcohol, 41% cannabis and 39%tobacco.184

Users of illicit drugs other than cannabis commonlyuse tobacco, alcohol and cannabis, as well ascombinations of ATS, hallucinogens,phenethylamines, cocaine, heroin andbenzodiazepines.166 This is particularly true ofinjecting drug users.223

Harms: poly-drug use has been shown to be asignificant factor in heroin-related deaths. Alcoholwas detected at autopsy in 46% of fatal overdoses inthe New South Wales study, and benzodiazepines in27%.234 In a Victorian study,235 alcohol was detectedin 37% of cases and benzodiazepines were presentin 44% of cases.

Mixing cannabis and alcohol is believed to greatlyincrease the acute risks associated with intoxication.In Victorian research examining autopsies for fatallyinjured drivers, alcohol was present for 27% andcannabis was the next most frequently detectedsubstance at 15%.236 The risk of asphyxiation isincreased through the mixture of alcohol andbenzodiazepines. Darke et al. observed that, despiteclear differences in the psychoactive properties ofamphetamine and heroin, there was a tendency forinjecting drug users to move intermittently betweenusing them.237

�

�� !��"�#��$��%��

In DUMA in 1999, one in three police detaineestested positive to multiple drug use (use of two ormore drugs from amphetamines, benzodiazepines,cannabis, cocaine and opiates) and 13% testedpositive to multiple drug use not includingcannabis. Predictors of multiple drug use were:being on government benefits, raising money fromillegal activities, and having been arrested in theprevious 12 months.238

('0'&0�� >�� )��)-�

Injection of drugs is a major risk factor for seriousadverse health effects. The 1998 NDSHS estimatesindicated a national population of around 300 000people who have ever injected illicit drugs and110 000 people who had done so in the previous12 months.175 Again, these self-report figures arelikely to underestimate true prevalence. There aretwo national monitoring systems that detailinjecting behaviour. The Australian Needle andSyringe Program (NSP) Survey is an annual studythat has been surveying NSP participants in everyjurisdiction since 1995. Its focus is the prevalenceof hepatitis C and HIV/AIDS, using tested bloodsamples and patterns of injecting behaviour.239 TheIllicit Drug Reporting System consists of threecomponents: interviews with injectors, interviewswith key informants who have regular contact withinjectors, and examination of extant data sourcesrelated to illicit drug use.201 These two surveysidentify high rates of unemployment, incarcerationand low education levels. Most injectors are maleand the average age of first injection was 18.

In the 2001 IDRS it was noted that in response tothe shortage of heroin throughout 2001, someinjectors had switched to injecting stimulants—ATSin most jurisdictions, and cocaine in NSW.201 NSWwas the only jurisdiction where a substantialproportion had last injected cocaine. An increase inthe reporting of cocaine use by injectors in Sydneywas first noted in 1998.240 It was initially thoughtthat this rise was largely associated with the practiceof speedballing, or mixing cocaine and heroin,240

but more recent data has demonstrated that cocaine-only injecting is also becoming more common.201

Extensive poly-drug use is the norm amongstinjectors. The 1998 NDSHS data were used to showthat 26% of those injecting in the past 12 monthsinjected both ATS and heroin, and 18% reportedinjecting other combinations.175 Between a third anda half of injectors in all jurisdictions reported theuse of illicitly obtained benzodiazepines in thepreceding sixmonths. Many of those who obtainedbenzodiazepines illicitly also obtained them onprescription.201

It is very rare for people to switch back to othermeans of administration after they have beguninjecting and this usually occurs only in the contextof vascular damage.166 In a study of more than 300ATS users in Sydney, Ross et al. noted that while78% of users first used the drug by non-injectingmeans, only one-third continued to use this methodwith the remainder making a transition toinjecting.241 In a Perth study, opiate usersconsidered smoking of heroin a waste of moneybecause there was a perception that more of theproduct was required.242

Trends: the NDSHS shows widespread injecting ofATS as well as heroin and it has been demonstratedthat between 1980 and 1995 the predominant drugof first choice was amphetamine; after 1995, heroinbecame cheaper and more widely available.175 In the2001 IDRS, ATS were most frequently (38%)nominated as the last drug injected, followed byheroin (35%), morphine (12%), cocaine (7%) andmethadone (5%).

The rate of injecting drug use has greatly increasedsince the 1988 NDSHS survey.175 Other multipleindicators point to an increase in the use of illicitdrugs, particularly those that are traditionallyinjected. Encouragingly, however, data fromAustralian needle and syringe programs show thatneedle sharing has declined from 31% in 1995 to16% in 2000.243 However, 49% of injectors reportsharing equipment such as spoons, filters, andtourniquets, which can also result in viraltransmission.184

Harms: injecting is associated with a very high risk ofBBV infections, increased risk of overdose amongheroin users, and dependence. Injecting can causeharms to the community as well as to individualdrug users if use results in improperly discardedsyringes.244

Health complications associated with injecting illicitdrugs can result from contaminants in the drugs,use of non-sterile injecting equipment, blood-to-blood contact with contaminated blood fromanother drug user, and failure to inject in a sterilesite or an appropriate body location.245 Non-communicable and communicable infections areassociated with injecting—the major non-communicable infections are: thrombophlebitis(inflammation of the vein); bacteraemia (bloodpoisoning) and septicaemia; and endocarditis(inflammation of the endocardium, which lines theheart).245 Injection of temazepam from capsulescauses significant harms to the circulatory system,including abscesses, blocked veins, and tissuedamage due to lack of circulation.246 Temazepam gel

��

caps have recently been moved to Schedule 8 of theNational Drugs and Poisons Schedule in order toreduce their use by injectors.247

The prevalence patterns of BBVs among injectorsvary greatly. HIV prevalence among thosepresenting at needle and syringe exchanges is lowby world standards, at less than 2%.243 In 1999,only nine of the 196 new AIDS diagnoses weresolely attributable to injecting drug use. Thepercentage of AIDS diagnoses attributable solely toinjecting has remained relatively constant at around9 to 12% since 1992; 9% of AIDS deaths in 1999were believed to be in cases where AIDS wasobtained through injecting drug use.184

Hepatitis B was first identified as a virus in 1965,long before HIV/AIDS and hepatitis C. In Australia,there are 150 000 to 180 000 hepatitis B infectedindividuals with an estimated 1200 deaths perannum.248 Hepatitis B is efficiently transmitted byblood, vertical transmission and by sexual contact,and people who inject drugs are at high risk ofinfection. The prevalence of hepatitis B amonginjectors, in a 1994 national study, was 18.9%positive (anti-hepatitis B core EIA positive) and waspositively associated with duration of injecting.249

Of the three BBVs, hepatitis C is by far the mostprevalent amongst those who inject drugs. To theend of 2000, over 160 000 Australians had beenexposed to hepatitis C of which 80% are estimatedto be related to injecting drug use. Mathematicalmodelling has indicated 16 000 new infections in2001.250 Just over 50% of people presenting atneedle and syringe programs in 2000 had hepatitisC.243 Clients of Australian needle and syringeprograms who had injected for six years had aprevalence rate of 65%; for three to five years, 30%;and less than three years, 20%. This demonstratesthat the length of time spent injecting is a criticalrisk factor.

Hepatitis C is a major public health concern due tothe possible sequelae of serious health consequencessuch as cirrhosis of the liver and hepatocellularcarcinoma, which can result from chronic infection.It is estimated that of one hundred individualsdiagnosed as hepatitis C antibody positive,approximately 50% will progress to chronicinfection over a 15 year period. For those whoprogress to chronic infection, 50% will developcirrhosis over a 25 year period and of these, 25%will develop liver failure and 10% liver cancer; bothconditions usually fatal within two years.251

Infection carries with it a high probability of long-term carriage and infectiousness and treatmentoptions are restricted. The potential health care costs

are large and will probably exceed those of HIV/AIDS in the coming decades.

The social costs of hepatitis C infection among IDUsare difficult to quantify. Chronic hepatitis Cinfection frequently results in overwhelmingfatigue, malaise, abdominal pain and headaches thatcan result in significant life disruption.252

Discrimination against those with hepatitis C occurswithin the general community, amongst themedical profession, and sometimes within familiesthemselves.253

�� 0��-��!�! �� 1��

Table 4.3 summarises data from the 2001 NationalDrug Strategy Household Survey regarding theprevalence of drug use in those older than 60.22

Rates of use of both licit and illicit drugs aresubstantially lower for the elderly. However, thecumulative effects of lifetime tobacco and alcoholuse, and acute problems associated with loss oftolerance and interactions of alcohol with prescribeddrugs, reduce some benefits accrued due to reducedprevalence.

Harms: patterns of harm reflect these drug usepatterns. Smoking is by far the leading cause ofdrug-related deaths in the elderly, with almost15 000 deaths per annum (see Table 4.1)254

compared with just over 1000 alcohol-relateddeaths. There were a total of only 33 deathsestimated to be related to the use of illicit drugs in1998, mostly from the consequences of hepatitis Cacquired earlier in life. A pattern of light drinkingprotects against heart disease and is thought toprevent over 5000 premature deaths per year in theelderly. The optimal level of intake has beenestimated to be between one and two drinks per dayfor men and one or fewer drinks per day forwomen.255

The elderly are more vulnerable to harms fromalcohol due to lower tolerance, increased sensitivityand increased likelihood of interaction with othermedications;256 even relatively small amounts ofalcohol can cause problems. It is possible, in anelderly population, for consumption to remainsteady while one’s ability to tolerate that level ofconsumption decreases, leading to alcohol-relatedproblems without any increase in consumption.256

This results in a significant number of alcohol-related fall injuries.256 Elderly persons admitted togeneral hospital wards drink more than theircounterparts in the community. One study found asmany as 21% of inpatients aged 65 to 74, 10% of

�

�� !��"�#��$��%��

those aged 75 to 84, and 5% of those above 84,drank at hazardous or harmful levels.257 Foremergency ward patients over 65, 19% of malesand 6% of females drank at these levels.257

Elderly people experience a high rate of social andemotional problems associated with bereavementand social isolation. Some elderly people, especiallywomen, develop harmful drinking patterns inresponse to these adverse circumstances later inlife.256, 258, 259 Such late onset drinkers respond betterto treatment and experience less harms than doelderly people with more entrenched drinkingpatterns.258

The majority of alcohol-related harms in the elderlyappear to be in those with an established drinkinghistory260 as, to some degree, alcohol damage iscumulative.259 Some of the harmful effects ofalcohol that tend to be associated with early-onsetdrinking include cognitive deficits and variouspsychiatric disorders. There are high frequencies ofdetectable cognitive dysfunction in elderly currentand ex-problem drinkers,256 and alcohol useincreases the risk of cognitive deficits.261

Differentiating alcohol-related dysfunction fromother types of dementing problems is quitedifficult.256, 261 Cerebellar/cortical atrophy is oftendetectable in elderly problem drinkers.256 A historyof heavy drinking in elderly men is associated witha five-fold increase in risk of psychiatric disorder.262

The interaction of prescribed drugs with alcohol is asignificant issue for the elderly since manycommonly prescribed medications in elderlypopulations interact adversely with alcohol.256 OlderAustralians living in the community are more likelythan younger Australians to be heavy users ofprescribed psychotropic drugs, mainlybenzodiazepines, for the treatment of sleepingproblems, anxiety and depression.263 Continualbenzodiazepine use over one decade was found tooccur in 16.6% of a sample of Australians aged 75and over.264 The use of benzodiazepines has beenassociated with increased risk of hip fracture,265, 266

and increased risk of motor vehicle crashinvolvement.267 WHO recommend prescribingbenzodiazepines cautiously to older populations,and to choose shorter duration benzodiazepinesbecause they are less likely to accumulate in theblood, which increases the risk of harmful sideeffects.203 There is a significant literature oncomplications associated with various types ofprescription medications. Australian studies havetypically reported that around 15 to 20% of allemergency department admissions for the elderlyare drug-related.268

��'�#� � �� $�� >=;�<�'+'�1==:��

$��%��!�� &��!� �,4 "��!� �,4) )

��( �� :=%1 ?%47�� /� �� 0 :%2 =%@

��$�� @2%= 1>%>

< � �� 2@%@ >@%1

*�� ?5%5 >?%=(��.��$�� $ �� 4%= 3%3

(��.��$�� $ ��/�� 0 ?%> 1%>

�� *�� 2%@ :%>

�� :1� =%? =%2*�� /�� 0 5%1 ?%?

�"��

��"�� -��!�! ��+�� 1��

For every drug type considered, Indigenous peoplehave higher rates of risky use than do non-Indigenous people, with correspondingly higherrates of drug-related harm.

('6'&�� -��-)*-� ��)-�

In 1994, the National Drug Strategy (NDS)conducted a national survey among 2993Indigenous people in urban areas—populationclusters of �1000 people, in which about 67% ofIndigenous people reside—referred to as theNDSHS-Urban Aboriginal and Torres Strait IslanderPeoples Supplement (UATSIPS).269 In the surveyreport, the results were compared to those of thegeneral population in the 1993 NDSHS. No nationalsurvey has been undertaken since and these data stillprovide the best base-line estimates of theprevalence of substance use among Indigenouspeople.

Tobacco

The 1994 NDSHS-UATSIPS reported that similarproportions of the Indigenous and non-Indigenouspopulations had ever smoked; however, thepercentage of Indigenous people who had smokedin the previous 12 months (54%) was 1.9 timesgreater than that among non-Indigenous people(29%).269 Indigenous people also reported taking upsmoking at an earlier age than did non-Indigenouspeople. Numbers of cigarettes smoked were loweramong regular smokers in the Indigenouspopulation.

Local and regional reports on the prevalence oftobacco smoking from various communities inNSW,270–273 NT274, 275 and of school-aged children inNSW276, 277 and WA278, 279 show some variation byboth region and gender but overall they confirmthat prevalence of smoking among Indigenouspeople is about twice that among non-Indigenouspeople.

Alcohol

The 1994 NDSHS-UATSIPS confirmed the results ofmany other smaller studies showing that fewerIndigenous people are current drinkers; those whodo drink do so less frequently, but more do so athigh risk levels.269 Three times more Indigenouspeople stated they were ex-drinkers compared withnon-Indigenous people. A pattern of episodicdrinking associated with acute consequences wascommon, with 70% of male drinkers typicallyexceeding six standard drinks and 67% of females

drinkers typically exceeding four standard drinks,compared to 24% of males and 11% of females inthe non-Indigenous population.269 Those who weremore socioeconomically disadvantaged were morelikely to engage in episodic risky drinking (and tosmoke cigarettes) than other members of theIndigenous population.280

Studies of alcohol consumption among Indigenouspeople have also been conducted at local,272, 281, 282

regional,283 or Territory levels.275 The results showconsiderable variation. Some of this is likely to be amethodological artefact but as with smoking, thestudies suggest geographic variation that is hiddenin the aggregate 1994 Survey results. This is alsosuggested by a study of regional variation in percapita alcohol consumption in the NT.284

Illicit and injecting drug use

In the 1994 NDSHS-UATSIPS, 54% of Indigenouspeople reported ever having used at least one illicitdrug, and 29% that they had used an illicit drug inthe previous 12 months (compared to estimatedpercentages of 42% and 18% respectively amongnon-Indigenous people). Forty-eight percentreported ever using cannabis and 22% had used it inthe previous 12 months (compared to 36% and13% of non-Indigenous people). For many,cannabis was the only illicit drug used. Of the otherillicit drugs in the previous 12 months, much lower(and therefore less reliable) rates were reportedbeing 2% for hallucinogens, 1.7% amphetamines,0.6% ‘designer drugs’, and 0.4% heroin.269

The 1994 NDSHS-UATSIPS also found that moreIndigenous people than non-Indigenous people hadeither ever injected (3% compared to 2%) orcurrently injected (2% compared to 0.5%) illicitdrugs.269 Taking this as a baseline estimate, Gray etal. estimated likely changes between 1994 and 2001in WA.285 They examined changes in hospitaladmissions for all conditions caused by drugs otherthan alcohol or tobacco, opioid caused conditions,psycho-stimulant caused conditions and drugpsychoses, hepatitis C notifications and drugoffences. On the basis of increases ranging from73% to 370%, they conservatively estimated that theprevalence of injecting drug use had increased bybetween 50% and 100%, the percentage ofIndigenous people who had ever injected wasprobably between 4.5% and 6% and that thepercentage of current injectors was between 3% and4%.

Studies among non-random samples of Indigenouspeople who inject drugs have raised concerns aboutthe young age at which injecting commences, and

�

�# �� !��"�#��$��%��

about the safety of injecting practices. The medianage at which people reported first injecting was 15in Western Australia, 16 in Brisbane, and 17 in theLower Murray.285–287 In the Lower Murray, 48%reported ever sharing needles and 28% that theyhad done so in the previous 12 months.287 Amongthose in the Brisbane study, overall 39% reportedhaving shared a needle in the previous month, butamong those aged less than 20 years 63% had doneso.286 In the Western Australian study 43%acknowledged ‘normally’ sharing needles.285

Use of inhalants

In the 1994 NDSHS-UATSIPS, 7% of Indigenousrespondents reported ever having inhaledsolvents—either petrol (4%) and/or other inhalants(5%) such as glue. This was about 1.75 times thepercentage reported among non-Indigenous people.The percentages reporting inhalant use in the 12months prior to interview were similarly very lowin both populations (0.8% compared to 0.7%).269

Rates of use, mostly occasional, are much higheramong young people.276, 278

The sniffing of petrol, as opposed to other volatilesubstances, is largely concentrated in smallcommunities in Arnhem Land, Central Australia andthe Goldfields region of WA.233, 288 In 1985,Freeman estimated that 49% of a population of 105people aged 10 to 14 years at Amata in SouthAustralian were either occasional or chronicsniffers.289 Similarly, in a non-random sample of 58males aged 13 to 32 years (31% of the malepopulation) in an Arnhem Land community 38%were current sniffers and 31% were ex-sniffers.290

More recently, it has been reported that in the year2000 in the Anangu Pitjantjatjara Lands in SouthAustralia, 6.2% of the total population and 12.2% ofthose aged between 10 and 34 years were engagedin petrol sniffing.291

Kava

The use of kava is confined to a small number ofcommunities in Arnhem Land. In a study of druguse patterns in the Northern Territory, Watson andher colleagues found that among a sample of 1764persons aged �15 years 6.9% drank kava (10.5% ofmales and 3.6% of females).275 The percentageswere higher among this Arnhem Land sample, butwould be less among the total NT Indigenouspopulation and considerably less among theIndigenous population of Australia as a whole.

Poly-drug use

As with non-Indigenous populations, there is acorrelation between heavy tobacco smoking andheavy drinking among Indigenous people.269, 275

Petrol sniffers in Maningrida were also more likelyto be cigarette smokers, heavy drinkers and lightkava users than were non-sniffers.290 In Albany WA,14% of 105 young people aged eight to 17 yearsand 48% of 15 to 17 year olds were ‘frequent poly-drug users’.278 High frequencies of poly-drug usehave also been found among non-random samplesof injecting drug users in both WA and SA.285, 287

('6'.�#��-��+��-)*-� ��-)-�� )-��)-��+�� -

Indigenous Australians are acutely aware of many ofthe health and social consequences of excessivealcohol and other drug use. In the 1994 NDSHS-UATSIPS, 95% of the urban Indigenous populationsampled regarded alcohol as a primary concern,63% stated that alcohol or alcohol-related violencewas the most pressing social concern, and 66%claimed it was the cause of most drug-related deathsin the Indigenous community.269 The report of theRoyal Commission into Aboriginal Deaths inCustody stated that ‘Alcohol is having a devastatingeffect on the Aboriginal people of Australia’ (vol 2,p299).292 This view was also put strongly in thereport ‘Too Much Sorry Business’293 and has beenmade in the context of Cape York Communities, byPearson.294

Mortality

Tobacco-related mortality is significantly higheramong Indigenous than among non-IndigenousAustralians. A WA study found the relative risk oftobacco-caused deaths for males to be more thandouble, and three to four times higher for females,compared with their non-Indigenouscounterparts.254 In the NT it was estimated that 23%of Indigenous male, and 17% of Indigenous female,deaths were attributed to smoking.295 A number ofreports implicate alcohol as a major cause ofexcessive Indigenous mortality.296–305 Unwin et al.estimated in WA, for the years 1983 to 1991, therate of alcohol-related deaths to be five times higheramong Indigenous males and between four and sixtimes higher for Indigenous females. The leadingcauses of alcohol-related deaths among Indigenousmales tend to be alcoholic liver cirrhosis, alcoholdependence syndrome and road injuries. AmongIndigenous women, they were alcohol dependence,cirrhosis and assault. In 1983 to 1991 it wasestimated that alcohol caused 9.6% of Indigenousdeaths in WA.306

�%��

There are no national data available on mortalityrates among Indigenous Australians related either toillicit drugs or volatile substance abuse.

Morbidity

Various local studies describe the impact of alcoholon clinic or hospital admissions within Indigenouscommunities;307 the excess of health problemsamong drinkers within Indigenous communities;281

the association of problems such as recurrentseizures with levels of alcohol consumption;308

proportions of consultations or hospital admissionsfor alcohol-related problems that are about twice ashigh among Indigenous than among non-Indigenous patients;309, 310 and higher rates ofalcohol-related hospital admissions.311

At the State or Territory level, Hunt cites NSWHealth Commission data showing Indigenous tonon-Indigenous hospital admission relative risks(RRs) of 9.3:1 for ‘alcoholism’, 3.0:1 for cirrhosis;RRs for other conditions known to be associated tosome extent with alcohol consumption ranged from1.5:1 for road traffic accidents to 16.7:1 forassault.312

For WA between 1989 and 1991, the Indigenous tonon-Indigenous RRs were 9.3:1 for males and12.8:1 for females (23:1 in 1994) for alcohol-caused mortality. The main causes of alcohol-relatedadmissions for males were assault, alcoholdependence and fall injuries; and for females wereassault, alcohol abuse and alcohol dependence.306

Gray et al. reported significant increases in bothhepatitis C notifications and crude hospitaladmission rates for conditions caused by drugsother than alcohol or tobacco among Indigenouspeople in WA between 1994 and 2000.285 In alarger WA study, Patterson et al. examined theincidence of first-time hospital admissions for illicitdrug problems between 1980 and 1995. Theyfound a dramatic increase during those years, fromaround half non-Indigenous rates to almost double.Largest increases were in admissions involvingamphetamines, although increases were also seenwith opiates, hallucinogens, cocaine andcannabis.313

There are no population-based studies of morbiditydue either to petrol sniffing or kava use. However,reported health problems associated with petrolsniffing include burns, weight loss, fitting,psychomotor impairment and psychosis.233 In thecase of kava, clinical observational studies havedocumented problems such as weight loss, scalinessof the skin and allergic reactions among heavyusers.314, 315

Violent crime

Indigenous people are frequently victims of alcoholand other drug-related violence and crime. Almosthalf those Indigenous people interviewed for the1994 NDSHS-UATSIPS reported that they had hadgoods stolen or damaged, more than a quarterreported being physically abused, and more than athird being verbally abused or threatened bysomeone influenced by alcohol.269 In WesternAustralia as a whole in the period 1994 to 2000,crude arrest rate for offences commonly associatedwith alcohol (liquor licensing, drink driving,against the person and good order) were 17.7 timesgreater among Indigenous than among non-Indigenous people; offences against the person were20.2 times greater.316 It has been estimated thatbetween 60 to 80% of violent crimes by Indigenouspeople in remote areas of Western Australia arealcohol-related.317

Family violence and child abuse

Family violence and sexual assault is common-placein drinking families and communities. Indigenouswomen are almost 40 times more likely than non-Indigenous women to suffer so-called ‘spousalviolence’.318 If they live in the country their risks ofviolence are even greater.318 While not all suchviolence is alcohol-related, much of it is.319, 320 AllIndigenous submissions to the Aboriginal andTorres Strait Islander Women’s Task Force onViolence cited alcohol as a major contributing factorto violence.321

An inquiry in WA, completed in 2002, found that‘… family violence and child abuse occur inAboriginal communities at a rate that is muchhigher than that of non-Aboriginal communities’(p xxiii).322 The authors of the report found that thecauses of family violence and child abuse arecomplex but implicated alcohol and other drugs inthe causal network.

Impact on families

Few Indigenous people have escaped the effects ofdrugs, particularly alcohol, on family andcommunity life. Even those who drink moderatelyor not at all belong to families in which harmfuldrinking has led to neglected children, under-nutrition and lack of parental supervision.321, 323 Formany young people reared in heavy drinkinghouseholds their nearest adult models are peoplewho spend much of their lives drinking,intoxicated, in gaol, or connected to violence andcrime.321, 324 These environments pose seriouschallenges for children’s schooling and socialdevelopment, and there is some evidence that boys

�

�' �� !��"�#��$��%��

may be more severely affected because of therelative absence of positive male models.325 In fact,many Indigenous people who have quit drinkingdid so because of the impact of their lifestyles ontheir children and grandchildren.326

Injecting drug users in Western Australia reportedthat their drug use caused strains on their partnerand sexual relationships, conflicts with their parentsand children and breakdown of friendships. Theyalso acknowledged the anguish that their drug useand its consequences caused to members of theirfamilies.285

In their review of petrol sniffing interventions,d’Abbs and MacLean cite various studies thatdocument the problems that petrol sniffers cause totheir families and communities. These includeparental distress, and intra- and inter-familialconflict.233

Suicide

Although Indigenous suicide is commonlyassociated with incarceration,292, 327 for more than adecade before the publication of the report of theRoyal Commission of Aboriginal Deaths in Custodyhigh frequencies of alcohol-related Indigenoussuicides had been documented in variouspopulations.281, 309, 328 Then as now, Indigenoussuicides are more likely to occur outside policecustody. Recent data from South Australia, WesternAustralia and the Northern Territory indicate thatage-specific Indigenous suicide rates are higher thanthose of the non-Indigenous population for all agegroups up to that of 40 to 44 years.329

Hunter’s detailed analysis of Indigenous suicide inthe Kimberley region of Western Australia over athirty year period demonstrated that more thanthree-quarters of all suicides between 1957 – 1989were heavy drinkers from families in whichdrinking was the norm.325 Hunter also showed thatthere had been both a dramatic increase in thenumber of mainly young men who had taken theirown lives and changing patterns of suicide overtime. Indigenous suicide is sharply concentratedamong young males and much more likely toinvolve hanging than any other method used by thegeneral population.329

As discussed below, the relationship betweensuicide, alcohol and/or other drug misuse andmental health problems is complex. Higher rates ofthese problems among Indigenous Australians areall associated with the same underlying socialdeterminants. However, the misuse of alcohol canexacerbate mental health problems and increase theprobability that a person with such problems willcommit suicide.

Social and emotional wellbeing

In the years 1998 to 1999, the Indigenous to non-Indigenous age standardised hospital separationratios for mental and behavioural disorders were 2.0for males and 1.5 for females. In the case of mentaldisorders due to psychoactive substance use, theseratios were 4.1 for males and 3.5 for females.330 In1997 to 1999, in Queensland, South Australia,Western Australia and the Northern Territory, deathrates from mental disorders among Indigenouspeople were twice those among non-Indigenouspeople; 78% of those deaths were attributed topsychoactive substance use.330 Such hospitalisationsand deaths are likely to be a small proportion of awider prevalence of mental health problems that areeither treated in primary health care settings or notat all.

Many of the other problems associated withsubstance misuse—such as assaults, suicide, familyviolence—are, in many cases, indicators ofunderlying mental health problems. However, notall mental health problems among Indigenouspeople are directly attributable to substance misuse.As with non-Indigenous peoples, many people withmental health problems also have substance misuseproblems that often reinforce and exacerbate eachother.331, 332

Culture and tradition

The relationship of alcohol and other drugs toIndigenous culture and tradition is complex. Someperceive a basic incompatibility between Indigenousculture and the use of substances such as alcohol.326

In a similar vein is the belief that drinking leads to a‘loss’ of culture.321 Against that view is the evidencethat traditional culture continues within heavydrinking communities and conversely, that in thosecommunities where dispossession and threats toculture from European settlement are less evident,alcohol and other drugs remain powerfullyattractive.174, 326 We need to look more broadly,therefore, at the multiplicity of factors thatinfluence drug use in these communities.

Economic costs

Indigenous communities have much higher rates ofun- and under-employment, poverty, poor housingand other social infrastructure. In thesecircumstances, the diversion of proportions ofincome significantly higher than the nationalaverage to alcohol and tobacco—and probably forother substances as well—has severe consequencesfor both users and their kin.269, 321

�(��

��#�� -��!�! ��$��21/��1��

Almost a quarter—23%—of the Australianpopulation, a total of 4.4 million people, is bornoverseas, 9% from mainly English speakingcountries and 14% from non-English speakingcountries.333 Australian research generally suggeststhat young people born overseas and from at leastsome non-English speaking backgrounds are lesslikely to use illicit drugs.123, 334, 335 However,generalisations cannot be made to all culturalgroups in all areas. For example, pockets ofproblems have been identified and there are highlyvisible problems of heroin use in some areas amongSouth East Asian youth.336, 337

There has been much discussion about the use ofalcohol, cigarettes and other drugs amongstdifferent immigrant groups, the mostcomprehensive of the Australian studies being thoseconducted by the Drug and Alcohol MulticulturalEducation Centre (DAMEC).338–342 These studiesshow that there are differences in the use of alcohol,tobacco and other drugs amongst some ethnicgroups, but these differences may be due toconfounding variables such as socioeconomic andmarital status.

Table 4.4 compares people from CALD populationswith the general population using 1998 NDSHSdata.

Table 4.4 shows that that in 1998, people with aCALD background were less likely to consumealcohol, smoke tobacco, or use any illicit drug thanother Australians.

Some methodological difficulties are found whenaddressing the use of alcohol and other drugs inethnic communities. It is very difficult to accuratelydefine differential membership of ethnic groups forresearch and statistical purposes. Two standard ways

by which ethnic groups are defined is by country ofbirth and by whether they or their families originatefrom a country where English is not the firstlanguage. There are problems in that the second andsubsequent generations will be registered in formal,large scale surveys and in the census by country ofbirth.343 This, however, implicitly assumes thatbeing of Chinese background is a definingcharacteristic irrespective of whether people areborn in China or Hong Kong or Australia. Somestudies have shown that alcohol use and mortality,where at least a proportion of deaths can beattributed to alcohol, tobacco and other drug use,can be consistent into the second generation,344–350

but other studies have shown there to be changesinto the second generation.343, 351

An important confounder is socioeconomicstatus,352–357 which has a clear demonstratedrelationship with both alcohol and cigarette use.358–

363 Rather than factors to do with the country ofbirth, it may be socioeconomic variables that betterexplain alcohol, tobacco and other drug use. Thesocioeconomic status of immigrants may welldiffer, depending on migration policy. For example,the more recent shift towards business migrationand a reduction in family reunion is reflected inmore affluent immigrants coming to Australia.333 Ithas also been shown that socioeconomic variableschange towards those of the Australian born thelonger immigrants are resident in Australia.351

A further potential confounder is religion and alsothe extent of religious engagement.364, 365 Inaddition to nominated identification with a certainreligion, the extent to which one engages in theceremonies of that religion is important.344, 365–367

The two key variables that emerge from theliterature as being important predictors of alcoholand cigarette use are acculturation and education.Acculturation increases the likelihood that alcoholconsumption will remain similar to that of one’sethnic group.368 Education predicts lower cigarettesmoking for African-Americans as well as

��'�'�� $�� :554�<�'+'�/�@=0��/� � �� 0

$�� 9�!� �� :!� �� 7�� ) ) )

�� 24�/:=0 >1�/5=0 @=�/4:08��.�� >1�/2@0 24�/>@0 :1�/>:0

�� 52�/>:0 ?�/250 2�/:40

�� 43�/@30 :>�/3>0 4�/110

�� % 4? :2 ?

�

") �� !��"�#��$��%��

Caucasians, the suggested mechanism being thateducation assists access to health messages.369

Ethnicity and drug use does not appear to be asignificant public health concern. Most identifiedstudies report lower rates of use of various drugs inmost ethnic minorities.184 It is possible that thedifferences in the use of alcohol, tobacco, and otherdrugs amongst some ethnic groups may be as muchdue to different sociodemographic profiles (e.g.socioeconomic status, marital status) as culturalfactors. If this is the case, then ethnic-specificprevention programs may be more appropriatelytargeted at the broader structural determinants ofhealth and the general risk factors for drug-relatedproblems rather than drug use-specific approaches.

��%�� !�� 3��!�

Rates of substance use are strikingly higher inpatients diagnosed with a mental illness of someform.370 There is evidence of two-way causalrelationships such that persons with pre-existingmental health problems are more likely to use arange of drugs for purposes of self-medication371

and also that substance use can exacerbatesymptoms of some mental disorders.178, 372 Extendeduse of stimulants such as amphetamines and cocainecan precipitate a psychotic episode in someone withno previous history of such conditions. It alsoappears that heavy cannabis use can precipitatepsychotic symptoms, especially in persons with apredisposition towards schizophrenia.373 Heavy useof alcohol has been found to exacerbate pre-existinganxiety disorders and elevate the occurrence ofagoraphobia and social phobias.178 Retrospective,prospective and experimental studies all indicatethat alcohol can reduce fear and anxiety in theshort-term but over days and weeks, drinkers’tendencies to experience these unpleasant emotionsare exacerbated.178 Similarly, there have beenreports of people who have become dependent onbenzodiazepines experiencing panic attacks for thefirst time.374 A recent major review conducted forWHO has found that the strong association betweenalcohol dependence and depression significantlyinvolves alcohol dependence preceding depression,more often than the reverse.375 A significant healthconcern is the dramatically elevated rates ofsmoking seen in mentally ill patients, especiallythose with a psychotic illness.376

The major source of information on patterns ofoccurrence of mental health problems and substanceuse in Australia is the National Survey of MentalHealth and Well-being (NSMHWB) conducted in

1997. This survey was conducted in three stagesinvolving: a large-scale population survey, a secondstudy on children aged four to 17 years, and a finalcomponent studying low-prevalence mental healthdisorders. The data have been analysed andpresented by a number of different scientists.179, 377

Seventeen percent of Australians will have a mentaldisorder (using ICD-10 diagnostic criteria) in any12 month period according to the NSMHWB data.Although substance use can be defined as a mentalhealth disorder, this figure does not includesubstance use disorders.377

The NSMHWB presents information on theprevalence of substance use disorders in people withco-occurring mental health disorders. These figurescan be seen in Table 4.5 below:

��'�6��$��

�� &�� ;��) �)

'�� ::%= 3%3�� ?%: :1%:

'�� 14%= :3%=

�� 3%: ?%3

'�� 23%= :>%=

Table 4.5 shows that men were more likely thanwomen to have substance use disorders or co-occurrence of other disorders with substance usedisorders. Women were more likely than men tohave anxiety disorders and affective disorders.

People with psychotic illnesses demonstrate elevatedrates of substance use. Extensive research has beenconducted to identify the Australian prevalence ofsubstance use in people with a psychotic-spectrummental illness,196 and to compare this to anappropriate age-matched control group. (Table 4.6)

��'�,� � �� /��E�<'++#-0��

$�� <��) )

�� >=%= 12%=

�� 1:%@ :@%=

# �� 11%> >%4

�� 5%1 :%27�� @%: =%>

"��

Table 4.6 shows that rates of every category ofsubstance use were higher for those with apsychotic illness than for the general population.

Individuals with antisocial personality disorder aresignificantly more likely to have alcohol abuse anddependency problems and are significantly morelikely to be aggressive whilst intoxicated.378

��'�� 0��-��!�! ��

The rates of drug use amongst those who arearrested have been extensively assessed as part of theDUMA project.202 DUMA compiles self-reportinterview data and urinalysis results from detaineesin a range of police lockups across Australia. Thesedata are used to provide information about thenature of drug use in those involved in crime.202 Itis a highly sensitive instrument and can providerapid information on changes in drug consumptiontrends amongst police detainees. DUMA tests foramphetamines, (including phenethylamines),benzodiazepines, cannabis, cocaine, methadone andopiates.221 Urinalysis detection of drug use isskewed by the fact that different drugs remaindetectable for differing degrees of time followingdrug consumption. ATS can be detected up to twoto days after use, as can methadone. Cocaine andopiates can be detected for two to three days only.Benzodiazepines vary greatly and can be detectedfrom two to 14 days after use. Cannabis is especiallypersistent and can be detected up to 30 days afteruse in some circumstances.221 Alcohol use is nottested but is part of the self-report questionnaire.

Table 4.7 shows the proportions of DUMA male andfemale respondents who tested positive to a rangeof drugs, in 2001.

��'�+�(� �� )��1==:��

� �� &�� "��) �)

�� 2= 2@- �&��& �� C 1: 22

�� @? @?

7�� CC :? 14

C ��

CC �� 41A�� 34��

Data on the prevalence of substance use amongprisoners (rather than detainees) is available fromthe Drug Use Careers of Offenders (DUCO) study, arandom sample of 2135 sentenced male inmates inQueensland, WA, Tasmania and the NorthernTerritory, conducted in 2001. The following ratesof drug use were reported (Table 4.8).

��'�5�(� �� 7��

� �� :!� ��) �� )

�� 55%2 5?%4�� 5>%3 4=%?

8�� 5=%= 41%4

�� 41%4 ?:%5

�� 4=%> >1%?�� >?%5 @@%2

�� @4%= 31%4

+�� @1%@ 13%:

+ �� 33%@ 14%1- �&��& �� 3:%> 12%>

�� 21%2 :?%@

) �� 1=%: :=%@

!�� :5%4 �3%:

Table 4.8 shows that rates of drug use in thispopulation are much higher than in the generalpopulation. The rate of use of illicit drugs otherthan cannabis is particularly high: over half of thispopulation were regular users of illicit drugs otherthan cannabis.

Harms: it is difficult to study and quantify the ratesof drug use and drug-related harm experienced inAustralian prisons. Obtaining access to inmates,representative samples and reliable estimates ofbehaviours are all intrinsically more difficult withinthe prison system.380 Illicit drug use in prison is aserious concern not only because the drugs used areillicit, but because of the risk of spreading BBVdiseases. Although drug injecting occurs in prisons,the most frequently detected drug in prisons iscannabis.200

Injecting drug use is common in Australian prisons,as it is worldwide.380, 381 Risk behaviours for BBVs(such as sharing needles) are also very common. Ona worldwide basis, rates of HIV and hepatitis C arehigher in prisons than in the general population.381

Roughly a third of Australian prison entrants testpositive for hepatitis C.380 Less than 1% of prisonerstest positive for HIV.380 This is reflective of the lowprevalence of HIV in the Australian population and

�

"� �� !��"�#��$��%��

in Australian injecting drug users, and the highprevalence of hepatitis C in injecting drug users.

A review of studies of injecting in Australian prisonsreports that over 50% of all injecting drug users ineach study reported injecting drug use while inprison.380 About 25% continue to inject while inprison and about 90% of those who continue toinject share injecting equipment. Access to cleaninjecting equipment is very limited. As many as66% of those who enter the prison system havehepatitis C.382

Rates of transmission of hepatitis C withinAustralian prisons are not known. Identifying thelocation of viral transmission down to a prison isparticularly difficult as the majority of inmates areon short sentences and are transient, making it verydifficult to identify in later years if any BBVinfections occurred in the prison or in thecommunity.380

��(�+!�� 3� �� !!��

('<'&�"�)��)-��

One of the major impacts of drug use—both legaland illegal—on the community is crime, both interms of economic costs and social costs, whichinclude loss of amenity in areas where street druguse and drug-related crime is perceived to beprevalent.383 The quantification of the extent towhich drug and alcohol use is associated with crimevaries according to the way it is defined. Aparliamentary report noted that if only illicit drug-related crime and intoxicated crimes were included,approximately 10 to 12% of crimes were drug-related.2 A broader definition, which includescrimes committed by drug users as well as drug-related crimes committed by non-drug users,estimates that 70% of all crime is drug-related,2 andthis figure is frequently cited. It does not, of course,indicate the extent to which drug-related crimes arecaused by drug use.

Recent Australian data indicate the extent to whichpersons in police and correctional custody are drugusers, and the types of crimes that they commit. TheDrug Using Careers of Offenders (DUCO) study, a2001 study of prisoners in Queensland, WA,Tasmania and the Northern Territory, found that ina sample of approximately 2000 male prisoners,almost three-quarters of the sample were regularusers of illicit drugs, and 55% were regular users ofillicit drugs other than cannabis. Over 80% usedalcohol regularly. The most serious offence (MSO)

of the majority (58%) was a violent crime and theMSO of 18% was a property crime. All of the othercrime types were MSOs for fewer than 10% of thesample.379 These figures, however, reflect the natureof offences for which people are incarcerated morethan they do the offending patterns of thepopulation.

The Drug Use Monitoring in Australia (DUMA)study is a monitoring study of police detainees thatcollects data every three months, in police lockups.Between 1999 and 2001, over 5000 male andfemale detainees were interviewed in four policelockups: two in Sydney, and one each on the GoldCoast and in Perth. In this sample, two-thirds had alifetime prevalence of use of heroin, ATS and/orcocaine, and around 40% had used at least one ofthese drugs in the past month. Their most seriouscurrent offences were property (33%), driving(27%), breaches (21%), violent (19%), drug anddisorder (11% each).

The nature of the relationships between use ofspecific drug types and offending is considered inmore detail below.

Alcohol

Australian and international studies have establishedthat alcohol is significantly associated with crime,particularly violent crime.384 The causal link inalcohol consumption and violence is beyonddispute. Many studies suggest that alcohol isinvolved in 40 to 50% of violent crime and a lesserbut substantial proportion of other crimes. Meta-analysis of experimental studies show that alcoholconsumption results in a measurable increase inaggressive behaviour, which may account, at least inpart, for the relationship between alcohol andviolent crime—but this is not the only mechanism.The relationship is a complex one and there aremultiple contributory and causal mechanisms,including the characteristics of the drinker, theeffects of alcohol, the drinking environment, andthe cultural expectations surrounding alcohol andviolence. Alcohol tends to increase the likelihoodthat people will respond aggressively or violently,and increases tendencies towards risk taking.385

Most alcohol-related violent crimes are committedby men—particularly young single men—and thiselevated rate is not entirely explained by higherrates of drinking. Alcohol-related crime is associatedwith deviance, power concerns and attitudes, andexpectations that aggression will be more acceptableif alcohol is involved. Violent crimes aredisproportionately associated with drinking inlicensed premises, and alcohol-related criminal

"��

behaviour is more likely in social settings wheremore than one person is intoxicated.385

The Australian National Alcohol Indicators Project(NAIP) has reported twice on alcohol-relatedviolence. The research group found that in 1997,124 deaths were attributable to alcohol-causedviolence, 4381 years of life were lost prematurelydue to alcohol-caused violence and 26 882 hospitalbed days were attributable to alcohol-causedviolence.195 In a later report it was estimated that in1998/99, 8661 people were admitted to hospitalfrom alcohol-caused assaults in Australia. Inaddition, 62 534 serious alcohol-related assaultswere reported to the police. Of the hospitaladmissions, 74% were male and two-thirds wereaged 15 to 34 years. Non-metropolitan areas hadhigher levels of alcohol-related assault thanmetropolitan areas. Most jurisdictions showedrelatively steady trends in alcohol-related violencefrom 1995/96 to 1998/99. It was concluded thatlevels of alcohol-related violence had not declinedin Australia despite community concern, theproliferation of Alcohol Accords and theintroduction of harm reduction strategies intolegislation.386

It is clear that the community is significantlyimpacted upon by alcohol-related crime. The 2001NDSHS found that 26.5% of the communityreported having experienced alcohol-related verbalabuse, 4.9% had experienced alcohol-relatedphysical abuse, and 13.7% were put in fear by aperson under the influence of alcohol.22 Studies ofSydney street crime have concluded that 77% ofstreet incidences (which includes offensivebehaviour, assault, offensive language, maliciousdamage, domestic violence and noise complaints)involved alcohol.384 Detailed analysis of NSW policerecords demonstrates that a wide range of assaultsand offensive behaviours are alcohol-related andthese are often associated with drinking, inparticular in licensed premises.384

Illicit drugs

The nexus between illicit drug use and crimeinevitably raises the question of which comesfirst—illicit drug use or crime. This is critical forpolicy in terms of whether crime control or drugcontrol measures should be pursued when dealingwith illicit drug use amongst the criminal justicepopulation. The criminological literature suggeststhat crime comes first, but the picture is complexbecause drug-using criminals are not ahomogeneous group.387 DUMA data shows thatwhile there are significant correlations betweendetainees charged with a violent and/or a property

offence and use of heroin and/or amphetamines,and the correlation between heroin and propertyoffending is highest and consistent across differentlocations, the overlap between illicit drug use andcriminal activity is not perfect. Most police detaineesdo not test positive to an illicit substance. However,there is support for an escalation model: drug-related offenders start committing crimes beforethey start using illicit drugs but use of such drugsbecomes part of the lifestyle and can lead todependency that increases the need to commitcrimes to support their drug use.387

There are high rates of crime amongst injectingdrug users. The IDRS found that more than half of asample of almost 1000 injectors had engaged in atleast one criminal activity in the previous month;most commonly drug dealing (39%) and propertycrime (20%). Almost half (44%) had been arrestedin the previous year, usually for the same offencetypes.201

The strongest link between illicit drugs and specificcrime categories is in regard to heroin and propertyoffending.193 Injecting drug use, in general, is anexpensive practice and heroin use is particularlyexpensive. Heroin-using property offenders commitmore crime than non-using property offenders, andthose who spend the proceeds on obtaining drugstypically commit further crimes.175 The price ofheroin makes it very difficult to sustain a pattern ofdependent use from a full-time income229 let alonefrom unemployment benefits, which is relevantgiven that there is a very high rate ofunemployment among heroin users.201 A 1998Australian review229 reported that in a sample of202 heroin users, 70% were involved in acquisitiveproperty crime and that illicit income accounted for82% of income in the week before the study. Theauthors estimated the total cost of heroin-relatedcrime as between $535 million and $1.6 billion peryear.229 However, more than half of heroin usershad committed crimes, most typically propertycrimes, before commencing use of heroin.229

Similarly, in the DUCO study it was found that, onaverage, offending commenced before the firstillicit drug purchase.379

A survey of incarcerated property offenders388 alsofound a strong link between opiate use and propertyoffending; and in DUMA it was found that 93% ofproperty offenders had tried illicit drugs, 85% hadused them in the past six months, 53% stated thatthey were addicted to illicit drugs, 41% blamedillicit use for their offending, and 26% stated thatthey were ‘hanging out’ for illicit drugs at the timeof their offence.2 However, recent analysis of

�

"� �� !��"�#��$��%��

DUMA data reveals that although it is commonbelief that most crime, particularly property crime,can be accounted for by ‘drug dependent’ persons,property offenders were more likely than others toreport that they were dependent on drugs; morethan half of them did not report being dependent.389

In terms of violent crime, concern is oftenexpressed about the association betweenaggression/violence and use of ATS.220 A specificextension to DUMA interviewed detainees at theEast Perth site who had used amphetamines in thepast year. These detainees reported strongassociations between criminal behaviour, beingaggressive and/or using physical force whilstintoxicated or withdrawing from amphetamines,but it is not clear whether there were causal linksbetween these activities.221 As noted above, the2002 IDRS found that methamphetamine usersreported aggression among other side effects ofuse.201

While it has often been alleged that steroid users areprone to violence there is only weak indirectevidence to support this claim. While it appears tobe true that steroid users experience higher levels ofaggression than normal, there is no evidence thatsteroid use is actually associated with increased riskor rates of violence.207, 210, 212 Petrol sniffing is alsoassociated with violence and anecdotal reportsfrequently note very aggressive behaviour whilstintoxicated.233

The use of illicit drugs is associated with the cost oflegal sanctions to users, their families and theircommunities. Most of the research in this arearelates to the legal and social costs of cannabisconvictions.173

Almost 80% of arrests related to illicit drugs arerelated to consumption rather than sale orprovision.184 The majority of criminal justiceresources allocated to drug use are utilised by minorfirst-time cannabis offenders, who, apart from theircannabis use, are in all respects a non-criminalsection of the community. An Australian reviewfound that the recording of a criminal convictionfor experimentation with cannabis use, or even thecultivation of small amounts of the drug, was a costfar out of proportion to the seriousness of theoffence.390 Other research suggests that many of thesignificant and demonstrably real harms associatedwith cannabis use are the direct consequences ofinvolvement with the legal system.391 A cannabisconviction can lead to negative employmentconsequences, further problems with the law,negative relationship consequences, andaccommodation consequences.173

('<'.��2)�+��--�)+

There is a significant link between sexual assault andalcohol,392 although, again, untangling causality isdifficult. Sexual assault and alcohol consumption, byboth victim and perpetrator, co-occur at a highfrequency—typically, either both victim andperpetrator were drinking, or neither was drinking.Approximately half of sexual assaults are committedby men who had been drinking, and approximatelyhalf of victims report that they were drinking at thetime—drinking alcohol lowers an individual’scapacity to resist sexual assault.392

A number of risk factors for sexual assault have beenestablished:

� general, heavy alcohol consumption,

� expectancies that alcohol increases aggressionand reduces sexual inhibitions,

� agreement with stereotypes that women whodrink are available, promiscuous or appropriatetargets,

� belief that alcohol is an excuse for sociallyunacceptable behaviour.392

Alcohol also enhances aggressiveness and increasesthe chance that a perpetrator will misperceive awoman’s behaviour as indicating consent oracceptance of sexual advances. At the proximal level,alcohol reduces the ability to evaluate risk, andmotor impairments reduce the ability to resisteffectively.392

The use of intoxicating drugs in sexual assault (so-called ‘date rape’) has occasioned some recentcommunity concern. The Australian Federal Policereported more than 70 cases of drug-facilitatedsexual assaults, which usually involved Rohypnol(the best-known brand of flunitrazepam), or similarsubstances. Rohypnol now contains a blue dye tomake drink spiking more difficult; however,flunitrazepam remains available under the brandname Hypnodorm (which does not contain dye).Flunitrazepam was placed on Schedule 8 of theDrugs and Poisons Schedule in 1998, partly due toconcerns about its use in rape cases. GHB has beenreported internationally as used in cases of rape andsexual assault. It dissolves easily in water, is notvisible in solution although it does have a salty taste,and rapidly produces stupefaction. There is a verysmall difference between the recreational dose andthe dose required to stupefy.165 There are somereports about ketamine being used in the same wayas GHB.165

""��

This area is a distinct gap in the literature. It is notpossible to make any sort of definitive commentsupon which drugs are used, the frequency of suchsexual assaults, or any other comment about the useof intoxicant drugs in rape.

Domestic violence

In the Women’s Safety Survey conducted in 1996,6.2% of Australian women reported that they hadexperienced physical or sexual violence from a maleperpetrator and 23% of married women or womenin de facto relationships reported violence(undefined) from their partner at some stage duringthe relationship.393 Being a victim of violence hassignificant flow-on effects other than the physicalharm suffered. Women who have experienceddomestic violence experience a range of problems ata higher rate than the general population; includinganxiety, depression, sexual and gynaecologicalproblems.394

Substance abuse, especially alcohol abuse, isstrongly associated with domestic violence. It hasbeen estimated that between 25% and 50% of menwho were physically abusive to their partners hadsubstance abuse problems.395 If other preconditionsfor domestic violence are met, and in men who arealready predisposed towards domestic violence,alcohol intoxication increases the risk of violence.395

Alcohol consumption also plays a role in increasedrisk of victimisation. Intoxication increases: the riskof being a victim of domestic violence, the risk thatthe partner will be drinking at the time of violence,and the chances of physical injury.395 A recentparliamentary committee, however, asserted thatthere were multiple causes of domestic violence andthat attributing causality or responsibility solely toalcohol consumption was inappropriate andinaccurate.2

('<'0��)*+��-��,�� ,

There is no doubt that many in the Australiancommunity believe that problems associated withsubstance use and abuse are associated withsignificant harms to the social fabric. Loss ofamenity, noise, litter, vandalism, aggression, pettycrime, violence and road safety issues have all beenfound to be concerns to the community that areassociated with alcohol use.396 These harms areparticularly relevant at the community level becausethey are personally experienced.

In terms of illicit drug use, the Capital Cities LordMayors made a statement on illicit drugs.

Users of the ‘hard’ drugs tend to gravitate to the major cities,both because of the ready availability of the drugs in cities andbecause of the attraction, to them, of the drug ‘culture’ inwhich they find comfort. The Councils of major cities inevitablyface special problems in dealing with the situation in whichdrug users find themselves. They face resentment that builds inthe community from awareness of the drug problem and itsnegative impact through crime. They must also deal withdisturbance of the normal trading environment for businesseswhich arises when illicit drugs are used and traded openly andtheir impact on normal amenities in the streets for residentsand visitors. There is also resentment at the substantial costsborne by the community as a result of the drug problem.(p61).383

A common theme, noted in this statement, is loss ofamenity. There is a perception that the communityhas lost access in part to its public places, because ofperceived threat or perceived presence of substanceuse or some of its undesirable sequelae such asdiscarded syringes, high crime rates, publicdrunkenness/intoxication or public violence.383

Highly visible drug use and high perceivedavailability of drugs in a community have beenshown to increase the risk that youth willsubsequently engage in harmful drug use.10 Thevisibility of a local drug use scene, particularlyinjecting, is also often associated with the level ofconcern experienced at the local community level.For example, visibly intoxicated people loitering inorder to obtain drugs and the discarding of drug useparaphernalia has been associated with elevatedlevels of community concern in a given area.397

Equally, street dealing has been documented to havean adverse impact on local communities and nearbysmall businesses.398

('<'(��-�� +��-

Families often feel powerless to deal with the druguse and related problems of other family members.They find drug use difficult and painful to accept,and many deny it until the evidence is too obviousto ignore at which point, not knowing what to do,they panic.399

Much of the literature on young people’s drug usefocuses on the family transmission of drug use, andchildren in families with dysfunctional drug use. Analternative view, however, is that parents, and otherfamily members, are the victims of stress associatedwith family members using drugs in ways that aredisturbing to the rest of the family.400 In one of thefew studies to examine this perspective, 50 closerelatives of identified problem drug users wereinterviewed and 35 individual problematicbehaviours posed by the drug users identified.

�

"# �� !��"�#��$��%��

These ranged from physical violence and verbalaggression, to stealing, lying, being manipulativeand/or behaving in an embarrassing way in front ofothers.401 A further study, in which 270 affectedfamily members were interviewed,402 found thatfamily members generally adopt one of three broadmechanisms to cope with such behaviours in asignificant other: tolerating, engaging andwithdrawing. The authors speculate that thesecoping positions may represent universal ways ofresponding to dysfunctional drug use and associatedproblematic behaviours, but that particular copingpositions may be more naturally adopted accordingto the nature of the relationship between the druguser and their significant other. (See also section8.2.2. ‘Assisting parents concerned by youth illicitdrug abuse’).

There are a number of agencies assisting families todeal with dependence or overdose deaths in theirfamilies, and representing their views. A concern isthat low numbers of parents access support servicesor education groups. The failure to attract parents inneed may be due to a difficulty in anticipating thebarriers that parents might experience in accessingservices.403 Most of the Australian programsdesigned to support parents and other familymembers affected by the drug use of a familymember do not appear to have been formallyevaluated.

('<'6��-�� 1�+��

The precise contribution of licit and illicit drugs towork-related problems (either productivity orsafety) is largely unknown.404, 405 Equally, thecontribution of the after-effects of alcohol use andprescription drug use is poorly defined.404, 405

Absenteeism, productivity, accidents and injuries,job satisfaction, employee turnover, the socialclimate of the workplace and the image of thecompany have been identified as critical factors thatmay be influenced by the level and pattern ofemployee alcohol and other drug use.405, 406 There isa considerable amount of literature on alcohol andother drug use, and work but systematic reviews ofthe area consistently conclude that the empiricalevidence is poor and does not well inform decisionmaking as to beneficial responses.405, 406 Overall, thescientific literature regarding the prevalence andnature of alcohol and other drug-related harm inthe workplace is equivocal and fragmentary.

The drug making the most significant contributionto drug-related harms in the workforce isundoubtedly alcohol.404, 405 The contribution ofillicit drug use to workplace-related harms is oftenoverstated.404, 405 Alcohol intoxication is inherentlymore disabling than intoxication with most illicitdrugs.404, 405 There is also a higher baselineprevalence of alcohol use.

Patterns of use and harm in the workforce willreflect those of the community where the workforceis located.407 English et al. give an aetiologicalfraction of zero for illicit drugs and 0.07 for alcoholand occupational injury in Australia.408 The impactof illicit drugs on workplace safety and productivitywould appear to be small and insignificant based onavailable evidence.405

There are consistent associations demonstratedbetween alcohol and drug use and higher rates ofabsenteeism; however, the evidence base is notstrong enough to demonstrate causation.405

('<'7��-�� )��

Alcohol

The most significant drug in regard to road traumais alcohol. Driving whilst under the influence ofalcohol was responsible for 418 road deaths and7789 hospitalisations in 1997.409 The cost of a roadfatality was estimated at $750 000, and the cost ofan episode of hospitalisation $132 000. Thisequates to a total cost of around $1.3 billion in1997.

It has been estimated that between 1990 and 1997,31% of all driver and pedestrian deaths werealcohol-related. These statistics do not includepassengers injured by drunk drivers.409 The majorityof alcohol-related road deaths occur between 10p.m. and 2 a.m. on Fridays, Saturdays, andSundays.409

Nationally, there was a downward trend in alcohol-related road deaths between 1990 and 1996. Themajority of these declines occurred in the first fewyears and were broadly mirrored by declines in percapita alcohol consumption at a national level.409

There was no such decline evident for non-alcohol-related road injuries.

Young people are greatly over-represented inalcohol-related traffic morbidity and mortality. Ofall alcohol-related serious road injuries, 52% occurin people aged 15 to 25 years.409 The average age ofa person killed on the road in an alcohol-relatedcrash is 27.5 years.

"%��

Cannabis

Cannabis reduces driving ability under controlledconditions; in simulators it reduces motor skills,reaction time and coordination. However, cannabisconsumption alone also decreases driving speed andrisk taking behaviour.215, 216, 410 There is substantialepidemiological evidence that cannabis use is notassociated with an increased risk of fatality, and ithas been speculated that this is due to a morecautious driving style in those affected.166

It has been hypothesised that the lack of anyapparent evidence that cannabis smokers are morelikely to be in road crashes is related to theirincreased levels of caution. A study by Robbemeasured real world driving performance on avariety of tasks, and reached the conclusion thatcannabis smokers exercise greater caution thannormal and are able to compensate for its adverseeffects when driving.411 It has been noted that whilealcohol-affected drivers over-estimate their abilityand do not attempt to compensate, cannabis-affected drivers under-estimate their ability and doattempt to compensate by driving more slowly,paying more attention, and being more cautious.412

Other reports have also found that low doses ofcannabis produce a lesser degree of relativeimpairment in driving skills under controlledenvironments than does a blood alcohol level (BAC)of 0.04, which is legal. A high dose of cannabisproduces a decline in skills and road-tracking abilitythat is less than that seen with a BAC of <0.08.412

While the overall impact of cannabis use in isolationon driving safety appears to be very small, there aresome circumstances where it is likely that cannabisconsumption increases risk. This includes crisissituations where the driver would be underconsiderable stress, prolonged monotonous drivingand operation of heavy machinery.412

The majority of scientific reviews in this areaconclude that cannabis use alone does not appear tocontribute to motor vehicle accidents;166 however,cannabis use in conjunction with alcohol use ishighly dangerous.

Drug combinations

A significant contribution to research on the role ofcannabis in traffic accidents has been made bytechniques utilising culpability analysis—theretrospective examination of fatal road crashes toascertain the fault attributable to each of the driversinvolved. A major study413 provided a great deal ofuseful information. Taken in conjunction withautopsy data on the presence of drugs in the

bloodstream, these methods can be used to examinethe effects of drug use in fatal road crashes. Thesestudies demonstrated that alcohol is by far thestrongest contributor to road crashes, even aftertaking into account the prevalence of alcohol useversus the prevalence of other drug use. The groupof drivers having both cannabis and alcohol in theirbloodstreams was also significantly more likely tobe in a crash. Overall, the drugs or drugcombinations found to be significantly associatedwith crash culpability were alcohol, alcohol andcannabinoids (cannabis), and alcohol andbenzodiazepines. Benzodiazepines taken alone alsoapproached significance. No other drugs or drugcombinations were found to be associated withsignificant culpability, including cannabinoids only,no drugs or alcohol, stimulants only, stimulants andcannabinoids, benzodiazepines and cannabinoids, orany other combination of drug type.215, 413

Drug driving by police detainees414

A specific analysis of DUMA data from 1999 to2000 looked at the extent of drug use among trafficdetainees, who were defined as any detainee whowas being charged with a driving-related trafficcharge. Generally, it was found that their drug usepatterns were similar to those of other DUMArespondents, but that 55% tested positive tocannabis and 37% to multiple drug use. Comparedto traffic detainees who tested negative to all drugclasses, drug positive detainees were younger, withless education, and more likely to have beenarrested in the previous year particularly for traffic-related offences. The report noted that thesedetainees were not representative of the drivingpublic, and concluded that roadside screening,random testing and compulsory blood testing wereall needed as strategies to control drug driving.

‘Dance drugs’ and driving

A major review conducted on English and Europeanliterature in an attempt to assess the use of so-called‘dance drugs’ (as defined by the authors to includeamphetamines, cannabis, LSD and ecstasy)concluded that there was little evidence to indicatethat this was a significant problem at present. NoAustralian literature addressing this question wasfound.244

Stimulants and truck driving

Between 25 and 50% of truck drivers are believed touse stimulants of various kinds to stay awake412 andone Australian study found that 21% of a sample ofdead truck drivers had stimulants (licit and illicit) intheir bloodstream.412 There have been concernsraised in the past that persistent stimulant use can

�

"' �� !��"�#��$��%��

exacerbate fatigue, with the driver rapidly becomingfatigued once the stimulants have worn off. As it isfatigue that is believed to be a major causal factor intruck driving fatalities, this remains an issue ofsome concern.412 Generally, culpability analysisreveals that truck drivers are less likely to be at faultin a fatal accident than car or motorbike drivers andso, overall, this population appears to demonstrate agood road safety record.412

"(��

��

� � � � � �

#) �� !��"�#��$��%��

#��

�� "��!!��

The link between poor health and disadvantage is now beyond dispute. While the precise meansby which deprivation causes ill-health is still debated, few disagree about the importance of thisrelationship. Similarly, there is a clear relationship between alcohol and other drug use and socialfactors, such as unemployment, low income and insecure housing. A variety of evidence hasassociated drug-related harm and poor health with low social cohesion. There is a complexrelationship between these broad social determinants and individual risk and protective factors,which means that some individuals do better than others on all health indicators, including drugmisuse, despite their material deprivation. Furthermore, these relationships are more pronouncedfor some drug types where they have more adverse outcomes than others. In fact, there isevidence that income is positively related to tobacco use among young people, and to volume ofalcohol use among the general population. Overall, the evidence base for the social determinantsof drug use is such that researchers and policy makers need to plan and implement a wide rangeof interventions that acknowledge some of the social origins of poor health and risky healthybehaviours at all levels—from the macro-social to the individual.

such as mortality, years of life lost and morbidity.This study concluded that a diverse range ofmeasures and determinants of health wereassociated with inequalities by socioeconomicgroup, ethnicity and gender.420

While there is general agreement about the impactof socioeconomic variables on health, themechanisms by which this occurs and the strengthof these relationships is disputed.421, 422 For instance,studies of the association between individualincome and health show a consistent relationshipbetween income level and morbidity and mortality,with those on the lowest income experiencing thehighest mortality rates and lowest health status.These relationships have been confirmed for theUnited States, Britain, many European countries andAustralia among males and females at all lifestages.423 Those claiming that this individual incomeinequality translates to poorer health for the wholepopulation, however, have been challenged byrecent discussions of the relationship betweenincome inequality and population health, with onereviewer claiming that new data on incomeinequality for 16 Western, industrialised countriesshow that ‘…the association between incomeinequality and life expectancy has disappeared’(p1).424 This has renewed debate about the natureof the relationship between inequality and health.421,

422, 425 In particular, it has focused attention on theway in which the contextual effects of inequality—

"��+��

Explorations of the links between social conditionsand health have a long history dating to at leastmedieval times in Europe, expanding under therubric of ‘social medicine’ in the 17th and 18thcenturies and including ground-breaking work bypeople such as Percival, Snow and Engels in the19th century, which illustrated the impact ofspecific features of urbanisation andindustrialisation on the health of the poor, inparticular.415

More recent work in this tradition has arisen sincethe publication of the Black Report in 1980: a UKGovernment sponsored review of thesocioeconomic determinants of health thatdemonstrated clear differences in the health profilesof occupational groups with those in higheroccupational categories experiencing better health.Since then, researchers have attempted to explorethe connections between social factors and health,and policy makers have struggled to conceptualisehow these findings could be translated into healthpolicy.416–419

One recent very comprehensive example in thistradition is a UK study that explored the impact ofsocial factors such as income distribution, belowaverage household income, education,employment, housing, homelessness, public safety,transport, ethnicity and sex, on health indicators

�

#� �� !��"�#��$��%��

such as low social status, for instance—rather thanincome inequality, per se, may influence healthoutcomes.421, 425

"�� !��

There are various models of the social origins ofhealth, but most postulate interconnected levels ofsocial determinants of health from the most broadlysocial to individual pathophysiologic/biologicpathways, including: macro-social factors (politicaleconomy, the cumulative effects of historicalfactors, social institutions, culture); distal socialconnections (neighbourhood and community);proximal social connections (family and friends);individual characteristics (socioeconomic,psychosocial and behavioural); geneticcharacteristics (human biology and genetics); andpathobiology (pathological biomarkers).415, 418, 420,

426–428

One of the challenges in these models isdetermining the way in which broad socialcategories such as class, for example, impact onhealth. The most common means of measuring theeffects of class on health and other outcomes is toconvert class to an index of socioeconomic status(SES). SES is used in most empirical research bycombining assessments of income, education andoccupation levels; then sorting the outcomes intolow, medium and high categories, often withreference to geographical areas of residence. SEStends to cluster in communities to create anenvironmental risk; as an individual risk factor,however, it can be influenced by governmentpolicies through, for example, employment andwelfare programs.419 Although SES can be used toindicate levels of economic inequality, it is generallya more narrowly defined concept than class, but it isnot uncommon for the two terms to be usedinterchangeably in the research literature.

Much of the research on the social determinants ofhealth and illness has focused on the effects ofsocioeconomic gradients on health. The term‘gradient’ in this context refers to an increase ordecrease in an outcome variable that relates towellbeing and is linked to an SES measure such asincome, education or occupation.429 Thosesupporting the argument for an unambiguousrelationship between material factors and healthpoint to the international evidence base thatsuggests that it is not the total wealth of a country,but inequalities in the distribution of wealth withinit, that determine health status.429, 430 Here the

concepts of material or absolute deprivation andrelative deprivation can be differentiated. Forinstance, countries with low levels of economicdevelopment may suffer from absolute deprivation,while relative deprivation exists within and betweendifferent classes or ethnic groups within developedcountries. Thus it is argued, it is not the wealthiestdeveloped countries that invariably have the besthealth outcomes but rather those countries that havemore egalitarian social structures.431 The effects ofabsolute and relative deprivation can be observed inthe socioeconomic gradient in health. Societies withinequitable income distributions tend to have steepsocioeconomic gradients in health, whereassocieties with more equitable income distributionstend to have shallower gradients.430 Thus, forexample, a middle class person in a society with asteep socioeconomic gradient may have worsehealth than someone of lower SES in a country witha shallow gradient. This phenomenon has beenobserved across all SES levels and research from theUnited States has demonstrated that this variabilitycannot be solely explained by differential access tohealth care, for example.430

However, the relationship between equality andhealth may not be as straightforward as previouslythought. A recent study of occupational classmortality in European nations found no correlationbetween greater equality in income and equality inhealth.432 Similar findings that refute the associationbetween income inequality and population healthhave been reported for Denmark, the United Statesand Japan (cited in 424). This evidence has led torenewed interest in the way in which the broadsocial environment, rather than simply income,influences health outcomes.421

The social environment can influence health andwellbeing through structures such as social supportsand levels of social cohesion. The concept of socialcapital has increasingly been identified as amechanism by which to explain relationshipsbetween social factors and health outcomes.433 Socialcapital has been defined by Kawachi as ‘thosefeatures of social organisation—such as the extentof interpersonal trust between citizens, norms ofreciprocity and density of civic associations—thatfacilitate cooperation for mutual benefit’(p1187).434 The influence of social capital ispotentially broad, impacting on governmentfunctioning and democracy, prevention ofdelinquency and crime, promotion of successfulyouth development and the ability to decreasesocioeconomic health disparities. Social capital, sodefined, has also been found to influence individual

#��

health, even after controlling for variables such asincome, education level and risk behaviours such assmoking.419

The effects of gender and ethnicity are not separatefrom class but form part of the web of socialrelations that create or deny access to wealth and theavailability of life chances. With respect to gender,for example, in addition to sexually specific diseasessuch as breast, cervical and prostate cancer, manydiseases occur in different patterns in women andmen. Although female life expectancy in thedeveloped world is greater than that of males andthe causes of death are similar, more women thanmen report illness and seek treatment. Menexperience higher rates of life threateningconditions and risk factors such as alcohol abuse anddangerous driving, while women have higherprevalence of painful non-lethal conditions such asmigraines and arthritis. Women reportedly suffermore psychological distress but men commit suicideat greater rates.435

Factors that interact with gender and impact onhealth include differences in exposure to risk factorssuch as smoking; the nature of gender-linked work;lower exercise rates among women than men;higher participation rates by men in risky activitiessuch as dangerous driving and contact sports;higher levels of physical aggression in men buthigher rates of victimisation in sexual and domesticassault in women. Many of these differences interactwith other social variables such as class andethnicity.435

Ethnicity may also impact upon health if a particulargroup has a history of both relative poverty andinformal and formal social exclusion.436 Socialexclusion may include: exclusion from civil societyby law or regulation (such as the detention ofasylum seekers in Australia); limited or no access tosocial goods for groups with particular needs (e.g.bilingual education for Indigenous children);exclusion from social production because of beinglabelled deviant or in need of control (e.g. injectingdrug users); and economic exclusion from normalsocial consumption (e.g. restricted visas for certainclasses of immigrants). Gay and lesbian people—particularly the young—may feel isolated by theirsexuality, so much so that they are a risk group forsuicide.437 Social exclusion thus adds to the burdenof other social determinants that may beexperienced by ethnically diverse populations, suchas the disproportionate concentration of somegroups (such as Indigenous peoples) in low SEScategories.

It is not simply broad social categories such as class,gender and ethnicity that have an impact on health.Religious affiliation and observation also appear toinfluence health outcomes. Membership ofparticular religious groups (such as the Mormonsand Seventh Day Adventists) and frequent religiousattendance appear to confer health advantages, withthe more devout enjoying lower mortality rates fora number of major causes of death.428

The links between these social determinants ofhealth and the pathways in ill-health at theindividual level are complex. For example, highstress levels can impede immune systemfunctioning while impoverished social networks,low-self esteem, high rates of depression, anxiety,insecurity and a poor sense of control all impact onquality of life.431 Some societal level factors such aslack of access to education can affect life chances.430

In turn, the stress associated with such inequalitycan impact on emotional wellbeing at the individuallevel.438 Negative emotional states impact onpsychological health manifesting in problems suchas depression, anxiety, anger and hopelessness, andon physical problems such as coronary heart diseaseand possibly cancer, although the causal factors forthese associations are complex.438

Social epidemiology has primarily studied SES andthe characteristics of individuals (psychological andbehavioural) associated with risk, protection,vulnerability and resilience.415, 439 Despite the socialdeterminants of health that confer poor healthoutcomes on categories of people, some individualsin those categories maintain good health andwellbeing. Researchers are exploring how at eachlevel, social factors, significant life events andtransitions can either ‘increase vulnerability and theprobability of an adverse outcome (risk factors) orreduce the risk or promote resilience and increasethe probability of positive developmental outcomes(protective factors)’ (p217).439

The relationship between macro-social factors andrisk and protective factors is complex. The outcomesof broad social determinants may also feed backand, in turn, influence those broader socialdeterminants. For example, risk and protectivefactors have been demonstrated to undermineattainment of individual capacities for cognitive andphysical performance, and to undermine theeffectiveness of education and economic access.From this perspective, risk and protective factorsmay also be important predictors of subsequentvariation in class mobility at the individual level.

�

#� �� !��"�#��$��%��

6'0'&��)-��+�� -��

Australian research on the social determinants ofhealth is growing.440–443 Research by Turrell andMathers shows that Australians at the lower end ofthe socioeconomic hierarchy suffer more ill-health,and that health differences by SES are apparent at allages. Morbidity and early mortality are moreconcentrated in people with lower SES, andAustralian Bureau of Statistics (ABS) data show thatthe morbidity gap is widening for certain conditions(coronary heart disease, lung cancer and motorvehicle accidents in men aged 25 to 64). On theother hand in women, colorectal cancer, breastcancer and melanoma are associated with higherSES.444

A 1992 Brisbane study examined the Person YearsOf Life lost by SES and found that Australians in thelowest SES lost 50% more years of life than peoplein the highest SES group.443 People fromdisadvantaged groups are more likely to go tohospital and seek medical consultations becausetheir health is worse than that of more advantagedgroups, but they are less likely to take advantage ofpreventive care and screening services.444 Inaddition, Kawachi, Kennedy and Wilkinson foundthat as unemployment tends to clustergeographically, this leads to impoverishedneighbourhoods with commensurate problemsincluding lack of role models, difficulty findingwork through contacts, high crime and delinquencyrates, as well as restricted access to good housing,prejudice from others and non-conformist attitudes;resulting overall in marginalised sub-cultures.445

Single parent females are also increasinglyconcentrated in poorer neighbourhoods, andadolescent employment rates in low SEScommunities are 80% of those in high SEScommunities.419

Less is known in Australia of the way in whichgender and ethnicity impact upon health, althoughthere is a clear gendering of life expectancy withwomen living almost six years longer than men.The fact that the gap between Australian male andfemale life expectancy has narrowed in the past twodecades indicates the social rather thanconstitutional nature of these changes.428, 435

As indicated above, ethnicity can affect health ifrelative deprivation is implicated. This means thatAustralian ethnic and minority groups with historiesof long-term social exclusion and materialmarginalisation are more likely to suffer the worsthealth. The most extreme example, of course, is thatof Indigenous people whose health is universally

regarded as dismal. The starkest indicator of this isthe much lower life expectancy —between 15 and20 years—among Indigenous people. Muchresearch and countless government inquiries havelinked poor Indigenous health to the colonial past,which has left a legacy of disadvantage and relatedill-health.325, 446, 447 Apart from the injusticesentailed, it is necessary to understand the history ofcolonialism that has shaped the government policiesand institutions that confront Indigenous people,and is at the root of the social inequalities faced bythem today. This inequality is reflected in a numberof key social indicators. Fewer Indigenous peoplehave post-secondary qualifications, are employed,or occupy professional, managerial oradministrative occupations, than the generalpopulation. Median individual income amongIndigenous people, aged 15 years and over, was74.1% of that among their non-Indigenouscounterparts; and among Indigenous families,median income was 68.2% of that of non-Indigenous families.448 While there is considerableevidence of the absolute material deprivationsuffered by Indigenous people, much less researchhas explored the impact of material inequality onpsychosocial factors.

The social and economic circumstance of migrantsis more complex, depending upon their country oforigin, English language competence, andrecognition of qualifications and skills in Australia,among other factors. The SES of immigrants maywell differ, also, with migration policy. Forexample, the more recent shift towards businessmigration and a reduction in family reunionmigration is reflected in more affluent immigrantscoming to Australia.333 In general, migrants inAustralia have lower levels of morbidity andmortality than the general population, largely dueto the screening processes of the post-warimmigration program. The longer their residence inAustralia, both socioeconomic variables and healthstatus tend to converge towards those of theAustralian born.351, 449

6'0'.��+�� -�� )-��

The evidence from both overseas and Australia isunequivocal in identifying social factors asdeterminants of health status. Where debate exists,it is about the relative importance of particularsocial factors and about the links in the causal chainfrom the macro-social to the individual. AsWilkinson points out with respect to the first issue,it is not simply that exposure to poor materialenvironments leads to ill-health but also that relativeincome, which dictates one’s social position, is also

#"��

influential.450 It is factors such as low social status,poorly developed social networks and lack ofcontrol over one’s work environment that may bemore potent contributors to chronic stress andhence ill health, than simply lack of materialgoods.421 These different emphases on absoluteversus relative deprivation have been identified as‘materialist’ and ‘psychosocial’ perspectives in theliterature (p54).415, 438

Others have developed this argument and claim thatthe focus on material deprivation has neglected theway in which cultural influences mediate healthoutcomes through psychosocial factors.438 Eckersley,for instance, writes about the way in whichsocioeconomic factors are amplified or moderatedby cultural determinants, in much the same way aspsychosocial pathways are posited for the socialdeterminants of health. Culture mediates the wayindividuals encounter and make sense of the world,and their experiences of inequality are alsocultured.438 This conflating of culture with socialcategories such as ethnicity or class has beenattributed to the dominance of epidemiology, withits origins in medicine, in the social determinants ofhealth debate. Anthropology, on the other hand, ismore likely to define culture as the system by whichpeople make meaning of their lives.451

As Eckersley points out,438 the importance of cultureto health appears in some of the classicepidemiological studies, ranging from the analysisof the association between exposure to Westerninfluences and increased coronary heart disease inJapan, to explaining low mortality rates in Roseta,Pennsylvania as a consequence of social cohesionand egalitarianism. However, this work has beenovershadowed, Eckersley and others claim, by thecontemporary focus on socioeconomic inequalities.What is the impact of modern Western culture onhealth? Eckersley suggests that the relationshipbetween culture and wellbeing is interactive andinter-related. Four facets to Western culture thathave positive and negative effects on wellbeing are:consumerism, individualism, economism (in whichsociety is viewed primarily as an economic systemwith choices based on economic considerations),and postmodernism (characterised by the loss ofover-arching truths in favour of relativism,ambiguity and fragmentation).438 Harm to healthand wellbeing may occur through the promotion ofanti-social values, moral ambivalence and thecontradictions and tensions between cultural idealsand social realities. In this way, ‘cultural influencescan interact with structural conditions to modifytheir social effects’ (p61).438

Eckersley applies this analysis to the health andwellbeing of young people and contrastsepidemiological emphasis on socioeconomicinequality to postmodern sociological literature andits focus on the cultural qualities of contemporarylife. It is the structural changes to family, work andeducation, he suggests, that has had more impact onyoung people.438 He cites research that indicates thatamong young people there is a general pessimismwhich suggests that they may be more influencedby individualisation and insecurity related to theunpredictable nature of modern life. Young peoplefrom privileged social backgrounds are concernedabout failure and their uncertain futures, whilethose from disadvantaged backgrounds may regardthe risks they face as ‘personal and individual, ratherthan structural and collective’ (p66).438

With regard to the second area of debate aboutcausation, rather than simply describing associationsbetween social factors and health—such as thestrong gradient of health by income, for example—calls have been made for researchers to isolate causalmechanisms by which these associations occur, suchas ‘the causal contribution of income on health asopposed to income being a proxy/correlate fornumerous other variables’ (p28).416 However,rather than enter into these increasingly arcanedebates about causation, Najman428 suggests:

…we might usefully think of causes that are closer to (andmore distant from) a health outcome. Rather thanconceptualising the causes of disease in binary terms (somethingis or is not a cause), we can more usefully think of causalpathways with some causes distant from the outcome, e.g.poverty, others at an intermediate point, e.g. cigarette smoking,and others more proximate, e.g. cellular abnormalities (p75).

The point here is the emphasis on the many levels atwhich the social determinants of disease occur.Overstatement of the degree of methodologicalrigour required for any analysis of the relationshipbetween social factors and health, at any level,should not inhibit attempts to demonstraterelationships between social determinants andhealth status.

6'0'0��+�� -��)��)-��

While the evidence for the social determinants ofhealth in general is now extensive, and models ofthe social origins well developed, less research hasbeen devoted to the way in which socialdeterminants impact upon drug use. However,there is a considerable body of work thatdemonstrates an association between social

�

## �� !��"�#��$��%��

determinants such as unemployment, homelessnessand poverty, and health-damaging behaviours,including drug misuse.363, 436, 452 Indeed, accordingto Jarvis and Wardle, the link between deprivationand health behaviour is strongest for alcohol andother licit and illicit drug use.363 Similar links havebeen found between social determinants includingunemployment, poverty, family disadvantage andcommunity and cultural factors, and crime.9 Therelationship between drug use and crime will bediscussed in a later section.

The relationship between deprivation and healthbehaviour has been consistently demonstrated forcigarette smoking, which is more frequent amonglower SES groups, people living in rented dwellings,those without private transport, the unemployedand those living in crowded accommodation. Thereis also a gradient by education level and by maritalstatus, with those who are divorced, separated orlone parents more likely to be smokers. Thestrength of this relationship led Jarvis and Wardle topropose a general law in Western, developedcountries that, ‘any marker of disadvantage that canbe envisaged and measured, whether personal,material, or cultural, is likely to have anindependent relationship with cigarette smoking’(p242).363

Disadvantage is not the only dimension in researchon social variables that is associated with alcoholand other drug use. There is also a literatureindicating that for some population groups,consumption is powerfully influenced by theamount of disposable income. An Australian studyof schoolchildren, for example, found that personalincome was a strong predictor of the number ofcigarette packs smoked in a week.453 Beingemployed was also found to be a strong predictor ofthe volume of alcohol consumed by respondents toa large Western Australian survey.454 There isevidence from international and Australian data thateconomic downturns are associated with asignificant reduction in per capita consumption ofalcohol, as well as in the prevalence of seriousalcohol-related harms—the recession in Australia inthe early 1990s was associated with reduced alcoholconsumption, reduced alcohol-caused deaths455 andalcohol-related road crashes.409, 456

There is little research that elucidates the precisemechanisms by which social factors such as income,employment and education influence excessivealcohol and other drug use. Rather, health-damaging behaviours related to poor diet,inadequate exercise, cigarette smoking, excessivedrinking and illicit drug use, appear to be

embedded in a complex network of socialdeterminants and risk and protective factors.457

These behaviours are also mediated by culturalinfluences.438, 451, 458 An important question iswhether or not broad social determinants such asclass, gender and ethnicity maintain anydetermining influence on drug use after controllingfor more proximal risk and protective factors.Available evidence supports the view that much ofthe influence of these variables is mediated by riskand protective factors.459

Drug use in Australia is not an isolated phenomenonbut one of a range of risk behaviours with commonsocial determinants, common risk and protectivefactors, and common outcomes. Class, gender andethnicity are the major determinants of inequality inhealth outcomes but other influences are found inthe sociocultural and physical environments.Developmental research has examined the directinfluence of gender, ethnicity and class on youthdrug use and also the potential for these factors tomediate and moderate the effects of otherdeterminants of youth drug use.419

Levels of risk and protective factors tend to differconsiderably by class, gender and ethnicity.However, the determinant relationship between riskand protective factors and youth drug use, inparticular, tends to be more similar than differentacross class, gender and ethnicity. Although there issome evidence that authoritarian parenting mayhave less adverse influences in some cultures, otherrisk factors tend to operate in similar ways.419 It ispossible that gender, for example, may influencethe impact of some risk factors. Bond and colleagueshave speculated that females may experience greaterdepressive symptoms in response to similar levels ofpeer and family conflict.460

Class or SES does not generally appear a strongpredictor of youth drug use in Australian follow-upresearch.123, 461 However, there is evidence for apossible threshold effect such that conditionsassociated with extreme poverty may result in verysevere drug use trajectories.10 In a graphic exampleusing the New Zealand Dunedin birth cohort,Fergusson, Lynskey and Horwood demonstratedthat when the focus narrowed on the 2.7% of youthexperiencing the most severe behaviour problemsearly childhood development was characterised byin-utero insult (maternal smoking and drinking),birth complications, problems in infant care (lowbreast feeding, low infant care and poor parenting)and social instability through childhood (familybreakdown and parental changes).125 The familybackgrounds of these children were characterised by

#%��

high levels of social disadvantage (teenage parents,low education, sole parents) and socialdisconnection (low rates of church attendance, andparent mobility).

Fergusson and his colleagues also found thatincreasing exposure to unemployment wasassociated with increasing risk of psychiatricdisorders (which included nicotine and othersubstance dependence), independently of pre-existing family and personal factors.462 Conversely,there is some evidence that being employed canimpact negatively on substance use. For example,Spooner cites research showing that employedyoung people have higher levels of alcoholconsumption.419

Developmental research prior to the 1990s tendedto find males had higher rates of adjustmentproblems and were more likely to become involvedin harmful drug use.10 Through the 1990sAustralian data suggested rates of involvement indrug use for girls had tended to equalise with thosefor boys.153 Although these changes have beenmoderately large, the factors influencing the trendsare poorly understood. One possible explanationmay involve a tendency for families to useincreasingly similar child rearing practices for boysand girls. The trends do suggest the potentialimportance of cultural influences in thedetermination of youth drug use behaviours.419

The literature on risk and protective factors withregard to drug use and young people has beenreviewed in this document. One Australian studyreported that for adults being married, or in a defacto relationship, appeared to exert a protectiveeffect on both volume and pattern of alcoholconsumption.454 This finding concurs with a bodyof literature around treatment outcomes thatconsistently reports that persons with familysupports and stable relationships have betteroutcomes for treatment.5 However, a review ofsocial and demographic factors in the adultpopulation could not be located and it isrecommended that this should be a future researchpriority.

With respect to ethnicity and drug use the evidenceis not clear-cut. Australian research generallysuggests that adults and young people born overseasand from some non-English speaking backgroundsare less likely to use drugs.123, 334, 335 However,generalisations cannot be made to all culturalgroups in all areas. For example, pockets ofproblems have been identified, such as heroin useamong South East Asian youth in south-westernSydney336 and in Melbourne.337

The Victorian Department of Human Services hasrecently published a study of the involvement ofethnic communities in Victoria with illicit drugs.463

A major finding was that SES—manifest in highyouth unemployment and low levels of literacy—more than ethnicity per se, was the major contributorto high risk behaviour and drug use in culturallyand linguistically diverse communities. Familyfactors were identified as important mediators.These factors included lack of discipline for youngpeople, unrealistic pressures on children to succeed,lack of communication in families, lack of effectiveparenting skills and supervision, and generational/cultural conflict.

Higher rates of drug use among some immigrantcommunities may stem from family isolation,337

family disruption associated with traumatic refugeeexperience,464 and/or loss of parental control overadolescents due to differential acculturation and rolereversal.465 On the other hand, having rules andgood parental supervision have been found to beprotective against substance use among adolescentsfrom some ethnic communities.336

The adverse health and social impacts of drug use,particularly alcohol, among Indigenous Australianshas been attributed by most researchers to a broadrange of social determinants. A comprehensivereview of the relationship between excessive levelsof alcohol misuse, related harm and structuralfactors is provided by Saggers and Gray.174 Based ontheir own research and a detailed review of theliterature, they describe the similar patterns ofalcohol misuse and related harm among Indigenouspeoples in Australia, New Zealand and Canada. Theyshow that there is no evidence for the biologicaldetermination of these common patterns, and thatsocial patterns of misuse are not explicable in termsof individual pathology. They also show that arange of cultural factors play a role in observedpatterns of alcohol misuse. Some cultural factorsmediate the impact of broader political andeconomic factors and others, such as thedevelopment of destructive drinking patterns, feedinto a continuing cycle of poverty and disadvantage.However, the diversity among these Indigenouspeoples precludes explanations based solely oncultural factors. The common patterns of alcoholmisuse are related to the common experienceamong these Indigenous peoples of colonialism,dispossession and economic exclusion and theircontinuing consequences.

A concrete example of the determining role ofcolonialism and dispossession in Indigenous alcoholmisuse is provided by Hunter.325 He describes

�

#' �� !��"�#��$��%��

patterns of ill-health and alcohol consumption, andrelated harm, among Indigenous people in theKimberley region of Western Australia. He linksthese to the colonial history of the region and to theforced exclusion of Indigenous people from theirtraditional lands following the Conciliation andArbitration Commission decision granting equalwages to Indigenous workers in the pastoralindustry, which came into effect in 1968.

Although it is not generally their main focus, anumber of studies link substance misuse amongIndigenous Australians to these social indicators. Ajoint publication by the Australian Bureau ofStatistics and the Australian Institute of Health andWelfare (ABS & AIHW) summarised the results ofthe two most comprehensive national surveys ofIndigenous people that both linked cigarettesmoking to such social indicators.269, 280, 466, 467

An extensive analysis of NATSIS [National Aboriginal andTorres Strait Islander Survey] data revealed that both Indigenousmales and Indigenous females aged 15 and over who hadcompleted at least year 12 at school were less likely than thosewho left school earlier to report that they smoked. Indigenouspeople in forms of employment other than CommunityDevelopment Employment Projects (CDEP) (a work for socialsecurity entitlements program) were less likely to report thatthey smoked than those in CDEP scheme employment, theunemployed and people not in the labour force (Cunningham1997). Similarly, in the NHS [National Household Survey],Indigenous adults aged 18 years and over from non-remoteareas were less likely to report smoking if they were employed(49%) than if they were unemployed (63%) or not in thelabour force (55%) (p53).280

As summarised by Ivers, two smaller studies havelinked—or partially linked—tobacco smokingamong Indigenous Australians to workforce andemployment factors.468 The first, a study of 306Indigenous and 553 non-Indigenous people in twoVictorian towns, found that 66.9% of Indigenousmales and 24.0% of non-Indigenous males werecurrent cigarette smokers.273 When the analysis wasconfined to those not receiving a pension, benefit orallowance, the proportion of Indigenous males whowere current smokers was reduced to 39%.However, a similar relationship was not foundamong Indigenous females. In the second study,Hogg reported that among a sample of 273Indigenous people in New South Wales, theprevalence of tobacco smoking was 54% amongthose who were unemployed but only 27% amongthose who were employed.469

The ABS and AIHW publication also linked alcoholmisuse among Indigenous people to education,employment and income.

An analysis of Indigenous drinkers aged 18 years and over inthe NHS showed that those in the high risk category were lesslikely than low risk drinkers to have a higher educational degreeand more likely to have left school before the age of 15, to beunemployed or not in the labour force, to earn the majority oftheir income through government pensions, to earn less than$10,000 per annum … Although the numbers of people ineach category are small, the patterns are consistent in suggestingthat high risk drinking among Indigenous people is morecommon among the socioeconomically disadvantaged (p55).280

In a study in Albany, Western Australia, Gray et al.found that among 105 Indigenous people, agedeight to 17 years, 15% were ‘poly-drug users’(occasional users of some combination of tobacco,alcohol and cannabis) and 14% were ‘frequentpoly-drug users’ (frequent users of somecombination of tobacco, alcohol and cannabis andoccasional users of volatile substances or otherdrugs). Among those aged 15 to 17 years, theproportion of frequent poly-drug users was 48%.278

Logistic regression analysis showed that amongchildren aged eight to 14 years, those who weredisaffected from school were 23 times more likelyto be poly-drug users; and that among those aged15 to 17 years, those who were unemployed were13.5 times more likely to be ‘frequent poly-drugusers’ than those who were employed, in training,or still at school.279

These socioeconomic factors are indicators of acomplex network of structural determinants ofsubstance misuse among Indigenous Australians thatalso includes institutionalised racism and thecultural and psychological impacts of colonialismand dispossession. Interventions that focus oneconomic inequalities alone will not prevent themisuse of alcohol and other drugs amongIndigenous Australians. However, without greaterfocus on remedying these up-stream socialdeterminants, the effects of other interventions willbe critically circumscribed.

6'0'(��+�� -/��)��)-��

Interventions for drug misuse are based on a widerange of factors, some of which have little to dowith the evidence base for health-damagingbehaviours. Given what is now known of therelationship between social determinants and healthstatus, including drug use behaviour, researchersand policy makers need to demonstrate that therecommendations they make for interventions canbe linked to this research base.

#(��

Attempts have been made to model the wayinterventions may be targeted at each level at whichthe social determinants of drug use impact.6, 7 Forexample, Lenton has adapted Holder’s ‘alcoholprevention conceptual model’, which conceptualisesprevention activities as being targeted at a numberof the levels, from the macro-social (global, stateand nation) to the individual. Lenton added to thisthe ‘conditional matrix’ of Strauss and Corbin,470

which can be represented as a series of concentriccircles where each level corresponds to a differentaspect of the environment. The resulting model forthe prevention of alcohol and other drug problemscontains levels that allow for differential preventionactivities, mechanisms of action and context (Figure1.1). Importantly, influences flow both down themodel, from distal to proximal levels, and are fedback up the model from the proximal to the distal.6

The Public Health Systems model outlined inChapter 1 illustrates how there are a number ofpolitical and economic mechanisms of actionoperating at global, national and state levels that canimpact on drug availability, use and harm. Theseinclude: treaties, conventions, enforcement andpolicy coordination at the international level; theenforcement of policies, laws and regulations, taxesand excise on substances, and health and welfarespending at the national level; and, in addition tothese other factors, licensing and education policy atthe State level. There are also a number ofprevention options available for intervening at theselevels.6 For instance, the Health InequalitiesResearch Collaboration was established by theCommonwealth Department of Health and Ageingto create new knowledge leading to the reduction ofhealth inequalities through research on policies andinterventions.440

In terms of interventions that take account ofdevelopmental research on children and youngpeople, there appears to be some consensus aboutthe importance of developmental pathways inleading to problematic behaviour, investing inchild-centred institutions and policies, andidentification and manipulation of multiple risk andprotective factors at different social levels andtransition points (p72).457 Specifically, Cashmoresuggests it is necessary to:

• intervene early in life and early in thedevelopmental pathway,

• aim at reducing the accumulation of risk atmultiple levels (child, family, community,society),

• use a coordinated approach that takes intoaccount common risk and protective factors,

• make any interventions acceptable andaccessible to the participants, including thechildren and the young people themselves,

• target transitions and prepare and supportchildren and their families through transitions;and evaluate preventive interventions to learnwhat makes a difference and why (pp219 –220).439

With respect to drug use, to take just one of theabove it seems clear that a number of risk andprotection factors are common to a range of adversedevelopmental outcomes and social problems thatare frequently dealt with as separate domains,including school failure, teenage pregnancy,domestic violence and substance misuse (p221).439

One model pioneered in the US,10 and trialled inboth the UK and Australia,459 focuses on communityorganisation or mobilisation and:

… involves the systematic identification and measurement ofrisk and protective factors in a selected community (utilisingmainly official data and a standard questionnaire completedby adolescents), and the selection, implementation andevaluation of appropriate evidence-based interventions by acommunity prevention board.

The extent to which risk and protection factors thatare inherently complex can be identified andmeasured is still problematic.457

"��2 ��

There is now incontrovertible evidence linking poorhealth to disadvantage. Although debate willcontinue about the precise means by whichdeprivation causes ill-health, few disagree about theimportance of this relationship. Similarly, there is aclear relationship between alcohol and other druguse and social factors, such as unemployment, lowincome and insecure housing, although much ofthe influence of these variables is mediated by riskand protective factors. A variety of evidence hasassociated drug-related harm and poor health withlow social cohesion. There is a complex relationshipbetween these broad social determinants andindividual risk and protective factors, which meansthat some do better than others on all healthindicators including drug misuse, despite theirmaterial deprivation. With respect to tobacco andalcohol, there is also a literature identifying somepositive associations with income and employment.The issue of the extent to which alcohol and drug-

�

%) �� !��"�#��$��%��

related problems are located within thedisadvantaged sectors of Australian society will bediscussed in the next chapter. In general, theevidence base for the social determinants of druguse is such that: researchers and policy makers needto plan and implement a wide range ofinterventions that acknowledge the social origins ofpoor health, and how poverty and associateddisadvantage maintain this poor health and riskybehaviours at all levels—from the macro-social tothe individual.

%��

�� #��!!��

This chapter examines current evidence for the developmental influences that lead to subsequentpatterns of drug use that have a high likelihood of harm. Follow-up research studies provideinformation relevant to the antecedent factors that lead to drug use. Risk factors were defined onthe basis of their tendency to independently predict involvement in early and heavy youth druguse. Protective factors were defined as influences that modify the effects of risk factors while notdirectly predicting drug use. Based on these definitions, effective harm minimisation strategieswere defined as protective factors. The research summarised in this chapter highlights some ofthe major risk and protective factors influencing drug use over the course of development.

Inherited vulnerability (for males), maternal smoking and alcohol use, extreme socialdisadvantage, family breakdown and child abuse and neglect were amongst the earliest riskfactors that increase the likelihood that children develop behavioural and adjustment problemsand subsequently become involved in harmful drug use. From the age of school entry, earlyschool failure, childhood conduct disorder, aggression and favourable parental attitudes to druguse all appear to be risk factors for drug use problems. From adolescence, low involvement inactivities with adults, the perceived and actual level of community drug use, availability of drugsin the community, parent-adolescent conflict, parental alcohol and drug problems, poor familymanagement, school failure, deviant peer associations, delinquency and favourable attitudes todrugs were all identified as risk factors for harmful drug use. Community disadvantage anddisorganisation, positive media portrayals of drug use, adult unemployment and mental healthwere further factors strongly associated with harmful drug use. The evidence was unclear as tothe role of childhood Attention Deficit Hyperactivity Disorder, intelligence, anxiety anddepressive symptoms in the prediction of harmful drug use.

Early age protective factors included being born outside Australia, having an easy temperament,social and emotional competence, and shy and cautious temperament. Protective factors inadolescence included family attachment, parental harmony and religious involvement; and inadulthood, well managed drinking environments, and marriage.

Little is known of the risk and protective factors influencing the subsequent course anddevelopment of drug use in later life. The research at this stage is limited regarding the factorsinfluencing Indigenous youth involvement in drug use.

leading to drug-related harm. The main strategy inthis chapter is to update previous literature reviewsby examining a selection of well-conducted studiespublished over the past decade. Earlier literature isincluded where it is relevant to historical trends inrisk processes.

#��+��

Earlier sections of this report identified patterns ofdrug use that exact high levels of economic andsocial harm. Regular tobacco use, alcohol abuse andalcohol dependence, frequent adolescent cannabisuse, illicit drug use and poly-drug use have eachbeen documented to lead to harm. In this section,we examine current knowledge of the predictors ofthese harmful patterns of drug use, drawing bothon longitudinal research and intervention studies toarrive at judgements regarding the major influences

�

%� �� !��"�#��$��%��

#��$�� 4��$�� !��5�� $��

There is a growing emphasis in Australianprevention programs upon a ‘developmentalpathways’ approach. This approach, emphasised inAustralian mental health471 and crime preventionstrategies,472aims to direct evidence-basedinvestment to modify the early developmentalpathways that lead to later problems. The approachhas emerged through the synthesis of a range ofscientific endeavour but draws in important wayson life-course development research, communityepidemiology and preventive intervention trials.473,

474 In common with the broad area known asprevention science, the developmental pathwaysapproach seeks to prevent health and socialproblems by identifying and then reducing theinfluence of factors that lead individuals and groupsto subsequently develop health or social problems.The approach is based on techniques developed toprevent health problems within the fields of publichealth and epidemiology.8 Although an ultimateunderstanding of underlying causal processes issought, in general this approach aims initially tounderstand the probabilistic relationship betweenearly indicators and subsequent problems.

7'0'&��)��)-��-��-1�� -

Drug use predictors are characteristics measuredbefore the emergence of drug use behaviour and arestatistically associated with an increased probabilityof subsequent emergence of the predictedbehaviour. Predictors are typically identified usingfindings from follow-up (or longitudinal) researchstudies; however, in cases such as genetic research,alternative methodologies may be relevant.

By increasing understanding of the major factorsthat can be reduced in order to prevent healthproblems, the identification of risk factors has provenuseful in public health. The terms predictor and riskfactor are often used interchangeably. However, inthe current document risk factors will bedistinguished as a specific class of predictors thatrepresent theoretically independent domains.Independence in this context relies upon, firstly, theability to empirically distinguish the predictor as aunique measurement domain, and secondly,evidence that the risk factor predicts the targetbehaviour after adjustment for other knowninfluences. Risk factors can also be furtherdistinguished by their potential to be modified

through intervention. This emphasis on malleabilitydistinguishes risk factors that have been emphasisedin the current document as potential preventiontargets.

Protective factors have tended to be loosely defined andare often empirically measured, as the opposite endof a risk factor. For the present purposes, protectivefactors were defined as a special class of predictorsthat act to moderate and mediate the effect of riskfactors.10 To avoid confusion, it has been furtherspecified that protective factors do notindependently predict drug use behaviour, but areimportant due to their capacity to reduce theinfluence of risk factors. Where a purportedprotective factor has been shown to independentlypredict drug use behaviour (i.e. prediction ismaintained after adjustment for other known riskfactors), the interpretation used in the currentdocument leads to it being reclassified as a riskfactor. By avoiding ambiguity, the ground is clearedto make an accurate assessment of the factors thatdirectly and indirectly influence the development ofdrug use.

Risk and protective factors are defined inrelationship to the outcome being predicted, thedevelopmental age, and the stage in thedevelopment of the behaviour being predicted. Thetiming of developmental events has importantconsequences and hence the age at which particularexperiences and events occur is an importantdeterminant of their developmental impact. Forexample, initiating alcohol use at the legal drinkingage appears to have very different developmentalimplications than alcohol use initiated in childhoodor early adolescence.

Harm minimisation has been an important frameworkguiding Australian drug abuse prevention effortssince the mid-1980s. Harm minimisation strategiesinclude efforts to ‘minimise the risk of harm from’(p123) drug use.475 Accepting that harmminimisation strategies do not necessarily influencedrug use but rather reduce the risk of harm fromdrug use, it then becomes clear that many harmminimisation strategies can be defined as protectivefactors.

7'0'.��-1��--�-

The current state of the science of preventionsuggests that risk factors influence the course ofdevelopment through their cumulative impactacross time. This means that there is no single riskfactor that lies at the heart of developmentalproblems. Rather, these problems can be regardedas having complex causes, or multi determination.

%��

The more risk factors that persist over longerperiods of time, the greater the subsequentdevelopmental impact.476, 477

From one view, the cumulative effect of risk factorsoperates somewhat like a snowball. According tothis view, risk factor exposure early in life canimpair the subsequent course of development andlead to a snowball effect, with subsequent riskfactors tending to adhere and accumulate as aconsequence of the earlier problems.478 So, forexample, mothers’ tobacco smoking may impedefoetal development resulting in cognitive deficitsthat then lead to poor school adjustment. Poorschool adjustment and school behaviour problemsmay lead on to social aggregation with other poorschool achieving youth. From this perspective, thesolution is to check the avalanche of risk byintervening at the top of the mountain, the earliestpoint in the course of development.

From a slightly different perspective, the cumulativeeffect of risk is more analogous to a snowstorm.According to this view, a child can withstandextreme weather for a brief period but over time thechances of illness through exposure increase. Forexample, a healthy child may withstand drug use inthe peer group and community for a period but,over time, if this behaviour is common the chancesof the child becoming involved in drug useincrease. Both parents who are unavailable and badexperiences with teachers may increase the chancesof the child becoming interested in drug use. Theprotective benefits of positive relationships withadults, observed even for children with damageddevelopmental pathways, are again suggestive of thepotential to protect health in bad environments byreducing further risk exposure and providingprotection (analogous to providing shelter in astormy environment). From this perspective,solutions lie in improving social environmentsthrough the course of development,479

strengthening the child’s capacity to survive riskyenvironments and enhancing protective factors toreduce the impact of risk.

In attempting to explore the development ofbehaviour, an understanding of the sequencing ofrisk processes across the life span becomesimportant. At one point in development, a riskfactor (such as early age alcohol use) can be studiedas a predictor of subsequent behaviour (e.g.cannabis use). However, at an earlier stage indevelopment the same factor may be seen as anoutcome to be predicted by prior risk factors (e.g.childhood behaviour problems). In what followswe have attempted to organise the developmental

research by working forwards through the lifestages. Beginning with early influences, we haveattempted to identify the chain of development(pathways) leading to subsequent harmful drug use.

Identifying the independence of risk and protectivefactors requires that their positive prediction of druguse be maintained after adjusting for otherinfluences operating at that age. Clearly until theinfluence of all alternative risk factors has beencomprehensively examined, accepting that riskfactors are independent remains tentative and opento revision as further evidence accumulates. In thecurrent report if a risk factor maintained its effectafter adjusting for a range of similar risk factorsoperating at the same stage of life and within thesame domain (e.g. family, peer group) it wasregarded to be independently predictive of druguse.

#��6�� $��

The major sources of evidence for identifyingdevelopmental influences on drug use have beenlongitudinal, self-completed, follow-up surveys.Depending on the stage of development, therespondents have either been children, adolescentsor adults. In some cases, such as in researchinvestigating early life influences, observationsmade by parents or professionals supplement childself-reports and are linked to reports collected laterin life. This mixing of different sources ofobservation is one of the strengths of the existingknowledge base in developmental ‘life-course’research. The capacity of follow-up research toclassify behaviour based on multiple observationsacross time represents a further strength.

A major problem in follow-up research is attritionof participants with higher levels of risk. The studiesselected for inclusion in the current review satisfiedcriteria suggesting the initial study participants wererepresentative of the population they were selectedfrom and the effect of selective attrition (loss tofollow-up) on observed relationships between earlyand late events had been appropriately considered.

Although the issue has received insufficientattention, it is generally agreed that self-reports ofchildren and youth are reliable and valid so long asresponses are perceived to be confidential andquestions are clearly worded and appropriate to thestage of cognitive development. There has beenconsiderable previous work attempting to synthesiseresearch into early childhood and adolescence.

Although there are some studies examining earlyadulthood, there has been little follow-up research

�

%� �� !��"�#��$��%��

examining developmental processes later inadulthood. We have not been able to find follow-upstudies examining Australian Indigenous and illicitdrug-using populations. In these cases, thediscussion centres on associations rather than riskfactors and the conclusions rest on weaker evidenceand hence are more speculative.

#�"�� 3��

A range of factors established prior to a child’s birthhave been shown to influence the likelihood of thatchild eventually becoming involved in drug use andof progressing to potentially harmful drug use.

7'6'&��+��-��

Summary: Being born or raised in a familyexperiencing extreme economic deprivation is a riskfactor for harm associated with drug use (1longitudinal study).

The effect of social disadvantage on thedevelopment of drug-related harm is complex.Work in the well conducted New ZealandChristchurch cohort has associated severe socialdisadvantage with risks to healthy childdevelopment at a very early age; leading to acumulative snowball of deficits that culminate inserious social and behaviour problems throughoutlife. However, there is also evidence from studies inthe United States that children from highersocioeconomic status (SES) may sometimes havesufficient affluence and freedom to seek out, andengage in, new and novel drug use experiences. Thetendency of the lifestyles of the wealthy andinfluential to lead social aspirations can result in thephenomena observed in Australia and other nations,where new patterns of drug use may be introducedand made fashionable through the involvement ofpeople from higher SES. Lower SES is associatedwith less money and influence and hence, all thingsbeing equal, drug use is more likely to lead toundesirable social consequences for those withlower SES. This means that as drug use epidemicsmove into more disadvantaged groups, anincreasing level of social costs may be observed.Problems associated with drug use can also explaindownward mobility through the impact of drug usein lowering educational achievement and increasingexposure to unemployment and incarceration.

The Christchurch Health and Development study inNew Zealand has documented some of the earliestpredictors of drug use problems. In that study, lowfamily SES from birth predicted more frequentcannabis use at 15 to 16,156 higher amounts of

alcohol consumption and a tendency for alcohol-related problems at age 15.125, 480 Children born intofamilies with a larger number of children tended todrink alcohol more frequently,125 and experiencemore alcohol related problems by age 15.480 Each ofthese findings was maintained after adjusting forother known risk factors in childhood andadolescence.

For the small minority (2.7%) in the Christchurchcohort evidencing poly-drug use, concurrent withthe most severe behaviour problems at age 15, earlyfamily backgrounds were characterised at birth andinfancy by high levels of social disadvantage andsocial disconnection. These children were morelikely to be born to teenage and or sole parents whohad low education and low incomes, and theseparents tended to be socially disconnected (lowrates of church attendance and/or parentmobility).481

However, evidence from other cohorts suggests thatin some groups it is not low SES, but high SES, thatpredicts entry to drug use. In a cohort of USstudents interviewed in the early 1970s, higherlevels of father’s education at age 16 was anunadjusted predictor of more frequent involvementin illicit drug use at ages 24 to 25 for males.162 In aseparate US cohort, females from higher SESbackgrounds were more likely to engage in lateemerging poly-drug use at age 16, in the mid1980s.135 In the follow-up of the AustralianTemperament Project (a birth cohort of around2400 followed from 1983, in Victoria) entry to age15 poly-drug use, or illicit drug use, appeared to beoccurring across the range of social backgrounds inthe late 1990s.461 In this study, SES appeared to haveno direct or interactive effect on age 15 poly-druguse or illicit drug use. There are other Australianstudies that show a similar trend, revealing that at avery broad level, drug and alcohol disorders are notassociated with the level of education.377

The finding that harmful drug use is inconsistentlyexplained by SES also arises in some other areas ofchild health research. A large follow-up studyinvestigating the development of children in Canadanoted that poor child outcomes were less wellpredicted by SES than by parenting. In that study,poor parenting was related to the SES gradient butalso varied considerably across different levels ofSES. At least 25% of children in the highest SESgroup showed developmental difficulties.482

The findings from the Christchurch cohort are inconcert with an earlier review of findings fromresearch conducted in the 1970s and 1980s thatconcluded that severe economic deprivation was a

%"��

risk factor for youth drug abuse.10 These findings fitwith a substantial worldwide literature showing thatfor nearly all health conditions, the level of harmcaused by that condition takes the form of agradient, whereby the lowest SES groups experiencethe most harm and the level of harm reduces as SESincreases.430 SES does not necessarily predictinvolvement in potentially harmful patterns of druguse, but it may increase the risk of experiencingsuch harm.

7'6'.��+,�*��1��

Summary: Being born or raised in a sole parenthousehold is a risk factor for more frequent druguse in adolescence (two longitudinal studies).

The last three decades have seen increasing rates offamily breakdown in Australia and other countries.Families experiencing breakdown tend to sufferhigh levels of social disadvantage.

In well-controlled cross-sectional research inWestern Australia, sole parent status appeared toincrease the risk of youth maladjustment,independent of levels of family conflict.483 It ispossible that sole parent households are at greaterrisk not because of the number of parents butbecause of the additional emotional distress andfinancial pressure under which those households areoften placed, due to dual working and parentingroles and financial pressures.478

In the New Zealand Christchurch cohort, earlyfamily breakdown was associated with heavieralcohol use at age 14484 years and poly-drug use atage 15 years.481 Family divorce and separationpredicted more frequent cannabis use by age 15years in the Victorian Adolescent Health Cohort(VAHC), a representative cohort of around 2000youth followed from age 14 to 21 years with lowsample attrition.123

The influence of family breakdown on drug useappears to be independent of child behaviourproblems, raising the possibility that a majorcomponent of the influence lies in the familymanagement of adolescents. In the analysis of theVAHC, the influence of family breakdown onfrequent mid-adolescent cannabis use wasmaintained after adjusting for markers of childadjustment problems, including age 14 involvementin antisocial behaviour, peer cannabis use, andearlier drug use (age 14 daily cigarette smoking andheavier alcohol use). The influence of familybreakdown on age 14 alcohol use in the NewZealand cohort was also maintained after adjusting

for age 8 conduct problems and earlier age of firstalcohol use.484

A recent evaluation of a parent educationintervention documented success in reducing earlyadolescent poly-drug use. The success of theintervention was associated with the targeting ofparenting groups that included a high proportion ofsole parents.485

7'6'0��)-��)-��+��

Summary: Being born outside Australia is aprotective factor, reducing frequent drug use inadolescence (1 longitudinal study).

A range of evidence suggests that being bornoutside Australia is associated with lower levels ofadolescent drug use166, 334, 335 and lower rates ofsubstance use disorders.377 There is evidence fromone Australian study that the effect of non-Australian birth may not be direct but rather aprotective (or mediating effect) resulting in lessfamily breakdown, less involvement with drugusing peers and lower rates of early age drug use. Inthe VAHC, fewer of the children born outsideAustralia were found to engage in frequent cannabisuse by around age 15. This effect of non-Australianbirth was no longer significant after adjusting forthe influence of peer cannabis use, daily cigarettesmoking, high dose alcohol use, frequent alcoholuse, antisocial behaviour and divorced or separatedparents.123

7'6'(�:� �� +)� ��-

Amongst the earliest influences on youth substanceuse are genetic factors that interact withenvironmental influences to shape individualdifferences in drug use behaviours. Althoughknowledge is advancing regarding the role ofgenetic influences on drug use behaviour, there isstill a great deal that is unknown. Available researchsuggests that it is unlikely that a single gene will befound to explain drug use behaviours.465, 486 It ismore likely that a combination of genetic factorsinfluence behaviour through their interaction withenvironmental factors. Childhood predictors of druguse considered likely to have some geneticcomponent include behavioural problems andtemperament. All things being equal, it is likely thatmore serious drug use disorders may have a highergenetic determination.477

�

%# �� !��"�#��$��%��

7'6'6��2��

Summary: Being male is not a risk factor foradolescent drug use. The historical tendency formales to engage in more adolescent drug useappears to have been cultural (time-series analyses).

Summary: Males can inherit a paternal gene that is arisk factor for childhood externalising behaviourproblems and alcohol abuse (and perhaps otherdrug abuse) (two longitudinal studies).

In many Australian studies, males are found to bemore commonly involved in illicit drug use.166

Although in the NDSHS lifetime use of injectingdrugs was more common for males, for usage in thepast 12 months there was no gender effect.175 Thesetrends are in line with evidence that in the late1990s an increasing proportion of injecting drugusers were female.

Internationally, there has been an increasing trendfor females to take up alcohol use487 and to escalatein greater proportions to dependent use.488 In recentAustralian cohorts, females have been demonstratedto engage in higher levels of potentially harmfulpoly-drug use relative to males. This was observedin the Victorian Australian Temperament Project(ATP) cohort for age 15 to 16 years poly-drug usein the 1990s.461 In the late 1980s, a similar trendwas observed in Brook’s US cohort for lateemerging poly-drug use at age 16 years, for femalesfrom higher SES backgrounds.135

Findings from a number of studies report that maleshave a higher tendency for alcohol abuse anddependence. In a cohort of 449 in the Midwest ofthe US, the co-occurring pattern of alcohol andtobacco dependence symptoms was followedthrough five waves of data collection over sevenyears, from age 18 to 25 years. The study had littleattrition, enabling the course of alcohol and tobaccodependence to be followed over time. Through thisperiod, three trajectories of alcohol-dependencewere identified. Chronic trajectories applied to 6%of the sample and involved stable, high levels ofdependence across all years. Adolescent limitedtrajectories applied to 17% of the sample. Neither ofthese trajectories was associated with higher rates oftobacco dependence. However, in a further 7% ofthe sample, high levels of alcohol dependence co-morbid with tobacco dependence were noted. Beingmale was an unadjusted predictor of the morecommon forms of chronic and adolescent-limited,but not the co-morbid form of, alcoholdependence. In this study, the behaviouralexpression of alcohol dependence was indirectlypredicted by family history but this effect was

mediated by favourable attitudes to alcohol use andindividual characteristics of personality disorder andschool maladjustment.120

In the New Zealand Christchurch cohort, males hada greater risk for age 15 alcohol problems125 and age16 alcohol abuse.484 In a Finland cohort, males werealso more likely to develop alcohol abuse by age 26years.489

In the New Zealand Christchurch cohort, thefinding that males were more likely to developalcohol problems at age 15 years was maintainedafter adjusting for age at first alcohol use, lowerfamily SES, parental alcohol/drug problems andparental perceived attitudes to alcohol use.125 Thefinding that males tended to develop alcohol abuseat age 16 was maintained after adjusting for age 14alcohol use and peer drug use.484

In work by McGue and colleagues in Minnesota,490

twin data have been examined and associationsidentified between parent and youth alcohol use,were found to have a larger genetic component forboys compared to girls. In this study, thebehavioural expression of early alcohol use wasargued to occur through a greater tendency towardchildhood externalising behaviour problems forboys.

The VAHC noted males tended to engage in morefrequent cannabis use by age 15 years. However,this effect was no longer significant after adjustingfor the influence of earlier drug use, antisocialbehaviour and the impact of parental divorce andseparation.123

Some studies have examined the possibility ofdifferent risk processes for females relative to males.There is some evidence that internalising disorders,such as anxiety and depression, may be moreimportant predictors for female alcohol problems.In Finland, high social anxiety in females at age 8years predicted higher rates of alcohol abuse at age26, while anxiety in boys was associated with loweralcohol abuse.489 In an adult cohort in the easternUS, depression was found to predict female but notmale progression to alcohol abuse and dependencetwo years later.491

7'6'7�� +,��)��)-�*��)�-

Summary: Maternal smoking and alcohol use priorto birth, and environmental tobacco smoke are riskfactors for impaired child development; thisimpairment initiates a pathway of poor childadjustment, leading to harmful drug use (seeChapter 3).

%%��

Summary: Drug use of parents and other familymembers from late childhood is a risk factor forearly age initiation of the same drug, throughmodelling (three longitudinal studies).

As documented in Chapter 3, maternal smoking,alcohol and drug use prior to birth have importantdevelopmental consequences for children,extending to an increased risk for harmful drug use.For the small minority in the Christchurch cohortwith severe behaviour problems and poly-drug useat age 15, high rates of maternal smoking anddrinking were observed prior to birth.481

In Minnesota, US, McGue et al. found that fathers’and mothers’ own reports of earlier age of alcoholuse predicted an earlier age of alcohol use for theirearly adolescent offspring.490 Analysis of twin datain this study suggested that family environmentinfluences on early age alcohol use were moreimportant for girls, while genetic factors were moreinfluential for boys.

In the Australian Temperament Project (ATP),mothers’ drinking habits when children were aged13 to 14 years predicted higher poly-drug use byage 15 to 16 years after adjusting for a range offactors including peer relations, child behaviourproblems, gender and class.461

Findings from the New Zealand Christchurch cohortsuggest that parent behaviours may influence thedevelopment of adolescent high quantity alcoholuse independently of the influence of earlychildhood behaviour problems, parent attitudes toalcohol, gender or social class. However, otherfactors such as age of first alcohol use and parentalapproval may be more influential in determininghow frequently alcohol is used.

In the Christchurch cohort, parents’ alcoholconsumption at participant age 11 years was notassociated with alcohol problems at age 15 years.125

However, parents’ alcohol consumption didcontinue to predict higher amounts of alcoholconsumption at age 15 after adjusting for age 8conduct problems, parental history of alcohol anddrug problems (prior to 15 years), male gender,low family SES, family size and parental attitudes toalcohol consumption.480 The effect of parents’alcohol consumption at age 11 on frequent alcoholuse at age 15 was no longer significant afteradjusting for age at first alcohol use, family size, age14 parental approval of adolescent alcohol use andparental attitudes favourable to alcohol use.125

Previous reviews have also noted that the risk ofearly initiation of use of a drug may be influencedby the number of members of a household,including siblings, who use that drug.10

#�#�+��

Through the period from birth until entry to school,evidence suggests that parents remain a centralinfluence on child development. Research suggeststhat adequate nutrition, security and early learningopportunities are important for healthy childdevelopment through this period. A number ofstudies have demonstrated that experience throughthis period influences the later emergence ofharmful drug use.

7'7'&�� +� ��+�� *)-�

Summary: Child neglect and abuse is a risk factorfor impaired child development and thisimpairment initiates a pathway of poor childadjustment leading to harmful drug use (onelongitudinal study).

Inadequate parental provision of infant care hasbeen shown to influence the subsequentdevelopment of drug use in children. For the smallminority in the Christchurch cohort who evidencedpoly-drug use concurrent with the most severebehaviour problems at age 15 years, earlyenvironments from birth were characterised by in-utero insult (maternal smoking and drinking), birthcomplications, problems in infant care (low breastfeeding, low infant care, poor parenting), and socialinstability through childhood (family breakdownand parental changes).481

A further analysis of the Christchurch cohort notedthat sexual abuse in childhood or adolescencepredicted higher rates of illicit drug use from 15 to21 years,492 and cannabis and alcohol abuse anddependence at age 16 to 18 years.156 These effectswere robust and persisted after adjusting for otherinfluences such as deviant peer relations, behaviourproblems and earlier involvement in drug use.

7'7'.�� +,�*��)�

Summary: Easy temperament in early childhood is aprotective factor for positive child adjustment andreduces the influence of other risk factors, leadingto lower rates of involvement in harmful drug use(one longitudinal study).

Findings from the ATP cohort associated a numberof early age temperament indicators with poly-druguse at ages 15 to 16 years.461 As infants at four to

�

%' �� !��"�#��$��%��

eight months, these children were reported by theirparents to be less regular in their eating andsleeping habits. From ages one to two, the childrenshowed less tendency to persist to complete taskssuch as eating or playing, and at age two to threethey were perceived as less cooperative and moreactive or restless. From ages three to four, and insubsequent years, the children tended to be less shyand more outgoing toward strangers. Generally,these effects were small and were no longersignificant after controlling for other influenceslater in life. Hence, once the effects of delinquencyand early drug use on poly-drug use wereconsidered, the effect of early temperamentdisappeared.461 This demonstrates that temperamentwas not a direct risk factor but mediated theinfluence of other adjustment pathways.

Temperament is considered to interact with‘transactional’ influences that modify the care thechild receives and the impact of that care on laterdevelopment.493 Children with extremes oftemperament may be more difficult to manage andhave a higher likelihood of developing adjustmentproblems.493 Temperament is considered to have agenetic component but appears to be heavily shapedby environmental experiences.

Easy temperament has previously been consideredto be a protective factor that underlies the resilienceof some children who maintain positive adjustmentdespite high levels of adversity in childhood.10, 494,

495 The findings from the ATP are also congruentwith this position in that temperament did notdirectly predict poly-drug use but it did mediateother risk factors.

#�%��!�� "��

From the years following entry to primary school,parents continue to exert an important influence onchild development. However, other factors,including relationships with teachers, adjustment toschool and experiences with peers, begin to play anincreasingly important role in the child’sdevelopment.

7'3'&��+

Summary: Early school failure is a risk factor foralcohol abuse (one longitudinal study).

There is some evidence that early failure in primaryschool may be a risk factor for the later emergenceof drug use problems. In a cohort in Finland, lowschool achievement at age 8 years predicted alcoholabuse at age 26 years, after adjusting for a range ofother factors including age 14 aggression, social

anxiety, inattention, pro-sociality, conductproblems, parental drinking and socialbackground.489 This finding was independent ofgender and social class. A review of researchcompleted in the 1970s and 1980s reached a similarconclusion.10

7'3'.� ��+��*��)�

Evidence demonstrates that childhood behaviourproblems through the primary school age period areimportant risk factors for the development of druguse problems. Current evidence suggests thatconduct disorder in childhood may be moreimportant than Attention Deficit and HyperactivityDisorder (ADHD) in predicting later drug useproblems.

Attention Deficit Hyperactivity Disorder

Summary: The role of ADHD in predicting youthdrug use is unclear.

Analyses in the New Zealand Christchurch cohortand in the ATP cohort suggest that highersymptoms of ADHD in childhood or earlyadolescence are a predictor for the subsequentemergence of potentially harmful drug use.However, in the Christchurch cohort, these effectsappeared to be mediated by conduct problems inchildhood. In some US cohorts, ADHD was notfound to predict youth drug use.

Analysis of the New Zealand Christchurch cohortdemonstrated that those with ADHD at age 8 yearstended to use higher amounts of alcohol andexperience more alcohol related problems, higherrates of daily tobacco use and more illicit drug useat age 15 years. However, these effects weremediated by conduct disorder such that the effect ofADHD was no longer significant after controllingfor age 8 conduct problems.480 In their WesternNorth Carolina cohort, Costello et al. noted noassociation between ADHD at age 11 and alcoholabuse at age 16 years.118

In the ATP cohort, youth who reported symptomsof ADHD at ages 13 to 14 years had an increasedlikelihood of age 15 poly-drug use after adjustingfor other factors including externalising behaviourproblems, peer relations, SES and gender.461

Once substance use behaviours are established inadolescence, these behaviours may be very stableand may mask the influence of earlier risk factors.For example, Brook et al. found no effect for eitherADHD or conduct disorder at age 16 on heavytobacco use at age 22 years, after adjusting for theinfluence of prior patterns of drug use.138

%(��

Conduct disorder

Summary: Conduct disorder in childhood is a riskfactor for higher levels of alcohol consumption inadolescence (one longitudinal study). The influenceof conduct disorder on alcohol abuse may bemediated by family vulnerability (two longitudinalstudies) to alcohol problems or by earlier agealcohol use (one longitudinal study).

In the Christchurch cohort, age 8 conduct problemspredicted higher frequency and amount of alcoholconsumption at age 14, after controlling for earlierage of alcohol use and changes of parents.484 In thesame cohort, age 8 conduct disorder predicted theamount of alcohol consumed, alcohol-relatedproblems and illicit drug use at age 15 years. Theinfluence on the amount of alcohol consumptionwas maintained after adjusting for family SES, malegender, family size and parental attitudes favourableto alcohol use.480 However, the effect on alcohol-related problems was no longer significant aftercontrolling for additional factors including age atfirst alcohol use and parental alcohol/drugproblems.125 These various findings suggest that thetendency for children with conduct problems todevelop alcohol problems may be mediated not byclass or gender but by starting alcohol use at anearlier age and through family vulnerability toalcohol problems.

In a separate analysis in the Dunedin cohort (alsofrom New Zealand), conduct problems at age 11years predicted poly-drug use and the use ofcannabis and/or glue for boys (but not girls) at age15 years.496 These effects were very unstable andneither relationship was significant in an adjustedmodel that included age 11 depressive symptoms.

In a sample of Minnesota twins, McGue et al.demonstrated that for adolescent boys (around age14 years) an earlier age of alcohol initiation wasassociated with inherited vulnerability to childhoodexternalising behaviour problems at age 11 years(conduct disorder and oppositional defiantbehaviour problems).490

Aggression

Summary: Aggression in childhood is a risk factorfor early adolescent poly-drug use (two longitudinalstudies) and adult alcohol abuse (one longitudinalstudy).

Childhood aggression appears to be an importantrisk factor for the later emergence of harmful druguse. In Finland, children rated by peers and teachersto be more aggressive at age 8 years tended to havehigher alcohol abuse at age 26 years, after adjusting

for a range of other individual level predictors.489

Analysing the ATP cohort, Williams and Sanson et al.also noted that children rated by teachers as moreaggressive at ages five to six years, or at ages 11 to12 years, were more likely to progress to poly-druguse by age 15 to 16 years.461

Brook et al., analysing a cohort from upper NewYork State, found that the tendency for youth toremain abstinent from alcohol use between ages 16and 22 years tended to be moderated by a variety ofprotective factors, including low anger at age 8years.163 In the same cohort, Brook et al. found thataggression at age 8 predicted more frequent poly-drug use at age 14, after adjusting for SES.135

Depressive symptoms

Summary: The role of childhood depressivesymptoms in predicting youth drug use is unclear.

Although it cannot be described as a risk factor,there is some limited evidence that depressivesymptoms in primary school age children maypredict the emergence of adolescent poly-drug use.In an analysis of the New Zealand Dunedin cohort,depressive symptoms at age 11 predicted poly-druguse and the use of cannabis and/or glue for boys(but not for girls) at age 15 years.496 Only the effectpredicting poly-drug use remained significant in anadjusted model that included age 11 conductproblems.

7'3'0�� ++��

Summary: The role of childhood intelligence inpredicting youth drug use is unclear.

In previous reviews, child intelligence has beenconsidered as a protective factor predicting positiveoutcomes for children growing up in difficultcircumstances.494, 495 Available evidence suggests thatintelligence in children, in some contexts, maypredict involvement in drug use.10 Perhaps this ismore likely to occur at the beginning of a socialtrend toward a new form of drug use. However,intelligence is not a consistent predictor of drug use.For example, in the New Zealand Christchurchcohort, intelligence at age 8 was unrelated to thefrequency of cannabis use at age 15 to 16 years.156

7'3'(� ��+��-��+�� +��

Summary: Social and emotional competence inchildhood is a protective factor, reducing theinfluence of risk factors for alcohol abuse (onelongitudinal study) and illicit drug use (onelongitudinal study).

�

') �� !��"�#��$��%��

Social and emotional competence in primaryschool-aged children appears to mediate theinfluence of later risk factors, such as behaviourproblems and school adjustment, while not itselfdirectly predicting drug use. Children in Finlandrated by peers and teachers to be more pro-social atage 8 were less likely to be involved in alcoholabuse at age 26 years.489 However, this effect was nolonger significant after adjusting for the influence ofage 14 social anxiety, conduct problems and schoolachievement.

Brook et al., analysing a cohort from upper NewYork State found that children at age 8 years whowere low on emotional control had higher rates ofillicit drug use at age 22 years.163 This effect was nolonger significant after adjusting for measures ofpersonality and drug use through early adolescence(at ages 14 and 16 years).

7'3'6�� -� �+�,

Summary: Shy and cautious temperament inchildhood is a protective factor, reducing theinfluence of risk factors for early adolescent poly-drug use (one longitudinal study) and illicit druguse in early adulthood (one longitudinal study).

In the social context of a high communityprevalence of harmful drug use, the primary school-aged child’s tendency to interact with strangers andto socialise with other children emerges as animportant predictor of potentially harmful drug use.In an analysis of the ATP cohort, children who atage seven to eight years were more inflexible andwho were less shy and cautious in approachingstrangers had higher rates of poly-drug use by age15 to 16 years.461 In the same cohort, children whowere more sociable at age nine to 10 years werealso more likely to progress to poly-drug use in themid-adolescent periods. Each of these effects was nolonger significant after adjusting for latertemperament and behaviour characteristics.

Brook et al., analysing a cohort from upper NewYork State, found that children who showed fearlesspersonality characteristics at age 8 years were morelikely to engage in illicit drug use at age 22 years.163

However, these effects were no longer significantafter adjusting for measures of personality and druguse at ages 14 and 16 years.

#�'�� %�

During the secondary school age period, the youngperson develops an increasing independence fromthe family and this entails identity choices thatinclude attitudes and behaviours relevant to druguse. Relationships with parents, adults in thecommunity, and in the peer group have all beenshown to influence the subsequent development ofdrug use. Behaviour and attitudes through thisperiod are, however, also influenced by theattitudes, behaviours and relationships developed atearlier life-stages.

7';'&� ��) �,�� +)� ��-

It is possible to overlook the influence ofcommunities on youth drug use by limitingconsideration to individual-level longitudinalresearch. In this section, other sources of data areconsidered, hence some conclusions are necessarilymore speculative.

Community opportunities for access to positivesocial activities

Summary: Low involvement in activities withadults in adolescence is a risk factor for earlyadolescent poly-drug use (one longitudinal study).

During the secondary school age period,relationships with adults in the community emergeas more salient predictors of the subsequentdevelopment of drug use. In the ATP cohort,children who were less involved in sport andcommunity activities involving adults, at age 13 to14 years, were more likely to engage in poly-druguse at age 15 to 16 years.461 These effects weremaintained after adjusting for other risk factors suchas peer relationships, behaviour problems, genderand SES.

Recent reviews have considered whether youthaccess to recreation influences youth drug use.419 Itdoes not appear that access to services influencesyouth drug use directly. For example, there is littleevidence that living in a rural location directlyinfluences youth involvement in drug use. In theVAHC, youth at age 14 years living in rural versusurban locations demonstrated no differences in theirtendency to progress to either frequent cannabis useat age 15 years or to daily use at 16 to 17 years.123

Other surveys have also found no difference in thelikelihood of alcohol and drug disorders betweenrural and urban areas.377

'��

Perceived and actual levels of drug use in thecommunity

Summary: The perceived and actual level ofcommunity drug use in adolescence is a risk factorfor adolescent use of that drug (previous review,one intervention study).

In previous reviews, young people’s perceptionabout adult drug use497 and the actual communityprevalence of drug use have each been found to beimportant through the adolescent period inpredicting the subsequent emergence of youth druguse.10 Evaluation of drug education programs hasdemonstrated that programs that reduce youthestimates of adult drug use are more effective atreducing subsequent youth initiation of drug use.498

Community disadvantage and disorganisation

Summary: Community disadvantage anddisorganisation in adolescence has been associatedwith adolescent drug use (previous review,associational evidence).

In previous reviews, community disadvantage anddisorganisation has been associated with adolescentyouth drug use.10 Vinson has noted the tendencyover the last decade for unemployment and familydisadvantage to increasingly cluster geographicallydue to low cost accommodation.499 The implicationis that certain neighbourhoods have experiencedincreasing rates of clustering of economicallydeprived households. Very high rates of youth druguse problems and other related behaviours havebeen associated with geographic localitiescharacterised by low SES, low income and poorquality housing.500–502

Availability of drugs in the community

Summary: The availability of drugs within thecommunity in adolescence is a risk factor foradolescent use of that drug (previous review, fourintervention studies).

The availability of drugs at a community level hasbeen noted to predict the subsequent involvementof youth in drug use.10 Community interventionresearch has demonstrated that enforcing laws thatprevent minors accessing tobacco reduces youthtobacco use.503–505 Introducing laws to prevent salesof alcohol to minors has also been associated with acommunity mobilisation intervention thatdemonstrated some reduction in frequent youthalcohol use.506

Media portrayal of drug use

Summary: Positive media portrayal of drug use inadolescence is associated with adolescent use of thatdrug (one intervention study).

Intervention studies have demonstrated thatincreasing exposure to anti-smoking messages,combined with a school smoking preventionprogram, was associated with reductions inadolescent smoking.507

7';'.��+,�� +)� ��-

During adolescence, family relationships becomemore complex. The family appears to maintain animportant influence on drug use.

Family attachment

Summary: Attachment to the family in adolescenceis a protective factor, reducing risk factors for earlyadolescent poly-drug use (two longitudinalstudies).

Attachment to the family through adolescence tendsto be a consistent predictor of youth drug use but itseffect is often mediated by other influences,particularly peer relationships. In the upper NewYork State cohort, Brook et al. found that lowattachment to parents at ages 14 and 16 yearspredicted more frequent poly-drug use at age 22years, after adjusting for other influences.136 In theATP, parent reports of less warmth andcommunication was an unadjusted predictor ofillicit drug use by age 15 to 16 years;461 afteradjusting for the influence of peer relationships, theeffect was no longer significant. In the New ZealandChristchurch cohort, Fergusson et al. also found thatlow parental attachment at age 15 was anunadjusted predictor of frequent cannabis use at age15 to 16.156

Parental harmony and parent-adolescent conflict

Summary: Low parental conflict (parentalharmony) from late childhood and in adolescence isa protective factor, reducing alcohol problems (onelongitudinal study).

Summary: Parent-adolescent conflict is a risk factorfor early age drug use (one longitudinal study, oneintervention).

In the Christchurch cohort, parental conflict prior toage 14 years did not influence the frequency ofalcohol use at age 15, but did relate to increasedproblems with alcohol at that age. The effect onalcohol-related problems was not a direct risk factorbut appeared to mediate other influences such as the

�

'� �� !��"�#��$��%��

child’s age of first drug use. The effect was nolonger significant after adjusting for the child’s ageat first alcohol use, male gender, low family SES,parental alcohol/drug problems and favourableparental attitudes to alcohol use.125 In the samecohort, parent conflict in childhood was anunadjusted predictor of frequent cannabis use at age15 to 16 years.156 From these findings, it can beinferred that parental conflict was not a directpredictor of adolescent alcohol use but may haveinfluenced other family factors. We infer from thisfinding that parental harmony in adolescence maywork as a protective factor, mediating the influenceof other risk factors such as early age alcohol use.

Through the secondary school period, increasinglevels of parent-adolescent conflict can emerge. In asmall US cohort, Brody and Forehand demonstratedthat mother-adolescent conflict predicted earlyinitiation of adolescent alcohol use.508 This findingwas adjusted for other influences including familybreakdown and parental conflict. An Australianstudy demonstrated that reductions in early-adolescent poly-drug use were achieved through aparent education intervention that reduced levels ofparent-adolescent conflict.485

Parental attitudes to drug use

Summary: Favourable parental attitudes to drug usefrom late childhood is a risk factor for early ageinitiation of the same drug (one longitudinalstudy).

Analyses of the Christchurch cohort revealed thatfavourable parental attitudes to alcoholconsumption at age 11 years continued to predictboth the amount of alcohol consumed and theproblems experienced with alcohol at age 15 afteradjusting for a range of other risk factors.480 In thesame cohort, parental attitudes toward alcohol use atage 14 also predicted more frequent alcohol use andalcohol problems at age 15 after adjustment forother risk factors.125 Parental approval of adolescentalcohol use at age 14 was not associated withalcohol problems at age 15, but did predict frequentalcohol use at age 15.125

Alcohol and drug problems in the family

Summary: Parental alcohol and drug problems earlyin their offspring’s adolescence is a risk factor forearlier age alcohol use and higher levels of alcoholuse later in adolescence (two longitudinal studies).

There is some evidence to suggest that parentalproblems with alcohol use may be a risk factor foryouth alcohol use problems. In the New ZealandChristchurch cohort, this effect was maintained after

adjusting for other influences such as parentattitudes and child behaviour problems. Earlierwork summarised in the current document suggeststhis effect may be partly based on geneticpredisposition in males.

In a cohort in Western North Carolina in the US,family history of alcohol problems was found topredict an earlier age of initiation to alcohol use formales but not for females. However, this analysisdid not adjust for other known influences.118

In an analysis of the New Zealand Christchurchcohort, parental alcohol and drug problems at age14 years did not predict the frequency of alcoholuse at age 15, but did predict alcohol-relatedproblems at this age.125 This relationship wasmaintained after adjusting for age at first alcoholuse, family size (at birth), parental approval ofadolescent alcohol use (age 14 years) and parentalattitudes favourable to alcohol use (age14). In thesame cohort, parental history of alcohol and drugproblems prior to age 15 predicted the amount ofalcohol consumption at age 15. This effect appliedafter adjusting for age 8 conduct problems, parentalalcohol consumption, male gender, low family SES,family size and favourable parental attitudes toalcohol consumption.480 As analyses were notseparated by gender it is unknown whether similarpredictors applied for both males and females in thiscohort.

In previous reviews, parental criminality orantisocial behaviour has also been linked to thesubsequent emergence of drug use problems inoffspring.509

Parental communication and monitoring

Summary: Parental communication in earlyadolescence is a protective factor, reducing theinfluence of risk factors for harmful youth drug use(three intervention studies).

Parental supervision of children (being aware and incharge of what children are doing) is one of themost important predictors of the subsequentinvolvement of children in delinquent behaviour.This has been demonstrated, for example, infollow-up studies where home visitor impressionswere used to measure supervision.510 Inadolescence, the term monitoring is used todescribe the change in parenting towardcommunication and negotiation strategies to remainaware of, and have some influence over, what theadolescent does.

'��

Evidence from three intervention studies suggeststhat parental communication practices may haveprotective effects in reducing subsequent risks foradolescent involvement in harmful drug use. Perryand colleagues506 found that an interventiondesigned to improve parent communicationregarding alcohol resulted in increasedcommunication, based on child reports. Theintervention was subsequently successful inreducing early age alcohol use.

In work by Spoth and colleagues, a parentingtraining program designed to assist parents toimprove their parenting skills was associated withincreased adolescent ratings of positive familyattachment. The intervention was successful atreducing early adolescent alcohol use.511

In an Australian parent education intervention,parents were exposed to a curriculum designed toenhance skills for communicating with adolescentsand resolving conflict. In families exposed to theintervention adolescents’ and parents’ ratingsshowed lower parent-adolescent conflict andadolescents rated parents to be higher in care. Theseadolescents showed a reduction in poly-drug use.485

Research demonstrating that poor parentalcommunication directly predicts youth drug usecould not be found. However, in many studies,effective family communication has been modelledas a protective factor reducing youth drug use.512, 513

Family rules and discipline

Summary: Parental rules permitting drug use inchildhood or early adolescence is a risk factor forearly age drug use (one longitudinal study).

In an analysis with the New Zealand Christchurchcohort, parental approval of adolescent alcohol useat age 14 years predicted a higher frequency ofalcohol use at age 15.125 This relationship wasmaintained after adjusting for age at first alcoholuse, parental alcohol and drug problems at age 14,family size (at birth) and parental attitudesfavourable to alcohol use (age14 years).

Work examining parenting has suggested that twoimportant dimensions influencing child outcomesare: nurturance/warmth and demands forresponsible behaviour. Families high in nurturanceand demands are defined as authoritative and thesequalities tend to predict better developmentaloutcomes for children. Rules for responsiblebehaviour are typically included in measures offamily demands.

In a sample of US students interviewed in the early1970s, maternal use of a permissive parenting style(high nurturance but low demands) was anunadjusted predictor of more frequent involvementin illicit drug use at ages 24 to 25 years, formales.1149 In the same cohort, mothers’ use of anauthoritarian parenting style (low nurturance buthigh demands) predicted more frequentinvolvement in illicit drug use at ages 24 to 25years, for females. This effect was maintained afteradjusting for other influences including adolescentdrug use.

7';'0��+

Completing secondary school

Summary: Not completing secondary school is arisk factor for early adult drug problems. However,it is unclear whether this relationship is explainedby earlier developmental influences.

Young people who leave school in the earlysecondary school years tend to have a greaterlikelihood of engaging in drug use. Retention inschool is itself predicted by earlier childhooddevelopment, including school adjustment andbehaviour problems. Academic achievement andfeelings toward school are closely related factorsthat have been found to predict involvement inillicit drug use but may be unrelated to alcohol use.

Several studies have associated lower levels ofeducation with higher levels of harmful drug use.For example, lower levels of education have beenassociated with more frequent cannabis use (withincannabis users).166 Persons of lower educationalattainment are more likely to develop drug useproblems.514

It is unclear whether the relationship betweenschool failure and harmful drug use relates to earlierdevelopmental influences. School failure has beenassociated with an earlier age of onset of druguse. 516 In unadjusted analyses in the ATP cohort,youth who reported low school bonding or feweropportunities for involvement at school, at age 13to 14 years, had an increased likelihood of illicitdrug use at ages 15 to 16.461

7';'(��+�� -��-

Summary: Relationships with peers who areinvolved in drug use in late childhood oradolescence is a risk factor for alcohol abuse andillicit drug use (three longitudinal studies).

Association with peers who engage in drug useappears to be an important risk factor through the

�

'� �� !��"�#��$��%��

early secondary school period, influencingsubsequent involvement in harmful drug use. In theChristchurch New Zealand cohort, peer drug use atage 15 years predicted alcohol abuse at age 16. Thiseffect was maintained after adjusting for malegender and age 14 alcohol use.484

In analysis of the VAHC, both perceived peercannabis use and classroom prevalence of cannabisuse at age 14 years predicted more frequentcannabis use by age 15 and, for males, dailycannabis use at age 16 to 17 years. These effectswere significant after adjusting for a range of factorsrelated to earlier drug use, antisocial behaviour andparental divorce and separation.123

In an analysis of the ATP, children who were morepopular with peers at 11 to 12 years, or moreinvolved with peers at ages 13 to 14 years, hadhigher rates of poly-drug use by age 15 to 16 years.These effects were maintained after adjusting forother factors. In the same cohort, mixing withdeviant peers in early adolescence was alsoassociated with illicit drug use in mid-adolescence.461

Externalising behaviour problems—delinquencyand conduct problems

Summary: Delinquency in adolescence is a riskfactor for alcohol abuse and illicit drug use (sixlongitudinal studies).

The influence of child behaviour problems, such asADHD and conduct disorder, on youth drug usewas described above. Problems involvingdelinquent or criminal behaviour and/or conductand oppositional defiant behaviours are sometimesgrouped together and discussed as externalisingbehaviour problems.

Longitudinal studies suggest that externalisingbehaviours in adolescence have a high level ofstability from childhood. However, externalisingbehaviour problems do emerge in some adolescentswith no prior childhood history. A number ofstudies suggest that externalising behaviourproblems in early adolescence represent importantrisk factors for the development of drug problems.

In Finland, age 14 conduct problems in boyspredicted alcohol abuse at age 26 years, afteradjustment for a range of other risk factorsincluding age 14 aggression, social anxiety,inattention, pro-sociality, low school achievement,parental drinking and social background.489 In acohort in the US that included children withalcoholic parents, externalising symptoms at age 13years predicted alcohol abuse at age 20.147

Steele et al., examining a cohort in the south east ofthe US, noted that externalising behaviours in early-adolescence (age 13 to 14 years) predicted morefrequent alcohol use at age 20 for males andfemales, and for males, more frequent cannabis andillicit drug use at age 20 years.517 In the NewZealand Dunedin cohort, McGee et al. found thatexternalising behaviour problems at age 15predicted higher levels of cannabis use at age 18years.122 In the ATP cohort, youth who reportedmore involvement in delinquency at ages 13 to 14had a greatly increased likelihood of age 15 to 16poly-drug use after adjusting for other factors. Inthe same cohort, delinquency and conductproblems at age 13 to 14 also predicted illicit druguse at age 15 to 16 years.461 Brook et al., analysing acohort from upper New York State, found that thetendency for youth to remain abstinent fromalcohol use between age 16 and 22 years tended tobe moderated by protective factors that includedlow delinquency at ages 14 to 16.163

Once drug use is established in adolescence it hasconsiderable stability. Brook et al., analysing thesame cohort from upper New York State, found thatconduct disorder at age 16 did not predict eitherheavy tobacco, alcohol, cannabis or illicit drug useat age 22 after adjusting for prior drug use.138

Internalising behaviour problems—anxiety anddepression

Summary: The influence of anxiety and depressionin adolescence on subsequent harmful drug use isunclear.

In a cohort in the US that included children withalcoholic parents, internalising symptoms at age 13years were not related to alcohol abuse at age 20.147

There is some evidence that anxiety in earlyadolescence may predict subsequent drug use infemales; however, anxiety in boys may result in lessdrug use. In Finland, for age 14 adolescents theeffect of social anxiety on alcohol abuse at age 26tended to increase involvement for girls but reducedthe risk for boys.489 In a separate cohort in the southeast of the US, internalising behaviours in earlyadolescence (age 13 to 14 years) also predicted alower frequency of cannabis and illicit drug use, formales.517 However, in this cohort, internalisingbehaviours in early-adolescence were unrelated tomore frequent alcohol use at age 20 years. In theNew Zealand Christchurch cohort, anxiety disorderat age 14 to 16 was an unadjusted predictor offrequent cannabis use at age 15 to 16 years.156

Depression in early adolescence has been anunstable predictor of drug use. In a sample of US

'"��

students interviewed in the early 1970s, males whowere lower on measures of depression at age 16 yearswere more likely to engage in frequent illicit druguse at ages 24 to 25.162 This effect was maintainedafter adjustment for other factors includingadolescent drug use.

In the ATP, children who had more depressivesymptoms at 11 to 12 years were more likely toprogress to poly-drug use at age 15 to 16 afteradjusting for other risk factors. In the same cohort,depressive symptoms at age 13 showed anunadjusted association with more illicit drug use atage 15.461

In the VAHC, symptoms for depression and anxietyat age 14 years tended to predict a higher frequencyof cannabis use at age 15, and more involvement indaily cannabis use at age 16 to 17. However, theseeffects were no longer significant after adjusting forother factors such as peer cannabis use, dailycigarette smoking, high dose alcohol use, frequentalcohol use, antisocial behaviour, and divorced orseparated parents.123 Simons et al. have argued thatadolescent depression may exacerbate vulnerabilityto peer influence and this may have been one of thereasons that depressive symptoms were no longersignificant in the VAHC once the influence of peercannabis use was considered.518

Brook et al. found that neither anxiety nordepression at age 16 years predicted heavy alcoholuse, or heavy cannabis use at age 22, after adjustingfor prior drug use. In the same study, depression atage 16 was also unrelated to heavy illicit drug use atage 22.138

It is possible that depression predicts some forms ofalcohol dependence but not others. In a cohort inthe Midwest of the US, depressive symptoms at age18 years were found to predict the less commonforms of chronic alcohol dependence and alcoholdependence that was co-morbid with tobaccodependence, through ages 18 to 25 years. However,depressive symptoms did not predict adolescentlimited forms of alcohol dependence.120

7';'6�� -�� -��1� �� )��)-��-� �+�,��-

Summary: Sensation seeking and adventurouspersonality in adolescence are risk factors for poly-drug use (three longitudinal studies, previousreview).

Adventurousness and a lack of fear in earlyadolescence appear to be risk factors for subsequentinvolvement in drug use. In the New Zealand

Christchurch cohort, it was found that the tendencyto novelty seeking at age 16 years predicted higherrates of cannabis and alcohol abuse and dependence,and also illicit drug use, at age 16 to 18. Theseeffects were maintained after adjusting for otherinfluences such as earlier drug use, deviant peerrelationships and disadvantaged familybackground.156 In a previous review of researchcompleted prior to the 1990s, sensation seeking wasidentified as a risk factor for youth drug abuse.10

Brook et al. found that the tendency for youth illicitdrug use to remain stable between age 16 and 22was reduced (moderated) by protective factors thatincluded fearfulness at age 16.163

In the ATP, children who were reported by theirparents to be less anxious and fearful at 11 to 12years, or less shy and withdrawn at age 13 to 14,had higher rates of poly-drug use by age 15 to 16.These effects were maintained after adjusting for arange of other predictors including peer relations,behaviour problems, SES and gender.461 Negativemoods and emotions in early adolescence have alsobeen shown to predict subsequent involvement inpoly-drug use in the ATP.461

7';'7��)��-��)��)-�

Summary: Favourable attitude to drug use inadolescence is a risk factor for use of that drug (twolongitudinal studies).

Attitudes towards drug use behaviour in earlyadolescence are associated with an increasedinvolvement in subsequent drug use. In the upperNew York State cohort, Brook et al. found that thetendency for youth illicit drug use to remain stablebetween age 16 and 22 years was reduced(moderated) by protective factors that includedintolerance of unconventional behaviour at age14.163 In the same cohort, tolerance ofunconventional behaviour at age 14 predicted morefrequent poly-drug use at age 22, after adjusting forother factors.136 In the Christchurch cohort,favourable attitudes to drug use at age 14 predictedhigher rates of illicit drug use from 15 to 21 afteradjusting for a range of factors, including early agedrug use, childhood sexual abuse and delinquentbehaviour.492

The family appears to be an important context forthe development of attitudes and behavioursthrough adolescence. Kandel and Andrews notedparents’ values to be of particular influence inshaping fundamental adolescent beliefs relevant toeducation, work and social relationships.519

Attitudes to health behaviours may be particularlymalleable in late childhood and early adolescence,

�

'# �� !��"�#��$��%��

when decisions relevant to involvement inbehaviours such as drug use are being made.520

Catalano and Hawkins summarised evidencesuggesting that adolescent identification andmodelling of parent attitudes and behaviours aretypically mediated by family attachment479 althoughnot all studies concur.519

7';'3��+��

Summary: Religious involvement in adolescence isa protective factor, reducing the influence of riskfactors for harmful drug use (previous reviews).

Religiosity appears to be protective againstdeveloping drug and alcohol problems.10, 521 TheChristchurch cohort found those with severebehaviour problems at age 15 years werecharacterised by disadvantaged early familybackgrounds and parental characteristics thatincluded low church attendance.481

#�(�1�� '�� #��

In the years following the completion of secondaryschool substance use tends to escalate. As was notedearlier, young adults have higher rates of drug andalcohol use relative to other sections of thepopulation. Longitudinal studies suggest that formany individuals early adulthood will be a periodwhere potentially harmful involvement in alcoholand other drug use will peak. Involvement in druguse through this period is strongly predicted bybehaviours developed in the secondary school ageperiod. However, there are some factors that havebeen observed to influence young adult drug use,including peer and spouse relationships and patternsof behaviour in educational, employment and socialsettings.

For those populations that are involved in drug use,there are more particular public health risks that areposed through early adulthood. These include risksof disease transmission through needle sharing andrisks associated with crime and violence. Knowledgeof special populations, such as injecting drug usersand Indigenous populations, is not availablethrough longitudinal research, hence the availablesources of information should not be ignored.

There has been less research examining factors thatinfluence stability and change in patterns of druguse through adulthood. It is known that duringadulthood many individuals decrease theirinvolvement in harmful drug use. Factors associatedwith increasing family responsibilities, experience

with the consequences of earlier drug use anddecreasing vitality are all thought to be implicatedbut there has been little research.

7'<'&�"�)��)-��*��)�

Summary: Frequent drug use in late adolescence isa risk factor for drug-related harm in adulthood(Chapter 3).

Available evidence summarised in earlier sections ofthis report demonstrates that drug use behaviour inadulthood is strongly predicted by drug usebehaviours developed in adolescence. In the periodfrom around age 18 years, heroin use may start toemerge in some populations (although the age ofinitiation has lowered through the 1990s); alengthy history of drug use with alcoholintoxication at age 12 to13 are typical in heroinusers.166

Risk periods for initiation of cannabis (marijuana)use are mainly over by age 20 years and initiationafter age 29 is very rare. For adults in their late 20s,‘higher proportions of users stop using marijuanathan start using it’.522 Late onset of smoking is quiterare. Smoking is a behaviour typically initiated inadolescence.523

7'<'.� ��) �,

Summary: A well-managed environment foralcohol use in adulthood is a protective factor,reducing the risk of harms associated with alcoholuse (descriptive associational studies, Chapter 14).

Evidence from a number of studies suggests thatwell-managed environments for the sale andconsumption of alcohol provide an importantcommunity contribution to reducing harmsassociated with alcohol use.

Drinking settings appear to influence violence.385

Alcohol-related violence is highly associated with asmall number of venue types, with the majority ofvenues in most studies reporting zero incidences ofviolence. A number of factors in drinking settingshave been shown to increase the likelihood ofviolence. For example, either overly relaxed oroverly aggressive staff conduct is associated with anincreased risk of violence.385 Large numbers ofpatrons binge drinking at the same time increasesthe likelihood of problems as do frustrating orirritating patrons, poor ventilation, smoky air,inadequate seating and/or inconvenient bar access.More ‘permissive’ environments (e.g. allowingaggressive behaviour, swearing, rowdiness, etc) areassociated with higher likelihood of violence.524

'%��

Crowding and congestion in licensed venues is alsoassociated with increased aggression,385 as is poolplaying, patrons milling about and patrons beingbored.385 One of the strongest predictors of risk forindividual drinking venues is the presence of barstaff who continue to serve obviously intoxicatedpatrons.525 Another significant risk factor is closingtime of hotels and pubs. This is believed to beassociated with a gathering in the street of a largenumber of intoxicated people from a variety ofdifferent venues.525

Clean, well-presented drinking environments areassociated with fewer problems.524 Positive socialatmospheres where boredom is low, and wherevenues serve full meals are associated with a lowerrisk of aggression.524

Alcohol-related crime is also more likely in areas ofa community with a high concentration of licensedpremises in the one area, and with the oftencorresponding effect of large numbers of drinkersleaving the venues at around the same time.385

Licensed premises are an especially high-riskscenario for drink driving.385

7'<'0�$�� -

Summary: Marriage in early adulthood is aprotective factor, reducing the risk of harmsassociated with alcohol use (two longitudinalstudies).

In general, in US studies conducted in the 1980s,living with a spouse in the years followingsecondary school appeared to reduce subsequentinvolvement in illicit drug use. In a sample of USstudents interviewed in the early 1970s, males andfemales who were married by age 24 years wereless likely to be involved in frequent illicit drug useat ages 24 to 25.162 This finding was maintainedafter adjustment for a range of other influencesincluding adolescent drug use. In a separate follow-up of a national US sample of students, initiated in1975, involvement in illicit drug use in the yearsfollowing secondary school was also lower forstudents living with a spouse.164

Australian national mental health survey data revealat a very broad level that substance use disorders areassociated with marital status and livingarrangements. Substance use disorders were presentin 13.1% of people who had never been married, incontrast to 4.5% of married people.377 Marriedpeople had lesser rates of disorder than thoseseparated or divorced, who in turn had lesser ratesthan never married people.377

7'<'(��+�,��

Summary: Unemployment in early adulthood isassociated with harmful alcohol use but it is unclearwhether its influence is maintained after adjustingfor earlier risk factors.

Unemployed people are significantly more likely tohave used hallucinogens, and to have used them inthe last 12 months, than people in the workforce.166

The prevalence of injecting drug use, cocaine useand use of LSD is also highest for those who areunemployed.166

In a sample of US students interviewed in the early1970s, males who were unemployed prior to age24 years were more likely to be involved infrequent illicit drug use at ages 24 to 25.162 Thisfinding was no longer significant after adjusting forother influences including adolescent drug use.

7'<'6�$� �+��+�

Summary: Having a mental health problem inadulthood is associated with harmful drug use. It isunclear whether mental health problems determinedrug use.

As was discussed in the earlier sections of this reportdealing with drug-related harms, drug use inadolescence appears to predict a variety of mentalhealth problems. Evidence presented in earliersections of the present chapter suggested that somemental health disorders in childhood andadolescence predict subsequent drug use. It is lessclear whether mental health problems in youngadulthood continue to predict drug use, aftercontrolling for earlier drug use and mental healthsymptoms.

In a cohort in Erie County in New York, aged 19and older, depression was found to have adifferential effect by sex in predicting alcohol abuseand dependence two years later. For males there wasno effect, while for females the effect wassignificant after adjusting for prior alcoholproblems, race and social class.491

In a cohort in the Midwest of the US, depressivesymptoms at age 18 were found to predict, betweenages 18 to 25, both the less common forms ofchronic alcohol dependence and alcoholdependence that was co-morbid with tobaccodependence. However, depressive symptoms didnot predict adolescent-limited forms of alcoholdependence.120

�

'' �� !��"�#��$��%��

#��)�,��!�� #"7�

During the period of retirement and old age, late-onset drinking problems emerge in somepopulations. In a recent review, it was argued thatin many cases drinking problems emerging in theelderly are a continuation of high levels of non-problematic social drinking earlier in life.256

Previous drinking levels may become associatedwith problems due to the impact of cognitive orfunctional impairments emerging due to alcoholand/or aging, or through greater vulnerability tothe effects of alcohol with age.

7'&='&��+��-�+�� +�--�� -

Based on clinical impressions, losing a spouse,loneliness and reduced social support have beenassociated with late-onset drinking problems in theelderly.256, 260

7'&='.��

Other psychological stresses associated with agingthat have been anecdotally linked with late-onsetdrinking problems in the elderly include a loss ofeconomic status, unrealistic expectations ofretirement and a sense of loss of role.258

7'&='0��+�� -

In their 1989 review, Schonfeld and Dupreereported that loneliness and boredom were the mostfrequently self-reported experiences of late onsetalcohol abusers.258

7'&='(�#��+�

Increasing age is a significant risk factor for anadverse reaction to medicinal drugs.268

#��!��$�� $��-�� $��

In recent years, the risk focused approach to theprevention of adolescent health problems hasemerged from the public health arena. This grewfrom recognition that a range of adolescent healthand social problems clustered in particularindividuals and particular social settings. Thus, ayoung tobacco user is more likely to be a heavydrinker, use cannabis, engage in risky sexualactivity, have higher rates of antisocial behaviourand, if female, experience symptoms of depression.So too, different social settings (e.g. schools and

local neighbourhoods) tend to vary markedly in therates of a range of problems, including tobacco use.

One important reason for this clustering appears tobe a range of social and individual risk factors thatpredispose to adolescent health and social problems.Previous reviews10 have suggested that anindividual’s sense of positive connection orattachment to family, school and communityprotects against a range of risk behaviours, as wellas promoting positive educational and socialoutcomes. Similarly, the existence of a clear set ofvalues, for example a belief in social order, appearsprotective against a range of risky behaviours,including tobacco use.

Earlier work completed in Victoria526 has, forexample, shown that such a framework can besuccessfully applied in the Australian context toassess risk and protective factors among adolescents.The relationship between tobacco and alcohol useduring the last 30 days and the presence of each ofthe risk and protective factors was determined fromsurvey data of approximately 9000 Victoriansecondary school students and is presented in Table6.1 to illustrate the range of factors linked to bothtobacco and alcohol use. The risk and protectivefactors listed in this table were included in theVictorian survey following US literature reviewsimplicating these factors as predictors of youthsubstance abuse.10, 479 Many adolescent healthproblems share important risk factors. For example,academic failure and school dropout are associatedwith antisocial behaviour, higher rates of substanceabuse, tobacco use and emotional problems.Similarly, factors such as family attachment andfamily conflict are linked to a broad range ofadolescent health problems beyond tobacco use.This observation from both cross-sectional andlongitudinal studies lies behind a growing push totarget risk and protective factors, with multipleoutcomes, rather than restrict focus solely to thosethat are ‘issue specific’.

The interested reader is referred to further reviewson the determinants of drug abuse in adolescence,10

and integrative theories of resilience such as theSocial Development Model.479

Table 6.1 makes clear that risk and protective factorsidentified in overseas research predict alcohol andtobacco use in similar ways within Australian youthpopulations. The present overview makes clear thatthere are various risk and protective factorsinfluencing entry to drug use and progression toharmful drug use. The conclusions reached in thecurrent review are similar to those reached fromearlier review work examining evidenceaccumulated to the early 1990s.10, 124, 494, 527, 528

'(��

��,��8� �� 2=��

�� !�"� �#�� !�"��

��"$��%�#��$�� !�� !� ��"#��

��$��%�$��$�� "� �!� �� #� !�#� �!��!��

��&�'�� " �� !�� !��#��!�!�

(�$��%�&��&� � !�� !��!��!�� #" !�� !� ��!��

(�$��%�)��%��&��)�� !�� #"� �� !�� !��""�

�� !�! �!��"#� # !��! �!��!��&��)��

��"&$��#��'�%�#��$�*��&)$�� ! "�� "��"��#� � "�#� �"�#!��"��"�

+''��'��&��$�� ! "�� "�� "�� "�#� �"�# ��"��

�,��'��&��$�� !� "��# �"��!��"�� "�#� �"�#!��"��"�

�#��$��%�#��$�*&��$�&�� # !� �!��!�� !� # �!� "��!��

-�,�&�$$��$��&)�� #�� #! !�� !��!�##�

�#��&$��#��'�%�#��$�+''��'��&��$�� !� "�� "��"�� # "��" �"�##��"��

�,��'��&��$�� !� "�� "��"�#!� � "�#� �"�# ��"��

��!�'�!(��)&��$$��%�#��$�.��%��'��$��)�� # ��"!��

(�� #��#�� !��

��&��/�� ! ��" �� # !�� !��"��!�#��

(��0�� #� � ��#�"� �! �� #"�

1��/�� # �!� �� " �!��!��

�,��&��)�� !� � �!��!�� !�� !��!� �

2��$�� !� ��"�� #� !� �!� ��!��#�

�� !# ��""�� # !�� !��!�##�

*��&��)�� !� ��"�� #� # !�� !��!�"�

��&��,��)��&��'�� !!�� #"� �" !�� !��!��

(��&��)�� #�� # !�## �!��!��

��!�'�!(��)&��$&$��#��'�%�#��$��% !� "�#� �"��"�� "�� "��"��"��"�

1�&��/�� "�" �"�"��"�!�� ! "�� "��!��"��#�

3�� ! "�� "� ��"�� ! "�� "��#��"�# �

��(��"$��%�#��$�-�,��$�� !� � �� # ��

��&��%�� !� � �� # ��!� ��"��

-�,��)��)��&)$�� ! !�� !� ��!�� !��" �!�!��!��#�

��$$��%�� !�� !�#"��!��#� �� !� � �!� !��!��

��$��% �# !�!� �!�"��!�� # "�� "� ��!�" �

��$$��%��$��% �� !�"� �"�#��!�! � �� !�" �"��!�"��

��(��"&$��#��'�%�#��$�+''��'��&��$�� "�� "��!�" � �� !�" �"��!�"��

�,��'��&��$�� !� "�#� �"�#!��"��#� �" "�� "��"��

Prevalence of the outcome behaviour (smoking or alcohol use) amongst those high on the listed risk or protective factor.

�

() �� !��"�#��$��%��

#��8�� 5��$�� 9�,��-0�� $�� :

The ‘Prevention Paradox’ was a term coined by theepidemiologist Geoffrey Rose to describe how oftenit is the lower risk individuals who collectivelycontribute the bulk of preventable illness in thecommunity due to their greater numbers.529 Theclassic example is that the numerous individualswith moderately elevated blood pressure collectivelyexperience more heart attacks than do the smallernumber of people with highly elevated bloodpressure. Kreitman11 first drew attention to whatappears to be a similar phenomenon in relation toalcohol. People who on average consume moderateamounts collectively experience more harmfulconsequences from drinking than do those classifiedas heavy drinkers. It was subsequently demonstratedthat this ‘paradox’ vanishes when a measure of‘binge drinking’ is used instead of average dailyconsumption.530, 531 Most of the harmful outcomesmeasured by Kreitman were those that would beexpected to be related to binge drinking, forexample, taking time off work, being injured, beingin a fight.

In the summary above, the key concept of risk hasbeen applied more broadly to a range of social,familial, contextual and developmental variablesfound to be predictive of adolescent drug use. Doesthe prevention paradox apply to these data? In otherwords, are the high risk individuals responsiblecollectively for most of the risky drug use or is it thebulk of the population who are at low or averagerisk who collectively generate the bulk of riskybehaviour? This is a central question for policy inthe area of drug prevention as it speaks to whetherstrategies should be targeted mainly at high-riskindividuals or the population as a whole.

To address this question, a data re-analysis wasconducted (specially for this project) of a majorVictorian survey of around 9000 secondary schoolchildren, across school years seven , nine and 11.526

Risk and protection factors were measured using aspecially adapted version of the scale developed inthe US to measure risk and protective factors,identified in the review by Hawkins andcolleagues.10 This scale allows a single summarymeasure to be derived indicating degree of risk andabsence of protective factors. Scores on this scalewere categorised into three groups: Low risk,Average risk and High risk. These categories weredefined on the basis of the obtained frequencydistribution of scores. Thus, Low risk was defined asone standard deviation or more below the mean

score and High risk as one standard deviation ormore above the mean.

Potentially harmful drug use was defined for theanalysis as follows:

� drinking five or more drinks on one day, at leastweekly;

� smoking cigarettes at least weekly;

� smoking cannabis at least weekly; and

� using any other illicit drug at least weekly.

Figures 6.1 to 6.4 report the percentages of studentsfalling into each risk group for each drug typeexamined.

The extent to which the prevention paradox appliedto adolescent risk and protection factors varied as afunction of age and drug type. In relation to thelegal drugs alcohol and tobacco, the paradox holdsfor the older children. That is, for year 11 children(and year 9 for smoking) the majority of those whodrink more than four drinks at least weekly (whosmoke a cigarette at least weekly) are Average orLow risk on the modified Catalano et al. scale(Figures 6.1 and 6.2).

For at least weekly use of cannabis and of otherillicit drugs, it is the High risk students whocomprise the majority, at all school years (Figures6.3 and 6.4).

Across all drug types, the younger the student themore likely it is that potentially harmful drug use isassociated with higher risk individuals (thoughnumbers are fortunately very small for illicit druguse in school year 7). It is recommended that thisissue be explored further in older age groups. Thetrend for cannabis, for example, indicates that theprevention paradox might apply to weekly use bythe late teenage years.

In summary, while adolescents with a high numberof developmental risk factors and low protectivefactors are more likely to use all types of drugs in apotentially harmful manner, it is the bulk ofchildren at low or average developmental risk whoengage in smoking and drinking regularly by year11. That is, the prevention paradox holds for thelegal drugs. This is not the case for the illicit drugsthough future analyses are likely to find that by lateteens the prevention paradox also applies to weeklycannabis use. These findings suggest that preventionstrategies for legal drugs need to be universal intheir application and relevance to young people. Aninvestment in targeted programs for high riskadolescents is also warranted, especially with a viewto the prevention of later use of illicit drugs.

(��

"�� ,�� ?��5��::��

"�� ,�� ?��5��::��

0

10

20

30

40

50

60

70

80

High risk

Medium risk

Low risk

Year 11 (n = 661)Year 9 (n = 519)Year 7 (n = 120)

School year

Pro

port

ion o

f w

eekly

sm

okers

(%

)

0

10

20

30

40

50

60

70

80

High risk

Medium risk

Low risk

Year 11 (n = 548)Year 9 (n = 264)Year 7 (n = 67)

School year

Pro

port

ion o

f bin

ge d

rinkers

(%

)

�

(� �� !��"�#��$��%��

"�� ,�#� �� ?��5��::��

"�� ,�'� �� ?��5��::��

0

20

40

60

80

100

High risk

Medium risk

Low risk

Year 11(n = 204)Year 9 (n = 116)Year 7(n = 14)

School year

Pro

port

ion o

f w

eekly

cannabis

use

rs (

%)

0

10

20

30

40

50

60

70

80

High risk

Medium risk

Low risk

Year 11(n = 131)Year 9 (n = 49)Year 7 (n = 33)

School year

Pro

port

ion o

f w

eekly

use

rs (

%)

(��

��

� � � � � � � � � � � � � � ��

� � � � � � � � � �

(� �� !��"�#��$��%��

("��

�� %��!!��

In this chapter, interventions aimed at improving developmental outcomes for children and theirfamilies are reported. Of the interventions beginning prior to birth, there is some evidence thatfamily home visitation is a feasible implementation strategy with disadvantaged families and canimprove developmental outcomes for children. There are a number of strategies developed fordelivery from the pre-school age period that now have reasonable evidence for theireffectiveness. There is also evidence that parent education programs, delivered through thisperiod, can reduce child behaviour problems. School preparation programs appear to holdbenefits in enhancing learning potential. These programs are being delivered from as early as sixweeks after birth. Some of the strongest evidence for effectiveness in reducing harmful drug usecomes from interventions delivered in the primary school years. Family intervention, parenteducation and school organisation and behaviour management programs all have reasonableevidence for their effectiveness in preventing the transition to harmful drug use. Much of theresearch has been completed in North America and many of the effective programs have not yetbeen implemented in Australia.

�� !�� $ �� 9�� !�� !��

� �� =� � � � �� *�� $��

+ �� '�� /�� 0 �� '�� /�� 0 ��

��:� ��04��'2+ �� '��

��

�� '��

�� '�� $�� '�� '�� $�� !��

� � ��

� �� $��*��06��2�� '� �� '� ��

��

�� '� �� )��$ ��

'��$�� < �� '�� '�� '� �� $��

>��7 *��

� �� $��

� #��

� ��

��

��

�

(# �� !��"�#��$��%��

%�� 3��

Prior to birth interventions have been developed toaddress a number of intervention targets. Majorintervention targets for these programs include:preventing teenage pregnancy, childbirthpreparation and reducing foetal exposure to harmfuldrug use. Although a number of relevant programshave been implemented, and in some casesevaluations have been conducted, there have beenfew studies that enable assessment of the long-termimpacts on reduction of harmful drug use.

3'.'& �� +�,� �� ,� �,�) �� )+ ��*+��-

Definition: The use of a broad range of programsdesigned to prevent pregnancy amongst teenagersand vulnerable mothers. Strategies include: delayingthe initiation of sexual activity, encouraging the useof contraception, reducing risky sexual behaviourand providing access to pregnancy termination.

Summary: Warrants further research ............... � 0/0

Due, in part, to the ready availability of pregnancytermination services, Australia does not currentlyhave high rates of childbirth amongst teenagers.There are regions of high disadvantage, however,that are exceptions to this general rule. A recentreview of overseas literature relevant to theprevention of sexual risk taking concluded thatpreventive strategies, such as school-based sexeducation and community mobilisation, have beensuccessful in delaying pregnancy amongst youngpeople with a high number of risk factors.14

Although there exists evidence for the successfulimplementation of these strategies, their outcome inpreventing pre-birth exposure to drug use and druguse problems in future generations is unclear. Thereis evidence that in some cases, these programs mayreduce harmful drug use in vulnerable youngwomen.14 Further research in this area would appearwarranted.

3'.'. #��+��-��

Definition: Reorientation of existing health servicesaimed at enhancing conditions for maternal andchild health.


Common themes for interventions to preventadverse impacts on children of parental drug useinclude: facilitation to encourage child bondingduring pregnancy; early identification of foetus orinfant at risk or manifesting the effects of drugexposure, including early intervention to encourage

maternal abstinence; and education and training offamily members, the community and the widersociety.532 Chabon et al. describe multiple assessmenttechniques, including review of medical and socialhistory, observation, interviews with case workersand others, and mother-completed questionnaires.Assessment is argued to be useful for identifyingdrug-exposed children, informing treatment plansand evaluating treatment.532

Universal programs to discourage harmfulmaternal drug use

Several child health services are provided inAustralia, with the aim of supporting expectantmothers and identifying conditions that mightundermine healthy child development. Theeffectiveness of these efforts in reducing harmfulmaternal drug use is unknown.

Health service screening and assessment

Two studies have provided preliminary evidenceaddressing the feasibility of implementing screeningand assessment interventions for pregnant women.In overview, results show some level of acceptancefor these services but designs have generally notenabled behavioural outcomes to be clearlyestablished.

Ettlinger used an alcohol screening processimplemented by a maternal child health nurse andother health professionals.533 If mothers scored overthe risk threshold, the nurse followed up withcontinued encouragement to not drink, briefintervention, or suggested community treatmentresources. The health message was complementedwith an educational pamphlet and a wallet-sizeddurable card listing a range of communityresources. This program was evaluated with deliveryto pregnant women (n = 149) enrolled in asupplemental nutrition program in a rural area ofVermont in the US. Of the mothers who wereassessed, 5% self-reported alcohol problems. Nursesreported an increase in mothers coming forwardand wanting to talk about alcohol risks. The authorsconcluded that training nurses about alcohol risk inpregnancy and giving them referral information washelpful in attempting to address these problems.

Chang et al. reported a randomised trial of a briefintervention for alcohol use in pregnancy.31 Aselected sample of women enrolled in prenatal carein Boston were the focus (n = 250). All drankalcohol in the six months prior to assessment, 43%drank while pregnant and 40% satisfied DSM criteriafor lifetime alcohol diagnoses. Of those invited toparticipate, 24% refused and these were more likelyto be from minority populations. The intervention

(%��

involved a comprehensive assessment of alcoholuse, or the same assessment plus a brief 45 minutesintervention by a health professional. The resultsshowed that both groups reduced alcohol use afterassessment (17% increased alcohol use), thoughthere was no significant difference between groupsin the size of the reduction. Of the 143 who wereabstinent before the assessment, the briefintervention group maintained higher abstinencerates (86%) post the intervention than theassessment only group (72%) (p=.04). Drinkingalcohol while pregnant was a strong predictor ofpostnatal drinking.

3'.'0 ��+,��-�� 8� � ��+9

Definition: Family home visiting involves aprofessional such as a nurse developing arelationship with a family over a period of time inthe context of offering support, information, adviceon infant health and development, and maternalhealth, and advocacy for service access. Relevantprogram targets in the antenatal period includereducing pre-birth exposure to harmful drug use,the family’s harmful drug use, and reduction ofearly risk factors for the child’s later involvement indrug abuse.

Summary: Evidence for outcome effectiveness ��

Olds et al. evaluated a program involving regularhome visiting by a nurse from late pregnancy untilthe child’s second birthday for low income,unmarried and adolescent women having their firstbabies.534 The program focused upon supporting themother, promoting positive attachment with thechild and teaching parenting skills. Follow-upassociated the program with reduced rates ofsmoking and alcohol use for the mothers duringpregnancy, leading to reductions in cigarette andalcohol-related cognitive impairment in the childrenas pre-schoolers. The children have been followedup to age 15 years, documenting reductions in theirearly initiation of smoking and alcohol use.535, 536

Conclusions—interventions prior to birth

There is some evidence that interventions prior tobirth can be implemented and have some potentialto improve maternal and child health. Althoughmaternal alcohol use is relatively common inAustralia, there has been little research investigatinguniversal strategies to reduce alcohol use inpregnancy. Future investment in health servicereorientation should focus on innovation to extendthe range of service delivery models underinvestigation. Existing evidence supports the viewthat early home visitation strategies can be

implemented with at-risk families and when wellimplemented, this strategy can result in a variety ofrelevant early intervention outcomes.

There appears to be a role for drug treatmentservices to cooperate with other sections ofgovernment to ensure that quality home visitationservices are targeted to all mothers experiencingdrug use problems. The last 10 years have seen alarge increase in the use of methadone treatment,with the consequence that an increasing number ofchildren are being raised by parents who areundergoing this form of treatment. Investment iswarranted to ensure positive child developmentwithin these potentially vulnerable families.

Many of the program strategies for the early yearsmay have application for Indigenous populations.These applications are likely to be more successfulwhere strategies are tailored to local communitiesby encouraging coalitions to take responsibility forimplementation and management.

%�� +�� )��

Major prevention strategies from birth through thepre-school years have included health servicereorientation, family interventions, parent educationand school preparation programs. There have beensome longer-term follow-up evaluations in smallsamples, linking these programs with reductions inadolescent behaviours associated with harmful druguse. The evidence does demonstrate that exposureto these programs has resulted in improvements inchildhood risk factors that have been linked to thedevelopment of harmful youth drug use.

3'0'& #��+��-��

Definition: Reorientation of existing health servicesaimed at enhancing conditions for healthy infantdevelopment.


Better designing services to address the needs ofinfants and children may be an important strategyfor encouraging healthy child development.

Universal programs supporting infant andmaternal health

In all States and in most local governments, servicesare run to support mothers and encourage thehealthy development of the infant. In a number ofcases, service models have been developed to reducethe impact of maternal smoking and to assistfamilies with problems associated with alcohol and

�

(' �� !��"�#��$��%��

drug use. We have been unable to locate researchevaluating the impact of these strategies on eithermaternal drug use or child development.

Paediatrician training

In view of the significant health effects posed tochildren by exposure to drug use (specificallyenvironmental tobacco smoke—ETS), public healthpolicies are needed to protect this vulnerablepopulation.537 The evidence of effectiveness ofinterventions that have been tested in controlledcircumstances does not provide much optimism;538

within the confines of the paediatric careenvironment, a recent systematic review of theliterature indicates that there was no evidence thatpaediatricians are effective in reducing parentalsmoking. However, the evidence was muchstronger for the capacity of paediatricians to increaseknowledge about ETS. One interpretation of theseresults is that paediatricians may have insufficientcontinuity of care for this task. A systematic reviewis currently being conducted that will hopefullyshed light on the evidence for interventions acrossall methods and sectors to reduce the exposure ofchildren to tobacco smoke by carers and families.539

Specialist programs for mothers experiencingproblems related to drug use

Copeland and Hall found that women withdependent children were more likely to be retainedin treatment where services provided specialistsupport for childcare and parenting.540

There is a range of drug treatment programs,including specialist support facilities for mothersexperiencing problems with drug use. There issome evidence that these programs may be moresuccessful at encouraging mothers to remainabstinent from drug use.541

Conclusions— health service reorientation

There appear to be opportunities for improvingearly child development through the reorientationof existing services, although strategies remainpoorly developed. Currently there are many parentsin drug treatment programs receiving no formalassistance for parenting and child development. Bystrategically coordinating and targeting serviceexpansion across the early years, it should bepossible to achieve advances in healthy childdevelopment.

3'0'. ��+,��-��

Definition: (see 7.2.3) Relevant program targets in thepostnatal period include reducing infant exposure toharmful drug use, the family’s harmful drug use,and reduction of early risk factors for the child’slater involvement in drug use/abuse.


Strategies for reducing family-level risk factors fordrug abuse through the early years of life wererecently reviewed for the NHMRC by Mitchell etal.478 That review concluded that the most wellresearched type of intervention targeting theantenatal and infancy period was professional homevisiting, with savings and returns to government ofaround $5 for every $1 spent on the program overthe first 15 years of the child’s life. Intensive homevisitation has been shown to be most cost-effectivewhen provided as a selective intervention to womenat increased risk, by virtue of factors such as youngage, poverty, lack of partner support, and drugabuse. There is evidence that this strategy may notdemonstrate benefits where it is applied moreuniversally to include low-risk mothers.

An Australian study found some benefits for parentsexperiencing drug-related problems through aprofessional home visiting program.542 Trends weredocumented amongst the parents for reductions indrug use and increased compliance withappointments. Interactions with their childrenshowed some trends toward increaseddevelopmental stimulation opportunities at homeand increased emotional responsiveness. Amongstthe children, small increases in cognitive scoreswere observed at six months but these were notmaintained at one year or 18 months. It may be thatintervention needs to be maintained in the yearsfollowing the child’s birth in order to sustain theinitial advantages gained through intervention.

Conclusions—family home visiting

Investment (in partnership with other sections ofgovernment responsible for infant, child andmaternal health) should monitor the impact andcost of existing home visiting services in order todetermine a framework for service expansion.Family home visiting has shown promise in bothoverseas and Australian trials/studies and existingevidence supports its cost-effectiveness.

((��

3'0'0 �� )��

Definition: One or more parents receiving informationand/or a course of instruction aimed at encouraginghealthy child development. Delivery strategiesinclude targeted, universal and combinedinterventions.


The period immediately following birth isconsidered a critical time for the development of astrong bond between the mother and newborninfant. In some cases, postnatal depression or otherfactors can make early mother-child interactiondifficult and, in this way, undermine thedevelopment of a positive relationship. The HUGSprogram was developed in Victoria as a groupintervention to assist mothers to understand andrespond appropriately to their infants’ needs.Trained facilitators lead groups of parents through asequenced curriculum consisting of six contentareas. Each content area is covered in one to threesessions, of between one and 1.5 hours, at a settime each week. Meager and Milgrom reported asmall trial of the program that demonstratedreduced levels of maternal depression at programcompletion, for mothers exposed to the programrelative to controls.543

Step By Step Childcare is a prevention programdesigned to assist parents with intellectual disability,to enhance the healthy development of theirchildren from birth to three years old. Parentsreceive individualised instruction in basic care tasksuntil they demonstrate performance of the task to aset criterion. Controlled evaluations havedemonstrated that exposure to the programenhances parenting skills and also leads toimprovements in child development outcomes. Forexample, Feldman et al. demonstrated that infamilies where parents received the parent-childinteraction components, significant improvementsin the child’s language development wereevident.544

Parent management training, based on cognitivesocial learning theory, is one of the most widelyused parent education techniques. Parentingprograms based on this theoretical approach aim to:reduce early child behaviour problems, improveparenting practices, and effect changes in theenvironment that will maintain and reinforcepositive child development. These targets representimportant risk factors for later adolescent drugabuse, delinquency and mental health problems.

Findings of a meta-analysis study generally supportthe efficacy of parent education programs as astrategy for reducing child behaviour problems. Inthis review, it was noted that compliance can belower in families that have a high number of riskfactors.545

Mitchell et al. report that these programs tend tohave large effect sizes (0.86), with improvementsobserved for around two-thirds of participants.478

Interventions are successful in enrolling aroundtwo-thirds of those eligible. Typically, theseprograms are timed to prevent problem behavioursbetween the age of two to eight years; and to reduceproblem behaviours, such as non-compliance andconduct disorder, early before they become resistantto treatment.478 Patterson suggests that familieswhere young children already have establishedclinical problems typically require around 20 hoursof direct contact program time.546

Intervention strategies to specifically address parentand family issues, such as marital distress, familyviolence, lack of a supportive partner, maternaldepression, parent substance abuse and life stressorshave been shown to improve the effectiveness ofchild outcomes. Programs such as Triple P PositiveParenting Program and some of Webster-Stratton’sprograms include specific program content toaddress these risk factors when necessary.547–549

Triple P Positive Parenting Program

The Triple P Positive Parenting Program is the mostcommonly implemented parenting program inAustralia. Triple P is a multi-level program derivedfrom more than 15 years of research and has beenimplemented and researched with a variety ofdifferent family populations. There are five levels ofthe program, provided to accommodate thediffering severity in disrupted family functioning orchild behaviour problems. At Level 1, universalmedia-based information campaigns are providedand at Level 5, individually tailored programs areprovided to address more severe dysfunction.

The program is well supported through trainingevents and a wide range of professionally developedmaterials. Recent research has included an outcomecomparison of the Level 5 Triple-P, the Level 4Standard, Self-directed Triple P and a waiting listcontrol group for families with children between 36and 48 months of age, with a behaviour problem(parent-reported) and at least one risk factor such asmaternal depression, relationship conflict, etc. Theresults are presented post-intervention and at oneyear follow-up.550 All three intervention programsprovided better outcomes (child behaviour and

�

�)) �� !��"�#��$��%��

parenting behaviours) than the waitlist comparisongroup and the Level 5 program was better than boththe Level 4 programs, with the ‘Standard’ programbetter than the ‘Self-directed’ program results. Atone year follow-up, however, there were nosignificant differences between the results of thethree types of interventions.

Although there is already an impressive amount ofTriple P research evidence and on-going research,further longer-term follow-up research would behelpful in terms of understanding the maintenanceof outcomes. Also, the publication of the results ofthe comparison design evaluation of the Level 4group Triple P program is awaited to inform policyand services as this is currently being widelyimplemented across Australia. The evaluations todate have generally been small efficacy trials underthe control of researchers. Future researchinvestigating program effectiveness in a range ofservice delivery and family contexts would bevaluable in guiding future investment in thisprogram.

Webster-Stratton’s parenting curriculum

A series of parenting programs developed by CarolWebster-Stratton have been developed for use inuniversal applications and evaluated for theirimpacts in targeted populations with high levels ofbehaviour problems. There are different versions ofthe program for delivery in pre-school populationsand for the early school-age period. Centralcomponents involve facilitated groups of parentsand use of videotape examples to stimulatediscussion and to convey parenting programcomponents. The program has been evaluated in sixrandomised trials as an indicated or treatmentintervention for behaviour problems in childrenaged three to eight years.551 The results have beenreplicated by other researchers and in appliedservice efficacy evaluations. The program was alsoused and evaluated as a selective preventionprogram with other population groups such as‘Head Start’ mothers, Hispanic mothers, andAmerican-African mothers, in day-care centres.552 Ineach of the evaluations, increases in positiveparenting practices for participants have beenassociated with reductions in child behaviourproblems.

Helping the Non-Compliant Child

Helping the Non-Compliant Child is an indicatedand treatment intervention developed specificallyfor parents with non-compliant children aged threeto eight years. The program is relatively intensive. Atherapist works individually with the parents and

the child in a clinic setting. The programincorporates coaching: where the therapistcommunicates via an earpiece providing guidance asthe mother plays with the child alone and thetherapist observes the parent-child interactionthrough a one-way mirror. Other componentsinclude: modelling, role-playing and homework foreight to 10 sessions. Small randomised trials havedemonstrated positive improvements followingexposure to the program, in parenting practices andchild behaviour.553 In one small longer-term study,26 families where parents had completed theprogram were followed up. In these families,children had been assessed prior to entry to theprogram to have serious behaviour problems thattend to be stable predictors of longer-termdevelopmental difficulties. When followed up inadolescence 14 years after program completion, andcompared to community subjects of similar age anddemographic characteristics, the children whosefamilies completed the program appeared essentiallynormal. There were no differences on measures ofdrug use, internalising and externalisingbehaviours, social competence, emotionaladjustment, relationship with parents and academicprogress.554

Parent-child interaction therapy

The parent-child interaction treatment program wasdeveloped for children with conduct problems,aged two to six years, and their families.555 This is agroup program that combines play therapy (usingmodelling, role playing, and coaching usingmicrophones) with behaviour management trainingin a clinic playroom setting to assist parents build awarm and responsive relationship with theirchildren and to manage their child’s behaviourmore effectively.

Small experimental trials have evaluated theprogram, finding significant improvements in childbehaviour problems and in the interactions betweenthe parent and the child, maintained to one to twoyears following treatment.556 Other studies haveshown improved school behaviours.557, 558 Onestudy showed that the parenting component had alarger outcome effect compared with the playcomponent.559

Conclusions— pre-school parent educationprograms

There is now a substantial literature to documentthe value of early childhood parent educationprograms. Reviewers suggest that these programsmay be amongst the more cost-effective of thecurrently evaluated early intervention approaches

�)��

designed to tackle child behaviour difficulties.545, 560–

563 There are very little data on the effectiveness andappropriateness of the programs with differentcultural groups, or with Indigenous families.

There are few studies that provide longitudinaloutcomes greater than one to two years. From thedata that have been collected, it appears that therecan be reasonable maintenance of outcomes,particularly where parent-related difficulties orsocial adversity does not complicate the child’sproblems. There has been some longer-termfollow-up work by Forehand and Long and theircolleagues, where very small samples have been re-examined in adolescence and found not to havehigher rates of drug use relative to matchedcommunity controls.554 Further research is needed,however, to establish whether these positiveindications are more generally maintained across thevarious program and family contexts over time.Research is also required to establish whetherpositive changes observed in small efficacy trials canbe translated to wider scale benefits through successin larger applications in real world delivery contexts(effectiveness).

The extent to which the benefits of parent educationin childhood need to be supplemented throughother strategies, such as school intervention, is notwell understood. Webster-Stratton and others havenoted deterioration of the positive effects of parentmanagement programs over time.552 Suchobservations suggest that the programs may needbroadening to actively address other risk factors andto extend application through the pre-school andearly school years, if their potential to reduce druguse and conduct problems is to be realised in lateryears. The feasibility and benefit of this type ofcoordinated approach is currently being investigatedin the ‘Fast Track’ evaluation reported below.

3'0'( ��+�� -

Definition: Programs aimed at better preparingchildren for the transition to primary school.


An important developmental pathway foradjustment difficulties begins with the transition toprimary school. Better preparing children forprimary school is a practical strategy for improvingthe transition for vulnerable families.

The Carolina Abercedarian Project was a NorthAmerican program that addressed risk factors forlater difficulties by providing support todisadvantaged parents, plus early childhood

education in the form of combined home visitingand pre-school centre-based intervention.564 Theprogram was initiated for children as early as sixweeks of age and was specifically designed toenhance cognitive, social, perceptual, language andmotor development. Evaluation at the end of thepre-school program found that participatingchildren had significantly higher IQ scores thancontrols, and at age 15 years follow-up they showedhigher scores on reading and mathematics.Additionally, participating mothers became bettereducated and were more likely to find employmentthan the non-intervention group.564

In Sweden, research evaluated a day-care programfor children from the first year of life.565 Onehundred and twenty-eight children born in 1975were followed from their first year of life.Participation in day-care (either a high quality day-care centre or a licensed family day-care home) wasfound to have positive effects on children’scognitive and socio-emotional development atfollow-up when they were aged eight and 13 years.Teacher ratings showed that participating childrenoutscored controls on several measures of socio-emotional functioning, such as confidence, abilityto concentrate, creativity and social competence.These positive findings were sustained even aftercontrolling for home background, child’s genderand intelligence. School performance was highestamongst children who attended the program beforeage one year and the researchers attributed this tothe importance of this very early period for braindevelopment.

The Perry Pre-school project offered four half-daysof structured pre-school experience combined withweekly home visits over one or two years fordisadvantaged three and four year olds. In a smallrandomised design evaluation (123 African-American children with low IQ scores at age threeyears and living in families of low socioeconomicstatus—58 in the program group and 65 in thecontrol group), participants in the program wereshown to have significantly improved outcomes inthe follow-up studies at ages three to 12, 14, 15, 19and 27 years.

Evaluation found several long-term program effectsto age 27 years, such as a lower incidence of druguse and teenage pregnancy, lower risk of highschool drop out, increased likelihood ofemployment, and reduced reliance on welfarecompared to non-intervention controls. Womensuffered substantially less from mental healthproblems compared to those who did notparticipate in the Ypsilanti/High Scope program and

�

�)� �� !��"�#��$��%��

males from the program had considerably fewerarrests than those who were not exposed to theprogram.566

The Child Health and Education Study, a largelongitudinal study from the United Kingdom,measured children’s academic achievement andcognitive development based on whether theyattended half-day pre-school, child care and/or playgroups.567 Improved cognitive development andacademic achievement in three and four year oldswas found in those children who attended any formof organised group pre-school program comparedto those who did not. Findings were independent ofthe child’s socioeconomic status and the type ofpre-school setting attended. Parental involvementwas suggested as the strongest factor in determiningpositive child development outcomes, such as bettervocabulary at ages five and 10 years, better readingand mathematics at age 10 and better interpersonalcommunication skills.

As a broad-based American early interventionprogram, Head Start (from age three to school age)and Early Head Start (from pregnancy to age threeyears) focus on improving early childhoodopportunities for disadvantaged families. Evidencefrom evaluation of Head Start programs supportsthe conclusion that parental involvement is acharacteristic of ‘good’ pre-school programs,leading to improved outcomes for children.568

Sustained gains in school and school attainmentwere suggested for white children. It washypothesised that external factors, includinginfluences from the school system, were responsiblefor reducing the effect of initial gains from the pre-school program, particularly among African-American children, as they became older.

Early Head Start offers centre-based, home-basedand mixed approach options for families to providechild development services and build family andcommunity partnerships. Evidence from evaluationof Early Head Start programs shows that althoughoverall effects are modest, and show variation acrossdifferent populations, there are positive impacts oncognitive, language and social-emotionaldevelopment and parenting behaviours.569

Conclusions—school preparation programs

Although many evaluations appear to have beenrigorous, many are based on very small samples.The studies completed to date do not answerwhether any valid pre-school experience wouldhave similar outcomes. A Pre-school CurriculumComparison Study found few differences betweenthe Perry Pre-school Program and another pre-

school curriculum although both of these, using achild-centred and teacher-initiated approach withopportunities for mastery experiences and childproblem-solving practice, had significantly betteroutcomes than a standard pre-school program.570

There have been a series of cost benefit analyses ofthe 27 year Perry Pre-school Program study. Theestimates of government savings vary from $6 forevery $1 invested in the program for a one yearprogram, $3 for a two year program,571–573 andmore recent estimates of $2 savings for every dollarinvested in the program.574, 575

%�� !�� "��

From the entry to primary school, children areincreasingly exposed to a broader range ofinfluences. The role of teachers and experienceswith other children increasingly shapedevelopment.

3'('& ��+,��

Definition: One or more parents, children and otherfamily members receiving information, a course ofinstruction and/or therapeutic assistance, together,during the primary school age period; aimed atencouraging healthy family development. Servicesmay be targeted to families in settings such asprimary schools and family service centres.


Although we have presented family programstargeting the primary and high school yearsseparately, it is important to note that many of theprograms in Australia actually bridge the transitionfrom primary to high school ages. Furthermore,several programs that were originally developed forolder or younger age groups have been adapted forwider implementation throughout schools. In othercases, programs were initially developed to targetfamilies with drug use problems and have sincebeen adapted to target a broader section of thepopulation.

Targeted approaches in primary school age

The Families and Schools Together (FAST) programtargets families where students are identified byteachers to have behaviour, learning, or attendanceproblems in late primary school. Families areactively recruited through home visits. The initialprogram is professionally led and involves eightweekly two hour meetings involving a number offamilies.576 The meetings require considerablepreparation and involve a variety of activities,including meals, contests and prizes. For a

�)��

component of each meeting, parents and childrenare separated into different activities. The lastmeeting is conducted as a graduation. Familynetworks are encouraged to maintain mutualsupport through a further two years of monthlyfamily meetings run by graduating parents. Themajor intervention elements aim to build socialsupport (social capital) for families within theschool context, improve parent confidence andempower parents, improve family relationships andincrease child competence.465 To date, evaluationshave been limited to examinations of process butoutcome evaluations are in process.577 The programhas been successfully trialled in Victoria, beginningin 1997.

DeWitt et al. assessed an in-school program(Opening Doors) aimed at preventing or reducingdrug use and other deviant behaviour in 167 high-risk youth during their transition from primary tosecondary school.578 A student and parent programran concurrently. All students in experimental andcontrol schools completed a screening test ondemographic and behavioural variables and high-risk students were selected. There was only roomfor 108 of the 170 identified high-risk students inthe experimental group. High-risk participants inexperimental and control groups completed a pre-test followed by a post-test one month aftercompletion and a six month follow-up test.Although this was short-term follow-up, programparticipants compared with the control groupreported less frequent: drinking, cannabis use, non-prescribed tranquilliser use, self-reported theft; andimproved attitudes toward school. They alsoreported less supportive attitudes toward alcohol,tobacco and cannabis use and less risky drinkingbehaviour. There were no program effects onpersonal/social competence measures. The authorslist possible reasons for success, including highretention rates during the 12 month study period,recruitment of community health care professionalsto work with teachers, a comprehensive approachwith a focus on promoting warm parent-childrelations, peer and teacher relationships, supportfrom whole school and emphasis on building lifeskills. There were, however, several limitations tothis study, including non-random assignment ofschools to test conditions; self-selection bias to enterprogram; failure of 42% of year 9s to participate inthe screening, possibly because of need for parentalconsent; and evidence only available by self-report.This approach is, therefore, promising but requiresa more rigorous evaluation.

Universal primary school approaches

In the last 10 years there has been steady growth ofthe knowledge base on involving families in childand adolescent health promotion. Several programsincorporate family involvement as a component ofprevention programs run in schools or incommunity settings such as youth groups. Perry andcolleagues involved in Project Northland inMinnesota579 and the Midwest Prevention Programteam associated with Mary Ann Pentz580 have eachevaluated school-based health education programsthat include components aimed at involvingfamilies. Each of these evaluations has demonstratedthat the incorporation of parent activities intoschool health education curricula can be a practicalprevention option.

In general, evaluations have not been successful inuntangling the specific contribution of familyinvolvement to these prevention initiatives.Evaluation of Project Northland did demonstratethat, by year 8 of school, intervention studentsexposed to the program from grade 6 showed asmall significant tendency to perceive better familycommunication relating to alcohol use.579

Perry and colleagues evaluated the impact of theSlick Tracy Home Team program, a set of activitybooks completed as homework tasks requiringparental assistance, over the course of fourconsecutive weeks in sixth grade (about age 11years).506 The authors noted that the strategy of‘seeding’ their program into school homework wassuccessful in obtaining high rates of parentalparticipation from a range of cultural backgrounds.At the end of sixth grade, families exposed to theintervention demonstrated increasedcommunication regarding alcohol use, and childrendemonstrated lower initiation of smoking andregular alcohol use.

A smaller number of programs have evaluateduniversal (whole population) family interventionstrategies. These programs used a range of strategiesto encourage healthy family communication. Themore intensive programs of this type providedprofessionally-led, sequenced groups for parentsand children. Typically, programs either targeted allparents or selected samples in primary schools

The Iowa Strengthening Family Program (ISFP)developed out of Karol Kumpfer’s successfulexperience running the Strengthening Americas’Families Program with high-risk families in drugtreatment and disadvantaged community settings.511

The ISFP targets all families within late primaryschool (sixth grade). The sessions are structured

�

�)� �� !��"�#��$��%��

such that children and parents are in separate groupsfor the first hour and combine to one group topractice skills for the second hour. Young people’ssessions in the ISFP focus on strengthening positivegoals, dealing with stress and building social skills.Parent sessions focus on communication,monitoring and conflict resolution. A comparativeevaluation suggested initially that the program wasabout as successful in discouraging alcohol use as aless intensive strategy of providing assistance toparents alone.581 However, four year follow-up hassuggested that benefits extend to reductions inyouth hostile and aggressive behaviour,582 andreturns of just over $9 for every $1 invested in theprogram have been estimated.583 The futureevaluation evidence for this program should bemonitored to further establish its potential forpreventing harmful youth drug use.

Targeted approaches in drug treatment settings

To break the inter-generational cycle of illicit drugabuse, interventions need to work effectively andearly to ensure the healthy development of childrenin vulnerable families and, in this context, parentsexperiencing illicit drug use problems represent animportant target. Program strategies have beenevaluated, suggesting the practical potential ofworking with parents in drug treatment. Typically,programs are intensive and incorporate casemanagement and home-based or group skillstraining for both parents and children.

Approaches that have been designed to assistparenting in drug treatment populations include theFocus on Families and the Strengthening Americas’Families Program. The Focus on Families programdemonstrated reduced parental drug use andimproved family management.584 The StrengtheningAmericas’ Families Program demonstrated increasedchildren’s protective factors, reduced substance usein both adolescents and parents and improvedparenting behaviours.585

In the case of the Focus on Families intervention,methadone treatment was supplemented with 33sessions of family training (each 1 to 2 hours)combined with nine months of home-based casemanagement. In their evaluation, Catalano et al.randomly assigned 144 parents and their 178children (age range three to 14 years) tointervention and control conditions, repeatingassessments at baseline, six months and 12 monthsafter the intervention.584 Of the families assigned tothe intervention, 74% were actively engaged. At the12 month follow-up, parents exposed to theintervention showed less drug use (with cocaineand heroin usage reduced by two-thirds) and

increased problem-solving skills for avoiding druguse. The intervention reduced the exposure of thechildren to a range of family-level risk factors.Household rules were clearer, domestic conflictreduced and the children were in less contact withdeviant peers. There was, however, little changenoted in children’s attitudes and the initiation ofsubstance use was not different relative to controls.The lack of measurable impacts in childhood mayhave been because many were too young forattitude and behaviour change to manifest.

Conclusions—family intervention in familieswith primary school aged children

There is early evidence that moderately intensivefamily intervention, using strategies such as Focuson Families and Strengthening America’s Families,can enhance treatment outcomes for parents in drugtreatment and reduce risk factors for children. Long-term evaluation research will be required to buildappropriate skills and ensure program effects can bereplicated in the Australian context.

Although universal family intervention in primaryschool is a promising prevention strategy, it is moreexpensive than universal parent education and holdssome risks if children with behaviour difficulties areencouraged to affiliate. Evaluation is required toestablish whether these programs can achieve betteroutcomes than those obtained through parenteducation alone. Evaluation of programs, such asthe Iowa Strengthening Families Program (ISFP),suggest that family intervention programs in lateprimary school may be a particularly useful strategyfor reducing risk factors for youth alcohol use.

3'('. �� )��

Definition: One or more parents receive informationand/or a course of instruction aimed at encouragingthe healthy development of primary school agedchildren.


Information relevant to the delivery of parenteducation programs for primary school agedchildren is summarised above. In overview, theseprograms have shown positive results reducingchild behaviour problems and enhancing parentfunctioning.

The programs aim to provide training andinformation to all parents in a school context. Spothet al. evaluated the delivery of the parent trainingprogram known as Preparing for the Drug FreeYears (PDFY) in middle schools in the US State ofIowa.511 PDFY is a five session, professionally-led

�)"��

program aimed at enhancing positive parent-childinteractions, parent-child bonding and effectivechild management. Parents are encouraged toprovide their children with opportunities forpositive family involvement, teach their childrenskills for such involvement, and reward them forthis involvement while providing appropriateconsequences for rule violating behaviour. Of theeligible parents, 57% were willing to be involved inthis parent training evaluation study. For thoseassigned to intervention, the program wasdemonstrated to be effective in increasing youngpeople’s intention to abstain from alcohol and inenhancing family bonding. Follow-up revealed thatbenefits in the form of reduced youth alcohol usewere maintained two years after the intervention.Recent estimates suggest a return of just over $5 forevery $1 invested in the program.583 The programhas not yet been trialled in Australia.

Conclusions—parent education for families withprimary school aged children

In an earlier section, a range of research wassummarised supporting the view that parenteducation programs appear to show promise insmall trials as strategies for addressing behaviourproblems in early and late childhood. In thissection, evidence was summarised supporting theview that parent education continues to provide auseful source of support for families through to thelate primary school age period. From this period,follow-up studies have been conducted intoadolescence, raising the possibility of evaluatingimpacts through to adolescent drug use.

The PDFY curriculum appears to be effective ataddressing youth alcohol use and has been widelydisseminated in the US. The relevance of thisprogram in the Australian context remains unclearas the program has not been evaluated for deliveryin Australia. In Australia, the more generic Triple Pprogram has been widely disseminated and showsconsiderable promise. It is unclear whether parenteducation programs need to focus specifically ondrug use through the primary school period; hence,a comparative trial examining drug use outcomesfollowing exposure to a program such as Triple Pversus the PDFY curriculum may be useful.

3'('0 ��+5*�-��)��)��

Definition: Delivery of a structured social health, drugeducation curriculum within the primary schoolusually by classroom teachers, but in some cases byvisiting outside professionals.

Summary: Evidence for implementation .........� 7/12

There is some agreement that the optimal time forintroducing school-based drug education programsis in the late primary school or early high schoolyears, when experimentation begins. Researchrelevant to these programs is summarised below inthe section on high school aged youth. In thissection, we consider the effectiveness of primaryschool drug education and discuss evidence forappropriate intervention activities.

Lloyd and colleagues have conducted a review ofthe effectiveness of primary school drug educationtargeted at children below 11 years of age, with theobjective of preventing illicit drug use.586 Therewere few studies pertinent because of the length oftime required for follow-up, but some promisingstudies included the Australian Illawarra DrugEducation Program,587, 588 Project Charlie in theUK,589 and a new Hampshire Study.590 Toumbourouand colleagues also summarise evidence for a rangeof programs delivered in grades 5 and 6.14

There is some evidence to suggest that drugeducation programs may be less effective wherethey are initiated too early. Results from a 15 yearschool-based matched-pair randomised trial of theimpact of a theory-based, social influencesintervention on smoking prevalence among youthwere published recently.591 The HutchinsonSmoking Prevention Project (HSPP) was conductedin forty Washington school districts, between 1984and 1999. Each district was randomly assigned toeither the intervention or control condition. Districtpairs were matched on prevalence of high schooltobacco use, school district size and location.Participants were children enrolled in twoconsecutive third grades (n = 8388) and follow-upcontinued until two years after completion of highschool, with only 6% lost to follow-up. Nodifferences were found in the prevalence of dailysmoking between students in the intervention andcontrol communities at year 12 or two years afterhigh school.

The HSPP has generated ongoing correspondenceon the effectiveness of school-based socialinfluences approaches to long-term deterrence ofsmoking among youth. The HSPP was elegantlydesigned and included measures of implementationfidelity; hence, its longer-term failure highlightedthe difficulties in maintaining behaviour changethrough drug education. Although the HSPPfindings introduce an important caution, it is asingle study, initiated at an early-age and conductedin small schools in primarily rural settings withprimarily white youth.592

�

�)# �� !��"�#��$��%��

Based on evidence over the past decade, it might becautiously argued that drug education curriculainitiated in early high school (year 7 in mostAustralian States), and reinforced with additionalprogram elements, provides a sound deliverystrategy. One study found programs were moreeffective when delivered in year 7 rather than grade6.593 Shope and colleagues found no benefit forgrade 10 health education impact from earlier sixthgrade exposure to health education.594

Attempting to translate the US research on early agedrug education is difficult, as grade 6 in the USoften represents the beginning of middle schoolrather than the end of primary school. Programsthat start in grade 6 in the US face fewer difficultiesin achieving an integrated sequence of follow-upactivities in subsequent years. The finding that drugeducation initiated in grade 6, or earlier, is moreeffective when it is supported by ongoing drugeducation activities in subsequent years is acommon theme in many studies. Several programsthat supplemented activities initiated in grades fourto six with further drug education activities in lateryears have reported reductions in drug use.595, 596

Many programs that provided drug education onlyin either fifth or sixth grade found either nobehaviour change,594 behaviour change limited toonly some sub-populations or behaviouraltargets,597, 598 or initial change that was notsustained.599, 600

The available evaluation work reveals an expandingknowledge relevant to the elements of primaryschool drug education that appear effective inencouraging behaviour change. Effective programshave tended to address social influences and thedevelopment of positive peer relationships.587, 588

The successful Illawarra Drug Education Program(IDEP) in New South Wales587, 588 targeted thedevelopmentally appropriate drugs, with alcoholand tobacco, in particular, seen as ‘gateway’ drugs.The IDEP was designed to reduce uptake and assiststudents in decision making and resisting peerpressure in relation to drug use. The program wasconducted over several weeks with grade 6 students;it included group work in which students designedart works and short dramatic sketches thatdemonstrated ways of coping with peer or mediapressure to use drugs. Parents were invited to attendthree drug education nights during this time, andduring the final session, students presented theshort plays they had worked on in class. In additionto this first phase of the program, a booster sessionoccurred in year 7, when students showed videos oftheir plays and discussed the program with other

students and then returned to their primary schoolsto induct the new grade 6 students into theprogram.

Students who received the intervention reportedgreater ability to resist peer pressure than controlsubjects, and were more able to use drugs at aresponsible and minimal level if they did engage indrug use. Despite positive program effects, year 9appeared to be a critical time when greaterexperimentation with drugs occurred. The authorsargued that this provided a strong case for boostersessions into the early years of secondary school.The authors concluded that life skills approachespresented to primary school children can impact onfuture smoking, alcohol consumption and illegaldrug use.

A British study, Project Charlie, adopted a broad-based life skills approach to preventing drug use.589

The program’s content focused on peer selection,decision making and problem solving and self-esteem enhancement, as well as providinginformation related to drug use. Although theevaluation sample sizes were small, results showedthat students who received the intervention weresignificantly less likely to have used tobacco andillicit drugs than controls at four year follow-up.

In some cases, process measurement has identifiedprogram elements that do not appear helpful inchanging behaviour. In one US study, fifth and sixthgrade students (average age 10 years) in twentyschools in North Carolina were randomly assignedto either an intervention or a waiting list controlcondition. The intervention was the Drug AbuseResistance Education (DARE) program. DARE istaught by uniformed police officers who areemployed in the US by local counties. DARE usesdiscussion, role playing, behavioural modelling andextended practice to reinforce the curriculum. The17 session program runs over four months andcovers knowledge and attitudes (specifically thenegative consequences of drug use), affectiveeducation (self-esteem and decision making skills)and alternatives to drug use (e.g. exercise and stressmanagement). Over 1000 students were surveyedprior to the intervention and 91% also completedmeasures repeated at the end of the intervention.Relative to the control students, the DAREparticipants increased their awareness of the costs ofusing alcohol and cigarettes and of the mediapromotion of these substances. The program wasalso effective at improving assertiveness skills and atencouraging more negative attitudes towards drugs.Peers were also perceived to have more negativeattitudes towards drugs. The problem for the DARE

�)%��

program was that none of these changes wererelated to either students’ intentions to use drugs orto their actual drug use.594

Why was DARE ineffective? It is possible that at age10 years, drug use is less guided by attitudes andmore by opportunity and social context factors. It isalso likely that by the late primary years, a small butimportant high-risk minority of students feelalienated from school and may be attracted to ‘bad’behaviours. Findings from the evaluation of DAREand similar programs have led researchers to theview that primary school drug education programsthat focus on knowledge, attitudes and values alonemay be of limited value. Although well intended,the DARE program failed to address important socialinfluence factors including the family and olderpeer drug use. Despite many similar findings, thepolice involved with DARE have few opportunitiesto engage with research and appear reluctant toabandon what appears, from their immediateimpressions, to be a positive program.

A program that has had some historical similaritiesto DARE is the Life Education program that has beenwidely used in Australian primary schools.Hawthorne reported a survey of drug educationactivities in Australian primary schools in 1992 thatshowed 43% were involved with the Life Educationprogram. Relative to schools that were not involved,these schools were three times as likely to useineffective ‘knowledge/attitudes/valuesclarification’ models of drug education. In an earliersurvey of students Hawthorne had foundassociations between prior exposure to LifeEducation and higher rates of youth substanceuse.601 In recent years, the Life Education programhas made a considerable effort to address theseproblems, including developing and trialling asecondary school curriculum.

Controversy has surrounded the question of whento discuss specific drugs with children. Someprograms have attempted to target programs tohigher-risk children and this strategy hasdemonstrated some success.578 However, moregenerally, it appears unnecessary to specificallytarget drug education activities. A number ofevaluations have revealed that effective drugeducation activities delivered to all students mayprovide either similar593 or enhanced impacts602, 603

for students with higher risk factors for harmfuldrug use.

Conclusion—school-based drug education inprimary schools

The available evidence supports the view thatprimary school-based drug education has the

potential for short-term impacts on developmentalrisk factors and drug use behaviour, but longer-term behaviour change is not yet assured. Theavailable evidence suggests that the curriculumshould focus more on building relationships andsocial-emotional skills than on drug use. Futureinvestment could examine process factors morecarefully (including relationships, skills andintentions) as a preliminary to long-term follow-up.

3'('( ��+�� -�� *��)��

Definition: Programs run in primary schools toencourage positive interpersonal relationships atschool, ensure effective discipline, and to maximiselearning.


Interactions with other students and with teachersduring the primary school years play a formativerole in the selection and development of peerrelationships, relationships with adults, bonding toschool and the development of identity and socialcompetence. A number of programs havedemonstrated that by altering the early primaryschool environment, long-term improvements inyouth development are achievable. A critical lessonemerging through this work is that encouragingchildren who are not experiencing adjustmentdifficulties to better assist children who areexperiencing difficulties can have profound impactsin healing childhood trajectories damaged throughearlier developmental phases. Many of the impactsfrom intervention through this age period havebeen shown to be moderated by class anddisadvantage, such that the most disadvantagedchildren experience some of the greatest benefits.

Improving primary school social environments

An important element of the school environmentrelates to classroom instructional practices. Efforts tomodify classroom practices have been evaluated inthe Seattle Social Development Project (SSDP). TheSSDP aimed to modify practices in grade 1 andgrade 7 classrooms in order to impact risk processesfor later youth antisocial behaviour. The evaluationof the SSDP utilised a true experimental design. Theintervention focussed on training teachers toemploy more effective classroom management andinstruction program components, in combinationwith a social competence curriculum, and parenttraining.

After one year of exposure to the SSDP, grade 7experimental students showed significant increasesin bonding to school, relative to controls. This effect

�

�)' �� !��"�#��$��%��

held for low-achieving students, as well. Inaddition, low-achieving experimental students,relative to low-achieving controls, had significantlysmaller increases in school suspensions andexpulsions. Evaluation evidence suggested thatdelivery of the program in early primary school wasassociated with the greatest long-term benefits.Long-term follow-up of the SSDP cohort has beenmaintained to age 25 years with 94% sampleretention. Evidence has associated exposure to theprogram with long-term improvements in schoolbonding;604 reductions in school failure, drug abuseand delinquency;605 and reductions in sexual risktaking behaviour.606 The program appears to offerdifferential improvements for the most vulnerablestudents from low SES backgrounds.604 Considerablework has been completed to test the process ofdevelopmental change, linking exposure to the SSDPto improved adolescent health outcomes. Theevidence supports the view that improvements inacademic achievement and, in particular, schoolbonding have mediated many outcomes. In thisstudy, improvements in school bonding in primaryschool led to better school bonding in secondaryschool. The SSDP was successfully replicated inWashington State in the late 1990s and it isunderstood that similar outcomes are beingdocumented.

Another successful attempt to reduce antisocialbehaviour in the primary school setting was theLinking the Interests of Families and Teachers(LIFT) program. The LIFT program was targeted toall grade 1 and grade 5 students. The program wascomposed of four elements: classroom social skillsinstruction for one hour twice a week for 10 weeks;playground behaviour monitoring (in the middle ofthe classroom instruction sessions, to reinforcesocial skills); six sessions of parent education (toassist parents to develop their child’s social skills);and parent communication (including a weeklynewsletter and a classroom phone answeringmachine for parents to leave a brief message for theteacher and to receive a daily message aboutclassroom activities and homework). Theplayground behaviour monitoring was based on avariant of the ‘Good Behaviour Game’. Studentsreceived individual rewards for positive socialbehaviours, but an additional reward was withheldfrom the group if negative behaviours were notreduced. This latter component was designed toincrease the sense of responsibility in the group, forassisting children to improve their behaviourproblems.

Evaluation was based on a randomised clinical trialinvolving 671 students and their families. Of the

full-time students enrolled in the selected grades,85% of parents agreed to participate. Retention tothe three year follow-up was 97%. Program fidelitywas high and family satisfaction with the programwas also high, particularly in grade 1. Immediateprogram impacts were: reduced physical aggressiontoward other students in the school ground,reductions in negative parent-child interactions, andimproved teacher evaluations of social interactionbetween the children. Effects on behaviour wereparticularly pronounced for the children who weremost aggressive at baseline. For the fifth graders,the three year outcomes were: lower rates ofdelinquent peer involvement, lower arrests and lessinitiation of alcohol and marijuana use. For the firstgrade students, there were large reductions (effectsize 1.5) in the growth of inattentive, impulsive andhyperactive behaviour problems. Effects wereequally large for the students who were mostaggressive at baseline. The authors proposed that theintervention success with the more at-risk studentsarose through removal of the aversive interpersonalpeer relationships that they were particularlyvulnerable to.607

Kellam and colleagues provided an earlydemonstration of the importance of children’sschoolyard relationships as a determinant ofdevelopmental outcomes. A study targetingaggressive/disruptive behaviour and readingachievement in junior primary school children wasconducted to test if early intervention onantecedents associated with later tobacco use couldimpact on the uptake of smoking.608 Five urbanareas over a range of socioeconomic levels weredefined and three or four schools, where thestudents were of similar socioeconomic and ethnicmix, were selected from within each area. Twodifferent interventions were tested: firstly, theclassroom game targeting aggressive/disruptivebehaviour (the Good Behaviour Game); andsecondly, an enriched curriculum aimed atincreasing reading achievement. The GoodBehaviour Game used some similar interventionstrategies to those subsequently adopted in the LIFTintervention. Students were rewarded, as a group,when disruptive behaviour levels were reduced,increasing the responsibility for children withoutbehaviour problems to assist those experiencingsuch problems.

Within each area, one school was assigned to theGood Behaviour Game intervention, one to thereading achievement intervention, and theremaining one or two schools acted as controls.Within each intervention school, one first gradeclassroom was randomly selected to receive the

�)(��

intervention and a second classroom served as an in-school control. Boys in the first intervention groupwere significantly less likely to start smoking thanboys in the control group. Although smoking rateswere reduced among boys in the readingachievement intervention, compared with thecontrols they were not significantly lower. Nosignificant differences in later smoking behaviourwere found for girls for either intervention.

A further example of a successful organisationprogram aimed to turn schools into caringcommunities of learners.609 Environments werefostered of supportive social relationships, a sense ofcommon purpose, and commitment to pro-socialvalues and responsiveness to children’s socioculturalneeds. The program was implemented in junior andmiddle schools from six school districts in the US.Similar schools from the same districts served as acomparison group. Data on problem behavioursover a three year period indicated the program wasassociated with significant reductions in studentdrug use and delinquency. Effects were strongest inschools that had the greatest fidelity ofimplementation.

The Fast Track program’ —enhancing primaryschool education for disadvantaged groups

Fast Track is a multi-component preventionprogram that aims to reduce adolescent drug useproblems, delinquency, risky sexual practices,school failure, mental health disorders, earlyconduct problems and harsh and inconsistentparenting. There is both a universal and a selectiveprogram that identifies families in the kindergartenyear (age 6 years in the United States) and providesan integrated program of seven previouslyresearched and evaluated interventions, from grades1 to 10, addressing a range of risk factors for pooradolescent outcomes.

Fast Track is being implemented in the United Statesas a multi-site implementation and evaluationproject. There are four different areas, all with highcommunity crime and poverty profiles. The siteshave families with different cultural characteristicsand ethnicity profiles and the program is adapted torespond locally to these differences. There are 54participating schools, randomly assigned tointervention or control.

The universal program provides the PromotingAlternative Thinking Strategies (PATHS)intervention for child social competence and healthypeer relationships (see Greenberg et al. for theprevious PATHS program research).610 This teacher-provided classroom curriculum aims to increase

emotional regulation and social-cognitive skills thatenable positive friendships and non-aggressivebehaviours.

All children are assessed for their risk factors in thekindergarten year, at mean age 6.5 years. Across allsites, the children scoring in the highest 10% forrisk factor scores are invited to participate in theselective intervention program. The proportion ofchildren, in different sites, participating in theselective intervention varies according to the level ofrisk in each school. Some schools may only have afew children and their families in the selectiveprogram, whereas other schools may have a highproportion.

Both the universal and selective programs aresequenced across school years to match the child’sdevelopmental progression. In grade 1 in theselective program, child and parent groups areprovided for two hours a week on weekends orevenings; these groups are more frequent duringthe first school grade. Groups provide a parentingprogram (using elements from other previouslyresearched parenting programs), child social skilltraining groups,611, 612 and academic tutoring.613 Theselective program also includes a home-visitingcomponent and weekly phone contacts as anindividual support component (using family-centred practice principles to build trustingrelationships and family self-efficacy). The selectiveprogram children also have academic tutoring,using a phonetics-based program, three times aweek during school hours.

Results of the evaluation of Fast Track have beenpublished for the end of grade 1, in 1999 and,more recently, for the end of grade 3, in 2002.614–

616 The end of grade 1 results illustrate the generaldirection of the outcomes.

For the universal classroom program at the end ofgrade 1 (for 6715 children—excluding the 845high-risk children from the analysis) in interventionclassrooms, aggressive child behaviours were lowercompared with control classrooms when thecurriculum was implemented with fidelity and highdosage.614

For the selective program, there were 445 childrenin 191 classrooms in the intervention group and446 children in 210 classrooms in the controlgroup. At the end of year 1, intervention childrenhad better emotional and social coping skills, betterreading skills and language levels, and more positivepeer relationships in school. Their parents hadbetter parenting skills, with more warmth andinvolvement with their child; they also were more

�

��) �� !��"�#��$��%��

appropriate and consistent in their parenting, withless harsh discipline and more positive schoolinvolvement. Parents reported high levels ofsatisfaction with the components of the program.The effect sizes were moderate. There was somelack of clarity about whether the program reducedchild disruptive behaviour as the results with thevarious measures used were mixed. Non-biasedobservers reported significant reductions indisruptive behaviour, but overall, the parent- andteacher-reported results had no significantreductions.615

The Fast Track program has been implemented inWestern Australia in one school district, withpositive preliminary results.

Classwide Peer Tutoring Program

Academic difficulty in the early school years is a riskfactor for poor adolescent outcomes, includingsubstance abuse. The Classwide Peer TutoringProgram was developed in the US (in Kansas) toimprove early academic competence for childrenliving in low-income areas. It is an instructionalmodel, based on reciprocal peer tutoring that can beused at any grade, and has been found to havelasting outcome benefits on academic competence,at least three years later. It is designed to beincorporated flexibly into school curricula.

Students are pre-assessed on Fridays on their nextweek’s work. From Monday to Thursday, they workwith an assigned partner for 30 minutes; partnersare assigned into two teams per class. Students taketurns tutoring each other on their spelling, mathsand reading, and work with reading comprehensionquestions. Points are awarded for both tutor andtutee. At the end of each week, students areindividually tested on the week’s work and pre-tested on the next week’s work.

The program has been evaluated with largepopulation groups, including a number ofintervention and control outcome studies. Theseresults show, for example, that students whoparticipate in the program in grade 1 have bettercomprehension (in that grade) than students ingrade 2 who have not participated in the program,and the students achieve 11% higher results onnational standardised assessments three years later,in year 4. Students are also 20 to 70% more likely tostay on task and remain engaged with all lessons.The program has been used successfully with manypopulation groups, including schools with manylow-income children.617–621

Conclusions—primary school organisation andbehaviour management strategies

The existing research includes many well-controlledstudies and suggests that interventions focusing onimproving primary school environments can makean important contribution to reducing risk factorsfor drug use. In cases such as the Seattle SocialDevelopment Project and the LIFT intervention,cohorts exposed to the interventions have beenfollowed over sufficient periods to enable exposureto these interventions to be linked to reductions indrug use. For other interventions, there aredemonstrations that exposure to the interventionscan reduce risk factors that have been shown toincrease the risks of harmful drug use. Althoughinterventions to improve primary school socialenvironments appear to be promising areas forprevention investment, there has at this stage beenlittle work published in Australia.

%�" 2 �� ;�� $��

There is increasing evidence that investment inpreventive programs in childhood can contribute tothe reduction of harmful drug use. The evidence forthis claim relies upon demonstrations that theseprograms can be delivered to disadvantaged andvulnerable families, and evidence that suchinterventions make a difference in improving socialenvironments for healthy child development. Inmany cases, evaluations have demonstrated positiveimprovements over one to two years in childbehaviour problems. In an increasing number ofstudies, follow-up has been completed intoadolescence, and these studies have linked thepositive changes achieved through earlier preventiveinvestment to reductions in harmful drug use andassociated behaviour problems. There aredemonstrations that investment in the years prior toschool entry may be important for ensuring a fullerrealisation of learning potential. Efforts to reformprimary school environments in the LIFT program,the Good Behaviour Game, and the Seattle SocialDevelopment Project raise the interesting prospectthat outcomes for the most disadvantaged childrenare greatly influenced by broader support andunderstanding within the school environment.Although childhood intervention appears to holdprospects for improving developmental outcomesfor disadvantaged children, there are few Australianstudies investigating these interventions. Animportant evaluation objective should includeassessment of the relevance of both universal andtargeted childhood interventions within Indigenouscommunities.

��

��!�� '�� !!��

A growing number of evaluations have examined outcomes of interventions during thesecondary school years. Of the interventions targeting the secondary school age period, schooldrug education has been the most commonly evaluated strategy. The evidence suggests thatshort-term reduction in both drug use and progression to frequent drug use may be achievablethrough this strategy, but the prospects for longer-term behaviour change are still unclear. Agrowing number of evaluations have examined the impact of community mobilisation strategies.There is evidence that programs utilising this strategy can influence a range of youth drug usebehaviours; however, impacts are not always superior to those achieved through the simplerstrategy of drug education. There are promising early indications that parent education strategies,delivered through the adolescent phase, may prove useful for reducing a range of drug usebehaviours. There were no published studies evaluating drug use outcomes associated withschool organisation and behaviour management strategies in secondary schools. The success ofthese programs in primary schools, together with early evidence for their feasibility in secondaryschools, supports the importance of evaluating their application in secondary schools. Healthservice reorientation programs are being trialled and evaluated for impacts on youth drug use. Inoverview, programs appear more successful where they maintain intervention activities over anumber of years and incorporate more than one intervention strategy. There are many potentialintervention areas where evaluations have not yet been reported. Conversely, there areinterventions such as community drug education and peer education that appear relativelycommon in Australia yet are rarely evaluated, despite evidence in some applications that theyhave the potential to exacerbate problems. In a small number of studies, outcomes relevant toillicit drug use have been evaluated. More intensive strategies, including family intervention andpreventive case management, are amongst the more promising strategies for preventing harmsassociated with illicit drug use amongst adolescents with a high number of risk factors.

�

�� !��"�#��$��%��

�� !�� *�� ?�� 9�� !��%��

��

�� 7 7 7 �� !��

'��$�� $ ��%'�� '��

� ��

�� !��

� � ��

D��

� ��

) �� 7 7 ��

��$�� 7 7 �� 7� � ��

� � �� 7 7 7 ��

��

�� $ ��

+ ��

�� 7 7 7 7

'�� 7 )�� 6��

*�� 7 ��

��

>��7 *��

� �� $��

� #��

� ��

��

��

'��1� ��

;'.'&��1��) �

For the current document, adolescence is defined asbeginning at the age when Australian children startthe transition from primary school into secondaryschool and continuing until the establishment ofeconomic independence. The gradually increasingindependence and mobility through adolescenceintroduce a greater range of social influences intothe young person’s life. Social changes, includingtechnological advances and free marketcompetition, mean that many adolescents face ahigher educational threshold in order to enteremployment. These changes underlie the currentsocial trend for adolescents to spend more years ineducation and to delay into the late twentiestransition from living with parents to independentliving.622

;'.'.�� )��

Definition: One or more parents receivinginformation and/or a course of instruction aimed atencouraging healthy family development. Deliverystrategies include targeted, universal and combinedinterventions.

Parent education may range in intensity from thedistribution of one-off messages using socialmarketing strategies, to self-help books or self-completed computer programs623 through tosequenced curriculum packages that may involveprofessional contact over multiple sessions.

Universal parent education approaches

Typically, between 10% and 50% of families can beencouraged to enrol in formal sequenced parenttraining programs when invitations are extended toall parents within a defined population (universalinterventions). In their work in the rural US, Spothet al. reported that parents who participated in parenteducation tended to have higher levels ofeducation,624 however in the Australian evaluationof Parenting Adolescents a Creative Experience(PACE), disadvantaged and sole parent families werealso successfully recruited.485

Secondary school parent education groups

An evaluation of the Australian PACE programtargeted parents of early- adolescents.485 Designed asa universal intervention, facilitated groups, based onan adult learning model, utilised a curriculum thatincluded adolescent communication, conflictresolution and adolescent development.625 Oneevaluation investigated the impact of seven week

PACE groups on a national sample of 3000 parentsand year 8 adolescents. The evaluation included pre-and post-intervention surveys separated by threemonths for 577 families (parents and adolescents),representing a 60% response for those sampledfrom 14 schools targeted for intervention and 14matched control schools. Although only around10% of parents were successfully recruited intoPACE groups, pre- and post-intervention findingsdemonstrated that benefits extended more broadlyacross families in the schools where PACE wasoffered. At the 12 week follow-up, parents andadolescents reported a reduction in family conflict.Adolescents reported increased maternal care, lessdelinquency, and less poly-drug use (the odds oftransition to poly-drug use were halved).

Analysis suggested that intervention effects mighthave extended to youth with a high number of riskfactors for drug use problems. The evaluationdemonstrated that the parents recruited into theintervention were more frequently sole parents; andtheir children reported higher rates of familyconflict and poly-drug use. At the post-test, familyconflict and youth poly-drug use had reducedmarkedly in these families. Evaluation suggestedthat the drug use of respondents was influenced bytheir best friend’s drug use. Improvements introubled family relationships appeared to affect awider group of families, not directly participating inthe PACE groups, through changes in peer-friendship networks and through the program’sefforts to encourage parents to assist other parentsin their school community.485 The fact that theintervention and control groups were not randomlyassigned, and the lack of long-term follow-up,suggest the need for caution in interpreting thesepromising early results.

Evidence that peer attachments may be risk factorsfor youth drug abuse has led to interventions toassist parents to better manage their children’s peerrelationships. Cohen and Rice evaluated anintervention that attempted to facilitate thisadjustment. The intervention failed to producechanges in adolescent initiation of tobacco oralcohol use. Parent participation was poor, and,even among those who participated, attempting toinfluence their child’s choice of peer group was notconsidered a practical target.626 Interventions forfamilies with adolescents must be carefully designedas there are many tensions between issues such asyouth requirements for autonomy and increasingfamily cohesion.

�

�� !��"�#��$��%��

Integrating multi-level parent education withinschools

Several research teams are currently active indeveloping multi-level family support programs fordelivery within late primary or early secondaryschool. For example, Dishion and Kavanagh report aprogram involving the integration of three levels ofsupport within school (early in US middle schoolwhen children are around 11 years old).627 At theuniversal level, all parents are invited to an in-school meeting, and written information and videoscovering key parenting skills (cooperation in thehome, supervision, problem solving andcommunication) are distributed. At the next level, afour hour ‘family check-up’ offers a familyassessment and motivational interviewing toencourage accurate appraisal of child risk behaviourand the use of appropriate parenting resources. Forfamilies where problems are evident, moreextended parent training is offered. The program isdemonstrating some evidence for positive impacts,but longer-term outcomes are unclear.

Work is underway in Queensland schools toevaluate an adolescent version of the Triple-PPositive Parenting Program. Toumbourou andGregg et al. are currently investigating the impact ofan integrated multi-level secondary schoolintervention, Resilient Families, which incorporatescommunication training for students, aninformation night for parents, sequenced parenteducation groups, and brief family therapy. Theproject aims to further explore the assumption thatcommunity-based interventions can generatebenefits beyond the minority of participants directlyexposed to the intervention.628

Tobacco

Summary: Warrants further research ...............� 4/5

A number of programs have been developed thoughthe Oregon Social Learning Centre to assist familiesmanaging child behaviour problems. Using sociallearning principles within an early-interventionframework, the aforementioned Tom Dishion andcolleagues evaluated a 12 week parenting skillsprogram aimed at families in which adolescents hadexhibited behavioural problems in late childhood.Externalising behaviour problems were measuredusing both videotaped interactions and mother-reports on the Child Behaviour Check-list. Exposureto a parent group component reduced youthinitiation to tobacco use one year later. Reducedparent-adolescent conflict was also associated withthese positive changes. In alternative interventionconditions involving adolescent groups or

combining adolescent and parent groups, youthtobacco use and other problem behavioursincreased. The authors argued the positive benefitsof reduced parent-adolescent conflict werechallenged in these cases by contrary peer influencepressures. Findings suggested the importance ofstrengthening the parental sub-system in familyintervention through the adolescent phase.636

One of the issues to be addressed in parenteducation is the problem of engaging parents. In arecent US project addressing youth tobacco use,households were screened by phone to identifyfamilies with children in the age range 12 to 14years. Of the 2400 families identified, 55% ofparents and adolescents participated in a baselinephone survey and then half the parents wererandomly assigned to participate in a program calledFamily Matters. The program involved mailing asequence of four booklets to parents, withdiscussion with a phone counsellor after eachmailing. Early booklets focused on motivatingparents to make changes, and later bookletsprovided instruction on implementing familychanges to improve communication and preventyouth substance use. Parents and adolescents werere-interviewed at the completion of the three monthprogram, and again 12 months later. Relative to thecontrols, smoking onset was reduced by 16.4% forthe youth in the families exposed to theintervention. The program had no impact on theinitiation of alcohol use.629

Parent education approaches have been developedand evaluated as a component within multi-levelprograms aimed at addressing broader communityfactors that can undermine healthy youthdevelopment. Such programs typically incorporatecommunity mobilisation activities as an adjunct toschool-based health education; parentalinvolvement is integrated in this context.580,593 It isoften difficult to separate out the contribution ofparent education in these programs. Pentz andcolleagues reported a program that combined parenttraining in adolescent communication with schooldrug education and community mobilisation,including a parent organisation program forreviewing school prevention policy and skills.Available evidence suggested that 73% of parentsparticipated in one or more of the programcomponents.630 The program may have had someimpact in reducing escalation (recent use in 30days) in tobacco and marijuana use, but did notappear to influence alcohol use. In later analyses,youth using cigarettes at entry to the study wereshown to reduce their use compared to similarstudents in the control communities.631

��"��

Although the Australian evaluation of the PACEprogram did not specifically target tobacco use,there was some evidence that tobacco use wasreduced where family relationships improvedwithin peer networks.485 The intervention reportedby Cohen and Rice failed to impact tobacco use.626

Available research suggests that parent educationtargeting the adolescent phase may be promising asa strategy for preventing adolescent tobacco use. Asadolescent tobacco use is predicted by family-levelrisk factors, and there are some promising findings,there also exists a strong theoretical ground toassume that parent education may prove to be aneffective strategy in reducing adolescent tobaccouse. Further research in this area would appearwarranted.

Alcohol


Project Northland provides a further example of acommunity mobilisation program that, in this case,aimed at reducing youth alcohol use byincorporating parental involvement as onecomponent within a wider set of school andcommunity activities. A delayed entry to youthalcohol use was associated with changes to locallaws and ordinances controlling alcohol sales tominors, improved family communication relating toalcohol use, and reductions in the perception thatyoung people drank alcohol.579 In the evaluation ofthe Midwest Prevention Program, there was noinitial impact on recent youth alcohol use (past 30days);630 however, the program did appear toreduce progression to higher levels of alcohol usefor youth who were alcohol users at the initiation ofthe program.631

The PACE evaluation reported that getting drunk onalcohol was reduced where peers had better familyrelationships.485 The interventions reported byCohen and Rice and Bauman and colleagues failedto impact the initiation of alcohol use.626, 629

Research is insufficient to enable any conclusion tobe made regarding the overall effectiveness ofparent education for preventing adolescent alcoholuse. As there are promising early indicators, furtherresearch in this area is recommended.

Cannabis


There have been no published evaluationsexamining the application of parent trainingprograms to specifically address adolescent cannabisuse. In the evaluation of the Midwest Prevention

Program, parent involvement was included as onecomponent within a wider set of communitymobilisation activities and there appeared to besome evidence for an impact on recent cannabis use(past 30 days).630 Subsequent analyses suggested theprogram also influenced youth who were cannabisusers at the initiation of the program.631 The PACEevaluation revealed that cannabis use was reducedwhere peers had better family relationships.485

Rates of youth cannabis use have been high throughthe mid-1990s in Australia and current evidencesuggests that family-level risk factors influenceyouth cannabis use. Given promising indications inthe available evaluations, there would appear to bemerit in developing and evaluating parent educationprograms for their impact on youth cannabis use.

Other illicit drug use


There is evidence that parent education may be apotentially useful strategy to assist families facing ahigh number of risk factors for harmful youth druguse. Evaluations of work conducted through theOregon Social Learning Centre suggest that parenteducation using behavioural and social learningprinciples can be useful in preventing furtherescalation of problems related to illicit drug use.

Targeting programs to parents in juvenile justicesettings

Much of the research examining parent interventionfocuses on efforts to prevent escalation orpersistence in problem behaviours. Intensiveinterventions based on behavioural, social learningprinciples, and involving 44 hours of behaviouralparent training, have been demonstrated to reduceoffending and incarceration compared with standardjuvenile justice contact.632 It is likely that success inreducing crime and antisocial behaviour would alsotranslate to reductions in harmful drug use.

Assisting parents concerned by youth illicit drugabuse

An Australian study investigated the interventionopportunity that can arise when parents initiallyrecognise adolescent drug abuse. Parents in thesesituations often experience considerable distress,which can undermine effective responding. In aneffort to provide a cost-effective method ofassistance, Blyth et al. developed an eight week,professionally-led, group intervention known as theBehavioural Exchange Systems Training (BEST)program.633 High rates of depression amongparticipating parents at pre-test (87% with high

�

��# �� !��"�#��$��%��

symptoms on the General Health Questionnaire)were observed to drop substantially over the courseof the intervention (down to 24% after eightweeks).634 A small evaluation incorporating awaitlist control group revealed differentialimprovements for those exposed to theintervention: in mental health, parental satisfaction,and assertive parenting behaviours.634 The impact ofthese changes on youth substance abuse is not yetknown.

In a recent analysis of parent changes achievedthrough participation in the BEST program,Bamberg et al. noted that further reductions in youthdrug abuse might have been achievable forapproximately one-third of families had additionalfamily intervention been provided at the end of theprogram.635 Future research should investigate thepotential to curb youth substance abuse by addingbehavioural training components as a follow-onintervention for parents who have completed theBEST program.

Including early intervention in the context ofuniversal parent education.

The PACE evaluation revealed that parents weremore likely to participate in the PACE groups if theywere sole parents or their children were poly-drugusers. The program appeared to positively affectearly adolescent poly-drug use.485

;'.'0� � �+)-�� -?�� )�� -�� )-��+��

As reported in the previous chapter, evaluationresearch in the US demonstrated that the Preparingfor the Drug Free Years curriculum appearedpromising as a strategy for delaying the initiation ofalcohol use when delivered to families with lateprimary school-age children. In this section,evidence was presented revealing that a similarprogram strategy, using the Australian PACEprogram, also showed promise as a method forreducing early youth drug use when deliveredwithin secondary schools. The Victorian StateGovernment is currently investigating thedevelopment and application of a program strategyrelated to that adopted in the PACE evaluation(called ABCD).

Although the model of sequenced parent groups isemerging as a useful program approach for parenteducation, there are alternative models, such asphone counselling, that are being trialled withpositive results. Future investment shouldencourage further innovation in service deliverymodels. Investment in at least one large randomised

controlled trial with long-term follow-up wouldalso appear warranted to ensure that the promisingearly indications for the Australian PACE programcan be replicated and sustained to achieve longer-term reductions in youth drug use. Funding forsuch an investment might be organised as apartnership between Australian Government andState agencies, targeting substance abuse prevention,crime prevention and mental health promotion.

More intensive parent education has also beeninvestigated for delivery in settings such as juvenilejustice and youth drug abuse treatment. In theseapplications, the approach appears feasible althoughfurther research will be required to establishwhether this strategy can prevent harmful drug use.

;'.'(��+,��

Definition: One or more parents, adolescents andother family members receiving information, acourse of instruction or therapy together, aimed atencouraging healthy family development.

Family intervention programs have been based on avariety of theoretical frameworks including familysystems theories and social learning approaches. Intheir application in families with one or moreadolescents, parents and offspring participate in theprograms together; the balance of program timedevoted to such components, relative to activitiesthat provide instruction to adolescents and parentsseparately, varies.

Tobacco

Summary: Limited investigation ........................... O

There is evidence that family factors influenceadolescent tobacco use. In the trial reported byDishion,636 family intervention, involving theparents and the child, was less effective at reducingtobacco use in high-risk youth than parenteducation alone. The components providedseparately to the at-risk adolescents were consideredto exacerbate antisocial peer involvement andthereby undermined the effective parent educationcomponents.

Alcohol


The evaluation of the delivery of the IowaStrengthening Family Program (ISFP) in lateprimary school suggested the program wassuccessful in reducing alcohol use and related riskfactors.581 The development and evaluation of asimilar strategy in early secondary school may beconsidered. However, it should be noted that early

��%��

secondary school is a period when there are highrisks for affiliation with drug using peers, and hencedangers to be managed in implementing the peercomponents entailed in the ISFP.

Cannabis


There have been no evaluations of the delivery offamily intervention programs in secondary school toreduce youth cannabis use. It should be noted thatearly secondary school is a period when there arehigh risks for affiliation with drug using peers, andhence dangers to be managed in implementing peercomponents of current family interventionprograms. Family therapy models are used inAustralia to assist families facing difficulties withyouth cannabis use, but there has been littleevaluation.

Illicit drug use

Summary: Evidence for implementation ......... � 4/4

A family intervention program with some elementsof preventive casework, the Targeted Adolescent/Family Multi systems Intervention (TAFMI), used anindividual-focused intervention to assist studentsand involve parents. The program was targeted tolate primary/early secondary school studentsevidencing poor school performance and drugmisuse or abuse. A therapist held weekly meetingswith the students and their families. Early meetingsfocused on setting behaviour and academicimprovement goals with the student. The studentand family were then assisted toward achievingthese objectives using evidence-based strategies.Behaviour changes for those exposed to theintervention were compared against a randomlyassigned control group who received usual schoolcounselling. Differences between the groups werenot observed until the second year, at which timedrug use was lower for the intervention group.637

;'.'6�� )�� -��+,��+�� -��-��,�)�� )�� >)�� +��>)-��-�� -

Family interventions targeting multiple risk familiesare typically conducted in settings such as drugtreatment, juvenile justice and school welfare.Family therapy is often approached reluctantly bygovernment, due to perceptions of undefinedlength and expense. However, considerable workhas been done to better quantify the investmentrequired. Several research teams have presentedevidence supporting the effectiveness of manualisedforms of family therapy in the treatment of youth

drug abuse.638,639 Santisteban et al. described activerecruitment methods that can be successfullyemployed in interventions aiming to attractdisadvantaged families or youth with specificproblems such as drug abuse.640

The Addicts and Families Project was historicallyimportant in its use of a well-controlled evaluationto demonstrate the effectiveness of family therapy asa treatment for youth drug abuse.641 The studydemonstrated that, after six months, supplementingmethadone treatment with family therapy reduceddrug use in around two-thirds of cases.

Functional Family Therapy (FFT) has providedleadership in the movement to disseminate brieffamily therapy using a clearly-staged (readilytaught) family counselling program, involving aslittle as eight hours of therapist contact. Evaluationsof FFT have demonstrated reductions in juvenilejustice expenditure.642, 643 The program hasdemonstrated evidence as a strategy for reducing re-offending amongst voluntary and court-mandatedadolescent offenders, and has also beendemonstrated to prevent offending amongst theyounger siblings of targeted offenders. In theirreview, Bry, Catalano et al. found reductions inrecidivism ranged from between 75% for less severeoffenders down to 35% for severe offenders.637 Aoset al. reported the net economic benefit at around $4for each $1 invested in FFT.644 Process evaluationhas suggested a critical program component mayinvolve reframing problem attribution away fromindividual blame, to focus on the concept of amismatch in family needs.645 If this is accepted, itsuggests that a critical transition in the escalationfrom delinquency to serious crime may be linkedwith hostility generated in family relationshipsthrough adolescence. Although FFT has beendisseminated widely in the US there has been onlyone brief training in Australia.

Although important, the family is only one of thesocial systems influencing adolescent development.Henggeler and colleagues have used randomisedcontrolled trials to evaluate the impact of theirMultisystemic Treatment (MST) program on seriousjuvenile offenders.646–648 MST has a large componentof preventive case management but is based onfamily systems principles and extends assistance byincluding effective intervention strategies toenhance individual competence, tackle peerrelationship issues, and to ensure access to work,education and community resources.

Exposure to MST reduced offending and re-arrestsrelative to usual juvenile justice practices.648 MSTappeared more effective than individual counselling

�

��' �� !��"�#��$��%��

in reducing antisocial behaviour for seriousadolescent offenders.646 MST begins with assistancefor the young person and their family, to establishtreatment goals and objectives. Clinicians are trainedin the application of evidence-based therapies andrewarded for achievement of agreed clientoutcomes. MST has been effective in engagingfamilies with multiple and complex problems.649

The net economic benefit has been estimated ataround $5 for each $1 invested in MST.644 MSTtraining has been disseminated widely in the US anda project has recently been established in NewZealand.

;'.'7� � �+)-�� -?��+,��

Investment in both brief family therapy and theMultisystemic Treatment (MST) model appearwarranted for juvenile justice and youth drug abusesettings. Since programs of this type have a trackrecord of dissemination in applications in the US,investment in these programs may be justifiable inAustralia without awaiting long-term follow-upresearch. The skills required to engage youth andfamilies in these interventions do not appear to bewell developed in Australia. For this reason,investment in overseas trainers may be warranted inthe context of at least one demonstration projectfocusing on process evaluation and training, toensure the achievement of critical programcomponents. Juvenile justice programs would belogical partners in this type of investment.

Investment in at least one randomised controlledtrial, including three to four years of follow-up,would appear warranted. Investment of this typewould be appropriate for research funding fromagencies such as the National Health and MedicalResearch Council, the Criminology ResearchCouncil and relevant State health promotionfoundations. Alcohol and drug sector funds couldseed research support by providing money fortraining and dissemination.

'�� 3��

Definition: Efforts to reduce drug-related harmthrough the delivery of a structured social healtheducation curriculum within the school, usually byclassroom teachers, but in some cases by visitingoutside professionals.

There have been two recent and comprehensivecritical literature reviews on drug education inschools, conducted in Australia for governmentdepartments.650, 651 These describe the evolution ofapproaches to drug education in schools since the

1930s to the present day. The reviews also presentthe factors that have been shown in experimentaltrials to maximise the effectiveness of programs inpreventing or delaying the onset of drug use and inreducing use. Other reviews in recent times presentsimilar conclusions.586, 652–654 The more successfulapproaches to drug education have a grounding inwhat is known about the causes of adolescent druguse, adolescent developmental pathways in relationto drug use, and the psychological theoreticalframeworks of social learning and problembehaviour.655 Because this body of evidence hasbeen well-established over several decades ofresearch, the authors sensibly caution thoseconsidering developing drug education programs tobase them on what is known rather than whatseems intuitive or ideologically sound. Poorlyconceptualised programs have historically beenineffective or, at worst, actually harmful, forexample by increasing drug use.650 Factorsassociated with effective substance use programs inschools are as follows.520, 656–659

Programs should:

� be research-based/theory driven,

� deliver coherent and consistent messages,

� present developmentally appropriate, balancedinformation,

� provide resistance skills training,

� incorporate normative education,

� educate before behavioural patterns areestablished,

� relate strategies to objectives,

� address values, attitudes and behaviours of theindividual and community,

� address the inter-relationship betweenindividuals, social context and drug use,

� focus on prevalent and harmful drug use,

� make judicious use of peer leadership,

� be delivered within an overall framework ofharm minimisation,

� incorporate broader social skills training and bepart of a comprehensive health educationcurriculum,

� employ interactive teaching approaches,

� ensure optimal training and support forteachers,

��(��

� provide adequate initial coverage and continuedfollow-up in booster sessions,

� be sensitive to cultural characteristics of thetarget audience,

� incorporate additional family, community,media and special population components,

� ensure fidelity of implementation, and

� be evaluated.

The explanation of these factors and the papers fromwhich they derive are presented in reviews byMidford, Lenton and colleagues.650,651 The authorsof both Australian reviews have also summarised thework of Tobler et al. into a useful table of ‘whatworks and what doesn’t’ in cannabis and other drugprevention programs in terms of content anddelivery features.660 Tobler and colleagues derivedthis information from a meta-analysis of 120school-based adolescent drug preventionprograms.661

Readers should refer to either review for the fulldiscussion and references of successful drugeducation approaches.650,651 The main points arereiterated here. Successful drug education programsuse the social influence approach, or multiplecomponent programs, with a large emphasis on thesocial influences rather than information-basedapproaches alone or those targeting affectiveeducation alone. Affective education approacheswere based on the assumption that youth who usedsubstances had personal deficiencies; by enhancingpersonal development with training in self-esteem,decision making, values clarification, goal settingand stress management, the use of drugs woulddecrease. These programs did not succeed inconsistently changing behaviour, perhaps becausenot all youth using substances suffer from personaldeficiencies. Indeed, some research has suggestedthat young people who engage in minor drugexperimentation may be better adjusted than thosewho maintain complete abstinence, while frequent/heavy drug users tend to be poorly adjusted.

The social influence approach was developed fromBandura’s social modelling theory662 and McGuire’ssocial inoculation/resistance training.663 Initialformulations of this approach were based on thebelief that young people begin to use substancesbecause of social pressures from a variety ofsources, including mass media, peers, family andthe image they have of themselves. To successfullyresist the adoption of undesirable behaviour, youngpeople need to be inoculated by prior exposure tocounter arguments and have the opportunity to

practise the desired coping behaviour. Otherformulations are more complex and consider acompetition for behavioural identification betweendifferent social influences such as peers andparents.479, 515 Peer subcultures are increasinglyassociated with new illicit drug use trends, whileparents may either attempt to limit drug use orintroduce adolescents to their familial patterns ofdrug use. In the context of complex and contrarysocial influence pressures, it is not surprising thatthere has been some variation in outcomes for socialinfluence programs across different social contexts.

Despite the challenges, variants of the socialinfluence approach have been shown to havebenefits in reducing antisocial behaviour, affiliationwith deviant peers and school behaviour problems;and increasing academic performance andcommitment to schooling. Booster sessions added atcritical points of developmental transition, acomplementing parenting component, andreinforcement of social messages at the broadercommunity level seem to strengthen the effects ofsocial influence school-based programs.14

;'0'&�%��-1�++-��

An historically important social influence approachby Botvin and colleagues is Life Skills Training(LST).664 This strategy brought an emphasis on thepersonal and social skills that underpin lifestyle andhealth risk behaviour more generally. In addition toaddressing tobacco advertising and social resistanceskills, the program deals with managing anxiety,communicating effectively, developing personalrelationships and asserting one’s individual rights.The program is more elaborate than earlier healtheducational packages, with teacher manuals,ongoing professional development of teachersduring implementation, student guides and arelaxation audiotape. Various formats have beenevaluated but a common delivery strategy consistsof 15 classes in year 7, with 10 booster sessions inyear 8 and five in year 9.

The LST program has been evaluated in ten separatestudies and seems to have reduced alcohol, cannabisand tobacco use into early adulthood.665 It wasevaluated in (a sample of close to 6000) year 7students allocated to three treatment conditions(two active and one control). Six years later, in1991, follow-up data were collected from 3597students (60% of the original sample of 5954).Significant reductions were found in the prevalenceof smoking (22% versus 33%) and drinking to thepoint of being drunk (34% versus 40%). Rates ofweekly poly-drug use were halved in the

�

��) �� !��"�#��$��%��

intervention sample (3% versus 6%). Reductions inthe group who received at least 60% of theintervention program were even more substantial.652

In a recent update study, students were completingsecondary school in grade 12. Self-report datacollected by mail from 447 of these individualsshowed that those who received life skills trainingreported less illicit drug use compared to the controlgroup.666 These findings hold promise that targetinggateway substances such as alcohol and tobacco canprevent illicit drug use and that long-termprevention effects may be possible. The limitationsof this study, however, are the small sample sizeand the predominance of white respondents (92%),along with the possibility that the sample was notrepresentative of high-risk individuals. Although theLST approach appears promising, replication effortscan face difficulties. Gorman has critiqued theevaluations of LST approaches and noted that whilesome evaluation studies produced significanteducation effects, most indicated no change due tothe intervention.667 In addition, those reportingsignificant change had conducted manycomparisons such that some changes may haveoccurred by chance. In some cases, analysesreported only on efficacy, regrouping interventiondata to focus on successful implementations. Somestudies had small numbers and collapsed variablesinto dichotomised scales that may have influencedsignificance. The LST programs may require 30sessions throughout secondary school and thissubstantial time commitment may be difficult forsome schools to accommodate.

;'0'.��)��-��

Use of peer leaders has been another strategyreported in the literature that was reviewed byMidford, Lenton and Handcock.650 Evidence showsthat peer educators can be important when theymodel attitudes unfavourable to drug use. Peerleaders need to be selected carefully and wellsupported with management skills fromprofessional teachers. Peer educators need to becredible with high-risk young people, have goodcommunication skills and show responsibleattitudes, but simultaneously be unconventional.Hence, the ‘good’ students selected by adults maybe inappropriate if peers don’t engage with them,while in poorly managed programs there is the riskthat ‘cool’ students will encourage favourableattitudes to drug use. In the meta-analysis by Toblerand Stratton, peer education produced similarresults to other interactive programs presented byteachers and other leaders.661 It may, therefore, be

the interactive programs which engage students inrole-play, discussion and games that are a key factorin effectiveness.

;'0'0��)��

Hansen and Graham ran a study where studentsreceived one of four curricula: information alone,information plus resistance skills training,information plus normative beliefs, or, all threecomponents.498 Students exposed to the programswhere normative beliefs were included significantlydecreased alcohol use after one year compared tothose in the alternative conditions. Students oftenover-estimated the proportion of their age groupwho used substances, and student surveys onsubstance use and feedback encouraged morerealistic assessments. In their conclusions, Hansenand Graham argued that normative componentsmay play a critical role in activating students toutilise peer resistance strategies.498 In the absence ofa normative component, resistance trainingappeared relatively ineffective.

;'0'(�� -�,��

Timing of education is critical according to someresearchers. The consensus is that the optimal timefor introducing youth preventive programs is lateprimary school or early secondary school, whenexperimentation often begins. This may also help tocapture higher-risk individuals who may leaveschool early. Junior programs should be generic asthe most effective programs for reducing cannabisuse at this stage are also effective in reducingtobacco and alcohol use.650 However, onset of usecan vary in different populations and with differenttypes of drugs so timing of programs should beadjusted according to prevalence data for use of aparticular drug. For instance, findings fromcannabis education programs in senior secondaryschool suggest that a more differentiated approachis required with older students.

Based on evidence over the past decade, it might becautiously argued that health education curriculainitiated in early secondary school (year 7) andreinforced with additional programs provide asound delivery strategy.593

The intensity of the intervention may also beimportant in its success. It appears that most of thesuccessful programs are intensive and long-term(including booster sessions). Of the 10 soundlyevaluated effective programs reviewed by Whiteand Pitts,654 eight had 10 or more sessions devotedto delivery of the program and booster sessions.However, intensity did not guarantee effectiveness,

��

an example being the DARE program that wasintense in number of sessions but ineffective.

;'0'6��+�,��+�� --��

Hansen and McNeal conducted a study into howdrug education programs are implemented byteachers when there is not a specific managementprogram in place.668 This was in response to data, in1996, that teenage drug problems were risingdespite significant advances in best practice of drugeducation. They observed teaching in 12 middleschools in Forsyth County, US, amounting to 2828minutes of instruction, equivalent to 232 distinctclass periods of which 146 were reported. Theyfound that the teachers observed lacked a generalunderstanding about the concepts underpinningexisting drug prevention approaches. They tended,more often, to focus on the knowledge aspects ofthe program, in particular those about health-related consequences of drug use, while payingminimal attention to resistance skills training, valueclarification and normative education. In addition,teachers tended to emphasise alcohol, tobacco andless so cannabis in their lessons. Cocaine wasdiscussed 20% of the time, despite having lowusage rates among students, and amphetamines andinhalants discussed rarely. The teachers observedwere very individual in their teaching approaches,the substances they focused on and the life skillsthey included. The authors concluded that if thesefindings were typical of schools elsewhere, drugeducation would fail to make a long-term impact onsubstance use behaviours. They recommended thatthere should be a focus on increasing teachers’conceptual understanding of drug use andprevention, and of normal patterns of drug useonset and experimentation. They also recommendedincreasing training for building allegiance toresearch-based prevention strategies, and foradoption of the program components that havebeen shown to mediate impacts on drug use. Itseems that teachers need support and access to goodmaterials to work with, and that this support from aschool organisational level would be important.

;'0'7��-)-�) ��-�+� �� -

Most of the studies evaluating drug education haveemanated from the US, where the focus is onuniversal strategies to prevent or delay the onset ofdrug use. There have been too few interventionstargeting young people at different stages of theirdrug use and from different social and culturalbackgrounds. However, some studies have

investigated the effectiveness of interventions fordifferent populations and tailored for differenttarget groups.654 In contrast to the many studies ofdrug use in people of school age, there is littleabout progression of drug use into youngadulthood and factors associated with vulnerabilityand resistance in this time period.654 Suchknowledge of the ‘where, when and why’ drugs arebeing used, and the meaning drug use has to theuser at different stages, will be necessary forformulation of targeted interventions.

;'0'3�#��+��)�� -�� *��) �,��*�+�-��

Mary Ann Pentz and colleagues examined the effectsof a school drug education program run in thecontext of the Midwest Prevention Program, acomprehensive community mobilisationprogram.580, 593 First, a 10 session health educationprogram focusing on drug use resistance skills wasdelivered to grade 6 and 7 students. Evaluationsuggested positive program impacts on mediatingfactors (attitudes, knowledge, skills and peerinfluence) and on initiation and escalation in use oftobacco, alcohol and marijuana use after the firstyear. A three year follow-up combined the drugeducation program with a parent organisationprogram for reviewing school prevention policyand training parents in positive parent-childcommunication skills in the context of communitymobilisation elements.593 The program appeared tobe effective at preventing escalation in tobacco andmarijuana use, but not alcohol use. Effects weremost prominent when delivery occurred in year 7.

Research reported by Perry and colleagues providesfurther insight into the effectiveness of healtheducation delivered within the context of widercommunity mobilisation efforts. It would appearfrom the Project Northland research that thecommon observation that educational impacts decayover time also applies to interventions run in thecontext of wider community mobilisation.669 Acohort exposed to a social influence healtheducation curriculum from year 6 through to year 9demonstrated lower rates of recent alcohol use andalcohol misuse. These effects tended to decline inthe years following the intervention such that therewere few significant effects by year 12.

Tobacco


Drug education addressing a single drug may bedifficult to implement given the competitivepressures within schools for a range of material to

�

�� !��"�#��$��%��

be integrated into the curriculum. There is anargument, therefore, that programs should begeneric; the most effective programs for reducingtobacco use at this stage may be effective inreducing cannabis and alcohol use.650

An early popular example of the application of asocial learning approach to tobacco health educationwas the Minnesota smoking prevention program.670

The program was delivered by teachers andincorporated a structured series of lessons anddevelopment of a teaching strategy that allowedpupil-led discussion. The program contentaddressed cigarette advertising, the physical effectsof smoking and refusal strategies. US findings werepositive in bringing short-term reductions insmoking but translation of the program to othersettings has not always met with a similar degree ofsuccess.671 An Australian adaptation of this program,with both teacher and peer-led components, wasimplemented to over 2000 year 7 students inWestern Australia, in 1981. At seven year follow-up, no persisting effect of the program was foundin males but in females there was a 50% lower rateof tobacco use in those who had been non-smokersat the outset of the program. This study was marredby a low response rate at follow-up (55%) butadjusting for differential attrition of smokers didsuggest an ongoing effect in females.

Two notable European studies have also producedlonger-term follow-up findings on the use ofinterventions based on social learning theory. TheOslo Youth Study was based on the Know YourBody risk factor assessment program. The initialstudy took place with 827 students in six Osloschools. The program was in part peer-led andfocused on developing social resistance skills,making a public commitment to being a non-smoker, and on broad discussions of the social,political and health aspects of smoking. Theprogram had a significant effect on smoking inmales at 12 year follow-up but not in females. It isnoteworthy that daily smoking rates in thecomparison populations at follow-up were veryhigh (close to 50%). In general, the size of all theshort-term health effects in this programdiminished over time.599

The North Karelia Youth Project focused on socialresistance skills in 10 classroom sessions, withassistance from peer leaders. Effects on smokingrates were clearest at follow-up at two years, where23% of participants were smokers compared with38% in controls. At eight year follow-up, rates were37% and 47% respectively.672

There have now been a number of systematicreviews conducted on school-based programsdesigned to prevent smoking in young people. Thefindings from several of these reviews weresummarised, in 1999.673 A synopsis of the resultspresented in that review is given below.

One review was restricted to randomised controlledtrials (RCTs) that targeted individuals up to 18 yearsof age; the findings are based on a qualitativesynthesis of this literature.674 Over 60 primarystudies were identified but only 11 were consideredto be of high quality. The 11 studies all containedcomprehensive programs that provided informationon smoking and its consequences, and hadcomponents on decision making skills withresistance/refusal skills training. Results from thesehigh quality studies found smoking prevalence was8 to 15% lower in the intervention groupscompared with controls at 12 months follow-up.Similar trends were also found in the lower qualitystudies.

Twenty-seven studies targeting tobacco and othersubstances were included in a second qualitativereview.653 The programs were grouped into fourtypes according to content: information/knowledge, affective (decision making, self-esteem), social influences/skills development, andcomprehensive (including information, decisionmaking and resistance skills training). Theproportions of positive, negative and neutraloutcomes were reported for each program type. Thesocial influence programs had 51% positiveoutcomes, 11% negative and 38% neutral. Thecomprehensive programs had no negative outcomesand equal proportions of positive and neutraloutcomes.

There were five quantitative reviews that presentedthe findings as pooled effect sizes with change insuccess rates that could be attributed to theinterventions. Bruvold performed a meta-analysis ofthe California School-based Risk Reduction programusing eight studies that had a control or comparisongroup.675 The program was targeted at nine to 14year olds and, of the studies used in the analysis, sixused a rational, didactic approach while the othertwo studies included life skills training. Resultsshowed the life skills programs were moresuccessful in reducing smoking behaviour post-testthan the rational programs.

In another quantitative study by Bruvold, 84 school-based studies (94 programs) were categorised intofour types: rational or information giving;developmental, focusing on increasing self-esteemand developing decision making skills; social

��

norms, with presentation of alternative behaviours;and social reinforcement, teaching recognition ofsocial pressures and developing resistance skills.676

Results were presented for post-test and first andsecond follow-ups; however, no definitions wereprovided for what first and second follow-uprepresented. The results found the socialreinforcement and social norms programs producedpositive results at each point, the effectiveness of thedevelopmental programs was mixed, and therational programs were not effective.

In the Drug Abuse Resistance Education (DARE)programs developed in the United States, speciallytrained law enforcement officers teach drugprevention curricula in schools. A meta-analysis ofeight DARE programs that had either a control orcomparison group found there was a small butsignificant effect on tobacco use post-intervention.677 However, in the two studies thatcontinued to follow participants the effect was notmaintained at one and two year follow-ups.

Rooney and Murray performed a meta-analysis of90 studies of social reinforcement programstargeting smoking prevention, in 11 to 18 yearolds, that included control or comparison groups.678

Results showed smoking levels could be reduced byapproximately 5%. These results were sustainablefor up to one year when information on the short-term consequences of smoking was delivered inconjunction with information on the socialinfluences that encourage smoking and training onhow to resist the pressures to smoke.

Another meta-analysis was conducted of 90 drugprevention studies (120 programs) targeting 11 to18 year olds, which all included a control orcomparison group.661 Forty-three of the programsspecifically targeted tobacco use and werecategorised by content, depending on whether theycontained an active component or were non-participatory presentations. When all the studieswere included, participatory programs producedsignificantly better results than non-participatoryprograms; however, when analyses were restrictedto the better quality studies, although a higher effectsize was produced by participatory studies it was notsignificantly greater than that for non-participatorystudies.

A recent meta-analysis of 207 international school-based drug prevention programs, including morethan half with social influence elements, providedstrong support for the efficacy of social influenceprogramming at one year follow-up.679

A recent study on the cost-effectiveness of a school-based tobacco use prevention program, using socialinfluences curricula, found the intervention to behighly cost-effective.680 However, there are reasonsto believe that the integrity and effectiveness ofprograms may be diminished when attempts aremade to translate successful experimental programsinto more naturalistic and real world settings.681 Noexamples of broad scale successful adaptationoutside of experimental trials were identified.Murray et al. described one attempt to use legislativeand financial incentives to schools to encouragedissemination of programs of known effectivenessin prevention of tobacco use. Over a five yearperiod, no discernible effect was evident.682

Findings from the evaluations and major reviewssummarised above demonstrate small but consistenteffects for delaying the initiation of smoking, over aone to two year time frame, following well-conducted school drug education. Availableevidence suggests that over longer time frames,effects diminish. Despite this general tendency,there is some evidence of reductions in smokingmaintained to eight years following the NorthKarelia program, better outcomes for females sevenyears after the Western Australian implementationof the Minnesota program, and better outcomes formales 12 years after the Oslo Youth program.

Alcohol

Summary: Evidence for implementation .............. �

As is the case for tobacco use, there is goodevidence that school-based health educationprograms targeting alcohol use and utilising sociallearning principles can be successfully implementedwith booster sessions in subsequent years. Morerecent programs have successfully incorporatedapproaches that include peer leadership, smallgroup discussion, student-led participation,homework tasks and role-plays. Caplan andcolleagues noted positive consumer feedback fromstudents regarding their social competencecurriculum.683

Maggs and Schulenberg used hierarchical modellingto examine the success of the school-based AlcoholMisuse Prevention Study (AMPS) in alteringnormative trajectories of alcohol use, misuse,reasons to drink and reasons not to drink acrossearly to middle adolescence.684 The AMPS(University of Michigan) involved seven lessons ingrade 7 and 8 using role-play to teach specificrather than global strategies to resist pressures to useparticular drugs. A randomised pre- and post-experimental control design was used. Longitudinal

�

�� !��"�#��$��%��

data from 971 students, collected across fiveoccasions between 6th and 10th grade, provided4178 person-time cases. These data were used toevaluate the effects of the program on drinkingbehaviour, reasons not to drink and reasons todrink. Results indicated that AMPS may alter ayoung person’s trajectory of drug use, but priordrinking experience modified this effect. Amongstudents who engaged in prior unsuperviseddrinking, exposure to AMPS was associated with areduced rate of increase in alcohol misuse and areduced rate of decrease in reasons not to drinkacross adolescence. This implied an earlyintervention effect for this program. The authorssuggested that a goal for prevention research maybe, therefore, not to determine whether a particularprogram worked but rather for whom the programworked and why. It appeared that reasons to drinkincrease across adolescence and AMPS seemed toslow down this tendency.

White and Pitts reviewed Project ALERT because itcompared the effectiveness of the program onyoung people who were non-users at the time ofthe intervention with young people who hadalready experimented with drugs.654, 685 In the short-term, non-users showed more program gains thanusers. Another study examined the effectiveness ofan intervention for young people at different levelsof risk, where risk factors included prior use ofsubstances, and found the program was equallyeffective for all groups.593

In the study by Hansen and Graham, studentsexposed to the programs where normative beliefswere included significantly decreased alcohol useafter one year compared to those in the alternativeconditions.498 Other reasonably well controlled USstudies have also reported small one year impacts inreducing regular alcohol use.686

A recent Australian study, the School Health andAlcohol Harm Reduction Project (SHAHRP), hasdemonstrated main group effects in consumption,harmful/hazardous consumption, and harmassociated with young people’s own use ofalcohol.687 This study had a goal of harmminimisation and was an evidence-based classroomprogram (13 lessons) conducted over two years.155

Over the period of the study (from baseline to finalfollow-up 32 months later), students whoparticipated in the SHAHRP program had a 10%greater alcohol-related knowledge, consumed 20%less alcohol, were 19.5% less likely to drink toharmful or hazardous levels, experienced 33% lessharm associated with their own use of alcohol, and10% less harm associated with other people’s use of

alcohol, than students who participated in otheralcohol education.687 These behavioural effects weremaintained and/or increased up to one year afterthe final phase of the program.687 The key focus ofthis program was on the development of utilityknowledge and harm reduction skills and strategies.In addition, young people were involved in thedevelopment and testing of the program.155 TheSHAHRP classroom-based approach is a relativelycost-effective way of impacting on young people’salcohol-related behaviours.687

In summary, compared to the evidence for tobaccouse, the short-term impact of secondary school drugeducation has been less consistently demonstratedfor youth alcohol use and there is less evidence forlonger-term impacts. The SHAHRP evaluationshows promising early impacts; for AMPS there areeffects for selected groups over three years, and forLST there are longer-term effects in reducingregular alcohol use and poly-drug use. Whenprograms run in the context of communitymobilisation are included results are less promising.The Midwest Prevention Program was not effectiveat three year follow-up in reducing regular alcoholuse.593 Project Northland showed small effects onuse and misuse in mid-secondary school, but thesewere not maintained through to the end ofsecondary school.

Cannabis


There appears to be little evidence for theapplication of specific programs addressing cannabisuse alone. However, available evidence does suggestthat reductions in cannabis use have been achievedthrough more generic drug education approaches.Midford, Lenton and Hancock examined a subset ofdrug education programs that had an impact oncannabis use behaviour and found that thoseprograms that were successful in reducing cannabisuse were similarly effective at changing tobaccosmoking and drinking. In summary, the mosteffective education programs for cannabis werefound to contain certain essential information aboutboth long-term and short-term consequences ofdrug use; were small in scale or managed in a waythat allowed ownership amongst those involved;had high fidelity of implementation; and wereinteractively presented. In addition, normativeeducation in the form of feedback from schoolsurveys of peer drug use was helpful becausestudents often overestimated drug use by peers.Additional program components included analysisof the media and other social influences promotingpositive attitudes to drugs. Interpersonal skills of

��"��

drug refusal, and intrapersonal skills such asdecision making, stress reduction and goal setting,were also important components. The otherelements, noted above, of booster sessions andinclusion of a family and community componentwere shown to further increase effectiveness ofprograms. By contrast, programs that have beenlargely knowledge-based, delivered by experts, lessinteractive, and that failed to include family andcommunity components have been less successful.The less successful DARE program, using policeofficers to deliver education and some life skills tomiddle school students, is usually used to illustratethis.688

Although there has not been extensive research inthe studies where cannabis use has been measured,findings have generally suggested that positiveimpacts may be achievable through drug education.Small reductions in regular cannabis use have beendemonstrated three years following the MidwestPrevention Program, and at six and 12 yearsfollowing the LST program.



Follow-up data from the 6.5 year long large-scale,randomised prevention trial of the LST preventionprogram, delivered during junior secondary school,was reported in 2000.666 In this latest study,students were completing secondary school in grade12. Self-report data collected by mail from 447individuals showed that those who received lifeskills training reported less illicit drug use comparedto the control group.

;'0';� � �+)-�� -?-��+��)��)��

There is good evidence that drug educationprograms produce changes in knowledge aboutdrug use and the consequences of drug use foryoung people who attend school. Althoughinterventions based purely on providinginformation appear insufficient to change eitherintention to use drugs or actual drug use, provisionof information may be a necessary condition foreffective prevention. Drug education programsbased on social learning principles have consistentlyshown short-term effects on both intentions andbehaviours. In general, the effects of theseinterventions diminish and even disappear by latesecondary school unless supplemented by additionalprogram input or supplementary strategies.Successful supplementary strategies have includedsocial marketing, community mobilisation, andparental involvement.

Existing evidence suggests that drug education canrepresent a relatively low-cost method for reducinginitiation and escalation of tobacco, alcohol,cannabis and other drug use. Programimplementation costs for each class cohort typicallyinclude teacher time for around eight to 15classroom periods in early secondary school, withan additional investment of eight to 15 classroomperiods later in secondary school. When properlyimplemented, this investment can reduce regulartobacco use by 5% to 10%.

The systematic review by White and Pitts showedthat effect sizes for most programs that wereeffective were small, with 3.7% of young peoplewho would have used drugs delaying onset of useor persuaded to never use.654 Evaluations must becontinued in the long-term to see if this delay inregular use or progression to further use continues.To shed some perspective, the authors note thatpharmaceutical trials are stopped once effect sizessmaller than this are reached, on the basis thateffectiveness is too high to ethically allow the non-treatment arm not to have access to treatment(example, aspirin trials in myocardial infarction). Itis also true that effect sizes measured mayunderestimate the potential gains, as authors seldomreport on fidelity of implementation or on numbersof participants who received the entire program asopposed to parts.

In a study of the ALERT and LST drug educationprograms in the US, the RAND Institute found thatthese programs reduced drug consumption(cocaine) by around 35% (25% of that came fromreduced initiation of use and 10% from reducedconsumption). The cost per participant wasestimated at around US$67 or 1.5% of national drugcontrol spending. This report found that for everydollar of resource used by the program, $2.4 wouldbe averted in social costs. However, it wasrecognised that considerable uncertainty surroundedthis estimate.689

Because there is still much to be integratedregarding drug education, implementation can bedifficult. Kim and colleagues report a failed attemptto disseminate health education in NorthCarolina.690 In this case, few students could recallreceiving the program, providing some indicationthat failed implementation may have explained theabsence of effects.

There are some warning signs that thedissemination of effective drug education in theAustralian context will not be a simple matter. Perryand colleagues evaluated the effectiveness of drugeducation in different nations as part of a WHO

�

��# �� !��"�#��$��%��

trial.691 Evidence from this study demonstratedeffectiveness for drug education in many nationsincluding the US; however, an Australianimplementation was not effective. It is clear that inattempting to implement evaluated programsfurther evaluation will be required to ensuresuccessful adaptation to the Australian context.

Given there are considerable risks through theimplementation of drug education programs, it isimportant that investment in these programs beaccompanied with a proportionate investment inevaluation. Because of the difficulties involved, it isreasonable that 20% of the program budget andresource requirements committed for drugeducation should be directed at evaluation.Evaluation designs should include behaviouralfollow-up, and randomisation to test new programelements against control groups, such as theevaluations of drug education programs in WesternAustralia which are being conducted through theNational Drug Research Institute.

'�� 3��$� ��!��!��

Definition: Programs run in primary schools to betterprepare children for the transition to secondaryschool encourage positive interpersonalrelationships at school, ensure effective discipline,maximise learning opportunities and maintaindrug-free environments.

Although there is a growing body of US research tosuggest that interventions aimed at improvingprimary school social environments can havepositive impacts, there has been less researchexamining secondary schools. The available researchis limited and inconclusive but does suggest thatsecondary school organisation and behaviourmanagement practices may influence youth druguse. Further research is required to develop a widerrange of program options and to better establish theintervention processes that underlie programeffectiveness.

One element of school organisation that appears ofparticular importance for preventing antisocialbehaviour is the policies and practices relevant tobullying and aggression within the school. Olweus’sevaluation of a multi-component anti-bullyingprogram in Sweden presented evidence ofsignificant reductions in violent and delinquentbehaviour.692

The Victorian Gatehouse Project provides aninteresting example of an intervention attempting tomodify early secondary school environments to

promote mental health.693 The Gatehouse approachaims to increase school bonding and to reduceexperiences of victimisation. A potentially importantcomponent of the Gatehouse project has involvedthe incorporation into the standard curriculum ofcomponents aimed at teaching stress-coping skills.Given the program focus on improving socialconnections and enhancing life skills, it is plausiblethat risk factors for youth substance abuse may bereduced. A rigorous evaluation of the Gatehousestrategy utilising random assignment of schools isproceeding.693 Early indications have associated theprogram with reductions in youth drug use butthese analyses have not yet been published.

A very different intervention strategy to thatdeveloped in the Gatehouse program involvedselecting students at risk and offering them specialservices within schools. Eggert and colleaguesevaluated a selective intervention for late secondaryschool students at high risk of drop-out.694 Staffidentified students who were offered placement in‘personal growth classes’. The personal growthclasses offered group support, friendshipdevelopment, and school bonding through smallteacher-student ratios and an emphasis on positivepeer relations. A specific skills training course wasalso offered based on four units: self-esteemenhancement, decision making, personal control,and interpersonal communication. Findingsdemonstrated that those exposed to the groupsreported improvements in school bonding, self-esteem and reductions in deviant peer associations.Program participants demonstrated less entry toharmful drug use. Program participation wasvoluntary and the evaluation was small and weaklycontrolled, hence more rigorously controlledevaluation of this program is warranted.

An interesting exploratory study points to apotential future direction for assisting students whohave a high number of risk factors for harmful druguse. Hastie and Sharpe described a quasi-experimental intervention on a small (n = 20),non-randomly selected sample of male students ingrades 7 and 8 in a rural school located in aneconomically depressed part of Canada.695 Morethan one-third of the cohort from which these boyswere taken was achieving school grades below a Dlevel. According to the school principal, most wereat-risk for school failure and early drop-out. Hastieand Sharpe describe a 20 lesson unit of modifiedfootball, with rules configured in such a way as topromote the likelihood of interpersonal conflict.The lesson content was described and seven types ofbehaviour identified and operationally defined (e.g.leadership, compliance, positive peer interactions).

��%��

Students were divided into four teams and feedbackwas provided before each match about performanceat the previous match with respect to the targetbehaviours. Teams also competed for a ‘Fair Play’award. Outcome variables included teachers’ ratingsof videotaped behaviours and student questionnairedata. The authors reported improvement in positiveinterpersonal comments, leadership, and positivepeer support. There were strong correlationsbetween students’ post-match questionnaires andthe raters’ scores on the videotapes. This study wasbased on a small convenience sample and littlestatistical inference was used to compare pre- andpost-measures. This type of approach does warrantconsideration as a means of developing social skillsand positive school connections for students insecondary school who may have a high risk for druguse problems.

There are innovative efforts to improve schoollearning outcomes in Australian secondary schoolsthat are likely to address developmental risk andprotective factors that relate to harmful drug use,for example, the Advocacy program that has beendeveloped and trialled in Victorian secondaryschools. The program emphasises the developmentof an ongoing one-to-one relationship between ateacher (the advocate) and a student. Teachersvolunteer to be advocates and undertakeprofessional development. They meet with studentsindividually at least once every two weeks for atleast 20 minutes. During this time, they focusmainly on helping students manage their learning.The essence of the interaction is that the advocatesare caring adults who listen to the students andsupport them to develop an individual approach tolearning, based around their strengths.

;'('&��*��/��+��+/�� *�-�� ++��)��)-�


Evaluations presented in earlier sections revealedthat, by altering primary school environments,longer-term improvements in youth developmentappeared to be achievable. There has been lessresearch investigating the potential to improvesocial environments within secondary schools;however, current indications suggest suchapproaches may be feasible. The impact of suchprograms on youth drug use is currently unknownbut should be investigated.

;'('.� � �+)-�� -?-��+�� -�� *��)��

In recent years, several strategies relevant to schoolorganisation and behaviour management haveemerged as having potential in secondary schoolsettings. The Victorian Gatehouse project providesan excellent model for improving secondary schoolsocial environments and is now being supported forwider dissemination. Very promising evaluationfindings are being reported in the overseas literaturefor school organisation programs delivered in theprimary school setting. Investment in this type ofapproach for delivery in the secondary schoolsetting would appear warranted.

'�"��$��

Definition: Youth peers of common identity providesupport or deliver a health message.

Peer influence is an important factor in thedevelopment of youth drug use and, therefore, itwould appear logical to attempt to use peerinfluence to reduce youth drug use. Peer leadershipin drug education has been evaluated. Generally,peer-led elements are incorporated as onecomponent in a broader set of activities, making itdifficult to identify the specific contribution of thepeer intervention. There are some risks with peerinterventions that need to be acknowledged. Poorlyimplemented peer interventions have the potentialto increase affiliations between youth with a highnumber of risk factors, reinforcing attitudes andbehaviours favourable to drug use.

Use of peer leaders as a strategy for reducing druguse in the context of school drug education hasbeen reviewed by Midford, Lenton and Handcockand was summarised above.650

Tobacco

Summary: Warrants further research ........ � -1, +2 /6.

Although it is difficult to clearly assess theimportance of the peer component, the availableresearch suggests that evidence of reduced smokinginitiation has been associated with exposure to peerprograms in school drug education. However, insome evaluations impacts were not achieved andthere is, as yet, limited evidence to demonstrate thatlonger-term outcomes can be assured through theseprograms. Furthermore, there is some evidencefrom the study conducted by Dishion andcolleagues that poorly conducted peer interventionscan exacerbate tobacco use and related problems.636

�

��' �� !��"�#��$��%��

Wiist and Snider described an innovative approachto health education in schools.696 The projectKNOW incorporated the elements of assertiverefusal, role-play, analysis of the media etc. Studentteachers, peer leaders chosen on the basis ofsociometric analysis, and model students werecompared as program leaders. Non-smoking rates inthe program run by peer leaders were higher (55%)than in the model student (42%), student teacher(41%) and comparison groups (48%) though thesedifferences were not statistically significant. Theseresults suggest that the peer leader program couldinfluence behaviour change but a larger evaluationwill be required to establish this.

A pilot study to evaluate the effect of a peerdeveloped, smoking prevention program wasconducted in Athens, Greece.697 The interventiongroup comprised 237 students in the first twogrades of secondary school; they were compared toa control group of 90 students from another school.The outcomes were based on self-report andincluded smoking behaviour, intent to smoke,knowledge and attitudes toward smoking. Arandomly selected group of 37 students from theintervention school developed anti-smokingaudiovisual material that they then presented to thewhole group. At the one year follow-up, exposureto the intervention had attenuated the overalltendency for increasing experimentation withcigarettes such that the increase was lower in theintervention group. However, the intervention hadno effect on intent to smoke in the future, attitudestoward smoking, and knowledge about the healthhazards of smoking and its addictive nature.

There is evidence to suggest that teaching skills toresist peer influence may be important interventioncomponents in drug education programs. Arandomised controlled trial was conducted in theNetherlands to assess the effects of different peer-led social influence programs, with and withoutboosters.698 Fifty-two schools from 15 district healthcentres were randomly assigned to a social influenceintervention, a social influence intervention with adecision making component, or the controlcondition. Half of the treatment schools wererandomly assigned to receive boosters that consistedof three magazines containing pertinent informationon all aspects of smoking, interviews with non-smoking celebrities and a competition. The socialinfluence intervention was five lessons lasting 45minutes each and delivered in weekly sessions bypeer leaders. The decision making component wasthe introduction of a manual in the first lessonoutlining five steps to making a decision:identifying a situation in which you have to make a

decision, looking at possible alternatives, weighingalternatives, making a decision, and implementingthe decision. The interventions were implementedin grades 8 and 9. Smoking prevalence wasdetermined via questionnaire prior to theintervention programs being initiated and thenevery six months up until 18 months later. The firstbooster was delivered between the six and 12month follow-ups and the second and thirdboosters were delivered between the 12 and 18month follow-ups. In the short-term (6 months),both interventions were effective in reducing theonset of smoking; however, at the 18 monthfollow-up, only the peer-led social influenceprogram with boosters remained effective.

A study was completed in Wales that included twointervention and two control schools, recruited inMid Glamorgan.699 The intervention and controlschools were matched on size andsociodemographic mix. Popular students in years 8and 9, aged between 12 and 14 years, wererecruited in the intervention schools for intensivetraining by specialist staff to promote smokingcessation and to prevent smoking initiation amongtheir peers. Self-reported smoking status wascollected via questionnaires at baseline, immediatelypost-intervention and three months later. Studentswere also asked to provide saliva samples foranalysis of cotinine levels to validate reportedsmoking status. With the exception of self-reportedsmokers, all saliva samples were tested at baseline.Only saliva samples collected from students whoreported a reduction or cessation of smoking weretested at the second and third data collection points.Baseline prevention of ever having tried smokingwas significantly higher in the intervention schools;therefore, changes in smoking behaviour werecompared rather than actual smoking behaviour.The intervention showed no overall effect inreducing smoking initiation among non-smokers orincreasing smoking cessation among regularsmokers. However, pupils in the interventionschools who reported being ex-smokers at baselinewere significantly less likely to have resumedsmoking than ex-smokers in the control schools.

A recent study was conducted in three secondaryschools in Athens, Greece, which had a similardesign to the study conducted in Wales.700 Twoschools were assigned to receive a peer-ledintervention and one was used as a control. In theintervention schools, working groups wereestablished to develop audiovisual material, in theform of a videotape and drawings, with an anti-smoking message. The members of the workinggroups had eight weeks of training from senior

��(��

child mental health professionals: on groupdynamics, enhancement of student self-esteem andsocial skills, and group activities for healtheducation. These students then became peer leaderswho presented the videotapes and pictures to eachclass. The article stated that the students had a meanage of 13 years and four months but no age rangewas given. Assessment of knowledge, attitudes andbehaviours around smoking were compared pre-intervention, immediately after the start of theintervention and three months later. Those in theintervention groups reported a decline in smokingbehaviour and intention to smoke immediately afterthe intervention but this was not sustained at thethree month follow-up.

There are ‘self-evident’ peer strategies that have atrack record of exacerbating youth problems. In thesmall study by Dishion and colleagues, the resultssuggested that interventions that increase contactbetween high-risk youth are at risk of beingcounter-productive for preventing tobacco use.636

The findings reported above suggest that peer-leddrug education can be implemented and withappropriate training and support that behaviour canbe impacted. The research over the past decade ismostly based on small samples and shows that manyinterventions were unable to sustain longer-termbehavioural changes. Study designs also need tofocus on testing a wider range of the peer processesthat might lead to behaviour change. For example,is it positive peer interaction or changes inperceived peer norms that are more critical?Interpretation of the research findings over the pastdecade should be considered in conjunction withearlier research summarised by Tobler,701 whichsuggests the effectiveness of youth substance useprevention programs can be improved by peer-ledintervention. One way forward is indicated by theNetherlands study reported above.698 The finding ofsuperior impacts for the combination of peerleadership and booster sessions in subsequent yearsechoes the results of Botvin’s drug education work,reported earlier.664 Peer programs hold somepromise for tobacco control, hence furtherinvestment in evaluation research is recommended.

Alcohol and cannabis use


When evaluating two forms of an alcohol educationprogram involving peer leaders and teachers,Wilhelmsen et al. found a more highly structuredprogram significantly decreased alcohol use, while aless structured program was associated withincreases in alcohol use similar to those in thecontrol condition.702 The implication of this

evaluation was that teachers need to be closelyinvolved in programs that incorporate peer leaders.

Given the evidence of negative effects for poorlyimplemented peer programs, investment is notwarranted unless it is accompanied by evaluation toenable outcomes and impacts to be accuratelymeasured. Evaluation should examine the short-term impact of peer education on risk factors suchas attitudes to drug use. Where short-term impactsappear congruent with risk reduction, longer-termoutcomes should be examined. The VictorianGovernment has made some investment in thedevelopment and implementation of peer-educationprograms to address alcohol and illicit drug use.Controlled behavioural evaluation of these programsis vital.

Illicit drug use


Peer programs have been extensively used inAustralia to address risks associated with illicit druguse. Typically these programs have involvedstrategies such as encouraging safe injecting andproviding information regarding support options.The impact of these programs on illicit drug use isnot yet known. This would appear to be a valuablearea for further research.

'�#�< �� !�

Definition: Provision or utilisation of recreationalopportunities outside the school setting to promotethe positive development of children and youngpeople.

For many young people, involvement in drug use isconsidered to be ‘recreational’ to the extent that it isa component of social and non-school recreationalactivities. Efforts to reduce drug use have includedefforts to modify youth recreation settings.

Tobacco

Summary: Warrants further research. ..............� 0/1

Participation in health clubs has been a popularapproach to smoking prevention in the UK.Typically, membership is for younger adolescentswho are non-smokers. Activities of the clubs varybut include recreational outings, newsletters,competitions and retail discounts. The clubs havebeen popular in attracting membership and the‘Grampian Smokebusters’ has been the subject ofmore detailed evaluation.703 Participation in theGrampian Smokebusters was very high, with the

�

��) �� !��"�#��$��%��

club successful in attracting over half the eligiblemembers in the community and over 97% havingheard of it. Surveys were conducted over a four yearperiod to evaluate the effect of the program. Theprogram produced no discernible effect on smokingrates at the end of four years.703 The authorscommented that the program was set within acommunity where many other influences came tobear on the likelihood of young people smoking. Inthis context, a single approach of this kind appearedunlikely to be effective.

Other initiatives, such as the use of local magazinestargeting teenagers and the use of other youth clubsas settings for health promotion, have not been thesubject of rigorous evaluation.

Alcohol


With sport being an activity that is intrinsicallyattractive to many teenagers, involvement in it haspotential as a vehicle by which a range of positivesocial behaviours can be modelled and reinforced.In Australia, however, there is a close and long-standing nexus between many male-dominatedsports (e.g. football, rugby, cricket) and excessivelevels of alcohol consumption. Recent researchconducted by the Australian Drug Foundationsuggests that young male players account for thebulk of the excessive alcohol consumption that takesplace in amateur football club settings.704 Whilstlocal sporting clubs are widely regarded asimportant community assets, because of theconnectedness they offer across a wide socialspectrum, unhealthy practices with respect toalcohol management give cause for concern. Forthis reason, the Australian Drug Foundation hasrecently introduced its Good Sports AccreditationProgram (GSAP) as a means by which clubs can berecognised at a community level for theirengagement in alcohol policy and practiceinitiatives. An intrinsic component of the GSAP isthe systematic involvement of community partners,such as local police, community health centres andregional sports assemblies, so that communityownership and awareness of the program iscultivated. While it is too early to demonstrateimpact of the GSAP on at-risk young peoples’alcohol and other drug use, this settings-basedapproach, with a strong emphasis on links betweenpolicy and practice, holds promise as a means bywhich a long established cultural norm may bemodified over the long-term.

Cannabis and other illicit drug use


There does seem to be some evidence that involvingat-risk young people in organised community-basedactivity programs, such as youth clubs, conferssome protection against drug misuse at both anindividual and a settings level. Schinke et al. carriedout a pre- and post-test controlled evaluation ofdifferent boys’ and girls’ club settings and foundthat the presence of crack cocaine was lowest insettings that were enhanced by a drug abuseprevention program.705 Compared to settings whereno such clubs existed, parental involvement in thelives of young people was greater in settings whereclubs existed, but not necessarily greater in settingsthat were enhanced by the drug abuse preventionprogram. A similar pattern emerged with respect tovandalism rates, which were highest in settingswithout clubs but lowest in settings with standardclub programs (i.e. not enhanced by a specific drugabuse program). This raises the possibility that clubsconfer protective factors by virtue of theirpromotion and modelling of positive socialbehaviour but may not be successful as quasi-educational avenues with respect to drug use. Therewere some significant methodological limitationson this study (e.g. limited statistical analysis ofgroup differences, limited follow-up time frame);however, this approach seems to warrant furtherinvestigation.

There have been some efforts in Victoria to organiseopportunities for drug treatment clients to beinvolved in football competitions. The effect ofthese interventions on drug use has not beenevaluated.

;'7'&� � �+)-�� -�� -?,�)��-��

The above review demonstrates the initiation ofevaluation studies investigating impacts throughyouth sport and recreation programs. None of thestudies conducted to date have strong enoughevaluation designs to conclude the programs hadpositive impacts. The evidence does suggest thestrategy has promise and hence warrants furtherevaluation. Investment may be particularlywarranted to further disseminate the Good SportsAccreditation Program in the context of evaluationfor longer-term effectiveness in reducing harmfulalcohol use. Youth sport and recreation strategiesmay be of particular importance to young peoplewho are not attending school.

��

'�%��

Definition: Strategies to develop positive socialrelationships between youth and adults who canprovide support and healthy role modelling.

Youth connectedness with caring and responsibleadults has been shown to reduce the risk of druguse and related problems for youth with a highnumber of risk factors. Programs that recruit andthen train adult volunteers to offer advice andfriendship to young people have been implementedand, in some cases, evaluated.

Mentoring programs typically involve non-professional volunteers spending time withindividual youth in a supportive, non-judgementalmanner while acting as role models. Tierney andcolleagues evaluated the Big Brothers/Big Sistersprogram.706 In a well-controlled evaluation,significantly less early age drug use was observedamongst youth exposed to mentorship compared toyouth randomly assigned to a waiting list controlgroup. Exposure to mentoring also demonstrated arange of positive impacts on risk and protectivefactors relevant to early intervention, includingimproved academic performance and familyattachment.

Tobacco and alcohol use


As currently designed, mentoring programs focuson youth with a high number of risk factors, andhence their use for more universal applications toprevent youth alcohol and tobacco use does notappear warranted. It is conceivable that there maybe future applications of these programs in targetingyouth with specific vulnerabilities to alcoholproblems.



The use of mentorship appears promising as aselective intervention for youth who may have ahigh number of risk factors for illicit drug useproblems.

;'3'&� � �+)-�� -?�� -��

The Big Brothers/Big Sisters program has beenimplemented in Australia through Jesuit SocialServices and appears worthy of further evaluation.Mentoring programs could be supported for furtherevaluation directed at innovations to broaden therange of models, and for the establishment oflonger-term outcomes within a randomised

controlled trial design. Mentoring programs may beparticularly successful where they are conducted ininstitutional environments, such as schools, andwhere there are high rates of drug-related harmsuch as in Indigenous communities and among out-of-school youth.

'�'�2 !!�� 3��

Definition: Adolescent health education curricula orinformation delivered in a community setting otherthan in schools.

Based on experience to date, there may beopportunities to deliver drug education curricula toyouth in a range of community settings. Theexisting evidence suggests that investment in thisdirection should be approached cautiously. As withall types of drug education, community-basedprograms have the potential to exacerbate problems.

Tobacco


We are aware that community-based healtheducation, including circulation of QUIT smokinginformation, has been a popular approach fortobacco control in Australia. However, there wereno studies identified evaluating this approach totobacco control. The effectiveness of this approachas a stand-alone strategy remains questionable.Rigorous evaluation is required to establishconditions whereby this strategy might beeffectively utilised.

Alcohol


Various programs have been run in communitysettings to address problems associated with alcoholuse, for example, alcohol education programs runin drug treatment programs and at communitycentres. Such programs may target youth withdrink-driving offences or entering drug treatmentfor alcohol related problems. We have been unableto locate evaluation studies assessing the impact ofthese programs in reducing problems associatedwith youth alcohol use.


Summary: Warrants further research ..........� -1, +1/2

Warning: Negative results in 1/2 studies

Sequenced drug education programs, includingsocial learning approaches and drug information,have been developed and evaluated targeting youth

�

�� !��"�#��$��%��

with a high number of risk factors for illicit druguse. In overview, these programs appear feasible toimplement; however, there may be risks inaggregating high-risk youth in selected drugeducation programs.

Fors and Jarvis used a quasi-experimental design toevaluate a community education strategy deliveredto a selected population of runaway/homelessyouth.707 A four session curriculum was deliveredalternately by peers and adults. Evaluation suggestedthat from pre- to post-test, those exposed to peer-led instruction were the only ones to show asignificant difference relating to drug knowledgetaught through the curriculum, and increasedlikelihood of assisting friends to use communityfacilities. These findings must be cautiouslyinterpreted as the analysis did not adequatelycontrol for pre-program differences or compareeffects across conditions.

Palinkas and colleagues reported a selectiveintervention targeting females, aged 14 to 19 years,who were pregnant or parenting and screened to beat risk of drug use.708 Recruitment mostly occurredthrough community health clinics. This reportexamined a 16 week curriculum focusing on socialand life skills. The intervention demonstrated nobenefit for social skills training and in some cases,three month follow-up outcomes were worse. Post-hoc analyses suggested the skills group increasedtheir level of socialisation with delinquent and/ordrug using peers. It may be that improving globalsocial skills in the context of prevalent drug use maynot be a useful prevention strategy.

;';'&� � �+)-�� -?��) �,5*�-��)��)��

Although drug education curricula are beingdelivered to youth in a range of communitysettings, there has been little evaluation. Theexisting evidence suggests that investment in thisdirection should not be further pursued unless itincludes adequate evaluation. As with all types ofdrug education, community-based programs havethe potential to exacerbate problems by increasingaffiliation with youth who have a high number ofrisk factors.

'�(��$��$��!��!��

Definition: Coordinated delivery of more intensiveservices tailored to meet a range of developmentalneeds; generally targeted to children andadolescents with multiple risk factors.

This strategy attempts to assist youth with a highnumber of risk factors and typically involvescomplex coordination across a range of servicetypes. Efforts to assess needs, identify relevantservices, coordinate service delivery and monitoroutcomes are important ingredients within a casemanagement model. Although various modelsappear to be utilised in health, counselling, youthwork and welfare, there has been less emphasis onevaluation. In what follows, a range of evidence-based models that appear relevant to earlyintervention are discussed but it should be notedthis review is selective and ignores a vast practiceoriented literature.

Tobacco, alcohol and cannabis


There have been no studies evaluating the impact ofpreventive case management on tobacco or alcoholuse. Preventive case management approaches appearbetter suited for targeting youth with a highnumber of risk factors and hence may be of limiteduse in more universal applications.

Illicit drug use

Summary: Warrants further research ................�1/1

The Multisystemic Treatment (MST) program wasdescribed earlier in this report (see Targeted FamilyIntervention). The MST program emphasises thedevelopment of service delivery objectives inconsultation with the young person and, wherepossible, their family. Service selection emphasisesevidence-based interventions and the use ofmonitoring to ensure progress toward agreedtreatment goals. MST service purchasing is managedby senior clinicians who are paid incentivepayments based on achievement of agreed treatmentoutcomes.649 Current evidence suggests that the MSTmodel has achieved behavioural improvements formandated juvenile justice clients in the US aboveprevious best practice. As clients entering juvenilejustice have a high likelihood of escalating toserious drug abuse, this preventive casemanagement model is considered relevant as atargeted intervention.

The Children at Risk (CAR) program targets high-risk youth in lower socioeconomic areas wherethere is significant poverty and antisocial activity.709

It involves collaboration between government andother services dealing with juvenile justice, drugabuse and community health. It aims to involveneighbourhood youths in services that offerattractive alternatives to drug abuse, gangs andcriminal behaviour. CAR programs require intensive

��

case management; coordinating family intervention,after-school activity, mentoring, tutoring,individual psychiatric assessment and counselling.Evaluation of this approach has not yet beenreported.

The Action Research Intervention and SystemImprovement Team (ARISIT) is a preventive casemanagement approach that has emerged in regionalVictoria.710 The ARISIT is a joint initiative involvingpartners in education and human services andinvolves a case management model usingcollaborative strategies across service sectors. Cross-sectoral training in the model is provided for schooland youth and community services staff in theregion. Service delivery is then monitored using anaction research approach that is directed by the goalof reducing developmental risk factors andenhancing protective factors. The support of ARISITat a regional level enables information on serviceimpacts to be used for planning future servicedevelopment. The ARISIT includes trainingmaterials and manuals but to date there has been noevaluation examining outcomes.

;'<'&� � �+)-�� -?�� -��

Preventive case management appears to be a feasibleapproach for assisting youth with a high number ofdevelopmental risk factors although there have beenno evaluations assessing impacts on harmful druguse. Variants of this strategy are being developed fordelivery in Australian settings and investment toassist further implementation and evaluation wouldappear warranted.

'��)�2 !!��! 3��

Definition: Campaigns to initiate or strengthen anexplicit strategy of coordinated community actionaiming to advance healthy youth development andprevent harmful drug use.

The costs and resources required to implementmulti-level community-based interventions areconsiderable; however, they provide theopportunity to target a range of risk and protectivefactors influencing youth drug use. Evidencesupporting the benefits of targeting a range of riskand protective factors at the same time isaccumulating, and efforts to implement andevaluate ambitious community prevention andintervention activities are now being reported.

In earlier sections of this chapter, the communitymobilisation activities within the Midwest

Prevention Program and Project Northland werementioned. These programs are important becauseof the apparent success they have achieved inimplementing complex community interventions.Project Northland found some early indication thatcommunity-level intervention programs caninfluence positively a range of risk and protectivefactors and also behavioural outcomes.579 The onsetof alcohol use was successfully delayed in theintervention school districts both in year 7 and year8. Significant reductions in the onset of tobacco andcannabis use were observed for students who werenon-users of alcohol at baseline. The MidwestPrevention Program reported results for one yearand three year follow-up. At one year, this programdemonstrated positive impacts on mediating factors(attitudes, knowledge, skills and peer influence)and on initiation and escalation in use of tobacco,alcohol and marijuana use.580 The three year follow-up demonstrated the program was effective atpreventing escalation (recent use in 30 days) intobacco and cannabis use, but not for alcohol.593

Further details of these earlier studies aresummarised elsewhere.14

To determine whether the intervention reduceddrug use among high-risk individuals, data from theMidwest Prevention Program for students who wereusing tobacco, alcohol or cannabis at baseline wereanalysed separately.631 These results showed that atsix months, the baseline users in the interventionarm of the program had significantly decreased theiruse of all three substances compared with thecontrols. Although statistically significant effectswere not sustainable over longer periods, theintervention program consistently demonstrated atendency to reduce the use of tobacco, alcohol andcannabis among baseline users across all fourfollow-ups, except cannabis use at 42 months.Despite the data on drug use in this study beingbased on self-report, this research indicatedcommunity-based intervention programs mayimpact not only on non-users at baseline but alsothose who have started to use drugs. Community-based intervention has the advantage of being ableto reach the unidentified high-risk population ofearly drug users in an anonymous fashion and at anearly stage when their patterns of drug use may bemore easily influenced.

A Centre for Substance Abuse PreventionCommunity Partnership Project was established inSanta Barbara County, California, in 1991. TheCalifornia drug use survey was administered, atinitiation and at completion of the project (in1995), to students in grades 7, 9 and 11 to examinechange in drug use patterns.711 Valid responses were

�

�� !��"�#��$��%��

only received from 46% of students so somecaution is required when interpreting the results.Drug use rates increased over the four year periodin which the study was conducted, but to interpretthese increases, the rates were compared with therest of California. Students in Santa Barbara reportedless frequent recent use of all the drug typesexamined, with the exception of cannabis use, ingrade 7 and poly-drug use across all ages.Significantly, more students in Santa Barbarareported no use of alcohol or other drugs in the sixmonths prior to the survey compared with otherCalifornian students. However, poly-drug use,defined as the use of any two substances ‘on thesame occasion’ in the six months prior to thesurvey, was more than twice as likely to be reportedby grade 7 students in Santa Barbara (17.1%) thangrade 7 students from the rest of California (8.1%).In grades 9 and 11, the percent of studentsreporting poly-drug use was higher in SantaBarbara, but only by 4% and 3.6% respectively.Most students reported that the program had aneffect on them and did not increase their interest insubstance use.

A community mobilisation program targetingadolescents on American Indian reservationsthrough a variety of activities, including festivalsand parent training, found that youth exposed tothe intervention showed a decline in alcohol andcannabis use.712 However, similar falls were alsoobserved in control communities, making itdifficult to attribute changes to the intervention.

In a separate study, a culturally sensitive skills- andcommunity-based approach aimed at preventingsubstance abuse among American Indian youth wasdeveloped and tested in five States of America.713 Aconventional theoretical model of life skills wastailored to the cultural prerogatives and everydayrealities of American Indian young people living inthe target western reservation setting. Students from3rd, 4th and 5th grade in 27 tribal and public schoolsin North and South Dakota, Idaho, Montana andOklahoma were surveyed on their use of tobacco,alcohol and cannabis. Youths were dividedrandomly by schools into three arms: skills trainingalone, skills training plus a community involvementcomponent, or a control arm which did not receiveany intervention. The skills training involved a 50minute session each week for 15 weeks that coveredinstruction, modelling and rehearsal in cognitive-behavioural skills associated with drug abuseprevention. The community involvementcomponent of the other intervention arm includedmobilisation of the youths’ families, teachers,school guidance counsellors, neighbours, law

enforcement officials and commercialestablishments to support drug abuse prevention.Flyers and posters were distributed to businesses,health and social service agencies, schools andchurches, and information meetings were held forparents, neighbours and teachers about the skillstraining the youths were receiving. Semi-annually,youths in the intervention arms received boostersessions. Six months after the interventions andevery year for the following three years, all youthswere retested with the baseline assessment tool.

Drug use in the last week was used as the outcomemeasure. Tobacco smoking was defined as seven ormore cigarettes in the last week, and alcohol andcannabis use was defined as four or more instancesof use. In general the skills-based interventionproved to be more successful than the skills pluscommunity-based intervention. At the six and 18month follow-ups, there were no differences seenin the use of tobacco, alcohol or cannabis in thethree study arms. No impact was seen on tobaccouse in any further follow-ups but the use ofsmokeless tobacco was reduced in the skills onlygroups compared with the skills plus community orcontrol arms. At the 42 month follow-up, the skillsonly group consumed less alcohol and used lesscannabis than the control group, but the skills pluscommunity group had intermediary rates that didnot significantly differ from either the control orskills only groups. Hence, in this case, the findingsdid not indicate an advantage for thesupplementation of school drug education curriculawith community mobilisation activities.

Studies within the American Indian communitiessuggest that caution is required when attempting toculturally tailor established intervention programsand that conceptual work may be necessary todevelop new prevention approaches for differentcultural groups.

Tobacco

Summary: Evidence for implementation ........ � 5/13

Definitive evidence is lacking but, on balance, thecurrent evidence provides some indication thatthere may be benefits in supplementing school-based drug education with additional communitymobilisation components.

A systematic review of community interventions forpreventing smoking in young people was publishedby The Cochrane Collaboration, in 2001.714 Bothrandomised and non-randomised controlled trialswere included that assessed the effectiveness ofmulti-component community interventionscompared with single component or school-based

��"��

programs or control conditions and reported onsmoking behaviour in persons less than 25 years ofage. As is customary for Cochrane reviews, bothpublished and unpublished data were sought. Astudy by Biglan and colleagues was the only onethat was unpublished at the time of the review andit has subsequently appeared in press. The reviewconsidered 57 studies and rejected 44 that did notmeet the inclusion criteria.

Of the 13 studies that were included, four hadrandomised either schools or communities to theintervention and control conditions. One studycompared a community intervention with a school-based intervention only group and a no interventioncontrol group. The interventions assessed werediverse and the populations to which they wereapplied varied from small rural communities tocommunities within large cities. Interventionsranged in length from three weeks to six years andoutcomes were measured immediately post-intervention in five of the studies, while in theother eight studies, length of follow-up variedanywhere from six months to 15 years. Attritionrates from the studies also varied considerably,ranging from 0% to 45%. Most of the interventionswere directed at students in late primary or earlysecondary school, with five studies targetingstudents from 11 years of age, three from 12 yearsof age and two from 14 years of age. Only onestudy reported younger children being involved inthe intervention and that was the Smokebustersprogram in the UK, which was directed at eight to15 year olds.

Lower rates of smoking were reported in theintervention condition in two of the nine studiesthat compared intervention communities with nointervention control communities.600, 716–723 Adifference in smoking prevalence was reported inone of three studies that compared multi-component interventions with school-basedprograms only.715, 716, 721, 724 One study reportedsignificantly lower prevalence of smoking in agroup receiving a multi-component approach,which consisted of a combination of media, schooland homework interventions, compared with agroup that had the media intervention only.725

Another study comparing a multi-componentcommunity intervention with a mass mediaintervention alone found the former had a lowerrate of increase in smoking prevalence.726

It is important to note that not all evaluations ofcommunity mobilisation programs have reportedpositive impacts on youth tobacco use. In anevaluation of a heart-health community

mobilisation campaign in Rotherham, in the UK,Baxter and colleagues reported higher smoking ratesamongst students exposed to the programscompared to control communities.716 The costs ofcommunity mobilisation programs are considerableand they would need to guarantee superior resultscompared with well-implemented school-basedhealth education alone before this approach couldbe recommended for wide-scale dissemination.Existing research has helped to establish thecomponents of community mobilisation programsthat will be required to address youth tobacco use.It is recommended that future investment bedirected at evaluation to further explore thebehavioural impact of these programs.

Alcohol


Evidence from three well-conducted evaluationssuggests that modest reductions in youth alcoholuse can be achieved through communitymobilisation. Initiation to alcohol use was delayedin Project Northland, and in the Midwest PreventionProgram, Project Northland and the NewHampshire Program the studies appear to have beenwell implemented, carefully evaluated and all havepositive outcomes relevant to early intervention.Although there have been important successes,translating these promising findings into assuredprograms is not in all cases successful. The studieswithin the Native American communities suggestthat caution is required when attempting toculturally tailor established intervention programsand that conceptual work may be necessary todevelop new prevention approaches for differentcultural groups. Evidence summarised in Chapter 10also demonstrates that community mobilisationprograms have been more broadly successful inchanging community factors that impact alcohol useamongst young people. For example, theCommunity Trials Project conducted by Holder andcolleagues successfully reduced minors’ ability topurchase alcohol, underage drinking, and harmsassociated with alcohol.727 Although reductions inalcohol use have been documented throughcommunity mobilisation, the available evidencesuggests that further evaluation should investigatewhether outcomes are superior compared to simplerinterventions that are limited, for example, to asingle component such as school-based drugeducation or enforcement of laws relating to youthand alcohol use.

�

��# �� !��"�#��$��%��

Cannabis

Summary: Evidence for implementation ..........� 3/3

In 1987, a longitudinal study of substance abuseprevention in a rural cohort of pre-adolescents andadolescents was begun in New Hampshire, in theUS.728 One hypothesis was that preventing cannabisuse required a community as well as a curriculumintervention, to establish a threshold of societaldisapproval. This study had two intervention groupsand one control group, and a very high responserate of over 80%. One intervention group receivedan anti-drug program as part of the schoolcurriculum. The other intervention group receivedcommunity intervention in addition to the school-based curriculum through parent courses and acommunity task force. Data from this study wereused in proportional hazards models to identifyfactors that were associated with trying cannabisand, separately, with regularly using cannabis.590

After adjusting results for age and gender, theyfound no differences between the three studygroups in the proportion of children who triedcannabis. After adjustment for both demographicand psychosocial variables, children in thecommunity intervention were at reduced risk forregularly smoking cannabis compared with thecontrol group, while the control and curriculumonly groups did not differ. The authors concludedthat the community prevention approach did notdeter children from trying cannabis but did appearto be successful in preventing them from becomingregular users.

Project Northland579 and the Midwest PreventionProgram also had some impact in reducing somepatterns of cannabis use.631



Beginning with the Turning the Tide initiative inthe mid 1990s, in Victoria, many Australian Stateshave implemented community wide programsaimed at better coordinating community responsesaddressing illicit drug use. These programs includeda range of elements, and community mobilisationcomponents have been included in some. Commonfeatures include efforts to bring togethercommunity members for consultations anddiscussions culminating in efforts to establishagreed local strategies. In many cases, theseactivities have resulted in positive servicedevelopment and, in some cases, better resourcecoordination. However, we have been unable tolocate evaluations that enable the effect of theseprograms on illicit drug use to be assessed. More

systematic evaluation of these programs, includingrandomised controlled community trials, wouldappear warranted.

;'&='&��)��,�� ) �,��*�+�-�� -

Community mobilisation requires considerablecoordination but available evidence suggests thestrategy can be carried out. Although the balance ofrecent evidence appears favourable, further researchwill be required to better establish the conditionswhereby community mobilisation can translate toreductions in harmful youth drug use. Further workwill also be required to establish that communitymobilisation programs can be effectivelydisseminated outside of the context of researchdemonstration programs.

Research is warranted to investigate the feasibility ofcommunity mobilisation in the Australian setting.The results published on community mobilisationinterventions for reducing youth drug use arepromising, but not assured. The costs of theseprograms are considerable and they would need toguarantee superior results, compared to well-implemented school-based health education alone,before this approach could be recommended forwide-scale implementation. It is vital to ensure thefidelity of program implementation if an objectiveevaluation is to be conducted. Communities ThatCare (CTC) is a well-structured training andconsulting program developed in the US thatincludes elements of community mobilisation. CTCconsists of a series of training activities designed toadvance community readiness and capacity todeliver evidence-based prevention programs withinlocal prevention coalitions. The program seeks toassist the formation of community coalitions and tofacilitate these coalitions to develop and implementprevention strategies carefully tailored to localconditions. The program has been developed forimplementation in Australia but has not beenevaluated for its impact on youth drug use.459

'��*��$��

Definition: Reorientation of existing health services tomodify developmental risk and protective factorsand for enhancement of service access for youngpeople.

There is tremendous potential for health services tomake an important contribution to broadercommunity agendas in preventing or reducingharmful youth drug use. However, there are notmany studies testing the impact of early

��%��

interventions involving health services on drug useby adolescents and young people. This area requiresfurther research.

There are many studies, mainly from the US, thatreport on the existing practice of primary healthcare professionals in preventive screening or healthpromotion offered for health risk behavioursincluding alcohol, drug and tobacco use.729–735 Somereport on interventions to increase health riskscreening practices by primary health careproviders.736, 737 As with school-based education,some recommend that preventive screening beginin early adolescence, before longitudinal studiesshow many youth have initiated substance use.738

However, these studies do not evaluate the effectsthat screening or health promotion advice offeredduring a traditional consultation has on thesubstance use behaviour of attending adolescentpatients. Adolescents and families want theirphysicians to address concerns such as substanceuse,737 and young people rate health care providersas the most credible sources of informationsecondary to parents.739 It is acknowledged thatscreening is only part of the solution to tacklingsubstance use in health services. Without knowingabout substance use, health care providers will notbe able to direct adolescents to appropriate servicesor institute interventions themselves. Typicalbarriers to physicians screening for sensitiveadolescent issues include: time limitations, lack ofknowledge and skill, inadequate remuneration,740

not knowing what to do with a problem oncefound,741 concern about possible negative impact onthe physician-family relationship, and quality ofmedical care and organisational issues.737

Effective training programs can help increase thescreening practices of physicians.736, 742 Otherapproaches have been to change the method ofhealth risk screening from physician interview toself-completion questionnaires that can be laterdiscussed with a health care provider. Paperny andHedberg evaluated a strategy for providingpreventive health services to adolescents usingcomputerised health assessments withindividualised educational videos, trained healthcounsellors and nurses. Settings for theseassessments included schools, universities, shoppingmalls and after-hours clinics on Oahu, Hawaii.743

Participants spent an average of 21 minutes on thehealth assessment and viewing multimedia,followed by 15 minutes with a health counsellor.One-third of participants required a furtherevaluation and counselling by the health nurse(average 8 minutes). Over two-thirds (71%) of the258 youth of mean age 17 years preferred this

method to traditional office settings and 92% feltthe time spent was acceptable. On exit interviews,adolescents recalled 81% of the documenteddiscussions by the health counsellor and 64% of theanticipatory guidance. Health counsellors identifiedproblems and provided counselling more often thanoffice-based clinicians on important health issues,including alcohol use. This method of healthscreening was also only one-fifth the cost ofstandard preventive visits.

Once substance use is detected, effectiveinterventions are needed to help the users at variousstages. A means of detecting appropriate types oftreatment is also needed.744 Available evidencereveals a range of programs have been developed:some have been delivered in health services, whilstothers have been integrated in other settingsincluding schools and universities.

A potentially important strategy for reorientatinghealth services involves moving the site of theservice delivery into schools. Integrating schoolnurses into schools has been shown to have benefitsin reproductive health.14 However, there is a paucityof studies evaluating this integration of healthservices into schools as a strategy for reducing youthdrug use. The Centre for Adolescent Health hasaimed to increase school students’ help-seekingskills and behaviour by developing the ‘HealthAccess Workshop’; delivered by primary health careproviders in conjunction with teachers. Thisapproach has not been formally evaluated, but pilotstudies indicate students receiving the program gainknowledge about the types of services available andhow to access them. Whether this translates toactual access requires examination.

Tobacco


There has been some evaluation of programsincluding elements of school drug education thatrely on the involvement of health professionals forcoordination and content. The overall evaluation ofthese programs suggests their impact in reducingyouth tobacco use may be positive but theevaluation designs generally do not enableassessment of the specific intervention benefits ofhealth professional involvement.

A program that incorporated lessons by the healthvisitor and the local physician was delivered tostudents from 6th to 9th grade in the town ofSteigen, Norway, where smoking prevalence rateswere well above the national average.745 Ademographically similar municipality was chosen asa control community. Students were recruited in

�

��' �� !��"�#��$��%��

grade 6 each year, from 1992, so that by 1995 allstudents in secondary school were participating. Theintervention consisted of two lessons at the end ofgrade 6, and 10 lessons each year in grades 7through 9. The education provided information onthe short- and long-term consequences of smokingand each student was asked to sign a contract toremain a non-smoker during the remainder ofsecondary school. Brochures were sent to parentsand students watched and produced educationalvideos. In grade 8, they were required to produce aprogram for the local radio station, write letters forlocal papers and prepare posters. There was anincentive, in the form of a chance to win a CDplayer, for those who remained ‘smoke free’ untilthe end of year 9. In 1995, the prevalence of dailysmoking was 80% lower in Steigen than in thecontrol community, and those who smoked dailysmoked 50% fewer cigarettes.

It is reported that the school physician led theproject until he moved in 1995 to another countyand that, following his departure, enthusiasm forthe project diminished. This intervention appears tohave been relatively successful in a high-riskcommunity but, apart from some evidencesuggesting leadership characteristics wereimportant, the value of having a medical authorityinvolved is not clear.

Health professionals, particularly those working at aprimary care level, have an accepted status andunique opportunity to advocate in localcommunities for action in relation to adolescenttobacco use and other health risk behaviours. Theyalso appear uniquely placed to play a role withyoung people with whom they come into contact.Relevant professionals include community andprimary care nurses, general practitioners, dentists,pharmacists, and school counselling and healthpersonnel.

There have been a number of reports of successfulattempts to use the influence of health professionalsto modify smoking behaviour in adults746,747 andsome preliminary studies of the potential role ofhealth professionals with youth, both in theprevention of smoking and promoting smokingcessation. Klein et al. reported findings from atraining program for paediatric residents in theUnited States.748 One curriculum element related totobacco use and was based on receiving andworking through smoking cessation guidelinesfrom the National Cancer Institute. However,substantial barriers to training were found,including non-attendance at scheduled lectures andfailure to work with homework materials. Few

differences were found between participants of theintervention group compared to controls in relationto smoking uptake or physician practice.

A more recent study of an educational interventionwith post-qualification dentists in the United Statesalso had largely negative findings.749 Clinicians fromseventy-seven dentistry offices received 1.5 hours oftraining based on smoking cessation guidelinesfrom the National Cancer Institute. Elementsincluded the current profile of youth tobacco use,the role of the clinician, and instruction on creatinga tobacco-free environment. A strategy of usinganti-tobacco prescriptions in non-smokers was alsotaught and encouraged. A small financial incentivewas given for each ‘script’ administered. A total of17 925 adolescents were followed up, with a 93%completion rate in experimental and control groups.No difference in smoking rates was found betweenadolescents attending intervention and controlpractices. It was noted that implementation of thestrategy by clinicians was sub-optimal. Only 64% ofparticipating practices issued prescriptions and only14% reached their target. This was a probableexplanation for the disappointing findings.

A more successful attempt to control tobacco usethrough the reorientation of health services wasreported by Lionis and colleagues, in Crete.750 Inthis intervention, health professional assistance wasintegrated within schools through the provision of ahealth examination and a broad-based healtheducation curriculum. The intervention was wellreceived by parents and was associated withreduction in a variety of cardiovascular risk factors,including significantly less initiation of tobacco usein the intervention group (6%) compared to thecontrol group (20%), after one year.

Several health service strategies have been reportedfor control of tobacco use, though considerablework will be required to establish models ofpractice and methods for overcoming theconsiderable implementation challenges. Furtherinvestment to encourage innovation and evaluationwould appear warranted.

Alcohol

Summary: Evidence for implementation .......... �3/3

There have been some effective interventions aimedat reducing harmful alcohol use in universitystudents. Marlatt and colleagues conducted arandomised controlled trial evaluating a briefintervention designed to reduce the harmfulconsequences of heavy drinking among high-riskcollege students.751 They followed up participantsannually, for four years post-intervention.752 All

��(��

students who were to attend the University ofWashington were mailed a questionnaire aboutfrequency of alcohol use and drinking-relatedconsequences before finishing the last year ofsecondary school. Fifty-one percent of the studentsreturned the questionnaires and consented tofurther involvement. Twenty-five percent of thesestudents were identified as being high-risk in termsof their alcohol use and consequences. A normativecomparison sample of students, including 33 high-risk students, was randomly selected in order totrack natural changes in drinking patterns of thecohort over time. High-risk students wererandomised to either intervention or control group.The intervention consisted of a single brief non-confrontational counselling session, withpersonalised individual feedback and motivationaltechniques. Each student was required to self-monitor their drinking pattern in a diary for twoweeks before the counselling session. Drinking levelwas compared to norms for same-aged peers andbecame the resource for the personalised feedback.Follow-up assessments over four years showed thatdrinking problems declined significantly over timeand the intervention produced significantdifferences in alcohol use and harmfulconsequences over the four years. High-riskstudents continued to experience more alcoholproblems than the normative comparison group,though significantly less in the intervention groupthan high-risk controls. Among high-riskparticipants, 67% of the intervention groupcompared to 55% of high-risk controls had goodoutcomes over four years. Most students, overall,showed a decline in problems over time indicating adevelopmental maturational effect.

An earlier study by Baer et al. tested three methodsof alcohol risk reduction with young adults.753

Volunteers were randomly assigned to receive a sixweek class and discussion group, a six unit self-helpmanual, or a single one hour feedback and advicesession with professional staff. The interventionswere based on a skill-based approach to reducingalcohol use, involving cognitive-behavioural self-management principles, challenging assumptionsabout alcohol effects, and brief motivationalinterviewing techniques. Results showed significantreductions in self-reported drinking post-intervention (approximately 40%) and maintenanceof these reductions at a two year follow-up.Comparable drinking reductions were rated acrosstreatments except in those who did not comply withthe self-help reading program. The conclusionsfrom this study, however, are preliminary becausethere was no control group receiving the

assessments alone and self-report may have invitedsocially desirable responses.

These studies indicate useful approaches to reducingthe harm of alcohol use in older adolescents andyoung adults by integrating health advice within theuniversity setting. In other studies, programs havebeen integrated within health settings.

Rickert and colleagues used random assignment toevaluate two alcohol health promotion strategies foruse with young people visiting a primary healthclinic.754 In one case, the youths were exposed to acomputer-generated instruction program on alcoholand cannabis use, and in the other, a physiciandelivered anticipatory guidance. At post-test, bothconditions resulted in significant increases inknowledge of alcohol and cannabis relative to acontrol group. Female subjects preferred thecomputer-assisted instruction while males preferredthe physician encounter. The impact of this one-offintervention on subsequent alcohol and drug usewas, however, not evaluated.

Oliansky et al. evaluated the effectiveness of briefinterventions in reducing substance use among at-risk primary care patients in three community-basedclinics.755 All adolescents were screened and a nurse-led educational intervention, including pamphlets,motivational interviewing and setting a contract forpersonal goals, was associated with reductions inself-reported alcohol use in the intervention groupcompared with controls. There was, however, noobjective measure of alcohol use.

There exists some evidence to support selectiveinterventions within schools. Werch et al. evaluateda three-phase intervention selectively targetingschools with high proportions of African-Americanstudents.756 Student volunteers were randomlyassigned into the control or an intervention thatinvolved three components: a self-instructionalmodule, a health consultation with a physician or anurse, and a follow-up consultation with a trainedpeer health educator (an 8th grade student).Instructional messages were tailored to the stage ofalcohol use exhibited by the young person. Relativeto the control group, receiving only untailoredalcohol information, participants in the programdemonstrated a number of benefits at follow-up.Participants perceived lower prevalence rates foradult drinking and considered themselvespotentially more susceptible to alcohol problems.Their intentions to stop or reduce drinkingincreased and there was a small significant effect forreductions in the quantity and frequency of alcoholconsumed. Participants evaluated the nurse contactparticularly highly.

�

��) �� !��"�#��$��%��

The screening and brief intervention model hasemerged as a promising approach for health servicereorganisation aimed at reducing alcohol problems.The further implementation of this model should beaccompanied with investment for training andbehavioural outcome evaluation.



There are no studies evaluating impacts on illicitdrug use following health service reorientation.There are interventions targeting youth with a highnumber of risk factors where health services areinvolved, but the effects of the health servicecomponent would be difficult to segregate. TheChildren at Risk (CAR) program, described onp. 132, provides an example of a strategyincorporating elements of health servicereorientation and preventive case management. Thisapproach has not been evaluated for its impact onillicit drug use

A study by Catron and Weiss has evaluated a school-based mental health service for high-risk childrenwith serious emotional and behaviouralproblems.757 Results indicated that school-basedservices were more likely to be accessed. Ninety-eight percent of children (mean age 9.6 years)referred to the school-based service enteredservices, compared to 17% referred to traditionalclinic-based community agencies. There were noresults on treatment outcomes or attrition, however.These results introduce the possibility that serviceaccess for youth with a high number of risk factorsmay be enhanced through adaptation of servicemodels.

;'&&'&��)��,�� +��-��-

Health service reorientation appears a promisingintervention strategy. There is now reasonableevidence that a variety of strategies can be used toimprove the accessibility and effectiveness ofexisting health services relevant to young people.Ensuring existing services maintain a preventionfocus and utilise effective methods of engagementwould appear a fundamental step in the process oftackling drug issues.

Two study teams have demonstrated outcomesrelevant to alcohol misuse. In two studies, Marlatt,Baer and colleagues demonstrated reductions inharmful alcohol use through early intervention withcollege students in the United States.751 Werch andcolleagues have demonstrated an impact with earlysecondary school students.756 Although the

indicators are very promising, to be confident thisstrategy can consistently achieve outcomes willrequire further research.

Further research investment is recommended toencourage program innovation and evaluation ofuniversal health service reorientation strategiestargeting youth drug use. This investment should bemade in partnership with agencies responsible forhealth and mental health.

'��&!�� !��0��

Definition: Includes provision of pre-employmentassistance, employment experience, training orintervention in a post-school training setting, withthe aim of advancing adolescent health.

Tobacco, alcohol, cannabis and illicit drug use


Although a number of adolescent alcohol initiativesfocus on secondary school populations, there appearto be important further opportunities to modifyalcohol and other drug use through the period fromadolescence to adulthood. The entry to theworkforce and to post-secondary education is animportant developmental transition influencingyouth alcohol and drug use. No evaluation studieswere identified examining impacts on drug andalcohol use of employment and training programs.

'�� !�-��$��

Definition: Use of the mass media to promote a healthmessage relevant to the prevention of harmfulyouth drug use.

There is good evidence that mass media strategiescan convey a health promotional message to a highproportion of young people. Radio appears aseffective as more expensive media. Less attentionhas been paid to media such as the internet andteenage magazines. There is no good evidence thatsimple ‘one-off’ media campaigns affect drug use inthe young. There is better support for the use ofmass media in combination with other strategiessuch as school-based health education orcommunity mobilisation.

Tobacco


The mass media have been commonly used intobacco prevention approaches. Relevant mediainclude television, radio, movies, internet andteenage magazines. The attractiveness of theseapproaches lies in their capacity to reach a large

��

teenage audience quickly, but there have beenconsiderable doubts expressed about theeffectiveness of this strategy.

The Cochrane Collaboration published a reviewcovering mass media interventions for preventingsmoking in young people, in 2001.758 The reviewincluded randomised trials, controlled trials withoutrandomisation, and time-series studies that assessedthe effectiveness of mass media campaigns ininfluencing the smoking behaviour of young peopleunder 25 years of age. Of the 63 studies that wereidentified as reporting information about suchcampaigns, only six met the selection criteria. Fivewere conducted in the United States and one inNorway. All six studies reported outcomes betweenone and two years post-intervention; only tworeported a significant impact on smoking behaviour.

In Norway, two counties matched for size, smokingprevalence and socioeconomic variables wereallocated to either a provocative media campaigntargeting 14 to 15 year olds or the controlcondition.759 Surveys were conducted at baseline, in1992, and post-campaign, in 1995. In the interim,the intervention county was subject to three annualmedia campaigns each of three weeks duration. Thefirst two campaigns were specifically targeted atgirls and the last was targeted at both genders. Inspite of the fact that only 63% of the interventiongroup had exposure to the media used to deliver thecampaigns, Hafstad reported that one year after thefinal campaign, their prevalence of daily smokingwas significantly lower than in the control group.There was considerable loss to follow-up in boththe intervention and control counties but afteradjustment for response rates, the findingsremained significant.

One unpublished study assessing a televisioncampaign targeted at students between 10 and 12years of age was included in the Cochrane review.758

It was reported that 18 months post-intervention,there was no significant difference in the smokingbehaviour of the intervention and control groups.

A variety of intervention conditions were assessed inone study, including a media campaign, a school-based programme, delivery of health information,no intervention, and a combination of the mediacampaign and school-based program.760 The studyshowed no consistent program effects on smokingbehaviour; however, there was more than a 50%loss to follow-up and the integrity with whichinterventions were delivered varied significantly. Anearlier study, comparing a media campaign onlywith a combination of a media campaign and aschool-based program, also found no significant

difference in smoking behaviour or intention tosmoke two years post-intervention.761 However, lossto follow-up was very high (71%) and thereforethis study makes little contribution to the weight ofevidence for the influence of mass media campaignson smoking behaviour.

Bauman et al. compared three mass mediacampaigns: first, eight 30-second radio messagesfocusing on the consequences of becoming aregular smoker; second, an additional 60 secondradio message inviting 12 to 15 year olds to join an‘I won’t smoke sweepstakes’ and recruit friends tothe program; and third, a television broadcast of thesweepstakes.762 A randomised design was used tocompare six intervention geographic areas with fourcontrols. Radio brought a modest shift on theexpected consequences of smoking and proved aseffective as television in shifting these views. Therewas little effect of the peer recruitment strategy. Noeffect was found on smoking behaviour.

Flynn et al reported a strategy from Vermont andMontana, US, of combining a mass media approachwith social learning-based school smokingprevention programs for school years five to 10.507,

763–766 The mass media intervention used radio andtelevision, with spots either 30 or 60 seconds inlength placed in programs likely to appeal to ‘high-risk girls’ and ‘high-risk boys’. The content of themass media campaign was consistent with that ofthe school-based intervention. Two groups werecompared: those receiving both mass media andschool components (one community in Vermontand one in Montana), and those receiving theschool intervention only (one in Vermont andMontana). The combined groups showed positiveshifts on attitudes to smoking and rates of peersmoking, and there was a 30% reduction in weeklysmoking rates in the mass media exposed studentscompared with those receiving only the school-based program. The effects persisted in exposedstudents two years after completion of theintervention.766

Murray et al. examined the effects of a Minnesotastate wide mass media campaign on smokingattitudes and beliefs in teenagers.718 The campaigntook place over a five year period in the mid-1980sand used a range of media: television, radio,newspapers and billboards. Comparison was madewith neighbouring Wisconsin where no similarinitiative had been taken. Student surveys indicatedthat the advertisements had great penetration withthose aged 10 to 16 years and a much greater recallof anti-smoking messages. However, there was littleshift in beliefs about smoking and no concomitantshift in smoking prevalence rates.

�

�� !��"�#��$��%��

Research on the effectiveness of different anti-smoking messages in the United States waspublished, in 1998.767 Data were collected from 186focus groups, involving more than 1500 childrenand adults, to evaluate the effects of different anti-smoking advertisements that had been aired andadditional concept advertisements. Alladvertisements were classified by their primarymessage and eight different categories wererecognised: industry manipulation, environmentaltobacco smoke, addiction, cessation, youth access,short-term effects, long-term effects, and romanticrejection. The anti-smoking messages that werefound to be most effective in reaching all audienceswere those depicting industry manipulation andenvironmental tobacco smoke.

Social marketing strategies targeting tobacco use arewell developed and a considerable level ofevaluation has been published. The evidencesuggests that these strategies have little impact onbehaviour when implemented alone. However, thework of Flynn and colleagues suggests that theseprograms can be effective when delivered incombination with other prevention strategies.

Alcohol


Harm minimisation introduces the requirement forrelatively complex messages about youth andalcohol use. Recommendations for levels of use varyby gender and there is discretion as to the age atwhich parents should introduce children to alcohol.This complexity introduces challenges for thedevelopment of effective social marketingcampaigns. The distinct nature of Australian policymeans that scientific knowledge is limited fordetermining how specific messages will impactbehaviour change. There is a particular need forAustralian investment in social marketing to beaccompanied by behavioural research. Theevaluations conducted to date demonstrate somesophistication in utilising behavioural research;however, there has been a tendency for research toremain unpublished or to be published in-houserather than in peer-refereed research journals.

The Australian evaluation research makes clear thatmajor campaigns have achieved their immediateobjective of increasing public awareness of keymessages. However, awareness of these messageshas been inconsistently associated with behaviourchange. As distinct from the tobacco researchreported by Flynn and others in the US, there hasbeen no use of geographic control groups.

The initial Australian Drug Offensive CampaignAgainst Alcohol Abuse was launched in 1988 andincluded five television commercials reinforcedwith notices in magazines, and on buses and trains.Key messages emphasised that: alcohol use couldharm the physical development of young people,drunken behaviour could result in socialembarrassment, responsible alcohol consumptionwas important, and parental modelling of alcoholuse influenced the behaviour of young people.

Evaluation included a pre- and post-surveyconducted with 2400 youth aged 15 to 17 yearsand a smaller number of parents. Youth reportedhigh awareness of the television commercials butthis awareness was weakly associated with actual orintended behaviour change. A small shift in parents’behavioural intentions was noted. As there has beenlittle research to demonstrate that parents caninfluence youth who are using alcohol to drinkmoderately, it is unclear what effect theachievement of this campaign goal might have. Theevaluation concluded that there had been someattitude change associated with the campaign.768

The campaign had three major stages: from 1988 to1990, 1991 to 1992, and 1992 to 1993. Prior tothe commencement and at various intervals, 10national quantitative household surveys wereconducted with samples of the target group of 15 to17 year old teenagers. Drinking on the last occasionwas one of the major outcome measures. It wasfound that there was a significant decrease in theproportion of teenagers consuming five or moredrinks on the last occasion during the course of thestage III campaign, but a further increase over theperiod February 1993 to March 1995 when therewas no national adolescent campaign activityundertaken.769

The year 2000 National Alcohol Campaign aimed toreduce alcohol-related harm amongst youngAustralians. The campaign targeted youth aged 15to 17 years and parents with children aged 12 to 17.Messages were conveyed via a range of media,including television, radio, cinemas, magazines,newspapers and a web site. The campaign alsoincluded media components specific to people fromIndigenous and non-English speaking backgrounds.

Key messages for young people emphasised‘drinking choices’, with ‘excessive drinking’portrayed as socially unacceptable and having thepotential for regretted consequences. The decisionto avoid excessive drinking was contrasted as havingsocial and health benefits. Behaviours recommendedfor avoiding excessive alcohol use included: eatingbefore drinking, monitoring the amount and pace

��

of drinking, and substituting with non-alcoholicbeverages. Youth were encouraged to avoid socialpressures to drink, but how to achieve this was notspecified. The parents’ role in educating youngpeople about the possible consequences of excessivealcohol use and the need to set limits wasemphasised by the campaign.

Household and phone surveys, with samples ofaround 1000 parents and young people, wereconducted before and after the launch phase.Awareness of the campaign was high, with 88% ofyoung people and 81% of parents reportingawareness. Young people reported the campaignhad prompted them to think about the potentialnegative consequences of drinking (94%) and 63%reported discussing the messages with friends. Ofparents, 51% reported responding to the campaignby talking with children about excessive alcohol useand by ‘keeping an eye on them’. Parents acceptingthat ‘teenagers learn to drink from the way theirparents drink’ increased from 64% to 71% over thecourse of the campaign. The evaluation noted somesmall reductions in alcohol use across the surveysbut was unable to associate these with thecampaign. Levels of potentially harmful alcohol useremained high at the end of the campaign.770

In 1997, the NSW Government ran the ‘Drinkdrunk the difference is U’ campaign. Evaluationsuggested high awareness and approval and therewas some intention to change binge-drinkingbehaviour reported. In 1998, the VictorianDepartment of Human Services conducted asummer/autumn events mass media campaign incinemas, promoting harm reduction. A smallevaluation suggested this campaign was widelyrecognised and well-accepted by Victorian youth.

Community-based health education, includingcirculation of pamphlets providing information ondrug use and harm minimisation information, isone of the most popular approaches to drugeducation in Australia. Various informationpamphlets have been circulated by the State andAustralian governments and through agencies suchas the Australian Drug Foundation. However, therewere no studies identified evaluating outcomesfrom this approach and its effectiveness as a stand-alone strategy remains questionable. Rigorousevaluation is required to establish conditions whereby this strategy might be effectively utilised forbehaviour change.

Mass media campaigns have been successfullyincorporated into wider community mobilisationcampaigns but we have not managed to locatestudies evaluating their specific impact on youth.

The alcohol industry invests heavily in mass mediacampaigns that include youth targets.

The evaluation work conducted to date, in Australia,suggests that the impact of major social marketingcampaigns have included public awareness andattitude change. However, there is little evidence toassociate this awareness with behaviour change. Alarge body of international follow-up research, andsome cross-sectional Australian research,demonstrates that early age alcohol use is a riskfactor for the subsequent development of alcohol-related harm. In contrast, there is little research tosuggest how teenage alcohol users can beencouraged to do so moderately. Given there isconsiderable uncertainty, the relationship betweenharm minimisation campaign elements and youthbehaviour needs to be carefully evaluated. Bylimiting the initial launch of campaigns to selectedgeographic regions, it would be possible to morerigorously examine behavioural impacts and toevaluate the impact of different campaigncomponents relative to control groups. This level ofevaluation would require a closer relationshipbetween research and service delivery but it isimportant given the scale of public expenditureinvolved and the level of harm associated withalcohol misuse.



Efforts to address youth involvement in illicit druguse through social marketing face the challenge oftargeting to the minority of youth involved in thisbehaviour. Donohew and colleagues reported anevaluation at pre-production of a televised, anti-drug, mass media campaign.771 An individual-levelrisk factor for youth substance abuse, sensationseeking, was targeted in this intervention. Donohewand colleagues reasoned that media appeal andmotivations for youth substance use would differfor those high and low in sensation seeking. Twotelevision message campaigns were developed for18 to 22 year olds, one for high sensation seekersand the other for low sensation seekers. Focusgroups were used to identify distinguishingcampaign features that would appeal to youth withthese characteristics. Youth were exposed totelevised messages that varied by media andsensation seeking format. Behavioural intention tocall a hotline was found to interact with the type ofmessage presented and high and low sensationseeking. A high sensation message (loud, vivid andchanging) led to higher behavioural intention tocall the hotline. However, evaluation did not extendto examination of behaviour change.

�

�� !��"�#��$��%��

An Australian social marketing campaign targetingyouth amphetamine use was run as part of theNational Drug Offensive, from 1993 through to1995. The campaign titled ‘Speed catches up withyou’ utilised a mixture of media to raise awarenessof harms associated with amphetamine use. A seriesof four household interview surveys, eachconsisting of around 1200 young people in the agerange 15 to 30 years, were conducted at key pointsover the course of the campaign. At the end of thecampaign, recall of television commercials wasevident for 70% but there was less recognition ofnewspaper (19%), billboard, radio or magazineadvertisements (10%). Of those reporting televisionexposure, 84% could accurately describe keymessages (e.g. ‘you never know what’s in it’).There was some evidence to associate this campaignwith behaviour change. During the early, intensivephases of the campaign, intentions to useamphetamines steadily decreased, from 56% downto 35%, amongst previous amphetamine users.However, intentions returned to baseline levelsonce the campaign finished.772

The National Illicit Drugs Campaign, launched in2000, represented an important strategy shift in itsacknowledgment of the role of the family inpreventing youth involvement in illicit drug use.The campaign aimed to deter the initiation orcontinuation of drug use by children byencouraging parents’ and carers’ communication.Campaign elements incorporated a range of media,including television, newspaper, magazine,billboards, a web site, and a phone contact line. Aparent booklet was distributed to all households inAustralia. Resources were also tailored for deliveryto people from non-English speaking backgrounds.

Evaluation included a series of telephone surveyswith parents (n = 1800) and other adults, and alsophone and face-to-face surveys with young people.Surveys were staged for completion before and afterthe campaign launch. Awareness of the campaignwas almost universal, at 97% for both youth andadults. Of parents, 68% were aware of the bookletand 47% reported they had taken some action inresponse to the booklet. The percentage of youngpeople who had discussed illegal drugs with theirparents rose from 26% to 44%, and those agreeingthat ‘advice from my parents has steered me awayfrom drugs’ increased from 57% to 71%. Similarpositive effects were noted for non-Englishbackground families. Although positive familycommunication is an accepted protective factor, thereport did not attempt to associate these impactswith changes in youth drug use.773

Illicit drugs social marketing campaigns have beeneffective at gaining awareness, improving familycommunication and impacting behaviouralintentions. Further evaluation should investigate thetranslation of these promising impacts intobehaviour changes.

;'&0'&� � �+)-�� -?-��+��1��

There is good evidence that mass media strategiescan convey a health promotional message to a highproportion of young people, and television appearsto be an important component. But there is no goodevidence that simple ‘one-off’ media campaigns canalter the long-term development of drug use in theyoung. There is better support for the use of massmedia in combination with other strategies such asschool-based health education or communitymobilisation.

The work of Flynn and others suggests thatevaluations can be designed to assess thecommunity-level impact of social marketingcampaigns on youth behaviour. Investment in socialmarketing in Australia is typically associated withlower levels of evaluation evidence, such as pre-and post-marketing consumer recall surveys. Giventhe large investment of public funds involved, thefailure to test social marketing strategies withinmore rigorous community trials must bequestioned.

Investment is warranted to enable rigorousevaluation of social marketing campaigns toestablish directions for achieving outcomes oncommunity-level youth behaviours. To ensurepublic health investment achieves behaviourchange, a stronger commitment to evaluation willbe required in developing service delivery contracts.Future service delivery could be introduced intogeographic regions in phases determined byrandomisation. Behavioural measurement couldthen be used to assess comparative impacts forregions exposed and not exposed to socialmarketing. Of the specific drug types that could bepotentially targeted through these approaches, itappears that both tobacco and illicit drug use maybe amenable to universal social marketingapproaches. Youth alcohol use messages withinharm minimisation programs are particularlycomplex with targets such as limiting the frequencyand amount of use and different recommendationsfor males and females. This complexity suggests arequirement for behavioural research to developfeasible campaign targets.

Future evaluations could trial the different campaigngoals of delayed use of youth alcohol versus harm

��"��

reduction in community intervention trials. Trialingdifferent prevention goals would offer an importantopportunity to examine the potential of socialmarketing to contribute to a reduction in drug-related harm.

'��/50��

Definition: Setting and enforcing laws and regulationsregarding the minimum age for purchasing or usingsubstances.

Legislation to restrict alcohol and tobacco sales tochildren and adolescents has been in place since theearly part of this century. When appropriatelyenforced, such legislation appears to be effective inreducing early use of both tobacco and alcohol. Theillicit status of drugs may influence the attitudes andbehaviour of some young people; however, beingconvicted for drug use can also have harmfulconsequences. Balancing these considerations hasled to proposals to move cannabis use from acriminal to a civil offence. Current evidencesuggests these changes have not increased cannabisuse amongst young people. Efforts to divert youthapprehended for drug offences into health andsocial services appear to be promising as earlyintervention approaches.

Tobacco


There is now reasonable evidence that it is possibleto reduce youth tobacco use by preventing retailersfrom selling cigarettes to young people (for asummary see Toumbourou et al.).14 Successfulstrategies include social marketing to ensureregulations are well understood, the use of minorsas confederates to monitor retailer compliance, andthe use of graded penalties (enforcement officerscan be reluctant to enforce very large fines) andpositive feedback for compliance. Recent evidencesuggests programs have been successful atrestricting sales to minors, and that such restrictionin availability translates to reductions in youthtobacco use. An important evaluation wasconducted by Forster et al. who described arandomised, controlled trial in 14 Minnesotacommunities where a combination of strategiesreduced illegal tobacco sales.504 After 32 months,rates of daily smoking in year 8 to 10 students inthe intervention communities were 10% comparedwith 16% in control communities. If restricted toone geographic area, there is a need to consider theextent to which sales might occur in othercommunities.774

Alcohol


While it is well established that raising the drinkingage reduces overall levels of alcohol related harm,775

changes to the uniform Australian drinking age of18 years is unlikely given a lack of communitysupport.186 It has been noted that strict enforcementof existing laws on minimum drinking age tends tobe an effective means of reducing harm withoutincurring controversy, and receives strong supportfrom the wider Australian community.186 A largemulti-site randomised trial of the enforcement ofminimum drinking age laws in the USdemonstrated a significant reduction in youth accessto alcohol.776 There is great scope for preventionthrough greater attention to this issue as there isevidence that compliance with these laws is at bestpatchy in Australia, rendering youth access toalcohol comparatively easy.186

Cannabis and illicit drugs

Summary: Warrants further research ................... �

Evidence that is more fully reviewed in Chapters 13and 14 suggests there is potential to further evaluatethe impact on youth cannabis and illicit drug use oflaw, regulation and policing strategies. There iscurrently some debate regarding the effectiveness oflaws that prohibit cannabis use. Such laws may havea deterrent effect for some youth, conveying amessage that cannabis use is not socially approved.However, criminal penalties also run the risk ofcreating harm and social alienation where youthdrug use results in a criminal offence. In balancingthese considerations, the possibility of reducingfrom criminal to civil penalties appears feasible. InStates that have relaxed criminal penalties forcannabis use, there is no evidence that cannabis usehas increased. In cases where youth are chargedwith illicit drug use, there may be opportunities toreduce escalation to harmful drug use throughdiversion programs. Such programs providetargeted entry to strategies such as familyintervention and preventive case management.

;'&('&� � �+)-�� -?+��/��)+�� +��

Setting and enforcing laws and regulationsregarding the minimum age at which youth canpurchase and use tobacco and alcohol appear to beeffective in delaying initial use. To be effective,strategies require a coordinated approach, includingsocial marketing to ensure regulations areunderstood, the use of minors as confederates to

�

��# �� !��"�#��$��%��

monitor retailer compliance, the use of gradedpenalties and positive feedback for compliance.Further research and evaluation are required tobetter establish the influence of laws, regulation andpolicing strategies on youth involvement incannabis and other illicit drug use. The applicationof programs to divert youth apprehended by policeand the courts into effective intervention servicesmay be a promising early intervention strategy butto be effective, such strategies will requireconsiderable coordination between services andadministrative jurisdictions.

'��"�2 �� ;��$��

Considerable progress has been made over the lastdecade in the identification and evaluation ofstrategies that can successfully prevent patterns ofadolescent drug use associated with later harms.However, there are few studies that have completedfollow-up over a sufficient period to demonstratethat reductions in early drug use translate tosubsequent reductions in harmful drug use.

The focus of research investment has been largelyskewed toward school-based drug education. Otherstrategies have received less attention. Evidencesuggests that there may be particular advantagesthrough the integration of more than one healthpromotion strategy. The enforcement of laws andregulations appears important in reducing early agetobacco and alcohol use but implementation mayrequire coordination and support at the communitylevel. Drug education campaigns conducted in thecontext of community mobilisation can besuccessful, though at this stage the effect sizes arenot necessarily larger than those achieved throughhealth education alone. The work of Flynn andcolleagues suggests that combining drug educationwith social marketing may offer advantages inreducing youth tobacco use.

Some evaluations have been reported showingpromise in engaging parents and families intointerventions and in demonstrating outcomesfollowing intervention. Existing studies mainlyprovide evidence for impacts and longer-termfollow-up will be required to establish whetherfamily interventions can reduce drug abuse.

Few studies have evaluated outcomes relevant toillicit drug use. More intensive strategies, includingfamily intervention and preventive casemanagement, appear promising for preventingharms associated with illicit drug use ininterventions targeting adolescents with a highnumber of risk factors.

��%��

��"�� (��!!��

A number of prevention strategies target harmful drug use as one component within a broaderset of prevention goals. Examples of such broad-based prevention approaches include crimeprevention initiatives, education strategies and health and mental health promotion strategies. Inthis chapter several broad-based strategies are reviewed, focusing on their relationship to policiesand programs aiming to prevent harmful drug use. Many broad-based strategies, such as crimeprevention and health promotion strategies, target drug use explicitly. Other strategies, such associal improvement and employment strategies, contribute to the reduction of harmful drug useby addressing common developmental determinants. In some cases, there may be conflicts andtensions between the goals of different prevention programs. Tackling crime throughincarceration, or through poorly designed health education programs, has the potential toincrease drug-related harm. Efforts to prevent harmful drug use need to be well integrated withbroad-based prevention efforts. The immediate targeting of drug use and developmental risk andprotective factors, articulated within various broad-based strategies, is summarised in the tablebelow.

= ��@�� !�� !��A 4�� / ��&��

5�� (�&�� (�&��

�� :@��

��

'�� @��:=�� 9� ��

�� 8�� 9��

+� � �� $��

*��+ ��

�� 8�� 8��

!�"�� 8��

��+ �� #��

��

( ��'�'��

) �� 8��

��

� -�� %

C -��$��

��' �� !��"�#��$��%��

(��+��

In other sections of the current document,community and health service strategies that havebeen developed and evaluated with the explicit aimof reducing harmful drug use are reviewed. In thischapter, several broad-based prevention strategiesare examined. There is a substantial degree ofinterrelationship between a number of these broad-based ‘prevention’ approaches and the prevention ofdrug related harm. This chapter reviews the overlapbetween these programs, the common risk factorsthat are addressed and, where possible, the questionof how much impact these programs have onharmful drug use.

(��=� � 3��$�� $��

<'0'&�� -�� +,�,��-

Evidence that the first years of life are critical inshaping later development has led to an increasedemphasis on investment in these years. Examplesinclude work in the US with Head Start and, morerecently, Fast Track (summarised in Chapter 7);work in Ontario in Canada flowing from the EarlyYears Study Group report;482 and in the UK with theSure Start program. Each of these programs ischaracterised by efforts to coordinate largeinvestments from different areas of government to‘guarantee’ that the early development of childrenwill not be compromised by failure to receive basicneeds for nutrition, security, or learningopportunities. In the case of the Early Years StudyGroup report, recommendations extended tochanges in pre-school systems to enhancedevelopmental potential for intelligence moreuniversally.482 The impetus for these investmentsincludes brain research that has graphicallydemonstrated the role of adequate nurture in earlybrain development, developmental research trackinglater life problems to failure in early development,and evidence that interventions in the early yearsmake a difference later in life. In later sections, theevidence for interventions early in life is reviewed.In overview, the existing intervention findings arepromising but there have been few studies withmore than one to two years of follow-up and theseexceptions have tended to be small studies.Although investment in this strategy should offerbenefits, the current evidence suggests thatreductions to drug-related harm may be limited ifprevention investment were to exclude risksemerging at other life stages. In Chapter 6, follow-

up studies were summarised that demonstrate thatrisk factors for harmful drug use arise in childhood,adolescence and later in life. Further evidencesuggesting that risk factors arise later in life emergesfrom well-controlled evaluation studiesdemonstrating that reductions in harmful drug usecan be achieved through preventive investment inprimary school and in later years of development.The Early Years Study Group report is an impressivedocument and has important implications forinterventions to encourage the development ofintelligence. However, the summary of evidence inthat report concluded that crime preventionoutcomes could not be obtained unless investmentwas made prior to the child’s entry to school. Thatconclusion was based on a highly selective reviewof evidence and is at odds with the present reviewand also with the conclusions of other reviewers.472,

777 Furthermore, there is evidence that whereinvestments in early years programs have beenapplied beyond families with high risk factors forchild neglect, there may be few gains to be madethrough such investments.574

<'0'.��)�� -��-

Various Commonwealth and State Governmentprograms aim to influence school and educationenvironments in ways that are likely to have directand indirect impacts on drug-related harms. Forexample, the Commonwealth Government school-based drug prevention policy emphasises preventionthrough education, information and safe andsupportive school environments.778 In each State,there are a range of important programs that aim toenhance school learning environments and promotestudent health. The Education for Resilience reportprovides a summary of the evidence for a range ofAustralian and international school-based programsfocusing on positive youth development.779 Thereport summarises the underlying principles ofprevention and intervention, and describes variousstructures, programs and strategies. The approachemphasised in that report is congruent with adevelopmental pathways approach to prevention.The approach emphasises the improvement ofschool environments through evidence-basedinterventions that focus on reducing risk factors andenhancing protective factors.

A summary of the current evidence for risk andprotective factors influencing harmful drug use waspresented in Chapter 6. It included the finding thatpoor academic achievement, beginning in primaryschool, is a risk factor for harmful drug use. InChapter 8, evidence was summarised demonstratingthat drug education programs in secondary schools

��(��

prevent harmful drug use. Failure to completesecondary school and unemployment after leavingschool are both predictors of illicit drug use.However, these predictors do not appear to bedirect risk factors in that their influence on drug useappears to be explained by earlier determinants suchas childhood behaviour problems, early academicfailure and an earlier age of drug use initiation (seeChapter 6). It is likely that education programs thataim to improve academic outcomes and that beginin primary school may also result in a reduction ofharmful drug problems. It is possible thatimproving employment rates will exert somebeneficial impact on drug use problems but therehas been less research on this question.

A range of Commonwealth and State programs aimto enhance school effectiveness and improve schoolenvironments.779 The evidence base for preventionstrategies focusing on primary and secondaryschools was summarised in Chapter 8. In overview,the evidence suggests that well-conductedprevention programs in primary schools may beamong the best investments for addressing thedevelopment of harmful drug use. These programsappear to reduce risk factors such as negative peerinvolvement, conduct problems, aggression,victimisation, early age drug use and low schoolattachment. They may also work to enhanceprotective factors by increasing pro-social andresponsible behaviours within peer groups, andbonding to teachers, parents and school. Outcomesfrom these programs may be relevant to a range ofrelated prevention priorities, including crime,violence and mental health.

The evidence is also very strong that investment ineffective drug education in secondary schools cancontribute to preventing harmful drug use (Chapter8). Effective programs in secondary school appearto reduce a range of risk factors that are moredirectly related to drug use. Risk factors that appearto be reduced by drug education programs includefavourable attitudes to drug use, estimates of theprevalence of peer and adult drug use, specificsocial skills related to drug use, and communicationwith and attachment to parents. Programsaddressing secondary school organisation andbehaviour management may also influence a rangeof important risk and protective factors althoughthere has been less work in this area.

<'0'0�#��+�-- �--�-��-

There is a strong relationship between homelessnessand substance use.780 Substance use problems cancause or contribute to homelessness, and vice versa.Australian research has identified that approximatelyhalf of homeless Australians have a drugdependence problem, with heroin being the mostcommon. Equally, half of Australian homelesspeople have been estimated to have a mentaldisorder. The Alcohol and other Drugs Council ofAustralia (ADCA) conducted a review of an earlydraft of the Commonwealth Homelessness Strategy,in late 2001. This review concluded that theHomelessness strategy should include three keypriority actions related to drug use and thehomeless. These are:

� provide support to families at risk of breakdownin times of crisis,

� fund specialist services for hard-to-reachhomeless clients,

� provide more low-cost private and publichousing.

No comment is provided on evidence for theeffectiveness of these interventions.

In a large representative survey of Victoriansecondary school students, Bond et al. examined therelationship between developmental risk andprotective factors for youth drug problems; anindex was designed to measure risk forhomelessness (e.g. running away from home andconflict with parents).781 The report found that thecumulative number of elevated risk factors anddepressed protective factors were associated withincreasing rates of homeless risk. This evidencesuggests that programs that are effective at reducingrisk factors and enhancing protective factors foryouth involvement in harmful drug use will haverelevance for efforts to prevent homelessness. It isalso apparent that efforts to reduce youthhomelessness will have unique components thatmay be unrelated to efforts to prevent drugproblems. Strategies that reduce homelessness mayalso have a positive impact on drug use problemsbut there also exists the potential for policyconflicts. For example, where accommodationpolicies lead homeless youth to aggregate withyouth with a high number of risk factors,involvement in harmful drug use might increase.Readers are encouraged to refer to the chapter onsocial determinants for a discussion of the nexusbetween homelessness, social estrangement andproblematic drug use.

�") �� !��"�#��$��%��

<'0'(� ��

A large proportion of this section is indebted to thework of Don Weatherburn of the NSW Bureau ofCrime Statistics and Research. He and his colleagueshave produced a literature review on the efficacy ofcrime prevention and mitigation in reducing illicitdrug use, which will be cited liberally along withrelevant content identified by the review processand consulted experts.397

Another key contribution is derived from the workof Ross Homel and colleagues on the Pathways toPrevention series.9, 472, 782 As this document isprimarily concerned with the primary prevention ofcrime in young people it is central to significantinteractions between crime prevention policy andthe prevention of harmful drug use. The seriesacknowledges that it is firmly grounded in thedevelopmental pathways approaches,8 whichfocuses on modifying early risk and protectivefactors through interventions directed at an earlystage in the developmental pathway leading tocrime. To the extent many of the determinants thatlead young people to engage in crime are commonwith those that lead to harmful drug use, thedevelopmental approach to crime preventionprovides an important model for advancing theprevention of drug-related harm.

Risk factors for violence are very similar to riskfactors for harmful youth drug use. These includefactors undermining healthy childhooddevelopment including inadequate parenting,troubled family relationships and poor schoolachievement.9, 472, 782 Early age involvement inalcohol and drug use is also a risk factor for thedevelopment of crime and delinquency. Lifestylesinvolving harmful drug and alcohol use are closelyinterrelated with crime.397 Accordingly, crimeprevention programs have a great deal of overlapwith programs preventing drug related harms.General recommendations for violence preventionprograms include the notions that programs addressmultiple risk factors and enhance protective factorsat the community, family, school and individual/peer levels. Programs are also recommended to bemaintained across the course of development.397

Hence the developmental pathways approach tocrime prevention is congruent with the Protectionand Risk Reduction Approach recommended in thepresent document. The development of mechanismsto link investment and program delivery in crimeprevention with efforts to prevent drug-relatedharm is warranted.

Broad crime prevention programs often includeearly intervention programs with disadvantaged andat-risk youth.397 Programs include a substantialfocus on increasing social inclusion and offeringyouth the possibility of participating fully in socialand economic life. Such programs attempt topromote the attachment of individuals andcommunities to mainstream social supports.782

These approaches, once again, overlap with thosethat are emphasised through the Protection and RiskReduction Approach to preventing harmful druguse.

There is also a broad literature on preventing crimefrom a variety of approaches other than thedevelopmental pathways approach. Traditionallabels for these include criminal justice, situational,community or social approaches.9 Criminal justiceapproaches refer to traditional deterrence andincarceration approaches, as emphasised by thecriminal justice and policing system. There ispotential for synergy with this category of crimeprevention. There is also potential for conflictwhere, for example, policies such as incarcerationlead to elevated rates of drug-related harms. Therelationship between criminal justice policies anddrug policy are considered more fully under thesections on supply reduction.

Methadone maintenance therapy (MMT) has beenshown by Cochrane reviews to be an effectivemeans of reducing drug-related criminal behavioursin opiate users,783 including property crime andactual heroin consumption levels. Significantly, thiseffect is demonstrated across cultural and ethniccontexts and varying study designs. Accordingly,methadone maintenance therapy can be considereda crime prevention initiative; further demonstratingthe potential for additive benefits through well-coordinated prevention investment.

(��=� � 3��$�� $��

<'('&�#��+�� -��-

Australia has a well-developed public health systemthat identifies and places priorities on areasconsidered to significantly affect population health.Initiatives such as the National Public HealthPartnership and the National Primary PreventionStrategy assist in the planning and coordination ofnational public health activities. A series of reportsto Commonwealth and State Health Ministersaddress national health priorities and these currentlyinclude cardiovascular health, cancer control,mental health, injury prevention, musculoskeletaldisorders and diabetes mellitus.

�"��

<'('.� ��-�)+��-��-�� +��

Cardiovascular health was identified as a target inthe first National Health Priority Area (NHPA)report, developed in 1996; and reported on in1999.784 The NHPA reports acknowledge theconsiderable overlap between the various priorityareas in the factors that contribute to risk and thebarriers to better prevention and care.785 Tobaccoand alcohol misuse are acknowledged as risk factorsfor cardiovascular disease and other heath problems.Monitoring demonstrates efforts are succeeding inimproving cardiovascular health in Australia.786 Inthe intermediate monitoring for this program,successes in reducing adult tobacco smoking and thefailure to reduce adolescent smoking rates have eachbeen noted.785 A common theme in national healthimprovement efforts is the recognition of theimportance of socioeconomic factors in behaviouralrisk modification.

A recent initiative of the Joint Advisory Group(JAG) in General Practice and Population Health isthe Smoking, Nutrition, Alcohol and Physicalactivity (SNAP) Framework. This initiative has beendeveloped by JAG in consultation with Chairs ofNational Population Health Strategies.787 The SNAPframework is intended to, ‘guide theimplementation of integrated approaches tobehavioural risk factor modification in generalpractice focusing on smoking, nutrition, alcoholand physical activity’ (p1). The program, whichaims to coordinate State and national efforts,emphasises a system-wide approach to support themanagement of behavioural risk factors in generalpractice. The program should influence the extentto which the evidence base for behaviour changerelevant to harmful drug use is applied withingeneral practice. Research and evaluation is includedwithin the broad outcome areas; however,evaluations of this program are not yet available.

<'('0� � �� -��-��

Several policies on cancer prevention are relevant todrug use. Some of these policies and approaches tocancer prevention include strategies to reducetobacco use and harmful alcohol use, and in thesecases, initiatives overlap with those relevant to druguse prevention. Initiatives relevant to preventingdrug use include: manipulating the price ofcigarettes and the promotion and advertising ofcigarettes, enforcing age limits on cigaretteconsumption, improving broad indicators of SES,improving opportunities in life through training

and employment, addressing social exclusion,assisting people to quit smoking, providing smoke-free workplaces, health education campaigns,developing healthy workplaces and healthy schools,and providing sensible drinking advice.783

Prevention of cancer in relation to cigarette smokingis almost entirely a function of encouraging peopleto cease use, and reducing the number of peoplecommencing use. Australia’s National TobaccoCampaign, a significant recent initiative, wasfocused on getting adults to quit, but does havesome preventive effects by helping to ‘de-normalise’ smoking.788 The campaign wasconducted by the Commonwealth, with advice fromcancer prevention organisations and other non-government organisations, yet its goal was theprevention of uptake of drug use and the reductionof existing drug use, which are two of the primarygoals of national drug policy. The effectiveness ofspecific anti-tobacco interventions is reviewedelsewhere. All initiatives encouraging reductions insmoking, or limiting drinking to withinrecommended consumption guidelines, can beconsidered cancer-prevention initiatives.

<'('(�� >)�,��

Some interventions that have been considered in thecontext of injury prevention policy are alsointerventions that are of relevance to drug policy.These include reducing drink-driving and providingalcohol treatment services.783 The National InjuryPrevention Advisory Council reports that itconsiders injuries from road trauma, suicide, falls,fire, drowning and assault to be among the manytypes of injuries it is concerned with preventing.789

These injuries all have some degree of overlap withdrug use, particularly alcohol and, accordingly,there is significant overlap between national injuryprevention approaches and drug and alcohol harmreduction policy.

Injury prevention strategies often encompassprograms aimed at reducing road injury,783 a fieldthat includes, as a substantial component, programsaiming to reduce the risk of drink-driving. Beingdependent on alcohol dramatically increases the riskof death through injury, including by falling, fire,or burns.790 In the elderly, fall injuries are especiallyassociated with problematic alcohol use.167

A systematic review considered the impact oftreatment for problem drinking on injuries.790 Itdetermined that treatment for problem drinkingmay reduce injuries generally and that treatment forconvicted drink-drivers would reduce subsequentmotor vehicle crashes. The studies on injury

�"� �� !��"�#��$��%��

prevention all demonstrated significant effects andlarge effect sizes. However, small sample sizes usedin the studies limited the ability to generalise fromthese results.

Alcohol is believed to be causative in many injurytypes including road trauma, falls, fire injury,drowning, assault and to some degree suicide.789

Population groups at high risk of injury includeyoung men, the elderly, and people in rural/remoteareas.789

A number of the strategies that have beenrecommended for implementation in injuryprevention initiatives also appear relevant to theprevention of drug-related harm. Night patrolschemes in Indigenous communities have beensingled out as an effective intervention worthy offurther support in Commonwealth policy,789

although the report did note that further evaluationof their impact on injuries is needed.

Other approaches that have received endorsementfor injury prevention include:789

� enforcing responsible service legislation toreduce alcohol-related injury,

� further work into the more widespread use ofbrief interventions by doctors to reduceproblem drinking,

� the continued use of Australia’s successfulrandom breath testing programs to reduceinjury related to driving under the influence ofalcohol.

Interestingly, none of the literature on injuries andfalls in the elderly mentioned either alcohol or anyother form of substance use as an issue.789

In summary, injury prevention strategies accept thatreducing harmful drug and alcohol use wouldreduce the rate of injury; as such, they directlytarget risky use of various substances. Therefore, theactual interventions intended to reduce injury inmany cases overlap with the types of interventionsrecommended in efforts to prevent harmful druguse.

<'('6�#��+��)��

In many social and health service settings, acommon strategy has involved the provision ofinformation aimed at encouraging healthybehaviours. Many of these generalist healtheducation initiatives include information relevant toavoiding harmful drug use. In other sections of thisreview, aspects of health education are examinedunder sections dealing with programs more

specifically targeting drug use behaviour. Relevantstrategies include school-based drug education,community-based drug education, social marketing,and self-help materials designed for drug treatment.In this section, we briefly examine evidenceevaluating broad-based health education programs.

Health education has been a popular healthpromotion strategy for quite some time. There isnow a reasonable evidence base, derived fromCochrane reviews on various topics, andunfortunately, health education as a general strategyis fairly ineffective in changing behaviour, althoughsome studies have shown small benefits.783 Whilsthealth education approaches tend to result inimprovements in knowledge and awareness, theevidence base for claiming that education onlyprograms contribute to actual changes in behaviouris fairly weak.783 There is some evidence thateducation only programs may have a limited effectin altering the behaviour of higher SES groups butnot lower SES groups.783

The evidence reviewed earlier on school andcommunity programs suggests there may be moresupport for health education as a component in abroader set of behaviour change interventions. Ifpoorly designed, there is the potential for healtheducation information to conflict with efforts toprevent harmful drug use. For instance, healtheducation materials that convey information abouthow to access new drug experiences or that lead toupward estimation of the level of perceived peerand adult drug use may be contra indicated. Thus,there is significant potential for effective investmentopportunities to be lost to ineffective healtheducation. However, it is rare to find societalgroups actively opposing health educationprograms. Those proposing ‘self-evident’ healtheducation investments as a method of preventingdrug use problems tend to find support from publicand private funding agencies. To address thissituation, a wider understanding of what is requiredfor effective behaviour change needs to bedisseminated to the broader community. There ispotential for government investment in healtheducation to advance the evidence by includingbehavioural evaluation in service delivery contracts.Directions for designing such evaluations werediscussed in the earlier section on social marketingprograms targeting youth.

�"��

<'('7��)�� +�� +-

Given the strong link between SES and many healthproblems, including drug use, there is someliterature on attempting to address the impact of SESon health differentials. Cochrane reviews havereached the brief and somewhat vague conclusionthat structural and legislative measures appear to bethe most effective means of reducing healthinequalities.783 The source article for this claimreported on a review of 129 interventions aiming atreducing socioeconomic health differences. It notedthat most interventions were not evaluated and thathealth education was the most commonintervention. However, health education aloneappeared ineffective and was only effective whencombined with personal/social support andstructural initiatives.791

The review also concluded that there is not yetsufficient evidence on which to base rational policyto reduce socioeconomic health differences.791

There is also no evidence on the impact of tacklingsocial exclusion on any measure of health status.783

Some places in the US have conducted trials ofincome supplementation approaches. However,unfortunately, there is no experimental evidenceavailable that allows us to assess the impact of suchinterventions on physical or mental health.783 Astronger integration between service delivery andresearch could achieve important gains. It would bepossible to evaluate the community impact ofgovernment investments aimed at redressing socialinequality and exclusion if such investments wereintroduced in phases, and assessments wereconducted to compare community benefits. Suchevaluation should be considered a priority given thepotential gain that could follow improvedknowledge of effective community investmentstrategies.

(�"�� ! �� $��

There are various strategies for promoting goodmental health that are also relevant to theprevention of drug-related harm. These includereducing social exclusion, reducing alcohol anddrug problems, assisting people to developparenting skills, improving social support networks,improving education and training opportunities,reducing homelessness, reducing unemployment,and promoting healthy work environments.783 Theevidence base consists of a number of literaturereviews, a review of Cochrane reviews relevant tothe public health agenda783 and documentation

supporting and describing government policystatements. The mental health promotion literatureis largely grounded in theoretical understandings ofthe promotion of health and the prevention of ill-health.792 Comprehensive literature that providedinformation on the actual impact of mental healthpromotion programs, on either mental healthproblems or drug use problems, could not belocated.

Papers reviewed include a literature review of earlyintervention in the mental health of youngpeople,793 the National Mental Health Strategydocuments,794, 795 The Cochrane Collaboration783 andpapers prepared for the National Co-morbidityWorkshop.179, 376, 796, 797

Mental health promotion is a broad termencompassing a range of programs that aim toimprove mental health and wellbeing.795 It,therefore, includes primary prevention of mentalhealth disorders, early intervention and treatmentapproaches.

Mental health promotion activities aim to protect, support andsustain the emotional and social well-being of the populationby increasing the protective factors that lead to positive mentalhealth outcomes across the entire intervention spectrum, thatis, before, during and after the onset of mental illness (p34).793

Interventions aimed at improving mental health andwellbeing could conceivably reduce drug-relatedharm. Mental health problems and drug-relatedproblems are associated with the same socialdeterminants and risk factors.797 Therefore,programs that target these shared factors anddeterminants may improve mental health andreduce the risk of developing drug-relatedproblems.

In addition, both current conceptualisations ofmental health diagnostic systems (DSM-IV and ICD-10) consider drug use problems to be a mentalhealth disorder. Accordingly, the National MentalHealth Strategy794 is itself a form of interventionaimed at drug problems.

Mental health promotion strategies target the samebasic issues as do broad primary preventionstrategies: that is, to target the structuraldeterminants of ill-health at a general level.793 Bytargeting factors such as socioeconomic gaps,unemployment, social capital, the physicalenvironment and social beliefs and values, it istheorised that such strategies will have positiveimpacts on both drug-related harms andpsychosocial disorders.419 Evidence from the presentreview confirms that more extreme poverty and

�"� �� !��"�#��$��%��

social disadvantage are risk factors in predictingdrug-related harm.

Cochrane reviews have determined that grieftherapy does not yet have sufficient evidence ofeffectiveness to justify its routine use as a mentalhealth promotion strategy.783 This is of relevancegiven the predominance of grief and loss as anantecedent to late-onset drinking problems in theelderly.

It is well established in the field that single factorshort-term mental health promotion interventionsrarely show any effect.419, 793 Accordingly, themental health promotion field is concerned withbroad-based, multi-level, multi-systemic approachesto the promotion of health and wellbeing. In sodoing, it is addressing some of the same centralissues as are addressed in drug policy. There areopportunities to improve the synergy betweenmental health promotion and prevention programsaddressing drug-related harm. One important stepforward is for drug policy to acknowledge thecommon relevance of the developmental pathwaysapproach to prevention.

<'6'&�� +�� +��+��-��,

What follows is a broad overview of the NationalMental Health Strategy,794 with specific commentson those elements of the strategy with a high degreeof relevance to preventing drug problems. TheNational Mental Health Strategy is firmly based in adevelopmental pathways approach and focuses ondevelopmental risk and protective factors for mentalhealth. Within the strategy, ‘prevention’ is definedas interventions that occur before the initial onset ofa disorder, to prevent the development of thedisorder. ‘Early intervention’ comprisesinterventions that are appropriate for, andspecifically target, people displaying the early signsand symptoms of a mental health problem ormental disorder; and people developing orexperiencing a first episode of mental disorder. Thetheoretical framework has been influenced by theInstitute of Medicine framework8 that is alsodirected at drug abuse prevention and crimeprevention.

The National Mental Health Strategy assumes thatthe determinants of mental health status are firmlylinked to a range of social determinants, includingincome, employment, poverty, education and accessto community resources. As is indicated in othersections, these factors are also linked to drugproblems.

These social determinants are seen to translate at theindividual level into risk and protective factorsinfluencing the risk of mental health problems. Suchrisk and protective factors exist at the genetic,biological, behavioural, psychological,sociocultural, economic, environmental anddemographic levels. The strategy asserts thatprevention of mental health problems is more likelyto be successful if it considers multiple risk andprotective factors and if it simultaneously includes arange of interventions targeting these multiple riskfactors.

The National Mental Health Strategy recognises thatthe majority of risk factors for mental healthproblems lie outside the traditional domains ofmental health treatment. The strategy recognisesthat an appropriate intervention approachencompasses intervention at various areas of riskand protective factors: the individual, family,community and society level. Interestingly, one ofthe main goals of the strategy is to reduce drug use(along with many other specified mental disorders).The strategy, therefore, recognises that many of therisk and protective factors and social determinantsfor problematic drug use and for mental healthdifficulties are shared.

Included in the strategy are the following broadprinciples of general mental health promotion:

� creating supportive environments,

� strengthening community action,

� using the media to shape attitudes,

� developing personal skills,

� incorporating promotion, prevention and earlyintervention within mental health services,

� education and training on the promotion ofgeneral health, and

� research and evaluation.

Although the National Mental Health Strategyattempts to address many of the risk factors for drugproblems, there is no evidence that describes theactual impact of this strategy, or of broad-basedprevention programs in general, on levels of mentalhealth problems or substance-related harm.

<'6'&�"��--��

Beyond Blue is an Australian Government fundedinitiative aimed at preventing depression, guided bythe principles of the National Mental HealthStrategy. The Beyond Blue Annual Reportacknowledges that prevention programs aimed at

�""��

preventing drug-related harm may contribute toalleviating or preventing depression. A preventionprogram run by Beyond Blue, a school andcommunity program in which elite athletes deliverdepression awareness and guide young people intodesigning and implementing a community project,mentions ‘drug awareness’ as one communityproject topic.798 Overall, drug-related harm is onlyperipherally acknowledged and is not included as anindex of the program’s success. Therefore, it is notpossible to adequately evaluate whether BeyondBlue will lead to reduced drug-related harm.

<'6'.��)��

Suicide is widely understood to be a complexphenomenon caused by the interaction betweenbiological, psychological, social and culturalfactors.799 Factors contributing to both suicide riskand risk of drug-related harm include alienation,physical isolation and depression.419 The CanterburySuicide Project confirms the link betweendepression and suicide; over 60% of participantswho made serious suicide attempts had beendiagnosed with depression in the month prior,compared to less than 6% of controls.800 Beautrais,in her review of risk factors for youth suicide,reports that young people are on average 12 timesmore likely to commit suicide if they are diagnosedwith an affective disorder, and 5.5 times more likelyif they have a substance use disorder.801

Subsequently, suicide prevention programs that aimto prevent depression by mental health promotioncould conceivably impact upon both suicide ratesand drug-related harms.802 Many suicide preventionprograms also include goals of reducing harmfuldrug use.803

The National Youth Suicide Prevention Strategy hasbeen the basis for suicide prevention programs inAustralia including parenting programs, school-based inventions, early intervention programs forhigher-risk populations, promoting mental health,and reducing access to means of suicide.803 Thestrategy acknowledges that young people who usedrugs are at higher risk of suicide but the resultingprograms do not target the reduction of drug-related harm as a relevant index for evaluation.Thus, it is not possible to tease out any preventiveeffect of these initiatives upon drug-related harms.Nor is it possible to comment on the impact of thisstrategy on the prevention of suicide.

<'6'0�� 5��*��,

There is very little evidence on the best approachesto treatment for those with co-morbid disorders.376,

804 Cochrane reviews have identified that at presentthere is no evidence of superiority for any form oftreatment in regard to co-morbid mental health andsubstance use problems.783 There is moderatelystrong evidence that co-morbid mental healthproblems and substance use problems exert multi-directional causality, that is, mental health problemsare likely to worsen substance problems and viceversa.197, 797

<'6'(��)��,?�� +��+��

Similar Foundations: Mental health promotion in thepopulation rests on the same principle as broad-based prevention addressing drug-related harm: thatof ameliorating the risk factors predisposing peopleto harmful drug use, which are largely a function offactors such as developmental risk factors,unemployment, socioeconomic disadvantage andtraumatic backgrounds.419

Treatment: There is a strong theoretical case thattreating mental health problems is likely to beeffective in reducing substance use, even in theabsence of treatment for substance use problems.We did not, however, identify any direct evidencethat treating mental health problems leads to areduction in substance use.

Direction of influence: There is no evidence that mentalhealth promotion programs reduce substance use inadults, or that reductions in rates of mental healthproblems have a positive impact on rates ofsubstance use problems.796 Mental health promotionprograms, as with drug use primary prevention, arelargely aimed at children because the peak age ofonset for both problems is late adolescence. Theearlier chapter summarising risk and protectivefactors for drug use showed that in some cases,child mental health problems (particularlyexternalising problems) increased the risk ofsubsequent involvement in harmful drug use. Wehave also summarised evidence in earlier sectionsshowing that early age tobacco and cannabis usecontribute to the emergence of mental healthproblems later in life. The available evidencesuggests that mental health investment should formone component in programs aiming to prevent earlyage or regular drug use in adolescence.

�"# �� !��"�#��$��%��

(�#�2 !!��!�� $�!��

Community prevention initiatives have beendeveloped to directly target drug use issues. Theseprograms are considered in detail in Chapters 8 and10. Community drug prevention programs arebased in a broader body of work addressingcommunity improvement. The literature in thisfield is categorised variously as communitydevelopment, improvement, and prevention.Community prevention programs have typicallyinvolved a wide variety of initiatives encompassingall three of the National Drug Strategy categories ofdemand reduction, supply reduction and harmreduction. Programs include aims that encompassbroad-based prevention and result in multi-facetedstrategies. Community prevention initiatives largelyfocus on changing adult behaviour and thestructural issues that support and maintain drugconsumption.

The Stronger Families and Communities Strategywas launched by the Commonwealth Department ofFamily and Community Services in April 2000.805

The strategy’s three main priorities were:

� early childhood development and the needs offamilies with young children,

� strengthening marriage and relationships, and

� balancing work and family.

Although none of these areas immediately addressesyouth issues, one of the eight principles underlyingthe strategy is prevention and early intervention.The strategy makes clear that this involves helpingfamilies early on to prevent later problems such asdomestic violence, youth suicide, homelessness anddrug addiction. The Australian Government also setup the Youth Pathways Action Plan Taskforce inSeptember 1999, with the goal of developing a fiveyear plan to help young people make the transitionto adulthood. As part of this plan, the Governmentis funding various programs, including theStrengthening and Supporting Families Coping WithIllicit Drugs program and the Schools DrugEducation Strategy. These broad investments incommunity strengthening are also reflected in Stateprograms that have a similar focus. Although thecontribution that community improvementinitiatives make to reducing drug-related harm isunknown, there are theoretical reasons to expectthey should enhance protective factors for positivechild and youth development and, more generally,improve wellbeing.

(�%�2 �� ;3� � 3��$��

Broad-based prevention strategies vary in theirdirect or indirect relevance to the prevention ofdrug-related harm. The developmental pathwaysframework, with its emphasis on reducingdevelopmental risk factors and enhancing protectivefactors, is an important underlying framework thathas the potential to improve coordination betweencrime prevention, mental health promotion anddrug use prevention efforts. There are furthersynergies with drug policy in the efforts toimplement brief interventions and screening inhealth and harm reduction frameworks withininjury and crime prevention policies. The Protectionand Risk Reduction Approach to preventionarticulated within the present document has thepotential to integrate these varying frameworks,providing an important basis for improvingcoordination between different preventionstrategies. However, there are potential areas ofconflict between different areas of policy,emphasising the need for good coordination. Acloser link between research and service delivery hasthe potential to strengthen prevention policy bybetter defining strategy combinations that areeffective in reducing drug-related harm.

�"%��

��

��

�"' �� !��"�#��$��%��

�"(��

��#�� )��!!��

Reducing the demand for both licit and illicit drugs has been attempted through universalstrategies such as mass media campaigns, selective strategies such as brief interventions by healthcare workers, and indicated strategies delivered by specialist treatment agencies. The evidence forboth direct and indirect impacts of these various strategies on levels of risky drug use and harm,at the population level, is reviewed in this chapter.

From a population-wide perspective, the level of investment in effective treatment programsneeds to be a key ingredient of comprehensive prevention policies. There is suggestive evidencefor population-level impacts of methadone programs on local levels of crime, and of alcoholtreatment on acute and chronic alcohol-caused health problems in the community. There is also amajor opportunity to intervene with family members of individuals experiencing serious alcoholand other drug problems both because this can increase treatment effectiveness, and in order tominimise the inter-generational transmission of mental health and substance use problems. Thereis strong evidence for some forms of treatment of drug problems, particularly for the legal drugs(for example, nicotine replacement therapy for smoking, cognitive behavioural therapy foralcohol problems). There is also some evidence for the value of matching different treatmentmodalities to different types and intensities of drug problems. Thus, there is a clear potential forincreasing the population-level impact of society’s investment in treatment programs on levels ofrisky drug use and harm.

Strong evidence exists for the effectiveness of brief interventions delivered by primary health careworkers for smoking and alcohol problems; providing a major opportunity to increase the uptakeof these strategies and hence their overall impact on population levels of risk and harm. (Andthere is strong evidence for-del) The effectiveness of interventions by health care professionals toreduce maternal smoking during pregnancy is also based on strong evidence. There is a strongrationale for such programs for the health of both the mother and child, and for interventions formothers who are engaging in risky alcohol and other drug use. Further research into theeffectiveness of interventions to reduce and prevent drug use during pregnancy is recommended.

There is a growing body of international research into the effectiveness of interventions to reducealcohol problems delivered at the community level. The evidence is quite strongly in support ofthose programs that effectively mobilise the community to support local structural policy change(for example, incentives for responsible beverage service on licensed premises). Limitedinvestment has been made in the conduct of rigorously evaluated community-based interventionsin Australia, whether directed at problems with licit or illicit drugs.

Interventions delivered in the workplace have an enormous potential to reduce risky drug useand actual harm, due to the ability to access subgroups at risk who would not otherwise seekspecialist help. Interventions found to be effective in other settings (e.g. brief interventions inprimary health care) have the potential to offer community benefits in the workplace but moreAustralian research is needed. There has been substantial investment in drug testing programs,especially in safety-sensitive work environments where there is a strong rationale for deterringthe use of licit and illicit substances. More research is required into the effectiveness of drugtesting in the workplace, especially where this includes risky use of alcohol.

More money has been invested in mass media campaigns on licit and illicit drug problems thanperhaps any other intervention in Australia. Evidence for effectiveness of this type of interventionis strongest in relation to tobacco use, especially when campaigns accompany other structuralpolicy changes such as increases in taxation. There is some evidence for the effectiveness ofcampaigns around the risks of alcohol use, again especially when these support other initiativessuch as drink driving law-enforcement. It is recommended that a degree of investment in such asocial marketing approach continues, utilising available evidence for best practice. Social

�

�#) �� !��"�#��$��%��

marketing strategies such as national drinking guidelines and standard drink labelling are alsorecommended on the basis that they facilitate other evidence-based strategies e.g. briefinterventions for controlled drinking and road safety campaigns. Examples of effective mediaadvocacy in Australia are provided.

There is a general need to investigate the effectiveness of the universal, selective and indicatedstrategies in special populations, Indigenous communities in particular.

�� !�� *�� 9�� !��%��

�� '�� /�� 0�� 9��

.�� !�� -�� +��$ �� 9

�� 9��

8�� 7 7 �� $ ��

;� B�� !�� <� � � <�� $�� 9� �� $��

� $ �� <� 7 7 <�� $�� 9� �� $��

8 ��$�� <� � � -� �� $� �� / %�%��09��

��

-�� 7 )�� $��/ %�%�� 0

��@�� !��+ ��

��

'�� 7 �� 7 8�� 6��$��

'�� <�� 9��

<�� <� � <� '�� ,�� 9�� <�'+'

$��@�� !�� 8� �� $�� <��

��!�� < �� !�� $

��

�� '� ��

� �&��& �� 8� �� '� � �� $��

�� E��

�#��

>��7 *��

� �� $�� #��

� ��

��

��

�)��!�� 3��!�

This section explores the impact of treatment ondrug use and drug-related harm as a genericapproach, rather than the minutiae of different typesof treatment. The literature base for this section isextensive, including reviews such as Cochrane andother systematic reviews.783, 804, 806–811 These reviewsgenerally have a high degree of evidence, allowingassertions to be made about the role and nature oftreatment.

Treatment is aimed primarily at people withproblems related to drug dependence. It should benoted that most people who use both licit and illicitdrugs (with the possible exception of smokers) arenot dependent,804, 811 nor are many frequent usersdependent.804 However, some doubt has beenthrown on the extent of so-called ‘recreational’heroin use, with most studies having recruitmentdifficulties.166 Dependence captures the notion of achange from voluntary to involuntary use;811

although, as has been demonstrated, dependenceexists upon a continuum of severity. In relation ofalcohol, signs of mild dependence are evident in asignificant number of the population of adultdrinkers who are not in treatment.812

‘Treatment’ is a generic term that covers a widerange of clinical interventions, including those thatmake contact with and engage users; those thatdetoxify; and those that manage the process ofwithdrawal from chronic drug use, pharmacologicaltreatment and psychosocial treatment.804 Clearly,different approaches will have different levels andrates of success with different individuals andpatterns of drug use, and variability is necessary toallow users to access a clinical intervention that iseffective for them.

&='.'&��*��?��

Summary: Evidence for effective dissemination...................................................................... ��

Nicotine replacement therapy (NRT) has beenshown to be effective and the evidence issummarised in a Cochrane review. NRT—includingpatches, gums, and inhalable nicotine—is associatedwith a small but consistent and statisticallysignificant increase in the likelihood of achievingabstinence, in contrast to controls.813

&='.'.��+��+?�� +��+��*+��-��


Treatment efficacy for alcohol use problems hasoften been judged primarily by subsequentconsumption levels. In relation to alcohol (but notnecessarily for other drugs), total level ofconsumption is an excellent predictor of thelikelihood of harm.806 Ancillary measures of theextent of alcohol-related problems in domains suchas family, work, relationships and health, have alsobeen applied. Furthermore, moderate or controlleddrinking outcomes have been examined,particularly in relation to drinkers with a lesserdegree of alcohol dependence.814

The literature base in regard to alcohol is very welldeveloped, and alcohol treatment research ingeneral is at the stage of determining best practicerather than attempting to determine if treatment canbe effective.806 Using this criterion, there are manyapproaches to alcohol treatment with demonstratedeffectiveness, and, equally, there are many that havefailed to show evidence of effectiveness.806 Effectivealcohol treatment options include: motivationalinterviewing, brief interventions, social skillstraining, community reinforcement approach,relapse prevention and some aversion therapies. Thelargest ever controlled study of different forms oftreatment for alcohol problems was the US ProjectMatch.815 Positive results were obtained from eachof the three main treatment modalities: motivationalenhancement, supportive psychotherapy, and 12-Step Facilitation Therapy. There was little evidenceof any advantage from any one of these modalities;nor was there much evidence of interaction effectssuch that some types of client did better withparticular treatments. There was, however, evidencethat more severely dependent drinkers withotherwise lower social supports did better with 12-Step Facilitation Therapy, which seeks to encourage

�

�#� �� !��"�#��$��%��

regular attendance at Alcoholics Anonymous (AA)meetings. AA is an enormous worldwide networkof self-help groups to help alcoholics achieveabstinence. Treatments with no evidence ofeffectiveness include insight orientatedpsychotherapy, confrontational counselling,relaxation training, general ‘alcoholismcounselling’, education and milieu therapy.806

Unfortunately, some of these approaches with littleevidence of efficacy are very popular and widelyused.

Pharmacotherapies for alcohol dependence includedisulfiram, naltrexone and acamprosate. There isgood evidence from treatment outcome reviews thatthe effectiveness of disulfiram is limited.806 On theother hand, naltrexone has been found to beeffective in reducing alcohol craving and drinkingdays, and to help to stop the resumption of bingedrinking.806 A systematic review of randomisedcontrolled trials, from 1960 to 1993, found thatacamprosate and naltrexone demonstrated bothsafety and efficacy in the long and intermediateterm respectively.783

&='.'0�� ++��)�5��+��*+��-��

Harm reduction approaches have become widelyincluded in treatment for illicit drugs in addition to,or instead of, the traditional treatment goal ofreduced or no use of drugs. Goals of treatment mayinclude reduced drug use, reduced risk of infectiousdisease and improved physical and psychosocialfunctioning.804

It is important to note that expectations of a quick‘cure’ are unrealistic.804, 811 Dependence is a chronic,relapsing condition requiring continuing care811 andtreatment needs to be seen as a long-termproposition with the goals being improved care andcontainment of problems rather than unrealisticexpectations of a complete cessation of problematicuse.804

There are various treatments that are used fordifferent types of illicit substance use problems.804,

806

These include:

� pharmacotherapies, such as methadone,naltrexone, buprenorphine anddexamphetamine;

� detoxification approaches, such as inpatientwithdrawal, tapered gradual withdrawal, homedetoxification, ultra-rapid opioid detoxification;

� counselling approaches, including briefinterventions, motivational interviewing,cognitive behavioural therapy, relapseprevention; and

� psychosocial interventions, including outreachprograms, residential rehabilitation, housingassistance, and employment assistance.

The effectiveness of treatment varies considerablyaccording to the type of: drug problem involved(patterns of consumption, user characteristics),drug involved, and treatment employed.804, 806

Dealing with dependence typically involves aconsideration of both the physical and thepsychosocial aspects of drug dependence.804 Forexample, it has been demonstrated that providingpsychosocial services to methadone clients providessignificantly better outcomes than providingmethadone alone.804 There is also evidence fromcontrolled trials that the provision of supplementalsocial services (medical screenings, housingassistance, parenting classes and employmentservices) improves treatment outcomes in adultsbeing treated for substance abuse.816 In a similarvein, the social functioning of the client beingtreated is a significant predictor of the outcome oftreatment. Professionals who are opioid-dependenthave a significantly better prognosis thanunemployed, poorly educated people using lesseramounts of opiates.811 Similarly, lack of treatmentsuccess is associated with low socioeconomic status,co-morbid psychiatric conditions, and lack of familyand social support.811 Taken together, these findingsemphasise the importance of considering a broadrange of social factors in responding to drug-relatedproblems. It should also be noted that there is a lowlevel of access to treatment among illicit drug usersin the criminal justice system: between 40 to 60%of DUMA 2001 respondents238 and 55% of DUCOrespondents379 had never been in treatment.

Opioids


The strongest evidence for the efficacy of illicit druguse treatment is found in the treatment of opioiddependence. Methadone maintenance therapy(MMT) has proven effectiveness in: retention intreatment, reductions in drug use, criminalbehaviour, mortality, and improvement in healthstatus, when adequate doses (> 60 mg/day) aregiven. Effectiveness is improved when treatmentalso addresses psychosocial issues.804 MMT is alsoassociated with a reduced risk of HIV infection.804

�#��

Other opioid substitutions that are less wellresearched than MMT but appear to be efficaciousare buprenorphine and LAAM.804 There is lessevidence in favour of naltrexone.804 Detoxificationand withdrawal from opioids without othertreatment is not effective and relapse into use iscommon.804

Treatment in general, is associated with significantsavings from the costs of crime. Extensive UK multi-site studies of treatment effectiveness reported anaverage reduction in total crime costs of £12 000per annum per participant.817 It was furtherestimated that for every pound spent on treatment,£3 savings from crime and criminal justice systemcosts was returned.

Psychostimulants


Pharmacological therapies for psychostimulants arewidely researched, mainly in the US for cocainedependency, but no broadly effectivepharmacological therapy has been identified.804

There appears to be no review literature, as yet, onthe use of dexamphetamine as a substitute for ATSdependence but an Australian feasibility studyconcluded that a multi-site randomised controlledtrial of the efficacy of amphetamine substitutiontherapy should be undertaken before the practicebecame too widely available.818 Other treatments forpsychostimulants include cognitive behaviourtherapy, which has been found in one randomisedcontrolled trial to be effective in moderating cocaineuse, and contingency management approaches,which have also been found to be effective inreducing cocaine use.804 Again, detoxification aloneis not effective.804

Cannabis

Summary: Treating cannabis dependence; evidencefor outcome effectiveness ................................. ��Brief interventions for illicit drug use; evidence forimplementation .................................................. �

There are few systematically developed treatmentsfor cannabis dependence.819 Some of the cannabisinterventions have been adaptations of treatmentapproaches developed for alcohol, based on thepremise that cannabis dependence is directlycomparable to alcohol dependence.820, 821 Themajority of the cannabis treatment packages thathave been subject to evaluation are cognitive-behavioural in orientation. A 12-step treatmentapproach for cannabis dependence has beenproposed,822 although its efficacy has not beenreported. A small number of self-help manuals have

been developed, some with an abstinence focus,823

some that handle both abstinence and non-abstinence goals,824 and some that focus on harmreduction with less emphasis on strategies forchanging cannabis consumption.825 Such materialsmay include suggestions on managing withdrawal,removing cues to smoking, self-monitoring,development of alternative responses and relapseprevention. As yet, none have been systematicallyevaluated although some have been developed withextensive piloting and feedback from experts andusers.826

A comprehensive review of treatment concludesthat there is insufficient evidence to demonstrateconclusively that most treatment approaches areeffective. Cognitive-behavioural therapy appears toshow the most promise. Brief interventions may bemore cost-effective than extended groupcounselling efforts.804

&='.'(�� 5��*�� +��+�� -)*-� ��)-��*+��-

Definition: Treatment programs designedacknowledging that there is considerable co-morbidity between drug use disorders and othermental disorders, particularly anxiety and affectivedisorders.


The issue of treatment for co-morbidity remains acomplex and debated area. A systematic review byThe Cochrane/Campbell Collaborations reportedthat there is, at present, no direct evidence tosupport the notion that any one form of treatment isbetter for people with co-morbid mental health andsubstance use problems; this includes the currentpush towards integrated treatment.783 Gowing et al.note that most controlled trials of treatment efficacyspecifically exclude people with co-morbid mentalhealth problems.804 Others note that co-morbidity istypically associated with a poorer treatmentoutcome.811 Given evidence that there is often atwo-way causal relationship between substance useand mental health problems, especially mooddisorders, there is a very strong rationale forcombined approaches. Clearly, the nature of suchinterventions will need to be tailored to particulartypes and severities of presentation, both for themental health and the substance use dimension.There has been an Australian trial of briefintervention for alcohol and drug use for personsattending a psychiatric ward of a major generalhospital.827 Screening of admissions for risky alcoholand drug use took place and patients were thenrandomly assigned to either a session of

�

�#� �� !��"�#��$��%��

motivational interviewing or given a booklet. Inpractice, the number of patients recruited for thestudy was too small for meaningful analysis.

&='.'6�� +��+,


The issue of treatment for elderly populations hasgenerally not been well addressed. There have beencalls for an ‘elderly specific’ approach to elderlyalcohol problems, since at least 1987.258 Alcoholtreatment effectiveness does not seem to varysignificantly between elderly and youngerpopulations258 although there is some evidence thatlate-onset drinkers are more likely to respond totreatment than early-onset drinkers.258 Thepredominance of depression, grief and socialisolation as antecedents to problem drinkingsuggests that specific strategies addressing theseissues may be required.258 Similarly, some haveargued that treatment must focus on day-to-dayissues such as loneliness, loss of independence anddeclining health.259 Such treatment has involvedaddressing strategies to cope with negativeemotional states, dealing with drinking cues, andincreasing social support.258 There is some evidencethat a supportive environment and social supportnetworks are factors in improvement followingtreatment.258

&='.'7�� )+�� +��+

The impact of treatment on harms at the populationlevel is more difficult to assess in relation to illicitdrugs than in relation to alcohol treatment. Thereare clear indications that MMT can reduce HIV riskbehaviour828 and local levels of criminal behavioursuch as burglary.804 The experience in Switzerlandwith heroin prescription has shown that effectiveopioid treatment can reduce the level of opiateoverdose in the community.829 The Director of theNational Drug and Alcohol Research Centre notesthat if there are approximately 100 000 opiatedependent people in Australia, then having aboutone-third of them in treatment in opioidreplacement therapy is a good public andpopulation health measure (Mattick, personalcommunication). The potential for reducing crimeat the local community level through treatment ishighlighted by a careful UK study of the averageannual costs of crime per illicit drug user in the 12months prior to their admission to treatment,roughly AU$45 000.817 This study estimated that,on average, two-thirds of these costs are saved pertreated individual per year.

Some authors have argued that, given that most ofthe population-level harms from alcohol use occurin non-dependent alcohol users (the so-calledprevention paradox), treatment will be of little usein preventing population-level problems withhealth, safety and public order. However, there isalso some evidence that programs targeting high-risk and dependent drinkers do, in fact, exert abeneficial effect on aggregate-level harms in thewider population.830 Thus, while the preventionparadox may well apply in relation to most alcohol-related harms, this literature suggests that treatmentmay engage and benefit enough high-risk anddependent drinkers to make a difference at theaggregate level. The example of a mass screeningand early intervention project in Malmo, Sweden,for early signs of alcoholic liver damage was alsoindicative of the potential of secondary preventionprograms to impact at the population level.831 Thereis still the confronting truth, however, that mosttreatment programs typically engage only a smallproportion of people with drug- and alcohol-relatedproblems and dependence. Even if one includesadvice from a general practitioner (GP), only one inthree people with an alcohol problem will receiveany kind of treatment from a health careprofessional in a 12 month period.832

&='.'3��-�� -

There are some major gaps in the literature atpresent. Some authors have identified a completelack of information on best practice in illicit drugtreatment for groups with specific needs, namelyIndigenous people, young people, prison inmatesand poly-drug users.804 There is also littleinformation on effective treatment options forpeople with psychostimulant problems, particularlyin relation to ATS, since the existing literature dealslargely with American cocaine research.804 Overall,while alcohol, cocaine and opiate treatments arewell researched, best practice in treatment in regardto other illicit drugs is not well established.807

There is also a gap in relation to the low uptake oftreatment by some culturally and linguisticallydiverse groups. There are Australian reports thatheroin users from an Asian or Middle Easternbackground are less likely to enter into MMT.833 Areport by the Victorian Department of HumanServices attempted to identify best practice inproviding drug and alcohol treatment to youngpeople of Cambodian, Lao, and Vietnameseorigin.809 Various strategies were recommended,including improved cultural awareness, recognitionof culture-specific needs, collaborative effort, familysupport and flexible outreach services. However, no

�#"��

direct evidence was provided that addressing theseissues has resulted in either improved service uptakeor improved service outcome. This is a clear gap inthe literature. Some ethnic groups have low uptakeof treatment services, but the reasons why, orstrategies that may address this problem, have notbeen addressed in the literature.

&='.';�� *� ��-��

One of the largest and most rigorous economicevaluations, to date, is the CALDATA study based inCalifornia.834 A study of 3000 alcohol and drugdependent participants under differing modes ofcare showed treatment to yield a 7:1 return on eachdollar spent. Most of these benefits were obtainedfrom decreased levels of crime. In this study, gainsat one year after treatment more than doubled thegains made during treatment (although patientsdischarged from the methadone programs alsoincluded in the evaluation had relapsed and had anegative benefit-cost Ratio [BCR]).

A longitudinal survey of drug users in theTreatment Outcome Prospective Study (TOPS)found that after treatment, drug users impose lowerrates of crime-related costs on society but they didnot earn significantly higher wages.835 The BCR forthe TOPS methadone treatment program was foundto be 4:1. Subsequent studies have found lowerBCRs for outpatient treatment but when gains fromlegitimate employment are included, they improvedto 4.3:1.

With regard to alcohol treatment, there is alsolongitudinal research demonstrating a 24%reduction in health costs for treated alcoholicsversus untreated alcoholics, over a 15 year follow-up period.806

Standard substance abuse treatments have beenconsidered to be very successful, yet at the sametime, their effectiveness is frequently challenged.This paradox is due to the fact that various programsproduce multiple outcomes, which can be valueddifferently depending on the point of view of theevaluation. Outcomes such as abstinence, reduceddrug use, delayed use, harm reduction, etc. arecommonly cited but may conflict when trying tomake comparisons between programs. It hasgenerally been proven that investment in the morehigh-risk treatment groups shows the highestreturns, especially if treatment can be non-residentially based.

&='.'<� � �+)-�� -

There is a strong consensus within systematicreviews that treatment programs for alcohol andother drug problems can be effective against a broadrange of indicators including levels of use, healthand wellbeing. There is a substantial evidence basefrom major multi-centre studies with extensivelong-term follow-up and rigorous methodologythat treatment is associated with reduced drug use,improved health (mental and physical), improvedsocial functioning and reductions in crime.836

However, not all treatments work equally well.806

Different drug problems necessitate differentapproaches804 and some areas of treatment are moreamenable to rigorous assessment of effectivenessthan others.804 Detailed investigation of this issuecan be found elsewhere.804, 806

�)��*��$�� ;3��$��

&='0'&��1� ��--��


A Cochrane review of smoking cessation programsfor pregnant women concluded that such programsreduced smoking, low birth weight and pre-termbirth; but no effect was detected for perinatalmortality. A systematic review of prenatalcounselling on tobacco use showed that counsellingcan reduce the incidence of low birth weight.468

&='0'.�� -�� ,��+��-��1� �� +��+��*+��-


Brief interventions by primary care practitioners forboth smoking and early stage alcohol problemsfeature in a strong literature, based on systematicreviews and meta-analysis, that demonstrates theireffectiveness.783 While the increase in the number ofpeople reducing their consumption in response tobrief interventions is small, this increase is highlyconsistent across numerous different studies. Giventhat brief intervention is inexpensive, takes verylittle time, and can be implemented by a wide rangeof health and welfare professionals, this is a highlycost-effective strategy with considerable potentialfor harm reduction from a wholesale application ofthe method. A recent evidence-based review ofthese programs found that compared to a control

�

�## �� !��"�#��$��%��

group who received no treatment, 10% morepeople in the intervention group ceased or loweredtheir alcohol intake to a risk-free level.837 In acontrolled study of mass screening and briefintervention with follow-up, for men in Malmo,Sweden, there was a significant decline in hospitaladmissions and mortality in the treated group over afour year follow-up period.831 An 80% reduction inabsenteeism in the four years following the study, a60% reduction in total hospital days over five years,and a 50% reduction in all cause mortality over sixyears was reported.831 These are substantial changesfrom a brief low-cost intervention that had a wideimplementation in the host community and arecomparable to the GP driven Quit for Life anti-smoking program in Australia that has been shownto be highly cost-effective.838

�)��2 !!�� 3��$�� !�

Definition: The community is a catchment area that istargeted with health promotion messages to reducedrug-related harm. Community interventions focuson structural policy change, with preventionprograms that aim to achieve policy, legislation andpractice change, to indirectly influence alcohol andtobacco consumption in the community.

Summary: Health promotion community-basedinterventions; evidence for implementation ......... �Community interventions focused on structuralpolicy change; evidence for outcomeeffectiveness ..................................................... ��

In recent years, there has been increasing interest incommunity-based prevention programs because ofthe emerging understanding of how environmentaland social conditions contribute to alcohol andother drug problems. Most of these research-drivenprevention programs have focused on alcohol andtobacco because the influence mechanisms andoutcome measures can best deal with open andprevalent behaviour. What can be identified fromseveral decades of practice are two complementaryprevention approaches.839 In one approach, thecommunity is a catchment area which can betargeted with health promotion messages designedto directly change the way individuals use tobaccoand alcohol. In the other approach, preventionprograms aim to achieve policy, legislation andpractice change in order to indirectly influencealcohol and tobacco consumption in thecommunity.839 Not surprisingly, a number of recentprevention approaches have combined these twofeatures in comprehensive community preventionprograms.

The New Zealand Community Action Project (CAP)was conducted in six cities over a two and a halfyear period.840 It comprised two main components:a multimedia campaign designed to encouragemales to drink moderately, and the use ofcommunity organisers to stimulate local discussionof alcohol policy issues. The program’s main impactwas to curb an existing trend of increasingly liberalattitudes towards alcohol. Another community-based program, the Rhode Island Alcohol Abuse andInjury Prevention Project, involved communitymobilisation, training in responsible beverageservice and enforcement of laws relating toalcohol.841, 842 This intervention appeared to reduceemergency room injury visits, assault, head injuryand motor vehicle crash injuries843 but had anumber of methodological limitations.844

Community alcohol programs ideally consist of anorganised, planned, community-wide intervention,whereby a wide range of stakeholder agencies suchas police, health services, drug agencies and localbusinesses get together and implement a range ofcomplementary interventions. The interventions areimplemented systematically (e.g. community inputinto local licensing regulations, media awarenesscampaigns, police action on drink-driving,responsible service policies) as part of a broadintervention aimed at changing the way thecommunity defines normative patterns and levels ofconsumption, and responds to problems of use. AnAustralian example of this was provided by theChristmas Collaborative Campaign, which wasundertaken by a coalition of virtually all relevantstakeholders in the Western Australian town ofCarnarvon, as a way of reducing alcohol harmduring the high-risk period leading to Christmas.845

This campaign comprised a number ofcomplementary initiatives involving mediamessages, promotion of responsible drinking inlicensed premises and structural changes to reducerisk. Local data indicated a reduction in harm ascompared to the previous year, including areduction in road crashes.

Aguirre-Molina and Gorman, in a comprehensivereview of community-based drug preventionprograms, found that programs with the greatestpromise: relied heavily on community action as themeans of achieving change; sought to empower thecommunity through involvement in all decisionmaking; were comprehensive in terms of targetsand strategies; drew on the public health model toidentify factors other than the individual as causingproblems; and drew on the best available research toguide interventions.846 Cochrane reviews of broader

�#%��

health behaviour have reinforced these findingsfrom community alcohol programs. These reviewshave shown that all community-based programsaimed at changing health risk behaviour are moreeffective in causing positive behavioural changeswhen the program involves extensive, multipleinterventions in a variety of settings and contexts.783

While broad community-based initiatives have beenheralded in recent years as offering betterprevention possibilities than single element,individual-focused strategies, this enthusiasm needsto be tempered with an appreciation of what hasactually been achieved. Gorman and Speer identifiedonly eight well-designed community alcoholinterventions that sought to change populationknowledge, attitude or behaviour.847 The majorityof these studies reported minimal program effects,which seemed to be due to difficulties in generatingcommunity involvement and an inability toinfluence community processes. The programs thatreported most success had circumscribed objectivessuch as reducing drinking and driving, or limitingalcohol use in specific locations. A recent well-controlled quasi-experimental study fromStockhölm, Sweden,848 focused on responsiblebeverage service on licensed premises, with acombination of community mobilisation, trainingof bar staff and stricter enforcement of liquor laws.A substantial and significant drop in violent crimeoccurred in the intervention sites. This interventionclearly straddles the boundaries betweencommunity action and supply control. It alsosupports the idea of having defined objectives andincluding community mobilisation strategies tosupport community employees such as police, inimplementing what may be controversialprevention strategies.

The largest and most methodologically rigorouscommunity alcohol prevention program was theCommunity Trials Project (CTP) conducted byHolder and his colleagues in six locations inCalifornia and South Carolina, over a five yearperiod.727 This project aimed to reduce harmsassociated with drinking, rather than drinking itself,and comprised five interacting components:community mobilisation; responsible beverageservice; reduction of drinking and driving;reduction of underage drinking; and reduction ofaccess to alcohol. Holder has emphasised theimportance of using community action to achievepolicy and structural changes that can have asustainable impact on future levels of harm. TheCTP had particular emphasis on enforcement ofliquor laws regarding service to intoxicated andunderage drinkers, as well as utilising opportunities

for local controls on alcohol availability. The CTPwas successful on a number of measures including:the reduction of alcohol-involved motor vehicleaccidents, rates of alcohol-related violence,significant community support for theinterventions, an increase in media coverage ofalcohol-related trauma and prevention policyinitiatives, and reductions in sales of alcoholicbeverages to underage decoys.839 However, thissuccess needs to be measured against the cost of theproject, which was considerable. Holder and hiscolleagues noted that over a four year period, thecost of local community staff in one interventioncommunity was US$360 000 and this did not takeinto account evaluation costs or expenses associatedwith intervention activities.849 Such comprehensive,well-resourced, long-term community projects havethe greatest chance of achieving meaningful changebut this level of funding would be difficult to obtainon a routine basis. However, it should be noted thaton the basis of only one of the above outcomemeasures, a net reduction of 78 alcohol-relatedtraffic crashes in the intervention communitiesresulted in estimated savings of US$3 112 590.

The Community Mobilisation for the Prevention ofAlcohol-Related Injury (COMPARI) was anAustralian demonstration project designed to showthat alcohol-related injury could be reduced bymobilising a whole community to take an activerole in changing individual drinking behaviour andthe environmental factors that influence alcohol-related harm.850 The project operated over a threeyear period in the Western Australian regional cityof Geraldton and during this period undertook 22major component activities involving: communitydevelopment, local networking and support,provision of alternative activities, health education,health marketing and policy institutionalisation.Many of the individual activities conducted byCOMPARI resulted in changes in communityknowledge and behaviour. For example, thedesignated driver intervention for young adults,known as Pick-a-Skipper succeeded in persuadingyoung drinkers to select non-drinking drivers as‘skippers’ before they began consuming alcohol.851

While the project was sufficiently successful tobecome the major alcohol and drug service providerfor the region, there was limited measurable impacton overall levels of harm.

In one of the most recent reviews of the research oncommunity prevention approaches to alcohol,Casswell identified the following commonthemes:852

� projects must be community driven,

�

�#' �� !��"�#��$��%��

� traditional experimental designs may limit thecommunity dynamic,

� successful projects incorporate a process ofreciprocal and respectful communicationbetween different community sectors andbetween the community and the researchers,

� social capital is needed for effective communityaction,

� financial capital is also critically important.

Sufficient time for development of the intervention,measurement of change and institutionalisation ofpractice, has also been identified as an issue by anumber of researchers.727, 849, 850 If communityprevention programs are to reflect these elements ofbest practice, they will be complex, expensive andlong-term. However, Casswell considered a numberof community prevention studies have shown thatcommunity mobilisation can create changes in thenorms about alcohol use and alcohol harm and, as aresult, can facilitate structural change within thecommunity that has a direct impact on harm.852 TheAguirre-Molina and Gorman846 and the successfulStockhölm study,848 however, also suggest thatcommunity mobilisation on its own will not beeffective unless it is focused on a specific structuralchange process (e.g. encouraging more responsibleservice of alcohol in public places). If researchunderstanding of community preventionapproaches to alcohol is to be meaningfullyadvanced in Australia, a research project tailored forlocal circumstances, but of similar scale to theAmerican CTP, should be undertaken in thiscountry. In this way, community preventionapproaches that are best suited to the Australiancultural context can be identified.

&='('&��++��)��) �,� �� -

Definition: Community groups engaged in a range ofactivities designed to prevent or reduce illicit druguse in their community.


Although most of the community literature dealswith alcohol and tobacco programs, there isgrowing interest in programs targeting illicit drugs.Community groups interested in local drugproblems have operated in WA since 1997 and havespread throughout NSW since 2000. In Victoria,local councils have initiated community networks inresponse to local drug concerns. These structures,which involve local citizens, are based on the ideathat since drugs are a community problem the

community should be part of the solution. TheGovernment provides some financial support,advice when requested, and assists with amonitoring and reporting framework. However,because grassroots involvement is critical, parents,local residents, young people, local businesses,police, local government, schools, youth servicesand community groups are usually all represented.Activities organised by these groups include supportservices, employment, recreation and monitoringyouth parties.853 This approach offers considerablepromise but at this early stage evaluation is limitedto anecdotal reports.236, 854

The Community Partnerships Initiative (CPI) hasbeen developed by the National Illicit DrugsStrategy (NIDS) within the CommonwealthDepartment of Health and Ageing. Its purpose is tocontribute to the prevention and reduction of youngpeople’s illicit substance use by mobilisingcommunities and fostering relationships betweengovernment and the broader community. This ispursued primarily by funding community groups toundertake preventive projects. The anticipatedoutcomes are as follows: the development of anAustralian community partnerships model forprimary prevention of illicit substance use; abenchmark of quality practice in communityparticipation and action on a significant publichealth issue; an increase in the capacity ofcommunities to develop effective preventionactivity; national dissemination of quality practice inprimary prevention of illicit substance abuseutilising various forms of media; and an increase inself-sustainable community action across Australia.The National Drug Research Institute was funded toevaluate the CPI over the first two fundingrounds.855

The heart of the CPI is the funding of community-based projects. Funding is provided on a one-offbasis for a period of up to two years, with projectsreceiving between $5 000 and $211 000. To date,24 projects were funded under the first round to atotal value of $1.9 million, 63 organisations werefunded under the second funding round to a totalvalue of $3.98 million, and 23 organisations werefunded under the third round to a total value of$1.7 million. A fourth funding round hascommenced. Projects include: communitydevelopment; training schemes; peer education;information dissemination; and resource productioninitiatives in rural and remote, regional,metropolitan and suburban settings.

In the evaluation, projects have been categorisedaccording to the major activities planned, or themesdeveloped, in the construction of the project. The

�#(��

categories emerged from a literature review asfollows.

� Knowledge, attitudes and values (KAV)approaches can be defined as those that seek toincrease young people’s awareness about illicitdrugs while changing their values and attitudesthrough examination of personal needs, valuesand decision making patterns.856 KAVapproaches form the most common approach inthe CPI. Thirty-five projects (10 in the firstround and 26 in the second) incorporatedelements of this approach. Whilst there is littleliterature available on the efficacy of a KAVapproach in a community-based setting, school-based programs show that programs thatcombine these elements are more successfulthan programs that focus on just one or twoelements.856

� Peer-based approaches incorporate theteaching or sharing of information, values andbehaviours by members of similar age or statusgroup.857 Peer-based approaches were wellrepresented amongst CPI projects, with 29projects (10 in the first round and 19 in thesecond) containing elements of this. Theliterature review noted that there is no clearevidence linking peer-based preventionprograms to behaviour change, although theseare very popular.

� Alternatives approaches can take the form ofbasic life skills, job preparation, recreationalactivities and physical adventure programs.856

They are designed to increase personalcompetence and promote an individual’s senseof control.856 Sixteen CPI projects (3 in the firstround and 13 in the second) offered alternativesto illicit drug use such as recreational, sporting,employment and other activities. They werefound to be the most effective type of programfor fostering behaviour change in onesystematic review.856

� Community action approaches seek to achievethe representation and active involvement ofcommunity sectors that are perceived to be ofinfluence in preventing illicit drug use byyoung people.858 As noted above, localcommunity involvement is a critical factor inlocal prevention programs. Twenty CPI projects(6 in the first round and 14 in the second)sought to foster community action in order toprevent illicit drug use amongst young peoplein their community. These community actionsincluded establishing community-basedadvisory groups and developing communityforums and action plans.

� Parent-based approaches: nineteen programs(4 in the first round and 15 in the second)aimed to provide information and/or supportfor parents of young people who are using, orare at risk of using, illicit drugs.

� Broad-based approaches were defined as thosethat incorporated more than one of the above.Broad-based approaches were relativelycommon in CPI projects—24 projects (7 in thefirst round and 17 in the second round) took anapproach to primary prevention thatincorporated one or more approaches.Knowledge and/or affective approachcomponents were present in 18 of the broad-based projects. The most common combinationwas knowledge and/or affective andcommunity action approaches (5 projects).

One immediate conclusion from this analysis is thatthere is a significant gap between practice and theevidence base in this area.

�)�"�� !��+�� !!��

Definition: In principle, mainstream substance misuseservices are available to Indigenous Australians. Inpractice, these are often unaffordable, inaccessible,inappropriate and unacceptable. There have beenattempts by mainstream service providers to addressthese problems. However, in response to theseproblems, the level of substance misuse within theircommunities, and as an expression of self-determination, Indigenous Australians themselveshave developed a broad range of interventioninitiatives.


Arrangements to address Indigenous substancemisuse, as in other areas of Indigenous health, arecomplex. Although Commonwealth, State andTerritory substance misuse policy is partlycoordinated through the Ministerial Council onDrug Strategy (MCDS) and the National DrugStrategy,13 at the time of writing there is no nationalIndigenous substance misuse policy. However, aReference Group for Aboriginal and Torres StraitIslander Peoples provides advice to theIntergovernmental Committee on Drugs (which isresponsible to the MCDS) and its various nationalexpert advisory committees. This Reference Groupgives Indigenous people formal input into nationalsubstance misuse policy for the first time, and iscurrently overseeing the development of aComplementary Strategy to Address SubstanceMisuse among Aboriginal and Torres Strait IslanderPeople.

�

�%) �� !��"�#��$��%��

Arrangements to address Indigenous substancemisuse also include numerous pieces ofCommonwealth, State and Territory legislation thatcontrol the availability of alcohol and other drugs.These pieces of legislation do not generally addresssubstance misuse among Indigenous people.However, to varying degrees, some—such as liquorlicensing laws—provide communities in general, orIndigenous communities in particular, with a voiceand opportunity for action on these issues.

Substance misuse prevention services for Indigenouspeople are provided by State and TerritoryGovernment agencies, Indigenous community-controlled substance misuse and health serviceorganisations, and to a lesser extent, non-Indigenous controlled non-governmentorganisations. The Commonwealth Governmentdoes not have a direct role in the provision ofservices but plays a key role in the funding of State/Territory and Indigenous community-controlledintervention services. In fact, in the 1999/2000financial year, the Commonwealth provided 58% offunds for projects specifically targeting Indigenouspeople with substance misuse problems, as well asproviding significant amounts of funding forIndigenous community-controlled primary healthcare services.859 In addition to providing directservices, individual community-controlledorganisations and umbrella organisations, such asthe National Aboriginal Community ControlledHealth Organisation and the recently formedNational Indigenous Substance Misuse Council, playan important advocacy role.

Demand reduction strategies among IndigenousAustralians have focused on treatment and healthpromotion, and—to a lesser extent—providingalternatives to substance misuse use. In addition,both explicitly and implicitly, various communitydevelopment projects have the prevention ofsubstance misuse as a goal.

&='6'&��

The 1994 NDSHS-UATSIPS found that most peoplewho had sought treatment had done so in primaryhealth or medical care settings: either fromIndigenous community-controlled health services orgeneral practitioners.269 As well as treatment—under their objective of providing comprehensiveprimary health care—the various Indigenouscommunity-controlled health services also provide arange of early intervention, prevention and supportservices. Indigenous community-controlled healthservices also undertake about a quarter of thesubstance misuse specific projects discussed underdifferent categories below.859

On the basis of reports by Indigenous people aboutthe role of advice from medical practitioners in theirdecisions to give up drinking alcohol, and theireffectiveness in other populations, Brady—amongothers—has been an advocate of brief interventionsfor Indigenous people.326, 860, 861 Despite there beingno evaluation of the efficacy of such interventionsamong Indigenous Australians, given theireffectiveness elsewhere they should probably beused by health care providers as the opportunitypresents. More broadly, Hunter, Brady and Hallhave prepared a comprehensive set ofrecommendations for the clinical management ofalcohol-related problems in Indigenous primarycare settings.862 Also, a book edited by Couzos andMurray for the Kimberley Aboriginal MedicalServices Council provides guidelines for educatingand counselling Indigenous patients with alcoholand tobacco problems and for addressing theseproblems in the context of other health problems.863

Apart from general primary care interventions,focused treatment projects are the most commonform of intervention in Indigenous communities. In1999/2000, there were 107 treatment projectsconducted in both residential and non-residentialsettings.859 The majority of these targeted alcoholalone or alcohol and some combination of othersubstances. Although many treatment projects arebased on the ‘12-step’ model—or an adaptation ofit—in recent years services have begun to explore awider range of approaches, including life skillscounselling and vocational training.

Evaluations of alcohol treatment projects have foundthat some produced no significant outcomes, whilstothers experienced moderate degrees of success.864–

869 In one case, such results were reported to be aconsequence of the fact that there were no agreedcriteria against which success could be measured.866

In others, project effectiveness was circumscribedby limited resources and the need for additionaltraining for both clinical and administrative staff.867

In the case of petrol sniffing, d’Abbs and MacLeanreported that the results of the Healthy AboriginalLife Team (HALT) project conducted in CentralAustralia showed that ‘… in the hands of a skilledcounsellor, orthodox counselling … can be effectiveif used with sensitivity and respect for Aboriginalperceptions and values’ (p vii).233

An important gap in the provision of treatmentservices—especially given the practical problemsassociated with home detoxification in Indigenouscommunities—is the limited number ofdetoxification facilities. The lack of detoxificationservices is particularly acute for those who inject

�%��

drugs;175, 870 and staff from services focusingprimarily on alcohol-related problems report thatthey often do not have the training to dealeffectively with illicit drug-related problems.285

&='6'.��

In 1999/2000, prevention projects (in the morenarrow use of the term) comprised only 21% of allIndigenous intervention projects and received lessthan 10% of funds directly targeted at reducingIndigenous substance misuse. Furthermore, 47% ofthese received only short-term funding.859 Alongwith other intervention projects, a list of currenthealth promotion projects is available on theIndigenous Australian Alcohol and Other Drugs Database(http://www.db.ndri.curtin.edu.au), andoverviews of the range of projects are available fortobacco,468 alcohol871 and petrol sniffing.233

A small number of, largely qualitative, evaluationsof projects targeted at alcohol have beenundertaken. The projects included: provision ofhealth education classes, sporting and recreationalactivities, and support for homeless people;872 abush tour by the band Yothu Yindi and anassociated television commercial;873 alcoholeducation and related programs for youngpeople;874–876 and community education andactivities.867, 868 As with treatment projects, whilstmost were well received by the communities inwhich they were conducted, the outcomes of theseinterventions were equivocal. These evaluations alsoidentified a number of process issues that eitherenhanced or constrained project effectiveness. Inparticular, they emphasised the need for adequateresourcing and the provision of staff training andsupport.869

In her literature review of Indigenous Australiansand tobacco, Ivers identified a small number ofinterventions specifically targeted at Indigenouspeople on which published information isavailable.468 These included development ofresources and radio advertisements as part of theWestern Australian Quit Campaign;877 the Jabbydon’t smoke project878 and the Yamatji smokingprevention project879 (both of which were alsoconducted in Western Australia); and theManingrida be smoke free project880 in the NorthernTerritory. As Ivers indicates, none of these projectshave been formally evaluated. However, somegroups have compiled or developed guidelines foruse in the development of smoking interventionprojects.881–883

As indicated previously, prevention of the use ofillicit drugs among Indigenous Australians is a

relatively recent phenomenon. In 1999/2000, theonly prevention projects concerning illicit drugswere four that targeted cannabis use as part ofbroad-based prevention efforts—none of whichhave published evaluation reports available. Giventhe growing use of illicit drugs, there is clearly aneed for the development of more interventionprojects in this area.

Petrol sniffing intervention projects have focused onyouth work,884 recreational activities,885, 886 generaleducation,887 employment,888 and substance specificeducation.889 The number of projects specificallytargeting volatile substance misuse is relativelysmall—with six operating in the 1999/2000financial year. A review of petrol sniffingintervention projects has been undertaken byd’Abbs and MacLean.233 Based on the limitedevidence available, they suggested that—subject tosome caveats—a number of these interventionsappear to be effective. However, they concludedthat:

The most effective long-term strategies against petrol-sniffingare likely to be those which broadly improve the health andwell-being of young Aboriginal people, their families andcommunities (p viii).233

�)�#�� !�� !��! ��


There is little evidence available on effectiveprevention strategies in elderly populations. Mostreviewed publications cited earlier mention theimportance of ensuring that health care staff areaware of the issue of alcohol-related problems inthe elderly and are skilled in their detection. Asdiscussed, detecting alcohol problems can bedifficult, and there is a reasonable body of literatureaddressing approaches for improved screening inhealth care settings.262

WHO recommend prescribing benzodiazepinescautiously to older populations, and to chooseshorter duration benzodiazepines because they areless likely to accumulate in the blood, whichincreases the risk of harmful side effects.203 Kirby etal. suggest that the continued prescription of long-acting benzodiazepines may reflect a greaterfamiliarity with the older longer lasting drugs, lackof awareness of the greater risks of using this kindof benzodiazepine, and the perception thatalternatives with shorter half-life are not aseffective.890 Preventing harmful benzodiazepine usemay involve addressing these issues with doctors

�

�%� �� !��"�#��$��%��

and patients, as well as promoting alternativetreatments for insomnia, depression and anxietyamong older Australians.

�)�%�8 �-�� !�

There are two major rationales for alcohol and otherdrug interventions in the workplace: to improveproductivity, and to improve workplace safety.404, 405

Other proposed benefits include improved publicrelations due to public perception that companiesare acting to reduce drug problems and improvesafety in the workplace.891 However, developingworkplace programs is complex and requiresconsideration of health, ethical, legal and industrialrelations issues. Little is known about the actualpreventive impact of different interventions.404, 405

&='3'&��)�� +��+�� -��,�--)�-

In the Australian context, approaches to workplacedrug problems are strongly influenced byOccupational Health and Safety legislation andindustrial relations issues. Implementation ofprevention policy cannot be considered in isolationfrom these issues.

It is clearly established, in general terms, thatAustralian employers are legally obliged to take alldue precautions to reduce the risk of any potentialsafety hazards in the workplace.892, 893 Based on thisgeneral principle, it is generally accepted thatemployers have an occupational health and safetyresponsibility to address drug-related hazards in theworkforce.891–895 In occupational health and safetyterms, an intoxicated person can be considered ahazard892 and employers have a clear legalobligation to protect staff from hazards related todrug and alcohol use, including of tobacco, byother employees.893

&='3'.�� 1�+��

Historically, alcohol and other drug problems in theworkplace have been dealt with through employeeassistance programs (EAPs) and employer policieson employees with problematic drug use.896 In thecase of alcohol, this has resulted in the growth andacceptance of EAPs as a worthwhile response toindividual employees with drinking problems.Employees are more willing to seek help becausereporting drinking problems does not threaten theirjob, and employers are more willing to retain suchemployees because use of the EAP indicates awillingness to change behaviour.

Although the majority of United States897 andAustralian898 employers offer and support EAPs,there have been no definitive evaluationsdetermining the effectiveness of EAP for treatingalcohol and other drug problems.897, 898

Nevertheless, a well-functioning EAP provides theopportunity for brief intervention,898 a treatmentthat has already been demonstrated to be highlyeffective for early-stage and/or less serious alcohol-and drug-related problems.

Ames considers that in addition to treatingindividuals with problematic alcohol and other druguse it is just as important to develop preventiveinterventions, arguing that primary preventionprograms in the workplace should address thephysical and cultural factors of the workenvironment that promote or facilitate problematicalcohol and other drug use.899 Primary preventionseeks to protect employees who are at risk ofdeveloping alcohol and other drug problems, and toavoid problems in the workplace that can be causedby employee alcohol and other drug use. Holderargues that workplace prevention needs to focus onproblems caused by use rather than on individualusers.896 In this way, interventions can dealcomprehensively with all underlying factors.Prevention interventions can address features of thephysical working environment that may encourageproblematic alcohol and other drug use, such ashazardous working conditions900 or aspects of theculture and organisation of the workplace like poorpromotion opportunities.901 In some cases, thewider community may be an appropriate focus forprevention of alcohol and other drug problemsbecause community norms, practices andregulations can determine consumption patternsthat then adversely affect the workplace.902

Many employers equate prevention of alcohol andother drug problems in the workplace with drugtesting and this strategy is increasingly beingadopted in Australia, particularly in the miningindustry. Because this strategy is both controversialand widespread, it is considered in some detailbefore discussion of other workplace preventionapproaches.

&='3'0��"�)��-� �

Definition: The most common workforce preventionmeasure has been drug testing. This takes two mainforms: urine testing for illicit drugs; and breathanalysis for acute alcohol intoxication.404, 405 Whilsturine testing has a long history in the US, suchprocedures are less common in Australianworkplaces.903

�%��

Random urinalysis for illicit drug use in theworkplace—overview and concerns

The testing of urine to detect illicit drugs has a widevariety of applications, including law enforcement,clinical practice and research purposes. Makkai notesthat testing technology, while not automatic and notalways definitive, is nevertheless generally robustand lends itself to authoritative conclusions.904 It hasalso been observed that drug testing can improvethe evidence base for policing and criminal justiceby providing more scientific and rigorous methodsfor monitoring and evaluating criminal justicepractice.905

Forms of drug testing in the workplace includerandom testing, pre-employment testing, and for-cause testing when an employer has a presumptionthat an employee may be drug impaired. Neither ofthe latter two forms have generated the controversythat has been directed at random drug testing.176

There are two main justifications for workplaceurine testing regimes: to improve productivity, andto improve safety.404, 405 Nolan points out that therehas been increasing pressure on employers todemonstrate that they conform with OccupationalHealth and Safety Regulations if they are not toleave themselves exposed to the possibility oflitigation in the event of work-related injuries orillness. Drug testing has been seen as a way ofmeeting those legal obligations.893

Some major concerns about random drug testinghave been identified.

� Urine tests detect past illicit drug use, evidentover varying time frames; they do not detectcurrent drug use that may impact upon workperformance.404, 405

� Urine testing can be highly invasive and manypeople find the testing procedureobjectionable.404, 405, 892, 894

� Less invasive options (such as private provisionof urine samples), whilst commonly used,allow for ‘cheating’ and are still considered tobe undignified and objectionable procedures bymany.891

� There are significant privacy issues that relate tothe extent to which an employer can legallyinquire into the activities that employees engagein their own time.893

� Testing regimes have been criticised frequentlyas a quick-fix approach that ignores underlyingcauses of occupational health and safety issues,particularly fatigue which is arguably moredamaging and is subject to far less control.893

Effectiveness of urine testing programs


Urine testing is not an accurate measure of impairedwork performance, which can reduce safety,because it measures recent use not presentintoxication.404, 405 In the case of cannabis, use canbe as long ago as 30 days. There is no evidence thatrecent illicit drug use is associated with reducedproductivity or safety impairments.404, 405

Effectiveness evaluations of workplace urine testingprograms and their impact have been scarce andgenerally poor in scientific terms.404, 405 In 1994, theAmerican National Academies Committee on DrugUse in the Workplace found that the preventiveeffects of drug testing had not been adequatelydemonstrated, because of the weakness of thescience.176 Other reviews have determined that thereis no scientific evidence of improvements to eitherworkforce productivity or workplace safety fromthe implementation of urine testing programs,although there are numerous anecdotal reports orweak, poorly or uncontrolled evaluations reportingbenefits.404, 405

Breath testing for alcohol intoxication


Alcohol is both the most widely used drug in theworkforce and the drug with the most deleteriousimpact on workplace performance and safety.404, 405

Alcohol breath testing is different from urine testingbecause it measures current intoxication rather thanpast use. It is well established that blood alcoholconcentrations of greater than 0.05 (or possiblybelow this in some circumstances) produce a dose-dependent deterioration in performance.Impairment occurs mostly in attention,concentration, perceptual processes, and motorcoordination and feedback.906 However, due to thepoor quality of published studies, systematicreviews have been unable to confirm whetheralcohol testing programs lead to improvements inproductivity or safety.

Pre-employment drug testing


Pre-employment drug testing has been shown tohave a small predictive effect on workplaceperformance but other predictors, such as level ofeducation, are a more accurate predictor ofperformance.404 Again, there are no well-conductedstudies identified in systematic reviews that candetermine whether or not improved productivity orsafety results from such screening programs.404, 405

�

�%� �� !��"�#��$��%��

There is some evidence that the presence of drugtesting policies have a deterrent effect in that aproportion of current and past drug users, and somewith strongly held personal ethical objections todrug testing, will not apply for jobs.404, 405, 907 Butagain, there is no evidence that this improvesworkplace safety or productivity.404, 405

Testing for safety-sensitive positions


It is generally accepted that for positions withspecial safety concerns (e.g. airline pilots), alcoholand other drug testing programs have a useful role,despite the absence of any well-controlled studies todetermine effectiveness.404, 903, 908 The high costassociated with errors outweighs ethical concernsabout the high number of false positive results andthe invasiveness of the test.

Other forms of testing, such as hair testing andperformance testing, are available but there is, asyet, no evidence of any benefits or improvements toproductivity or safety criteria as a result of theirimplementation.404, 903, 909

&='3'(�� /��)�� /�� *�� -


There is a long history of using education programsto reduce or stop alcohol and other drugconsumption, particularly in school settings.However, the research evidence indicates that justproviding factual information about the harmfulconsequences of alcohol and other drug use isineffective.896, 910 This does not mean that alcoholand other drug education should not be provided.Holder suggests that educational campaigns canincrease support for other prevention programs. Inthis sense, they can serve to orient employees towhy use of alcohol and other drugs can be aproblem in the workplace and prepare the groundfor more targeted prevention programs.896

Accordingly, contemporary workplace programshave an education component but emphasise thedevelopment of specific prevention skills. Briefinterventions have been recommended forapplication in many workplace settings given theirwide evidence base for efficacy.908

&='3'6��5-��)��+�� -

Health Promotion Programs (HPPs) provideinformation on health-related matters and teachparticipants how to improve or maintain theirhealth. They are not usually aimed exclusively atalcohol and other drug use but they might beconsidered alternatives to drug testing in that, ifsuccessful, they would reduce the need fortesting.911 HPPs are a relatively recent feature in theworkplace and there is not yet a substantial body ofresearch that indicates whether such programs canprevent alcohol and other drug problems.405

&='3'7�@��+�5�5��) �,��-

It has been argued that that one of the majorlimitations of single application workplaceprevention approaches is they do not acknowledgethe role of the broader community in shapingalcohol and other drug consumption norms.896

Undertaking prevention activities solely at theworksite may be of limited value in some settingsand a more productive approach may be to includecommunity-level prevention, in the expectation thatless harmful patterns of consumption, generally,will also benefit the workplace.

&='3'3��)��,��1�+��

There is no strong empirical evidence that anyparticular workplace alcohol and other drugprevention strategy delivers benefit in terms ofreduced consumption or lower levels of harm.There are, however, strong theoretical reasons whyinterventions proven in other arenas, such as briefinterventions, breath-testing and developingpolicies to reduce availability of alcohol, will beeffective. The design of alcohol and other drugpolicies in the workforce is clearly an area whereextensive further research is required in order toinform effective, evidence-based policyresponses.404,405 For example, there are strongtheoretical advantages for the use of accurate drugtests that detect current impairment, such as breath-testing for alcohol. Strong theoretical grounds existfor developing more accurate tests of currentimpairment from other drugs that might impairperformance. Until these are developed, it isarguable that testing in worksites with no specialphysical hazards is unethical. Where there areobvious physical hazards, ethical concerns may beoutweighed by the theoretical benefits of drugtesting.

�%"��

�)�'�� !�-��;��>�� !��!��0��3 -��!��

&=';'&�$�--��1�� ,��-��

The mass media has a substantial presence in allmarket-based societies and has been a powerfulvehicle for the promotion of the two licitrecreational drugs: alcohol and tobacco. In themain, this has involved product advertising but themedia has also been used for political lobbying.Large advertising and public relations budgets,coupled with considerable expertise, have allowedthe alcohol and tobacco industries to developsophisticated mass media marketing and advocacyactivities. However, public health agents haveincreasingly entered the fray by advocating andmarketing competing messages about alcohol andtobacco use. In Australia, the QUIT campaign is agood example. This well-known national anti-smoking campaign has a long history of effectivelyusing television advertising to deliver its message.912

In recent years, media prevention campaigns havebegun to match the sophistication achieved by thealcohol and tobacco industry. In the process, Bootsand Midford considered that much has been learnedabout how to give health messages an effectivepresence in the mass media market-place.913

Mass media marketing and mass media advocacy aresimilar activities, in that they both use the massmedia as a vehicle to achieve the same primary goal:that of creating change within the community.However, while change is central to both activities,the type of change targeted and methods employedare quite different. Mass media marketing strategiesinherently support and seek to replicate the productmarketing strategies that are associated with the freemarket economy and its focus on choice,consumerism, and individuals. A central concept incommercial marketing is the exchange of a productor service for money.914 However, this mutuallybeneficial exchange is less obvious in socialmarketing and can lead to a sense that consumersare being coerced rather than being offered a way ofsatisfying their own needs. Accordingly, thebenefits of change need to be emphasised ratherthan the negative consequences of the targetbehaviour. Hastings and Haywood illustrate thiswith an example from research conducted intoyoung people’s attitudes towards drinking inmoderation.914 The research found that girlsconsidered boys who were controlling their alcoholintake in a social setting to be more attractive than

boys who were not doing this. This hasconsiderable implications for marketing a messagethat is mutually beneficial for the marketer and theconsumer.

Advertising is the major method used in mass mediaprevention marketing and it may be paid, or unpaidas in community service announcements. It hasbeen used to: orient the public to an issue, such asdrink-driving; teach relevant skills such asmeasuring standard drinks in different beverages;and warn of consequences, such as smoking causinglung cancer. A second strategy involves creatingpublicity, such as getting a well-known identity tolaunch a prevention campaign. This can be useful tosupplement advertising or advocacy campaigns butis seldom used in isolation.913 A more recentmarketing strategy is ‘edutainment’, which involvesthe deliberate placement of educational messages inentertainment contexts. Here the purpose is to havecharacters with whom the target audience identifies,to model certain behaviour that social marketerswant replicated.915

&=';'.�$�--��-� �?��1-

Mass media campaigns—tobacco


Mass media marketing of drug-related health issuesis not a recent phenomenon. Groups such astemperance unions have been creating news andeven sponsoring mass media advertising since thelate 19th century. However, despite their longhistory, evidence of the effectiveness of early massmedia marketing campaigns is difficult to find.According to Backer and colleagues, pre-1971 massmedia health campaign evaluations mostly showedthat the campaigns had failed.916 Montagne and Scottindicate that such old style mass media campaigns inisolation mostly influenced knowledge and had littleimpact on behaviour.917 They also tended to targetbroad audiences, which reduced their ability tofocus on specific issues. Montagne and Scottconsider that old style campaigns were limited toreinforcing existing social attitudes and norms, suchas not drinking and driving. However, anti-smokingmedia campaigns seem to be an exception. In anumber of American States, campaigns that featuredanti-smoking advertising were followed byreductions in smoking, although it was not possibleto identify the particular effect of the advertisingcomponent.912 In Australia, the QUIT campaignused TV advertising as its main element and anumber of evaluations showed that smoking

�

�%# �� !��"�#��$��%��

decreased in cities where the campaign was run.918,

919 These findings were replicated in a British anti-smoking television advertising campaign whereinthe evaluation indicated that smoking was reducedby about 1.2%. The well-known harms associatedwith smoking, and the simplicity of the ‘do notsmoke’ message, may be factors in the effectivenessof these media-based campaigns.

Most recently, the Commonwealth has developedthe National Tobacco Campaign, (NTC) a majorinitiative launched in 1997 with follow-ups insubsequent years, aimed primarily at assistingsmokers, aged 18 to 40 years, to quit.920 This hasbeen the most intense and longest running anti-tobacco campaign in Australia. A major element ofthe campaign is the partnerships achieved betweenthe Commonwealth and all State and Territoryjurisdictions, as well as non-governmentorganisations.920

The evaluation has also been intense and the NTChas been called the ‘most comprehensivelyevaluated national health promotion campaignmounted in Australia to date’ (p4).921 To the end of1998, the evaluation consisted of a Benchmark andtwo follow-up surveys, as well as a survey of CALDpopulations and continuous tracking. Severaladditional studies were undertaken: a comparison ofteenage and adult surveys, a price discountinganalysis, a print media monitoring study, and aneconomic evaluation of the NTC.921 These studieshave been published in a two-volume evaluationmonograph.920, 921

The campaign used the stages-of-change model as atheoretical base and aimed to move smokersthrough stages to increase the likelihood that theywould consider quitting. The evaluation found thatthe first phase was successful and there was alsoevidence of a reduction in overall prevalence ofsmoking. Young people, although they were nottargeted by the NTC, were also influenced, althoughthere are no specific data pointing to reductions inteenage smoking. The economic evaluationsuggested that the NTC was excellent value formoney and could have been associated with savingsof up to $24m set against the approximately $9mspent by Commonwealth and State/TerritoryGovernments.921

The evaluators concluded that the initiative ofinvesting strongly and intensively in a NationalTobacco Campaign was the right one, but warnedthat the focus on campaign strategies, target groupsand advertising themes needed to be kept underreview and refreshed on an ongoing basis.921

Mass media campaigns—alcohol


In recent years, mass media marketing of alcoholprevention has been used as part of larger,successful community-based programs.726, 922 Thestrength of this approach may be to reinforcecommunity awareness of the problems created byalcohol use and prepare the ground for specificinterventions.922 Backer, Rogers, and Sopory claim,however, that these recent alcohol campaigns havebeen successful in their own right because they havebeen based on rigorous social science theories,formative evaluation research has been undertaken,and because the campaign objectives have beenrealistic.916

An evaluation of the Danish National Campaigns onAlcohol, presented in the mass media since 1990,provides an example of what can be achieved by awell-conceptualised mass media campaign. Thecampaigns had three goals related to ‘sensible’alcohol consumption and an overall goal ofreduction in total consumption. These goals wereoperationalised into four objectives that includedthe two mass media marketing objectives ofincreasing knowledge of, and the number of peoplewho followed, the national recommendedguidelines for consumption.923

Strunge reported that the campaigns reached theirknowledge objectives. A small percentage of surveyrespondents also indicated that the campaigns haddirectly influenced their behaviour. Twelve percentof people surveyed, in 1997, stated that they hadreduced their alcohol consumption, with themajority reporting that health concerns were themotivation for this reduction. Strunge concludedthat ‘it is possible to generate positive awareness ofalcohol information’, and that ‘a continuous effortis necessary to maintain and increase the effects ofthe campaigns’ (p79).923

The National Alcohol Education and Risk AwarenessCampaign is an Australian example of a nationalalcohol campaign.924 This campaign was based onformative research that found that risky levels ofdrinking were highest in the age range 18 to 35years. However, the research also suggested thatthese drinkers did not identify with messages aboutalcohol-related harm. The campaign was intendedto engage drinkers in a personally relevant way andto increase the likelihood that they would drink inways that were not harmful to themselves or others.The initial phase commenced in January 1995, withtelevision, cinema, print and interior bus paneladvertisements for up to five months. There was

�%%��

also a national phone line and involvement of State/Territory Government services.

The evaluation included a benchmark survey and aninitial tracking survey. Reports on these studiesshow that awareness of the campaign was high andthat messages were recalled. One in three said theadvertisements had ‘made them think’ about theirown drinking but there was, as yet, no evidence ofmajor behaviour change. It was evident thateffective communication with these drinkers hadbeen very difficult, but the evaluators maintainedthat the campaign had played an agenda-setting roleagainst which standard drink labelling would be setlater that year.924 (See 181).

Another example of the prevention benefits of amass media alcohol campaign is provided by thesophisticated evaluation of a drink-drive advertisingcampaign in Victoria.925 The alcohol advertising waspart of a broader road safety advertising campaignundertaken by the Victorian Transport AccidentCommission (TAC). Approximately 70% of theadvertising was placed on television, with theremainder on radio, in the print media, on outdooradvertising and in cinemas. Although the alcohol-related component was only one of four elements ofthe advertising campaign, the evaluation was able toidentify the relationship between the drink-drive-related advertising and alcohol-related trafficcrashes. Analysis of crashes at times when alcoholinvolvement was known to be high identified thatthere was a significant relationship betweenreduction of crashes and TAC drink-drivepublicity.925

Mass media campaigns— illicit drugs


Australian media campaigns have also been targetedat illicit drugs. An early anti-heroin campaign926 wasindependently evaluated using community andillicit drug user surveys. These surveys found thatthe advertisements had been seen and the messagesrecalled, and there was some suggestion in thefollow-up illicit drug user survey that the campaignhad been associated with changes in the proportionof people at higher risk of using heroin. The follow-up rate among illicit drug users was low (57%),however, and users of ‘hard’ drugs were the mostlikely not to have been re-interviewed.926

A current campaign targeting illicit drugs is aimedboth at users and their parents. The National IllicitDrugs Campaign (NIDC) is a two-part communityeducation and information campaign whichcommenced in March 2001. It is aimed atpreventing young people from experimenting with

illicit drugs. Stage one targeted parents andencouraged them to talk to their children aboutillicit drugs. Stage two targets youth. The role ofStage one was to provide information and supportto parents of eight to 17 year olds about the rolethey could play in preventing drug use amongsttheir children and teenagers.927

The campaign is based on extensive formativeresearch involving focus groups and in-depthinterviews with parents and other members of thegeneral community; including members of Non-English Speaking Background (NESB) communitiesand approximately 1000 telephone interviews withparents of 12 to 17 year olds. These investigationsencompassed parents knowledge of, and fears abouttheir children’s illicit drug use. The researchshowed that parents would welcome help withimproving their skills, knowledge and confidence intalking to their teenagers about drugs.

The final campaign elements include televisioncommercials, print and radio advertisements, aparent booklet delivered to every household andmade available in 16 languages in addition toEnglish, a telephone information line, and acampaign web site that includes information forparents and public relations activities. Radio andprint media materials were developed for NESBparents and consultations about the materials wereheld with the National Drug Strategy ReferenceGroup for Aboriginal and Torres Strait IslanderPeoples.

The evaluation components for Stage one consistedof quantitative pre-campaign benchmark and post-campaign evaluation surveys with parents of eightto 17 years olds, non-parents aged 18 to 69 years,and youth aged 15 to 17 years. Other elementswere a quantitative post-campaign survey withNESB parents and weekly tracking surveys.928

The results of the parent surveys indicated thefollowing: that campaign recall and recognitionwere high and key messages retained; around halfof the parents surveyed had read at least part of theBooklet and most of those (41% of all parents and29% of NESB parents) found it useful; and aroundhalf of all parents and 40% of NESB parents saidthey had taken some action—the majority bytalking to their children about drugs. Most said theyfound this easier than before the campaign. Most(57%) of parents and 38% of NESB parents hadspoken to someone (most often their children)about illicit drugs in the past two months.928

It was concluded that the parents’ component of theNIDC had been successful in achieving its

�

�%' �� !��"�#��$��%��

objectives, although there had been no increase inparents’ feelings of confidence between the twosurveys. Most of the parents, both general andNESB, who were aware of the booklet had read atleast some of it and/or found it useful. However,fewer NESB parents (47%) compared to otherparents (68%) had seen the booklet. The specificneeds of NESB parents were not discussed otherthan to say that they were at ‘an earlier stage ofawareness’ in relation to illicit drugs. Until thecampaign has been concluded, it is difficult toestimate how successful it will be, overall.

Mass media ‘edutainment’ to reduce alcohol-related harm

The Harvard Alcohol Pro sought to introduce theconcept of a ‘designated driver’ as a new socialnorm in the United States, and a key strategy wasthe use of entertainment television to promote thedesignated driver concept. The project staff spokewith more than 250 producers and writersassociated with all the leading prime-time televisionentertainment shows and convinced them tosupport the project objectives. Their efforts resultedin more than 160 prime-time television programs,including the notion of the designated driver in atelevision episode. Evaluation of the impact of theconcept has been undertaken using Gallup polls thatask respondents about their use of designateddrivers. DeJong reported that 64% of adultsreported that ‘they and their friends assign adesignated driver when they go out for social eventswhere alcoholic beverages are consumed’ (p26).929

&=';'0�$�--��,

Here advocacy refers to the promotion of healthypublic policy by influencing decision makers toaccept the merit of processes, policies or structuresthat bestow a health advantage. A major tool used inthis process is political lobbying, which is thepresenting of arguments in favour of a particularpolicy course to those making the policy decision.Another is coalition building, which involves thedevelopment of a common policy objective forgroups and individuals in a community. A third isthe use of mass media, typically the news media, tohighlight and advance a particular public healthissue. This last approach has been promoted byWallack930 and is commonly referred to as mediaadvocacy.

The major challenge of all forms of advocacy is tomove the debate from individually focused, simpledefinitions of problems to a level of complexsociopolitical conceptualisation, where the targetedhealth problem is seen as a product of the

interaction between the individual and theenvironment.

According to Wallack et al

Advocacy is necessary to steer public attention away from diseaseas a personal problem to health as a socialissue...(and)...advocacy is a strategy for blending science andpolitics with a social justice value orientation to make thesystem work better, particularly for those with the least resources(p5).931

The most successful public health policy reformershave based their advocacy on sound research dataand have utilised all three approaches to achievetheir objectives. The approach has been appliedsuccessfully in the areas of smoking control932 andalcohol.930

Advocacy is a political activity because it encouragessocial change via a political route. Advocacy cantarget the laws of Australian, State or localgovernment, policies of governments and privateinstitutions, or the actions of groups or industriesthat seek to oppose public health goals. Such changeis likely to challenge the status quo, and therefore,the concept and practice of advocacy often threatensthose in positions of power and with vestedinterests. The hostility generated by this challengeto the status quo is often a major barrier to advocacygoals.913

There is increasing recognition that successfulmedia advocacy is dependant upon theimplementation of coalition building and politicaladvocacy.

As Wallack notes

The reality is that mass media, whether public informationcampaigns, social marketing approaches, or media advocacyinitiatives, are simply not sufficient to stimulate significantand lasting change on public health issues. The power for changecomes from a broader advocacy that has widespread communitysupport. Coalition building, leadership development, andextensive public participation form the foundation from whichsuccessful advocacy and media initiatives can make a difference(p27).931

Media advocacy to reduce drug-related harm maytake a number of different forms. For example,media advocacy can be used to set a public agendaby heightening the profile of a drug-relatedproblem through the presentation of researchfindings; it can be used to espouse the benefits orsuccess of a program or intervention in order tosupport its refunding; it can be used to publiclyoppose or question the actions of members of thealcohol or tobacco industry when those actions are

�%(��

likely to increase harm; it can support the call forincreased resource allocation to address drug-relatedproblems; or it can highlight the inadequacies ofgovernment action to address drug problems.

In practice, media advocacy can involve manydifferent actions ranging from covert action, such asreleasing confidential information to the media, toovert actions such as issuing a media release relatedto concerns about an alcohol product (e.g. alcoholicice-blocks). Chapman and Lupton provide a list of66 advocacy issues, tips and discussion pointsillustrated with numerous examples that providepublic health workers with a comprehensive pictureof media advocacy in practice.933

Examples include

� Advertising in advocacy: careful use ofadvertising can support or even initiate news orcurrent affairs coverage as well as being anadvocacy tool in its own right. For example, alarge paid advertisement in the Australiannewspaper called on State Health Ministers tointroduce standard drinks labelling on alcoholcontainers. This advertisement was used togenerate media releases in each State with localorganisations available for interview;consequently, significant media coverage wasgained.

� Anniversaries: health promoters can oftencreate a ‘new’ newsworthy story out of a storythat occurred in the past by advertising ananniversary of an event or instituting a day ofremembrance. Such events will probably besignificant public events such as gun massacres,notorious chemical spills, nuclear accidents, orthe death of a famous person from a particulardisease or condition.

� Creative epidemiology: this is a term used todescribe the process of translating complexepidemiological data into media friendly terms.For example, if 18 000 people per year die inAustralia as a result of smoking, (on average)10 people die every day in Perth as a result ofsmoking. Large numbers can lose their impactand, therefore, it is often useful to localise andhumanise statistics.

� Letters to the editor: like advertising, thewriting of a letter to the editor of a newspaper isanother form of undertaking media advocacy. Itmay also result in further public debate andmedia interest. Such letters, however, must fitwithin the guidelines laid by the newspaper andare more likely to be published if well writtenand topical.

� Opinion polls: opinion polls can be a veryeffective part of media advocacy because theycan form the basis of a media release. The use ofpolls to support a view is, of course, a standardploy used by people for many years. Such pollsare often treated sceptically by the public but,nevertheless, they can be invaluable if usedcarefully. Even ‘quick and dirty’ polls of smallsample sizes which ask questions that producethe ‘right’ answers can prove effective whenreleased before a decision making process is tobegin, and when released by a respectedorganisation. Polls by opponents of the view, orthose of related issues, can also be valuable as anopportunity to present a case. A promptresponse will be required to ‘piggyback’ onsomeone else’s research.

Case studies of mass media advocacy

An example of the role that media advocacy can playto reduce alcohol-related harm is described byHawks934 and Stockwell and Single.935 They recordedthe process, outputs and outcome of attempts tointroduce compulsory ‘standard drinks’ labelling onall Australian alcohol containers. Ultimatelysuccessful, and the first such legislation in theworld, this minor policy achievement wasnevertheless complex and required considerableexpertise, time, and effort, as well as a mix ofadvocacy strategies to effect the desired change.Advocacy of policies that more directly threatencommercial interests (e.g. tax reform) has been lesssuccessful in Australia despite similar levels ofeffort.188

In Australia, the process of introducing the ‘standarddrink’ began in 1984 with the earliest publicationsreferring to the need to identify a standard unit ofmeasurement that would allow consumers toaccurately comprehend the alcohol content of liquorwithin alcohol containers. Early public discussionswere followed by an initial proposal forgovernment action and a lengthy period of researchinto both the need for public education and into thepublic support of the concept. The development anddebate of further submissions and a period ofadvocacy for change eventually resulted in thegovernment decision to act and the legislationpassed through parliament. The latter final eventoccurred in December 1995, almost twelve yearsafter the first action was initiated.

During this twelve year period, the year of 1994included extensive media advocacy from publichealth advocates who, by 1994, had established aformidable coalition that included a researchorganisation, the National Centre for Research into

�

�') �� !��"�#��$��%��

the Prevention of Drug Abuse; an advocacy agency,the Alcohol Advisory Council of Western Australia;and an industry group, the Winemakers Federation.Similarly, considerable political lobbying hadoccurred behind the scenes using the research datacollected to support the submissions developed.Likewise, two large industry groups representingbrewers and distillers undertook considerablelobbying activity, including the production of adocument entitled Standard drinks: Myths, facts and somesurprises, and the delegation of representatives tomeet with all relevant government ministersthroughout Australia. To counter industry lobbyingand to support the other advocacy strategies, aconcerted media advocacy strategy wasimplemented in 1994 in the lead up to, andfollowing, a meeting of the Ministerial Council onDrug Strategy (who were responsible forrecommending such action to the Government). Ahalf page advertisement supporting legislation of‘standard drinks’ was placed in Australia’s majornational newspaper by 19 individuals andorganisations, and a series of press releases wereissued during September and October, 1994, thatresulted in widespread media coverage. Publicconfirmation of government support for standarddrink labelling occurred on 30 September, 1994and the legislation passed in December 1995.

The experience provides important lessons aboutmedia advocacy. The first of these is that advocacyfor major change usually requires considerable timeand commitment. This is so even when healthadvocates hold ‘the high moral ground’ (as in thisexample where the opposition was identified by thepublic as tainted by the motive of financial profit).Furthermore, media advocacy and other advocacystrategies are essential to achieve change whenchange is likely to have powerful and financialopponents. Such opponents will almost certainlyundertake advocacy of their own. Additionally,media advocacy requires extreme care during bothits planning and implementation so as present aclear message that is supported by all members ofthe advocacy coalition. Finally, key individuals canbe critical to the success or failure of advocacystrategies, especially when they are at the centre ofthe flow of information and advocacy activity, orhave decision making power.

&=';'(� ��*� � ��--��1�� ,

The media activity undertaken as part of theCommunity Prevention Trials Project demonstrateshow mass media marketing and advocacy can beused to create prevention synergies. Holder andTreno described how media publicity was used aspart of this prevention project to highlight andsupport the specific prevention components thattargeted drinking and driving, underage drinking,responsible beverage service and alcoholavailability.922 As part of the drinking and drivingcomponent, local police departments were providedwith additional breath testing equipment and newpassive sensor devices, which provided anadditional aid in the detection of over-the-limitdrivers not presenting observable symptoms ofheavy drinking. Use of this equipment represented anew approach to the detection of drink driving andwas actively promoted to the media within theaffected communities because of its novelty andbecause what was being done in each communityhad national practice ramifications. There is amarketing aspect to this example because the mediacoverage was designed to increase the perceived riskof drink driving detection; but Holder and Trenoconsidered that the news coverage also encouragedincreased enforcement efforts by police because itindicated community support. In this respect, theinitiative should be considered media advocacybecause it sought to change institutional practice.These effects were confirmed in the projectevaluation and Holder and Treno drew threeconclusions from this media strategy.

1. Mass communication in itself is not enough toreduce alcohol-related trauma but can be usedeffectively to reinforce specific environmentalefforts to reduce high-risk alcohol-relatedactivities, such as drink-driving.

2. Local communication is best presented throughlocal news media and can focus public attentionon alcohol-related problems without having touse professionally produced material.

3. Media advocacy can be taken up by communitymembers if appropriate training is provided,which means that the capacity to use thisprevention measure is capable of beinginstitutionalised within the community.922

�'��

&=';'6��)��,��--�� -

Holder considers that mass media marketing andadvocacy are not likely to be sufficient to produce areduction in alcohol-involved trauma, even whenused together,936 although these strategies seem towork better in reducing the prevalence ofsmoking.912, 932 It seems that the two mediacomponents are most effective when they form partof a broader strategy that includes a range of otheractivities, such as community development andcommunity mobilisation, school and communityeducation, health promotion, policy developmentand institutionalisation, coalition building andpolitical lobbying. Thus, mass media marketing andmass media advocacy are valuable activities, as hasbeen well documented by alcohol harm reductionresearch projects, in the United States (e.g.Community Prevention Trial to Reduce Alcohol-Involved Trauma),727 Australia (e.g. CommunityMobilisation to Prevent Alcohol Related Injury)850

and New Zealand (e.g. Community ActionProject).840 It is within the framework of suchcomprehensive public health action that ‘bestpractice’ forms of mass media marketing andadvocacy have been, and will continue to be,successfully employed to reduce alcohol-relatedharm.

&=';'7�� +�� 1� ��)��+� �-�� -� �� 1�+�*�++� �


Australia is one of a handful of countries whosehealth authorities promote and regularly updatenational guidelines for low-risk alcoholconsumption. The National Health and MedicalResearch Council provided revised guidelines, inlate 2001.42 Earlier editions of these guidelines, in1987,937 and its revision published in 1992,938 havebeen among the most significant and influentialsource documents in the field. They are frequentlycited and have informed national media campaigns(e.g. Alcohol Go Easy) and a wide variety of briefintervention, road safety and health promotionmaterials commonly used by Australian healthprofessionals. It important to note that some ofthese related interventions are themselves evidence-based, as discussed elsewhere in this volume—inparticular brief interventions and some road safetycampaigns. The levels of drinking once defined as‘hazardous’ and ‘harmful’ (now ‘risky’ and ‘high-risk’) were the basis of the two other documents ofnational significance in the 1990s: the English etal.408 quantification of drug-related morbidity and

mortality; and the Collins and Lapsley939 estimates ofthe economic costs of drug misuse. English et al.408

used the NHMRC guideline definitions of drinkingrisk as the basis for many of their calculations onalcohol-related harm, while Collins and Lapsley,939

in turn, relied heavily on the English et al. estimatesfor their costing exercise. In short, the previousguidelines have been central to policy, practice andresearch in Australia in recent years.

Given the central role such guidelines have playedand the reliance made on them by other evidence-based strategies, it is hard to provide any evaluationof their effectiveness in isolation from other relatedstrategies and policies. A number of surveys haveevaluated the extent to which the messages theycontain are known to, and understood by, thepublic. Two key components of the messages are:the concept of a standard drink, and the number ofstandard drinks that can be consumed per day bymen and women with minimal risk to health. Asnoted above, in 1995 the CommonwealthGovernment approved a submission for theplacement of standard drink labels on all alcoholcontainers sold in Australia, in order to facilitatedrinkers being able to estimate whether or not theyhad exceeded recommended daily limits.935 Thisdecision was based on evidence from Australianresearch that such labels enabled beer and winedrinkers to make more accurate estimates of thealcohol content of commonly available beverages,and also to pour a standard drink from differentcontainers.935 A survey conducted prior to thelabelling decision also found that while mostdrinkers were aware of the concept of a standarddrink, few could define it accurately or apply itaccurately to estimate the number of standard drinksin commonly available beverages.940 Trackingresearch conducted by the CommonwealthDepartment of Health suggested that generalawareness of the concept of a standard drink hadrisen during the 1990s, but no formal evaluation ofthe labelling strategy has been conducted. It shouldbe noted that the developmental research onstandard drinks in the early 1990s used 9mm highlettering whereas the current requirements are forlettering of only a minimum of 1.5mm high; thismay well limit public awareness of these labels.941

The 2001 NDSHS found quite strong public supportfor providing larger labels on all alcohol containers,with 68% supporting this proposal.942 A generalpoint in favour of labelling alcohol containers as ameans of getting information out to drinkers is that,uniquely, the people who receive and can recall theinformation are those who drink the most.Therefore, this is a very efficient means of

�

�'� �� !��"�#��$��%��

providing information to risky and high-riskdrinkers.943 This fact has led to the proposal forconsideration of using alcohol container labels as ameans of conveying simple messages consistentwith the new guidelines.475 The 2001 NDSHS foundthat 71% of the large sample of Australians surveyedwere in favour of providing information about theAustralian Alcohol Guidelines directly on alcoholcontainers.942

In summary, we suggest that Australia’s nationaldrinking guidelines are a valuable basis for much ofwhat is done in the name of prevention, treatmentand policy on alcohol. Their wide disseminationwill assist numerous other evidence-based strategiessuch as brief interventions in primary health care,alcohol education in schools and controlleddrinking programs for early stage problem drinkers.The concept of the standard drink, and its accurateapplication, is also an essential part of applying theguidelines across all drinking settings. As such, widedissemination of the NHMRC guidelines and theircomponent messages and concepts, by whatevermeans, is an important aspect of a nationalprevention agenda relating to alcohol. This is not tounderestimate the difficulties of changing socialnorms supporting high-risk drinking patterns; it isto recommend that national drinking guidelines arean important ingredient in an overall nationalalcohol strategy and are recommended here forcontinued implementation.

�'��

�� $�� !!��

This chapter reviews the regulation and law enforcement of licit drugs. There is strong supportfor regulatory strategies for reducing the use of legal drugs and their related harms. A wide rangeof laws and regulations control or restrict the availability of licit drugs, particularly alcohol andtobacco. These include licensing restrictions, advertising controls, the regulation of sale andsupply of alcohol and tobacco to minors, and taxation initiatives. There are also regulations thatgovern the availability of pharmaceuticals, including restrictions on the provision of the over-the-counter (OTC) drugs and restrictions on prescription drugs that can be used recreationally. Othercontrol strategies include initiatives to limit the free availability of inhalants, particularly tominors.

Strategies that maintain or increase the price of legal drugs through taxation have the strongestsupport within the international research literature. The States and Territories of Australia have, inthe past, implemented hypothecated taxes on alcohol and tobacco, which direct revenue toearmarked treatment and prevention programs. These are now only feasible at the AustralianGovernment level. Unlike other taxation strategies, such hypothecated taxes tend to be wellsupported by the public. The related strategy of restricting price discounting is also wellsupported. However, marked price rises can lead to increased black market activity that can offsetsome of the health and safety benefits.

Other well-supported regulatory strategies include restrictions on advertising and marketing oftobacco products, restrictions on late-night trading hours of licensed premises, responsiblealcohol service supported by liquor law enforcement, voluntary local licensee agreements orAccords (also when supported by law enforcement), restrictions on sales of tobacco and alcoholto minors, and restrictions of alcohol availability in mainly Indigenous communities. There issome evidence to support defining the legal responsibilities of licensees towards intoxicatedcustomers through the development of model legislation, or Dram Shop Laws as they are knownin North America. There is a need for further research to test the effectiveness of controls onliquor outlets density, and restrictions on alcohol advertising.

Moves to limit the diversion of prescribed pharmaceuticals onto the black market have achievedsome success, but restrictions on the availability of performance and image-enhancing drugs havebeen less successful. Control of the supply of volatile substances to young people is very difficultand cannot be said to have been successful.

While tobacco regulatory strategies have been implemented more widely and effectively than isthe case for alcohol, in Australia, progress with reducing tobacco-related harm has slowed andnew initiatives may be required if there is to be further progress. New strategies are required, inparticular, to respond to the internationalisation of advertising through new media, both foralcohol and tobacco.

The distinction between ‘licit’ and ‘illicit’ drugs is somewhat blurred. Some drugs, such asalcohol and tobacco, can be legally sold to adults but not to minors; some prescription drugs canbe obtained legally but are used for non-medical purposes, or obtained with stolen or boughtprescriptions, by ‘doctor shopping’ or on the black market. Some communities have chosen tolimit or proscribe the supply of alcohol, whilst the use of alcohol and illicit drugs is prohibited inprisons.

The chapter is divided by drug type and reviews evidence relating to tobacco, alcohol, andpharmaceuticals, including volatile substances, and performance and image enhancing drugs.

��

�'� �� !��"�#��$��%��

�� !�� $ �� 9�� !�� !��

��( �� -�� 6��

� ��

)�� !��

�� $ �� 9��

+ �� 7 *�� 9��

��

#�� 7 *�� 7� � ��

��

*��( �� ( �� 9�#+7�� % ��

��

!�� G �� $�� 9 8�� 9�� /��0 �� 9�� 9�� 9 ��

��

+�� G �� '� ��8 �� 9�� 9�� 9�� <H�� <8�/��09��

��

7�� '�� 9�� 6��

7�� '�� ;�� ( � �� /<'#��#��<80�� $��9�� '��

( �� $�� < �� " �9�� 9�� 6�� 9� ��

�� 6��

( �� G �� ;�� $��

��9�� 6�� I��J9��

�'"��

�� !�� $ �� 9�� !�� !��

*��*�� 1.1�� '�� $ �� / %�%��0 ��

�� / %�%� �� 0 ��

�=.:��

��'��*�� ,�� 9� �� )� ��'�� '��9�� 9�� 9 ��

*�� )�� !�� !�� #�� <89��

��

� �� <8��'��#�9 )�� !�� I��J ��

�� +!�� K�� '�� I��J � � ��

��

( $�� L� �� 9�� '�� 9�� & ��

�� 6� ��

( $�� !�('�� $��& ��

( �� L�� * �� "�� K��

�� 9 ��

�� 6�� !��

�� &�� 7 !�� &�� 1===.=:�� 7��

>��7 *��

� �� $��

� #��

��

��

��

�'# �� !��"�#��$��%��

�� 3��

Definition: The impact of regulatory controls on theprice, promotion and availability of cigarettes has astrong empirical base. Controls on smoking inpublic places to protect non-smokers from theeffects of passive smoking are discussed as a harmreduction strategy in Chapter 14.

&&'.'&��-�� *��-� �� -�� -��-��


Advertising by tobacco companies has promotedimages of smoking likely to appeal to many youngpeople. This has led, in many countries, torestrictions or bans on tobacco advertising andsponsorship. Victoria introduced restrictions ontobacco advertising and promotion in 1987.Australia-wide, the Smoking and Tobacco ProductsAdvertisements (Prohibition) Act came into effect inDecember 1990, banning advertising in magazines.It was followed by the Tobacco AdvertisingProhibition Act 1992, which removed all tobaccoadvertising by 1995.

Although these measures have met with very strongsupport, there is relatively little good data on theeffect of these initiatives on adolescent smoking.Norway and Finland introduced similar legislationin the mid seventies. In Finland, there was no fall inthe prevalence of smoking following the ban onadvertising, but some change did take place inbrands smoked, as well as a shift to low tarcigarettes. In Norway, there was a reduction in theprevalence of adult smoking, and that for youngpeople appeared to follow this trend.

Cochrane reviews have demonstrated thatadvertising controls and restrictions on thepromotion of tobacco can reduce tobaccoconsumption in the general community,468, 783 butthat this can be undermined by media substitution iftobacco promotion switches from one media formto another. For advertising restrictions to besuccessful, broad restrictions on advertising andpromotion are required.944

&&'.'.�$�� -� �� -


It is now established beyond reasonable doubt thatoverall population-level consumption of tobacco isresponsive to the price of tobacco. Increases in pricecause decreases in consumption.783, 944, 945

Accordingly, it follows that taxation or other

legislative measures can be used by governments todeliberately increase the price of tobacco with theintent of lowering consumption.

The impact of these policies is not evenlydistributed across the population. Women andyoung people display the largest reduction inconsumption.783 There is concern, however, thatamong the poorest groups, increases in pricing cancause financial hardship for those who do notmodify their consumption.468, 783

Reductions in consumption associated with priceincreases appear to consist of a reduction in absoluteprevalence of smoking (through reduced initiationand increased cessation) as well as reductions inlevels of consumption among smokers.Significantly, there is tentative evidence that aroundhalf945 to two-thirds468 of the impact of priceincreases is on smoking prevalence rather thansimply on levels of consumption. Some evidencesuggests that large increases in price will have agreater proportional reduction in consumption thansmall increases in price.945 Overall, price increasescan be an effective means of encouragingpopulation-level reductions in overall consumptionand smoking prevalence.

The heightened impact of price controls on theprevalence of youth smoking has been taken toimply that sustained and substantial increases incigarette prices through taxation appear to be thesingle most effective method of achieving long-term reductions in population-level smoking.945 AnAustralian study of tobacco use by school-agedchildren in NSW and WA found a 10% increase inthe likelihood of being a smoker for every extrapacket of cigarettes the student’s weekly incomeallowed them to purchase when other key variableswere controlled for.453 The main limitation to thebenefits of price controls via taxation is thestimulation of black market sales. Recent customsdata indicate a substantial increase in black marketcigarette sales in Australia182 which may have beenstimulated by increases in tobacco excise ratesassociated with the introduction of the GST in July2000.

&&'.'0�#��+�� +��1��-��


Since 1969, State legislation has been in place torequire a health warning on cigarette packages.These warnings were strengthened, in 1985 and1992, with requirements relating to the size of thewarning, explanation of the warning, details of

�'%��

Quitline contacts and information on nicotine, tarand carbon monoxide contents. There is someevidence of small reductions in consumption inAustralian smokers after the introduction ofstronger health warnings on cigarette packets, in1995.946

&&'.'(�@��1� ��*�� )-�,


In 1982, a voluntary agreement was reachedbetween the Australian Government and tobaccomanufacturers on the levels of tar, nicotine andcarbon monoxide permitted in cigarettes.

&&'.'6��*��-�� )-��) ��-

It has been suggested that Indigenous communitiesshould be given the power to control or regulate thesale of tobacco for themselves,468 as some do withalcohol. This has not yet been attempted, butappears broadly consistent with the principles ofprevention programs in Indigenous communitiesand with the evidence that advertising andpromotion restrictions are effective in non-Indigenous populations. These two factors suggestthat such interventions are likely to be effective.

��1��

&&'0'&��-�� +��+�� -��,�) ��+�

Definition: The marketing of alcohol and tobacco iscarefully conceived and has often promoted imagesappealing to young people. This has led, in manycountries, to restrictions or bans on tobacco andalcohol advertising and sponsorship.


Australia removed all tobacco advertising by 1995.Although the marketing of alcohol has not been asstrictly regulated, industry codes prohibit productsand images designed to explicitly appeal to childrenand young people. As new products are introducedthrough marketing, such as caffeine drinks, theneed to reconsider marketing regulations isincreased.

Advertising is an important method used by thealcohol industry to promote its products. A numberof studies have investigated the impact of alcoholmarketing and distribution on adolescent health.Well-conducted longitudinal studies in NewZealand947, 948 have found that adolescents who have

greater exposure to alcohol advertising, and alsoenjoy the advertisements, are more likely to drink ina risky fashion as young adults. However, acomprehensive US review949 suggests these effectsare slight, and identifies six studies of advertisingrestrictions and young people’s drinking with eithernegative or inconclusive results. Responding togrowing concern about evidence for increasedalcohol use and related harms by young people, aWorld Health Organization technical advisory groupmet recently to review the latest evidence andprepare a draft WHO declaration advising membercountries to ensure young people are adequatelyprotected from alcohol promotions.950

One potentially successful harm reduction strategyis that of encouraging lower alcohol beverages.Consumption of low alcohol beverages may be aneffective method of reducing acute harm. In arandomised trial, the alcohol content of beverageswas varied at parties attended by young people. Thisstudy demonstrated that the amount of beverageconsumed did not alter in the low alcohol conditionand consequently youth BAC was lower uponleaving the party.951

&&'0'.��-�� -)��+,��+��+�� +��) �,

Definition: Reducing the availability and supply ofalcohol is a significant part of effective alcoholpolicy.

In compiling this section, we have drawn solely onreview studies.152, 186, 188, 524, 775, 952–961 The evidencebase cited here represents the collective results of anextensive, systematic, highly rigorous worldwideresearch agenda and many of the conclusionscontained herein can be made with a high degree ofcertainty. The strength of the evidence for eachindividual strategy is indicated.

A longstanding issue in alcohol policy, andparticularly in relation to supply reduction, is theissue of targeting the general population as opposedto only targeting high-risk groups of heavy drinkers,or alcoholics. Illustrating what is known as the‘prevention paradox’,11, 529 there is strongepidemiological evidence that the majority ofoccasions of acute alcohol-related harm affect themajority of drinkers, whose average intake can bedescribed as low-risk.530, 531

In fact, closer examination of this issue reveals that:

� a common pattern of occasional sessions ofheavy alcohol intake occurs among peoplewhose average daily consumption is low-risk,

��

�'' �� !��"�#��$��%��

� a significant proportion of alcohol intake inAustralia involves drinking at risky or high-risklevels for acute harm—estimated to be 51% ofalcohol consumed by the Australian populationaged 15 and over,151

� when risky patterns of alcohol consumption foracute and/or chronic harm from drinking arecombined, this comprises as much as 67% oftotal alcohol consumption.962

In each case, the above estimates are from the 1998NDSHS, which underestimates actual consumptionby over 50%.151 Australia’s new National AlcoholGuidelines now define risky and high-risk drinkingboth in terms of average daily consumption (forharms caused by long-term heavy drinking) andalso amount consumed on any one day (for harmscaused by the acute effects of alcoholintoxication).42 As such, the ‘prevention paradox’effectively disappears, at least for acute alcohol-related problems since these are almost entirelycaused by risky or high-risk sessional drinking.530, 531

The large contribution of risky and high-riskdrinking to total per capita consumption helpsexplain the close association sometimes foundbetween per capita alcohol consumption and ratesboth of acute and chronic alcohol-related harm.963

These associations have also been evident in theNational Alcohol Indicator reports on patterns ofalcohol-related harm and per capita consumption inAustralia, across both time and jurisdiction.151, 386, 456,

964, 965 It follows that measures which reduce theoverall consumption of the entire population arelikely to have a positive impact on risky and high-risk drinking and, hence, on the amount of alcohol-related harm in the community. It also follows thatmeasures that reduce the amount and frequency ofrisky sessional drinking will impact on totalpopulation consumption. This ‘whole ofpopulation’ approach is a substantial underpinningof supply reduction policies in regard to alcohol.For policy application, it has been pointed out thatjustifying supply reduction policies on the basis ofreducing per capita consumption of alcohol aloneinvites scepticism from those unfamiliar with theepidemiology of alcohol-related harm.966 A firmerfoundation is to seek to reduce risky and high-riskdrinking, whilst noting that such drinking patternscomprise the great majority of alcohol consumptionin contemporary Australia.

Within this framework, various approaches havebeen used to reduce the supply and availability ofalcohol to the community with the intent ofreducing overall levels of alcohol consumption and/or risky alcohol consumption. We briefly revieweach of the main measures below.

&&'0'0��+��+��2��


It is established beyond serious doubt that (all otherthings being equal) increases in the price of alcoholusually lead to an overall reduction in consumption,and decreases in the price usually lead to an overallincrease in consumption.775, 959 This has beendemonstrated by several systematic reviews of theinternational literature.775 Whilst the size of theeffect varies for different countries,959 differentbeverages (e.g. beer consumption is usually lessresponsive to price changes than is wine or spirits),the direction of the effect is highly consistent.775

There are also numerous studies reporting that priceincreases result in subsequent reductions in level ofalcohol-related harm.775 Some evidence exists thathigher beer excise taxes reduce the frequency ofyouth drinking and the frequency of heavy drinkingin youth.959 There is also evidence that beer excisetaxes are highly effective in reducing fatal trafficaccidents.959

For these reasons, many have proposed that taxationand other measures of influencing price (e.g. banson price discounting—see below) should be used tomodify consumption levels and reduce overall levelsof alcohol-related harm. In relation to taxation, theimpact on retail prices of a tax change cannot beguaranteed and, on occasion, may not be passed onat all to the consumer.

Collaborative international research hasdemonstrated that increasing the sale of low alcoholbeer is associated with a reduction in alcohol-relatedroad crashes.967 Other Australian research has shownthat lower levels of alcohol-related violence andhospital admissions are found in communities withrelatively high levels of consumption of low- andmid-strength beer as a proportion of all beerconsumed.968 Accordingly, recent moves by theAustralian Government to reduce taxation rates onlow alcohol beer are likely to result in a reduction ofoverall levels of alcohol-related harm, as a result oftaking market share from higher strength brands.

It has been found that drinkers will sometimes adaptto price increases so as to maintain their alcoholconsumption at the same cost, by changing tocheaper brands or types of drink. However, overallconsumption is still lowered even in the face ofsuch substitution.959

There is also a fear that price increases will reducethe beneficial effects of moderate alcoholconsumption, though neither the beneficial effects,nor their diminution in this circumstance, are

�'(��

supported by the evidence. There is a fairly broadconsensus in the medical community that low levelsof daily alcohol consumption are beneficial tohealth. The concern has been expressed that raisingthe price of alcohol will, therefore, reduce thebenefits. However, it is not certain that pricecontrols do reduce beneficial drinking. Heavier andyounger drinkers usually respond more to pricecontrols than other drinkers969 and these groups donot tend to experience any health benefits fromdrinking. In general, supply reduction measuresimpact more on young and high-risk drinkers.970

Furthermore, while the major health benefit ofmoderate drinking is thought to be a protectionagainst heart disease, there is no relationshipbetween population levels of alcohol consumptionand levels of heart disease.963

There has been significant attention from publichealth advocates in Australia to the relativities of taxrates across different beverage types.188 At present inAustralia, taxation on beer and spirits partly reflectsthe actual alcohol content of these drinks as they aresubject to an alcohol excise tax directly proportionalto the amount of alcohol they contain. Wine, bycontrast, is excise-free and is taxed only on itswholesale value, and its retail price through theGST. This has resulted in a major local market forcheap bulk (cask) and fortified wines, which havebeen shown to be strongly associated withcommunity levels of alcohol-related violence andhospital episodes.968 As a consequence of taxingpolicy, rates of taxation on cask wine per standarddrink, are about five times lower than on low andmid-strength beer.

As with the use of tobacco excise to reducesmoking, there is a concern that large increases inalcohol taxation will stimulate a black market foralcohol. There is also clear evidence that increasingthe price of alcohol is one of the least popular ofprevention strategies. The 2001 NDSHS found thatonly 20.5% of the general public supported the ideaof increasing the price of alcohol to reduce harm.942

Previous surveys of public opinion have found clearmajority support for hypothecated taxes, that is,specific taxes earmarked for treatment andprevention purposes.971

Hypothecated taxes on alcohol to fund treatmentand prevention programs

Definition: Hypothecated taxes collect revenue that isdirected towards a specific purpose or purposes. Ahypothecated tax on alcohol can be used to raisefunds for treatment and prevention purposes. NewZealand currently collects such a tax to fund anAlcohol Advisory Council. Between 1992 and 1997,

the Northern Territory collected a ‘harm reductionlevy’ of approximately 5 cents per standard drink inorder to fund the Living With Alcohol program.Constitutionally, only the CommonwealthGovernment can now raise hypothecated taxes onalcohol.


The full effect of this tax increase was passed ondirectly to the consumer. The levy was used to funda comprehensive range of measures includingmedia campaigns, increased alcohol treatmentfacilities and community-based prevention; knowncollectively as the Living With Alcohol program.Overall, there was a 22% drop in per capita alcoholconsumption over the four years following theintroduction of the levy. This was associated with asubstantial and significant reduction in alcohol-related road deaths (39%) and serious road injuries(35%).964 The authors of that report argue that theincrease in the price of alcohol due to the levy wasone factor contributing to these dramatic reductionsin harm. However, it is no longer possible for Stateand Territory Governments in Australia to introducesuch harm reduction levies since such taxes wereruled as unconstitutional in a landmark High Courtdecision, in 1997. Only the Australian Governmentcan introduce such policies.

&&'0'(��,-��+��+�*�+�,�� +-

Definition: Physical availability refers to the likelihoodthat individuals will come into contact withopportunities to obtain alcohol in their localenvironment.

Overall, physical availability has tended to increasein developed countries over the past 20 years.186 Ingeneral, it is well established that manipulating anyof the main aspects of physical availability (outletdensity, outlet trading hours, sales to minors, andservice to intoxicated customers) causes changes inpatterns of alcohol consumption and alcohol-relatedharm. However, as always, there are importantqualifications to each approach to controllingavailability, and local variation is an important issue.To a greater extent than with most supply controlissues, physical availability is primarily a local issue,though policies at the State or Territory level cancertainly facilitate opportunities for effective localregulation.186

A recent review of the international evidenceconcluded as follows: ‘Do reductions in availabilityreduce alcohol-related problems? While the answerto this question is usually ‘yes’, it is also sometimes

��

�() �� !��"�#��$��%��

‘no’, depending on the local context’ (p701).186

This conclusion applies to each of the examples ofalcohol availability outlined below, and reflects thereality that there are numerous variables affectingthe levels of alcohol consumption and related harmsin a community, of which level of availability isonly one. Furthermore, some forms of increasedavailability, for example, increasing the number ofrelatively small licensed premises that provide foodand high quality alcohol, may actually reduce harmby diverting customers from higher risk venues.

&&'0'6��)+�� -�,


At the local level, such as individual suburbs, thelevel of outlet density is highly predictive of levelsof alcohol-related harm.186 However, thisknowledge has not, as yet, been developed into anevidence base on how the manipulation or controlof outlet density can be used to reduce alcohol-related harm. The local impacts on one problem(e.g. violence) may be different to the impact onanother problem (e.g. car crashes) in relation to theareas where the reduction in problems will occur.

Some authors have raised concerns about basingpolicy on this literature base, as yet.775 The effects ofmodifying outlet density appear to vary accordingto the type of alcohol product; and, notsurprisingly, levels of alcohol-related harm varywith different types of outlet..775 However, theoverall evidence base remains clear that outletdensity is a powerful driver of levels ofconsumption and harm.186 There is a need todevelop and test a practical model for approvingliquor licences so as to maintain a balance betweenmeeting consumer demand and addressing publichealth and safety issues.

&&'0'7��)+�� )�-


Large, broad changes in trading hours (e.g. wholedays of sale added or taken away) are associatedwith significant changes in overall level of harm,although not necessarily with overall levels ofconsumption. There is recent Australian evidencethat even small changes (e.g. later hours) can beassociated with a significant local-level impact onalcohol-related violence.972 Recent research in NSWhas identified that licensed premises with thehighest levels of violence are far more likely to bethose that trade between midnight and 3am.973

A recent Scandinavian experiment involvingallowing nightclubs to move from one am closingto all-night trading resulted in a much higher levelof problems presenting to police and emergencyservices. 974 Some Australian licensing commissionstreat extended trading hours as a privilege that isgranted or withdrawn according to whetherlicensees are thought to be operating responsibly.This approach assumes adequate monitoring of alarge number of individual premises to be in place,whereas in reality information systems are ill-equipped for such demands and rarely include dataon alcohol harm incidents.975

&&'0'3��-�� -�*+��-��

Definition: Responsible service policies involve avariety of different approaches, usually aimed atreducing the chance that patrons will becomeintoxicated. These include ‘house policies’ such aspromoting food and non- or low-alcoholalternatives, as well as training staff to identify earlysigns of intoxication and delay or stop service asappropriate.

Summary: Evidence for outcome effectiveness;when accompanied by enforcement ................. ��Evidence is contra-indicative; withoutenforcement ......................................................... �

It is well documented that continued service tointoxicated patrons is a significant risk factor forpatrons experiencing alcohol-related harm.525

Generally, if responsible service programs aresupported by management, and implemented, theytend to be effective at reducing levels of intoxicationand in reducing the chance that drunk patrons willbe served.957 Various programs have been trialledwhereby bar staff have been trained in responsibleservice practice. Some training programs resulted inbehavioural change and some did not. This was notjust a function of the training program itself:behavioural change was also associated withmanagerial support and local enforcement of thealcohol service laws.524

The existence of laws prohibiting service tointoxicated customers on their own have nodeterrent effect in the absence of credible andvisible enforcement strategies. There is aninternational literature indicating that such laws arerarely enforced by police and are consequentlyfrequently ignored by alcohol retailers.186, 525 It iswell established that monitoring and enforcementof the laws is required to result in behaviour changein retailers leading to reduced serving of intoxicatedpeople.525 Profit is a powerful incentive for retailersto disregard the laws. As all Australian jurisdictions

�(��

have responsible service laws in place, encouragingresponsible service is usually more a matter ofenforcing existing laws than creating new ones.

&&'0';��-��) � �


Happy Hours and similar alcohol price promotionscause a rise in consumption and do so in a way thatis likely to cause increased risk of intoxication.959

Accordingly, such practices have been discouragedin many Australian localities, often through themechanism of a local community Accord (seebelow) or directly under instruction of a liquorlicensing authority. There is a sound theoreticalbasis for banning discounting of alcohol, and alsosome direct evidence that such interventionscontribute to harm reduction when part of anAccord package (see below).

&&'0'<%�� -��-�� )��8�'�'��-9

Definition: Negotiated agreements between police,licensees and local councils covering standards ofservice and promotion, for example, banningpractices such as heavily discounting drinks; oftenknown as Accords.

Summary: Evidence for outcome effectiveness;when accompanied by enforcement ................. ��Evidence is contra-indicative; withoutenforcement ......................................................... �

Accords are a uniquely Australian phenomenon. Theidea emerged from the pioneering work of RossHomel and colleagues,524 and originated inQueensland. There are now some publishedevaluations of these community-basedinterventions, though only one is published in thepeer reviewed literature.976 Evaluations of thesemethods suggest significant reductions in alcohol-related violence in the short term but difficulty insustaining gains after a few months.524 To be fullyeffective over the longer term, the policing oflicensed premises must incorporate elements oftraditional enforcement as well as the developmentof voluntary codes of conduct. These localapproaches can be seen within the broader contextof the global move over the past two decadestowards alternatives to centralised ‘command andcontrol’ approaches to regulation, and an increasedemphasis on local negotiation and monitored self-regulation.977

&&'0'&=�"��+��-

Definition: An established body of law and principlewhereby those harmed by intoxicated persons have,under certain circumstances, the right to sue thelicensee of the premises where they served alcoholto the intoxicated person concerned. These laws,which operate in US and Canada, are usually knownas Dram Shop laws.


The American literature suggests that there is somemodest deterrent effect of Dram Shop laws.960 Theunderlying rationale of deterring service tointoxicated customers is sound and there is nolikelihood of adverse consequences. However, theiruptake is limited in Australia by the widespread useof public liability insurance, and the low probabilityof litigation being successful.

To date, Australian courts and legislation have notacted to establish principles for civil actionregarding alcohol service,953 although it has beenshown that existing legal principles contain thepotential for successful civil liability action. Areview of Australian liquor laws determined that,whilst by 1996 there had been no successful civilcases based on liability for serving customers tointoxication, some out-of-court settlements hadbeen made. The review noted that Australianattitudes about individual responsibility and the useof alcohol may limit the scope of alcohol-relatedcivil liability.

&&'0'&&��-�� -)��+,��+��+� �� )-��) ��-

Definition: Indigenous communities have taken twomain approaches to reducing the supply of alcohol:declaration of ‘dry’ areas, and use of liquorlicensing legislation to extend controls onavailability. Other communities have establishedtheir own ‘wet canteens’. Some groups have alsolobbied for changes in licensing legislation becausethe legislation has been seen to be biased in favourof the alcohol and tourist industries, limiting theopportunity of communities to make decisionsregarding the sale and consumption of alcohol intheir midst.978, 979

Declaration of Indigenous communities as ‘dry’


Legal procedures enabling Indigenous communitiesto declare themselves ‘dry’ vary betweenjurisdictions. These procedures and their effects, inthe NT, WA and SA, have been reviewed by

��

�(� �� !��"�#��$��%��

d’Abbs.980 He found that the procedures can beeffective but that communities needed support toenforce them and the underlying policies mustpromote Indigenous control. These findings reflectthose of an earlier study by Larkins andMcDonald.981

Licensing restrictions in Indigenous communities


Indigenous groups in the NT and WA—often incoalition with non-Indigenous groups—haveutilised liquor licensing legislation to extend therange of restrictions on the availability of alcohol.Restrictions commonly include limitations of hoursof sale and banning the sale of wine in casks ofmore than 2 litres (effectively a price controlmeasure). Evaluations of restrictions have beenconducted in Halls Creek982 and Derby983 in WA,and Tennant Creek984–986 and Curtin Springs987 in theNT. Generally, restrictions have been found to beeffective in reducing consumption and keyindicators of harm, such as hospital admissions andpolice arrests. They have been most effective whenthey have been initiated by Indigenous people,conducted as part of broader strategies to addressalcohol-related harm, and have had widecommunity support.988, 989

Many communities in remote Australia haveattempted to exercise some degree of control overthe availability of alcohol by operating their ownlicensed ‘wet canteens’ or ‘clubs’. As Brady notes,these had their genesis in the 1960s, following therepeal of legislation prohibiting alcoholconsumption by Indigenous people. Initially theywere introduced on mission settlements in anattempt to teach Indigenous people to drink in a‘civilised’ manner.990, 991 Communities have takenvarious approaches to the operation of canteens,including greater or lesser restrictions on theamount of alcohol that can be purchased.

The Royal Commission into Aboriginal Deaths inCustody reported that many communities expressedconcern about the impact of canteens,292 and Martinhas identified potential conflicts of interest betweenattempts by community councils to controlconsumption and their dependence on canteenprofits as a source of income.992 As d’Abbs notes,the operation of canteens has both risks and benefitsfor communities and, echoing a recommendation ofthe Royal Commission into Aboriginal Deaths inCustody, stresses the need for communities to beassisted in minimising the risks.993

��!��

Definition: Various measures to limit the recreationaluse and misuse of pharmaceuticals have beenadopted.

The major drugs of concern are prescribedbenzodiazepines and narcotic analgesics; although itshould be noted that excessive use of OTCmedications by some young people is a source ofconcern. Another area of concern among youngpeople is the diversion of medications (primarilydexamphetamine and methylphenidate) prescribedfor ADD and ADHD. Most accounts of this diversionare anecdotal, and it is not known whether there arespecific interventions to limit it.

Monitoring and education to prevent ‘DoctorShopping’


One way to limit inappropriate levels of access toprescription drugs has been through the DoctorShopping project, which is managed by the HealthInsurance Commission (HIC). ‘Doctor shoppers’were defined as people who visited 15 or moregeneral practitioners, had 30 or more Medicareconsultations, and obtained more PBS prescriptionsthan appeared to be clinically necessary, in a singleyear. When the project commenced in January1997, HIC employed pharmacists in each Statecapital to visit people identified under this programand counsel them about overuse of prescribeddrugs. The aim of the project is to improve healthoutcomes for these patients, reduce unnecessaryvisits to medical practitioners, and reduceunnecessary use of PBS medicine.994 HIC data showthat in 1995/96, 13 240 people met the descriptionof a doctor shopper. By 1999/ 2000, this numberhad fallen to 8780. Of the total PBS medicinesobtained by doctor shoppers, 36% werebenzodiazepines, 15% were codeine compoundsand 8% were narcotic analgesics.

The Doctor Shopping project will move into itssecond phase following a review of the first threeand a half years of the project and will continue tofocus on PBS usage.994 No evaluation of theprogram, other than the reduction in number ofdoctor shoppers identified, appears to be available.

Rescheduling temazepam


An alternative approach to limiting the availabilityof prescribed drugs is by changing the schedulingunder which drugs can be prescribed. Moodaltering prescription drugs are frequently diverted

�(��

from medical to recreational use, often by injection.Benzodiazepines are among the prescription drugsmost frequently diverted in this way, andtemazepam has been one of the most preferred forinjecting because it has been available as soft,gelatine liquid filled capsules.995 The ‘heroindrought’ in Victoria from late 2000 has beenassociated with the a marked increase in theinjection of temazepam capsules.995

In the United Kingdom, temazepam capsules wererescheduled and banned from being prescribed, inJanuary 1996. It was found that despite substitutionof other benzodiazepines, there was less frequentinjecting and consequent net gain in health.995

Other approaches trialled before capsules wereremoved included education and information fordoctors and pharmacists in order to restrict access.However, these approaches failed despite asignificant and continued effort over time.996

In Australia temazepam capsules have been movedto Schedule 8 of the Drugs, Poisons and ControlledSubstances Act in order to reduce their use byinjectors. Schedule 8 refers to ‘controlled drugs’,whose manufacture, supply, distribution, possessionand use are restricted for the purpose of limitingtheir use or misuse and physical and psychologicaldependence on them.200 Temazepam (10mgcapsules, in the 25 pack size) now requires anauthority prescription under the PharmaceuticalBenefits Scheme247 and it is hoped that this willreduce their availability to injecting drug users. Therescheduling of flunitrazepam (Rohypnol) toSchedule 8 in 1998 (see below) was effective indramatically reducing the number of flunitrazepamprescriptions and its availability to the injectingdrug user market.996 This experience, along with theUK success in reducing the availability oftemazepam, suggests that the intervention will beeffective.

Rescheduling flunitrazepam


Flunitrazepam, as noted above, was placed onSchedule 8 in 1998 because of its association withdrug-assisted sexual assault.200 The 1999 IDRSreported that flunitrazepam availability in the blackmarket, and use by injectors, had declined since theprevious survey although it could still be obtainedillicitly.200 There is no current evidence as towhether the schedule move has reduced theincidence of drug-assisted sexual assault.

��"�? ��3��

Definition: Most Australian States make the sale anddistribution of inhalants and volatile substancessubject to criminal penalties in certaincircumstances.997

Whilst legislation varies from State to State, in manyit is an offence for a person to sell a deleterioussubstance to another person if there is reasonablecause to believe that the other person intends to usethe substance for the purposes of intoxication.997 Itshould be noted that, across Australia, there is nouniform definition of volatile substances. In someStates, the definition is according to the type ofproduct, whereas in other States the definition maybe according to the chemical components

There are various legal provisions relating to thecontrol of volatile substances across Australia.

� In South Australia, the Government has powerto make regulations with regard to theproduction, manufacture, distribution,packaging, sale, prescribing, possession andstoring of volatile solvents.

� In Western Australia, legislation allows forIndigenous communities to make bylawsagainst petrol sniffing and other forms ofinhalant abuse within their community landsand boundaries.

� In South Australia and the Northern Territory,similar provisions in relation to Indigenouscommunities to those in Western Australia havebeen enacted.997

Restricting volatile substances


There are divided views as to the efficacy ofinitiating laws aimed at suppliers and distributors ofvolatile substances and associated by-products.997 Inthe UK, it was found that laws aimed at prosecutingand restricting suppliers of volatile substanceproducts had not reduced volatile substance abusebut had resulted in a switch from glue to butane,which was more dangerous. Other possibilitiesinclude restricting the sale of products containingvolatile substances to people over 18, as withtobacco and alcohol. This approach has been takenin the UK with the sale of butane lighter refills, butthere is, as yet, no evaluation on the effectiveness ofthat measure.997

There are problems associated with restricting salesof volatile substance products to juveniles,including the huge range of products that can be

��

�(� �� !��"�#��$��%��

used, the possibility that if products can not bebought they will be stolen, and the need to workcooperatively with traders in the industry. InWestern Australia, a Retailers Resource Kit has beendeveloped which is aimed at supporting businessesto restrict the sale of solvents and to take positiveaction regarding the availability of solvents andother substances; but no evaluation of this approachappears to be available.997

Whilst all Australian States have adopted standardsfor the scheduling of drugs and poisons, thecompounds and products used by young peopletend to be either exempt from scheduling or locatedin the schedule subject to the least restrictions.There have been calls for butane and toluene,among other substances, to be scheduled in thisway, and for health warnings to be displayed onpackaging and containers. In the UK, healthwarnings on aerosol products have been positivelyreceived by marketers.997

Controls on the supply of petrol in Indigenouscommunities


Indigenous communities have used supplyreduction strategies to reduce petrol sniffing andrelated harm. In communities in Central Australiaand Arnhem Land, aviation fuel—which does nothave the same psychoactive effects as petrol—hasbeen substituted for petrol.233, 888 Again, this hasbeen most effective when introduced in conjunctionwith other interventions. However, its effectivenesscan be compromised when petrol remains availablefrom other sources.233 Another measure to reduceavailability has been to lock petrol supplies incommunities, but this has had virtually nosuccess.233

In the Ngaanyatjarra Lands in WA, and thePitjantjatjara Lands in SA, sniffing or supplyingpetrol for sniffing has been made illegal. It is alsoillegal to supply petrol for sniffing in the NT. Noformal evaluation of these measures has beenconducted, but anecdotal evidence suggests theireffectiveness is equivocal—particularly as petrol iswidely available and the sanctions themselves donot act as a strong deterrent.233

��#�� !��!��&��

Definition: PEDs include anabolic and androgenicsubstances (steroids), and hormone preparations.

Most of these drugs are available only onprescription and some, like DHEA, cannot beimported without a Commonwealth Governmentpermit. Considerable attention was paid to thesupply of these drugs in 2000/01 because of theSydney Olympic Games.168

PEDs seizures

Summary: Limited investigation ............................... O

The number of PEDs seizures has been increasingsince 1994/5. During 2001/02, Customs make arecord 1630 seizures of PEDs, which represented anincrease of over 28% on seizures in the previousyear.165 Steroids were the most frequently seizeddrugs; seizures generally involved small quantities,mainly for personal use. Many of the substances arelegally available overseas and are frequently orderedover the internet and mailed to Australia. Almost all(86%) seizures in 2000/01 were through the postalstream, reflecting this trend, but some largerseizures (e.g. 10kg, 2.4 kg, 250 vials) were alsomade.168

�("��

�� $�� !!��

In this chapter, legislative and regulatory interventions to prevent and reduce the use of illicitdrugs are reviewed. The chapter opens with a consideration of evidence that law enforcement, byreinforcing community values against illicit drug use, plays an important role in prevention. Inits discussion of law enforcement and regulatory programs in the community, this chapterdistinguishes ‘supply-side’ from ‘demand-side’ law enforcement to reflect the role that supplyreduction and demand reduction play in Australian drug policy. It is recognised, however, thatthese are somewhat artificial constructs, with overlapping domains. The control of illicit drug useis primarily aimed at users, to reduce their demand for drugs; and at suppliers, to reduce theavailability of illicit drugs. Australian programs and activities aimed at reducing supply anddemand are described and evidence for their effectiveness reviewed. For the most part, Australianand international research reviews, rather than primary studies, have been used. In the main,there has been limited research and evaluation in Australia in this area. Some US initiatives showpromise, particularly in targeted policing. Further investment is required in the development ofintegrated monitoring data sets, as well as research and evaluation, in both Australia and theUnited States is required.

�� !�� $ �� 9�� !�� !��

�� (� �� !�� 9� �� /� �� 0 �� (� �� '�� 9�� /� � �� 0

(� �� 7 �� /�� 0

�� <�� '�I# �� ' �J�� 1== �� $ �� $ ��

�� (�� 9�� G�� '��9 ��

�� 6� ��

�� 7� �'�� @==�� < �� $� ��

��

�� %%%

��

�(# �� !��"�#��$��%��

�� !�� $ �� 9�� !�� !��

$��@�� -�� K � M�� & �9� �� !��1==:�� @%1;�:�� /�!�(0

�� $� �� % �� &��

-�� K � M�� & �9��<�7� �� )�� ' �� 9��'��

��

<��+ �� $�� '��

�� ( �� 9�'�� 9��

�� 6�� - �� & �� J��

�� *��( �� G�� '� ��'� �� #��9�#�� K�#�� 9��

��

>��7 *��

� �� $��

� #��

��

��

�� 5�� !��!�� !!��

Several mechanisms have been proposed to explainthe way in which laws shape community values andopinions about illegal behaviours, including druguse. A major review of US illicit drug policyundertaken by the National Academies of Science,Engineering and Medicine, and the NationalResearch Council176 examined this issue in detailand found that the nature of the link between druglaws and other forms of social controls is a centralpoint of dispute. Generally, there appear to be twomajor mechanisms by which sanctions against drugusers may depress drug use prevalence in thegeneral community: by expressing social normsagainst drug users (declarative effects), and bydissuading people from using drugs due to fear ofapprehension and punishment (deterrent effects).

&.'.'&�"��+��-

Definition: ‘Laws against drug use may generatedeclarative effects by expressing social disapprovalof drug use and therefore symbolising andreinforcing social noms against drug use andhelping to shape individual beliefs and attitudes’(p191).176


If drug laws reflect community antipathy towards abehaviour, sanctions against that behaviour maygenerate moralising effects, but in conditions of‘normative ambiguity’ sanctions may also generatereactivity in an alienated population.176 Theempirical literature testing these propositions isscant, however, because it is difficult to disentangledeclarative effects from deterrent effects and theeffects of informal from formal social sanctions.International research relating to tax compliance,theft and drink-driving found that in all three cases

�(%��

the existence of legal sanctions had a preventiveeffect on intention to offend, both through thethreat of punishment and internalised shame.998

Australian research concurs. Braithwaite reportedthat compliance with the tax system is shaped bybroad and relatively enduring factors such asconfidence in the regulatory system.999 In a study ofnursing homes, Makkai and Braithwaite found thatrespect on the part of a sanctioning agent increasedthe likelihood of future compliance.1000 No studies,however, have isolated the declarative effect ofsanctions against use of illicit drugs from theirdeterrent effect.176

Informal social norms play a major role inregulating psychoactive drug use.1001 Drug use ispowerfully governed by social pressures, or theinformal social conventions in people’s everydaylives. These conventions vary dramatically and canbe highly conducive to using or misusing drugs, orcan exert a strong protective effect against druguse.176, 397 The interaction of informal norms andformal laws may be a more powerful source ofinfluence on crime than either in isolation.397 Theseeffects are as significant for the young as for adultmembers of the community.

&.'.'.�"�� -

Definition: One of the major explanations for crimeavoidance is deterrence theory which asserts that‘undesirable behaviour can be curtailed ifpunishment is sufficiently certain, swift, andsevere’. (p 585) 1002


Deterrence theory assumes that behaviour isinstrumental, that is motivated by rewards andpunishments. There are two major questions indeterrence theory: to what degree do legal sanctionsagainst drug use deter future drug use (generaldeterrence), and to what degree does being legallypunished for drug use deter future drug use(specific deterrence)?176 The elements of deterrenceare celerity (swiftness), certainty and severity ofpunishment, although there has been little researchon celerity.1001 In relation to crime in general, theperceived certainty of punishment appears to be themajor driver of deterrence but there is virtually noresearch on the deterrent effects of criminalsanctions on illicit drug use in Australia. It is unclearwhether US research can be applied to Australiagiven differences in drug policy.397

General deterrence

Some Australian research suggests that the legalstatus of cannabis use might be a deterrent againstuse, particularly for those who had never used thedrug, but the major element of deterrence does notappear to be fear of police or fear of arrest.1003

Studies of variations in cannabis law across Australiahave found that, contrary to the predictions ofclassical deterrence theory, there has been noincrease in use of cannabis in States such as SouthAustralia, which have moved towards the relaxationof criminal penalties for cannabis use.1004 Theseresults are in accord with those in otherjurisdictions that have changed the legal status ofcannabis, such as some US States; this kind ofresearch has not been undertaken for studies otherthan cannabis.397

There is also the question of who is most likely tobe deterred: those who have never committed anycrime, or those who have committed crimes butavoided punishment? It has been argued that thesecond category have, by definition, acquiredexperience with avoiding punishment and that thisavoidance of punishment is likely to diminishperceptions of the certainty and severity ofpunishment.1005 By this argument, deterrence mayhave its strongest effect on those who have nevercommitted any crimes.

Specific deterrence

Definition: Specific deterrence refers to the degree towhich being legally punished for a crime detersfuture criminal activity.


Specific deterrence appears to be unsupported bythe existing evidence. Given the high rate of re-offending among those convicted for illicit druguse, it is difficult to envisage that contact with thelegal system acts to deter further use.176 However,far more detailed data than is currently availablewould be necessary to adequately answer thisquestion.176 The strength (or severity) of legalpunishment appears to have very little impact ondrug use176, 397 and studies comparing prevalence ofuse with the existing legal sanctions find norelationship.176 Accordingly, there is a lack ofevidence to support the widely held notion thatharsher sentences will lead to a reduced likelihoodof drug use. ��

�(' �� !��"�#��$��%��

�� 5�� !��!��! ��

Demand-side drug law enforcement is focused atillicit drug users, whilst supply-side drug lawenforcement is focussed at sellers of illicit drugs.Demand-side drug law enforcement specificallyaims to disrupt illicit drug markets and, by sodoing, to encourage drug users to give up or reducetheir drug use, often by entering treatment.

The principal goal of drug law enforcement is todisrupt illegal markets.397 One effect should be, byreducing their availability, to drive up the price ofillicit drugs. Some have argued that drug lawenforcement is ineffective because illicit drugscontinue to be available and used; with all thecommunity disruption, crime and loss of publicamenity, as well as problems for individuals andtheir families, that that entails. However, the sizethat the drug market would be without enforcementcannot be known, and so the true measure of thesuccess of drug law enforcement cannot begauged.397

Different law enforcement strategies may be neededfor different stages of a drug epidemic. Caulkinsargues persuasively that interventions early in anepidemic can have a much greater impact on totaluse throughout the epidemic than laterinterventions. Law enforcement, because of itsimmediacy and drug specificity, can be highlyeffective in curtailing the growth of an epidemic inthese early stages, when distribution lines are lessrobust and can be more easily disrupted. Later in anepidemic there are more addicted users andsellers.1006 He applied this analysis to a comparisonof the possible effectiveness of law enforcementagainst the US and Australian cocaine andmethamphetamine epidemics. In the case ofcocaine, law enforcement is likely to be lesseffective in the US where the cocaine epidemic ismature, than in Australia where it is relatively new.In both countries, the methamphetamine epidemicsare relatively recent and liable to be stronglyinfluenced by law enforcement.1006

Evaluated approaches to demand-side lawenforcement include the combination of targetedlaw enforcement with community development, theuse of civil remedies to deal with drug problems,police ‘crackdowns’, and the role of lawenforcement in encouraging illicit drug users toenter into treatment programs.

&.'0'&� ��*� ��+�� ) �,��+��

Definition: These approaches operate within targetedsites and combine partnership development,concentrated law enforcement against drugoffenders, community policing to develop trust andcooperation between police and the community,and resources for community program andinfrastructure development.


‘Weed and Seed’ is a US strategy to: control violentcrime, drug trafficking, and drug-related crime; andto encourage the provision of a safe livingenvironment for residents in targeted areas. It isfunded by the Federal US Department of Justice andin 1999, included 200 sites nationwide, mostreceiving funding of about $225 000 annually.1007

Weed and Seed mobilises and coordinates resourcesin high crime communities, aiming to stabiliseconditions and promote community restoration.There are two key components of the strategy:concentrated law enforcement to Weed outoffenders, and community policing to develop trustand cooperation between police and thecommunity; and Seeding the community withinfrastructure and problem behaviour preventionstrategies tailored to its needs. Both aspects requirethat local residents and agencies work incollaboration with law enforcement.

The national evaluation of Weed and Seed involvedeight sites spread around the country and selectedfor their different Weed and Seed applications. Allsites shared high rates of violent crime related todrug trafficking and use, and most had seriousgang-related crime problems. Evaluation dataincluded local crime and arrest statistics, interviewswith key informants and program participants, andresident surveys.

Although the findings varied across the eight sites, amajority of sites demonstrated a fall in the rates ofmajor property and violent crimes in both years ofoperation. Drug-related arrests followed similarpatterns. Moreover, in most sites these crime ratesdeclined more, or increased less, than in the rest ofthe city or county. Five of the eight sites exhibitedat least some, if not substantial, evidence ofimprovement in residents’ perceptions of theseverity of crime and police effectiveness.1007

The evaluation demonstrated that the two majorelements of Weed and Seed—targeted lawenforcement and community mobilisation—werethe mechanisms for program effectiveness.

�((��

However, it was strongly suggested that fundingshould be concentrated in fewer sites for longerperiods to maximise the likelihood that self-sustaining interventions would be developed.1007

Longer-term and more extensive evaluations areclearly needed, as are cost-effectiveness studies, toestablish whether this program represents goodvalue for money. Taking the pre-existing differencesof evaluated sites into consideration, however, thecurrent evaluation suggests that this may well be apromising approach to the restoration ofneighbourhoods and the reduction of drug-relatedharm.

&.'0'.��-��+�� +��-�� +��)�� -��*+��-

Definition: Civil remedies are procedures andsanctions specified by civil statutes and regulations,used to prevent or reduce criminal problems. Theytypically aim to persuade non-offending thirdparties to take responsibility and action to preventor end criminal or nuisance behaviour.


The use of civil remedies to prevent or reduce signsof neighbourhood disorganisation relating to druguse is predicated on the notion that civil remediesare more accessible, through cost and reducedburdens of proof, to frustrated and disadvantagedcommunities than are criminal penalties.1008 Civilremedies tend to be proactive and preventive,focussing on improving the quality of life. Theyinclude pressures on property owners and managersto clean up properties, act on health and safetyviolations and evict problem tenants. Otherapproaches include bans, injunctions and restrainingorders to prevent potential offenders from criminalbehaviour.

A randomised field trial of the use of civil remediesfor drug control was held in Oakland, California,using a program called Beat Health, which wascreated by the Oakland Police Department, in 1988.This program sent teams to troubled sites, providingaccess to services, advice on citizens’ rights andresponsibilities, and, where appropriate, legal actionagainst owners of properties with drug problems.

The research included all problem sites referred toBeat Health for three months, in 1995. A carefulresearch design was used to ensure that: there wasrandom allocation to control or experimentalconditions, effects on commercial and privateproperties were differentiated, and that spatialconfounding to adjacent areas was minimised. Drug

dealing was reported as a major problem inapproximately three-quarters of the locations priorto the start of the research. In the experimentalgroup, Beat Health officers visited all but two of theproblem sites and used a variety of tactics to resolvedrug and disorder problems. In the control group,city blocks containing problem sites were patrolledand offenders were arrested. The major outcomemeasure was the number and type of calls byresidents for police service related to violent,property, disorder, or drug offences. It was foundthat the program was effective in reducing callsrelated to drug problems, although not calls relatedto the other offence categories. It was suggested thatthe use of civil remedies could be an importantapproach to the reduction of drug-relatedproblems.1008

&.'0'0��+��1�� -

Definition: Short intensive burst of police activitydesigned to move drug users and sellers away froman area.


Crackdowns are situational crime prevention—anapproach that considers the nature of criminalevents and the settings within which they occurrather than the motivations of offenders. Therationale for situational methods comes from arecognition that offences are not evenly distributedacross geographical areas but are found in clusters.Situational crime prevention methods have beenapplied to the policing of drug markets, frequentlyin conjunction with locally-based enforcementinitiatives or crackdowns. Although theseapproaches have been applied in the UK as well asin Australia, most of the research literature emanatesfrom the USA.1009

The international literature on police crackdownsand their impact is somewhat inconsistent. Whilesome studies report reductions in drug use, othersreport no effects; others report increased use.397

Some better-controlled studies have foundindications of reduced drug use followingcrackdowns, but this was accompanied by increaseddrug use in neighbouring areas. There is alsoevidence that crackdowns vary significantly in theireffectiveness, according to the local characteristicsof the area involved and means available to drugmarket participants to evade the effect ofenforcement.397

One well-known American example relates to theJersey City Drug Market Analysis’ experiment.1010

Arrest data and community knowledge were used to

��

�)) �� !��"�#��$��%��

map the street-level drug markets, which were thendivided into experimental and control groups. Thenew strategy was applied to the experimental sitesand involved information collection, intensiveenforcement for periods of a few hours to a fewdays, and increased police activity, thereafter, tomaintain gains. Control groups received no specialattention.

The major outcome measure was a comparison ofthe number of emergency calls made for service atthe hot-spot locations during the seven monthsprior to and following the intervention. There wasgood evidence that the new strategy reducednarcotics activities, although not violent or propertyoffences, in the localities around the intervention;and some evidence that disorder was also reduced inthe experimental sites.

Australian examples of effective police crackdownscan also be given. In South Australia, a problem-oriented operation aimed at street drug markets,called ‘Operation Mantle’, was successful in limitingthe growth of drug-related crime.1011 ‘OperationPuccini’ was a police crackdown designed to reducethe illicit drug trade in Cabramatta in 1997.1012 Theauthors of these reports found that the benefitsarising from this kind of activity were wide-ranging. Distribution networks can be disrupted anddealers become more difficult to access and,although some displacement may occur, it isunlikely to be total so that overall drug use will bereduced. The dispersal of offenders will contributeto less criminal activity, thereby improving thequality of life for the local community. Increasedpolice visibility also builds confidence in thecommunity so that people use public spaces moreand may be more willing to report illegalactivity.1012

&.'0'(��) � � �� -�4)� ��-��1�� -

The costs of these policing strategies include theneed to ensure there is adequate funding tomaintain policing at sufficient intensity for longenough; a possible increase in neighbourhoodcrime to fund the increased price of illicit drugsfollowing reduced availability; increasedopportunities for police corruption; and concernsthat the police may be seen as intrusive.1012 Policecrackdowns in Australia have also been found toresult in unsafe injecting practices and increasednumbers of syringes discarded by users in a hurry toavoid detection.336, 1013 Police efforts to suppressstreet-level trafficking may inadvertently give rise topredatory crime by user/dealers seeking alternative

sources of income.336, 1014 In addition, vigorous lawenforcement may unintentionally lead to marketsbeing dominated by more harmful criminal groupswho are willing to work in a higher riskenvironment.1013

Dixon and Coffin investigated ‘zero tolerancepolicing’ (ZTP) of illegal drug markets using policeoperations in Cabramatta as a case study.1015 ZTP ismodelled on a US program in which police focus ondisorder and street offences with expectation ofcrime reduction, and engage in intensive operationsin specific locations. With drug markets, policeattention is focussed on street sales with a view togaining information that will lead to identifyinghigher level dealers. Dixon and Coffin found that, ingeneral, whilst such crackdowns may producetemporary reductions in activity, attention to highrisk people and places does not lead to a sustainedreduction in drug- related offences.1015

The crackdown in Cabramatta (‘Operation Puccini’)resulted in increased public health hazards such as:unsafe storage, transfer and injection of drugs,diffusion of markets, harm to police/communityrelations, and the encouragement of greaterorganisation in the supply of drugs. There was alsoan increased likelihood of high-risk injectingpractices such as rapid injections, needle sharingand improper disposal of syringes. Furthermore,civil liberties were diminished. Dixon and Coffinconcluded that the US ZTP model was not effectivefor drug-related offences since the US had thehighest rate of addiction, drug-related healthproblems and drug-related crime recorded in theworld, and that the role of police might best be seenas regulating, controlling and shaping marketsrather than eradicating them.

In relation to displacement of crime from one ‘hot-spot’ to another because of targeted policing (orcrackdowns), the perception in the community thatcrime will move from one targeted area to anotheris not supported by the crime literature. Braga, in asystematic review of ‘hot-spots policing studies’reviewed nine evaluations—eight in the US and onein Australia—five had randomised experimentaldesigns and four had quasi-experimental designs.1016

He found that displacement was quite limited butthat there were often unintended crime preventionbenefits. Ratcliffe, in an evaluation of OperationAnchorage, a burglary reduction initiative inCanberra, also found that activities of the operationdid not significantly displace burglary to otherareas;1017 and Green has shown that targetedpolicing of drug dealing does not always lead todisplacement of the activity. One outcome,

�)��

moreover, is a wider diffusion of benefits tosurrounding neighbourhoods such that crime isreduced in neighbouring suburbs. 1018

&.'0'6��+��+�� )�� )��)-��-��


The benefits of demand-side drug law enforcementin encouraging illicit drug users into treatment havealso been noted. In one Sydney study of over 500heroin users, it was found that more than 60% ofthose in methadone treatment rated avoiding moretrouble with police and/or courts as an important,or very important, reason for entering treatment.Those not in methadone treatment were more likelyto say they wanted to enter treatment if they hadbeen imprisoned for a drug-related offence.833 Theauthors concluded that drug law enforcement had arole to play in heroin demand reduction, althoughethnicity and previous drug and criminal historywere also influential. The readiness ofamphetamine-dependent users to enter treatmenthas not similarly been assessed but is clearlyrelevant.

�� 5�� !��

Definition: The control of illicit drug use throughstrategies aimed at suppliers of illicit drugs.

The major sources of supply of the most prevalentillicit drugs used in Australia can be seen inTable 12.1.

&.'('&��)��-�� -�-��

The Australian Illicit Drug Report (AIDR) has beenthe major source of descriptive material. The threemost recent AIDRs are cited unless there are cogentreasons for referring to earlier editions.165, 168, 200

Other sources of descriptive data include reportsfrom Government and non-government agencies—predominantly Australian but also international(including WHO and UN documents)—whereapplicable.

Evaluative material is far harder to identify(Wardlaw, personal communication). Publishedjournal articles and books—both Australian andinternational—have been used, along with technicaland other reports, to obtain the most completecoverage possible.

An assessment of the strength of the evidence willbe made at the end of the section.

&.'('.��)-��+�� -� ��-��)��-)��+,��++��)�-

Australia absolutely prohibits the supply and use ofATS, heroin, phenethylamines, cocaine, LSD,mushrooms and GHB. In most Australianjurisdictions, the use of cannabis is now dealt withby civil penalty; however, supply of commercialquantities of cannabis remains a serious offence.

��'�� "��

� �� $�� 0�2

�� *�� &��

�8'C �� &�� /I�� J0

+ �� !�� )�� '�� )��

�� !�� '��

*'� , � �� '

+�� /I��J0��

,��9�� $�� '

,+-.,-* )��

L �� 9� �� '�� & ��

C �� $��

��

�)� �� !��"�#��$��%��

In this section, the range of programs designed toreduce the supply of illicit substances to theAustralian community (whether by importation ordomestic cultivation and/or manufacture) isdescribed, and literature attesting to theeffectiveness or otherwise of such programs isreviewed.

The National Illicit Drug Strategy (NIDS)(commonly known as Tough on Drugs) waslaunched in November 1997. Under this strategy,the Commonwealth has allocated $213 million overfour years to supply control measures.2 Thesemeasures aim to better resource law enforcementagencies to protect Australia’s borders, and thecommunity, from illicit drugs.1019

&.'('0�� ++��)�-

Australia is a signatory to three internationalconventions that prohibit the trading of a range ofillicit substances, including cannabis, ATS, heroin,phenethylamines and cocaine. The aim of theseconventions is to restrict the use of drugs to medicaland scientific purposes. The conventions requiresignatories to exercise certain controls over theimport, export, manufacture and use of thesubstances listed in the schedules or tables to theconventions.1020

The Australian Customs Service (Customs) and theAustralian Federal Police (AFP) are the key agenciesresponsible for implementing strategies to reducethe supply of illicit drugs entering Australia.1021

Border protection – Australian Federal Police

Definition: The AFP has developed programs andmethods to combat the illicit drug trade offshore;maintaining that operating offshore, while beinglogistically complex and expensive, must becomemore commonplace as the influences ofglobalisation permeate the region.1022

The AFP has an international network of 33strategically placed officers who provide intelligencelinks to most of the world’s law enforcementagencies. These links allow the conduct of nationaland international investigations associated with drugimportations.

The Law Enforcement Cooperation Program (LECP)is based on the international liaison officer network.It was established using $5.7 million of NIDSfunding allocated over a four year period, from July1, 1998. The LECP initially involved countries in theAsia Pacific region but is now expanding globally.The ultimate objective of this program is to

strengthen the capability of foreign lawenforcement to interdict drug traffickers byproviding assistance in the form of training and, insome cases, equipment. These aims are beingachieved by funding short-term attachments toAustralia and exchanges between Australia and othercountries for operational law enforcementofficers.1022

NIDS funding has also been used for the formationof 10 intelligence-driven, proactive mobile striketeams of AFP investigators, analysts and supportstaff. These are regarded as the cornerstone of theAFP role in NIDS and are designed to provide long-term targeting of major crime figures to identify,disrupt and/or dismantle syndicates at theirinternational source.

Border protection: Australian Customs Service

Definition: Customs reduce the supply of illicit drugsthrough intercepting illicit drugs at the border anddeterring people from importing or trafficking inillicit drugs.1023

Under NIDS, Customs has been allocated $62million over the period 1997/98 to 2001/02 forincreased surveillance in the Torres Strait and forcargo profiling and examination, cargo examinationfacilities, communication and IT capabilities,additional intelligence analysts, and increased searchcapacity.1021

&.'('(�� )�)��+)��

Summary: Border protection AFP; warrants furtherresearch ................................................................ �Border protection ACS; warrants further research . �

The AIDR reports on border detections for theperiod 1998/1999 to 2001/02.165, 168, 200 Theiroverall assessment in 2001 was that despite recorddetections of illicit drugs coming into Australiaduring 1999/2000, availability and use continuedto increase; although some of these drugs, such ascannabis and ATS, are primarily produceddomestically. Between 1998/99 and 1999/2000,heroin seizures increased by one-third and cocaineseizures by 145%. Detections of methamphetamineat the border more than doubled and there wasincrease in the weight of border phenethylaminesdetections. The number of seized cannabisimportations increased although there was adecrease in weight.200

From 1999/2000 to 2000/2001,methamphetamine and phenethylamines seizuresincreased markedly, but the weight of heroin and

�)��

cocaine seizures decreased.168 From 2000/01 to2001/02, the weight of Customs border seizuresfor methamphetamine, phenethylamines, heroinand cocaine increased, markedly so for cocaine.165

The latest AIDR identified a steady increase inborder seizures in both ATS and phenethylaminesover the last five years, whereas heroin and cocaineborder seizures have fluctuated.165

The AIDR notes that traditional South East Asianheroin producers continue to diversify intomethamphetamine manufacturing. There have beenlarge seizures of methamphetamine tablets in theregion, with concomitant concern that thesemanufacturers will target Australia. However, it isclear that in Australia most ATS are beingmanufactured domestically. There has been anincrease in phenethylamines detections, which canbe attributed in part to improved Customs and AFPtargeting of higher level phenethylamines importersas well as better links with overseas lawenforcement agencies.200

The AFP maintains that its mobile strike teams havehad ‘an immediate and quantifiable impact’ on drugtraffickers by reducing supply and disrupting theactivities of syndicates. To support this, the AFPpoints to a total of 1680 kg of heroin, 1520 kg ofcocaine, 736 kg of phenethylamines and 458 kg ofother ATS seized domestically and overseas,between November 1997 and April 2001. The AFPconcludes that, ‘the frequency of large seizures hasincreased markedly since the introduction of NIDS’(p8).1022 Customs also demonstrates that thenumber of detections and quantity of drugs seizedhave increased considerably since NIDS waslaunched, in November 1997.1021

These outputs have been evaluated both internallyand externally. The AFP recently published in its in-house magazine Platypus an economic evaluation of apreliminary performance evaluation model.1024 Themodel evaluated illicit drug investigations in termsof return on investment, assuming that ‘the mainbenefit to be derived from successful druginvestigations is a reduction of supply of illicit drugsto the community and associated reduction in thecost that society bears as a result of drug abuse’(p18). In the absence of better estimates, the streetprice of drugs seized was used as a surrogate of theeconomic value of harm associated with drug use,although it is noted that this may be a conservativeindicator of the total cost of illicit drug use.Factoring in the investment in drug seizure activitiesof both AFP and Customs, the return on investmentwas judged to be 5.2: 1, that is over $5 wasreturned to the community for every dollar invested

by the Government. It should be noted that theassumptions and findings of this study have not, atthe time of writing, been published in a peer-reviewed journal.

The Australian National Audit Office (ANAO) hasundertaken a performance audit of Customs drugdetection strategies for air and containerised seacargo, and small craft activities. The following areaswere examined: intelligence and law enforcementcooperation; air and containerised sea cargo; cargoexaminations and technology; small craft activities;Customs funding arrangements (including fundingfor NIDS initiatives); and governance, includingperformance reporting.1021 It should be noted thatthe audit did not include Customs work in relationto drugs in the passenger and postal streams. Aspokesperson for Customs has informed the authorsthat subsequent to the drafting of the audit,Customs has increased and enhanced itsinterventions.

ANAO’s overall conclusion was that theadministrative effectiveness of Customs’ drugdetection strategies was sound, but that drugdetection strategies would be improved whenoperational risk management was fullyimplemented and performance measures to indicatethe effectiveness and impact of drug detectioninitiatives were developed. ANAO found that whileit was not possible to accurately assess the quantityof illicit drugs entering Australia, or the drugmarket overall, traditional supply-side measures ofseizures and quantities were flawed indicators ofeffectiveness reflecting levels of law enforcementactivity and could not be used as indicators of theeffectiveness of agencies in reducing the supply ofillegal drugs. ANAO recommended that multipleindicators of both supply and demand be used forevaluation of effectiveness.

While there seems little doubt that increaseddetections, both in Australia and throughinternational cooperation in source countries, relateto initiatives made available though NIDS,1021,1022 itis unclear whether these increases occurred as aresult of: increased total importations, increased lawenforcement, both, and/or other influences. Therecent heroin ‘drought’ is a good case in point.There was considerable debate about the causes ofthe drought. Politicians claimed that the additionalresources made available to law enforcementagencies, and the substantial drug seizures that hadoccurred, presumably as a result of that extrainvestment, had ‘led to a heroin drought in capitalcities and a substantial reduction in heroin overdosedeaths’ (Ellison, p11).1025 At least one commentator,

��

�)� �� !��"�#��$��%��

however, argued that it was caused by a series ofpoor opium harvests in Myanmar and the marketingdecisions of crime syndicates to promote ATS overopiates in the Australian market, rather thanimproved and increased surveillance and lawactivity.1025 Bush relied on comments made aboutthe heroin drought by senior law enforcementofficers, as reported in the media. The AIDR claimsthat the heroin drought was related, but not whollyattributable, to successful law enforcement both onand off shore.168 Weatherburn et al. hypothesise anumber of possible causes of the drought: increasedseizures; arrests of major heroin importers anddistributors; and a (water) drought in the opiumgrowing areas of Myanmar.1026

What this debate does point to, as does the ANAO,is the lack of evidence on which to base claims forthe relative effectiveness of one strategy overanother. It should be noted that research has nowbeen commissioned to examine the dynamics of theheroin market in Australia, the relationshipsbetween the supply, demand and harm reductioncomponents of Australia’s National Drug StrategicFramework, and the effects and implications offundamental changes in Australian illicit drugmarkets.1027

Other research projects on the cocaine andphenethylamines markets have also recently beenfunded.1028

&.'('6�� +�#�� )��8�#��9

Definition: The NHSP is a joint project between theAFP and the Australian Government AnalyticalLaboratories. It is based on the notion that drugs canbe identified by their characteristics as beingmanufactured in a particular part of the world, oreven by a particular group or individual.


The purpose of drug profiling is to not onlyestablish the provenance of drug seizures at the levelof region, subregion, and manufacturing batch, butalso at the distributor level where the illicit drugmay be repackaged and concealed. Hence, theprogram focuses not only on the chemical profile ofthe seized drug but also on an holistic capture of allthe forensic evidence, including physicalcharacteristics of the seizure. This information isthen applied at an operational level, often tocompare two or more seizures, and at anintelligence level for trends analysis andinvestigative leads (Robertson, personalcommunication). In excess of 280 ‘signatures’ have

been analysed, resulting in findings that include theexistence of a number of subtypes of heroin withinthe one geographical location, and the presence ofSouth American heroin in Australia (p16).1022

An additional $4.7 million over four years wasallocated in the Commonwealth 2002–2003 budgetto extend the drug profiling program to cocaine andATS.

With respect to evaluation, this is essentially aresearch program but there has been no publicationof the data in the peer-reviewed literature so thatthe academic community is not able to assess theefficacy of the program. However, the Director ofForensic Services at the AFP states that the data havebeen applied at the operational level on a significantnumber of occasions and evidence is being given onthe data held in the program on a regular basis—largely in-house (AFP). A number of informationreleases and posters have been issued and widelydisseminated. A street level survey has beencompleted and reported to the relevant forensiccommunity and workshops held to shareinformation. Whilst there may be a need for furtherdissemination of information, it is considered thatthe program is on a par with the best in the world.(Robertson, personal communication)

&.'('7��-�� -)��+,��++��)�-�� )-��+��

Domestic illicit drug law enforcement is mainlyundertaken by State and Territory police services.Information about their activities has been primarilygained from the AIDR.165, 168, 200


Cannabis

Cannabis is cultivated on a large scale Australia-wide. A range of groups and individuals areinvolved, including organised groups said to beparticularly outlaw motorcycle gangs (OMCG).During 2001/02, 9801 kg of cannabis were seized,which was an increase on the two previous years. Ineach jurisdiction there were many seizures with noweight recorded, which makes it difficult toaccurately compare the weights and numbers ofseizures at the national level. Cannabis is readilyavailable and the price has remained stable, ordecreased, across Australia.

Among the initiatives used to detect cannabis grownin Australia is aerial surveillance, which some policeservices report has led to a decrease in the numberand extent of outdoor crops being cultivated.However, hydroponic cultivation of cannabis

�)"��

continues to increase, with reports thathydroponically grown cannabis has become themost sought-after cannabis among user groups. Thishas also given rise to observable theft of electricityby cultivators of indoor hydroponic cannabis, inseveral States and Territories.

Cannabis law reform

Definition: Policies or legislation designed to reducepenalties for cannabis possession and use, therebyalso reducing backlogs in the justice system. Theseinclude cautioning and/or diversion programs,which are described and discussed in a later section.The other major approach to regulating theavailability and supply of cannabis is throughchanges to its legal status.


Cannabis law reform has been a recent commonapproach to reducing cannabis-related legal harm inAustralia. At the time of writing, the NorthernTerritory, ACT, and South Australia have all alteredcannabis laws in some form or another; WesternAustralia is in the process of alterations. The reasonsfor proposing cannabis law reform are grounded inrecognition that while cannabis use can bephysically and/or mentally harmful, a proportion ofthe harms associated with its use are primarilyrelated to its illegal status. Aside from the harms tothe user, described earlier, enforcement of cannabislaws requires a significant commitment of time andfinancial resources from the police, the judiciary,and the prison system. The South Australian civilpenalty system, for example, demonstrated savingsof $1.4m p.a. over a criminal penalties model.1029

The widespread use of cannabis, and the existenceof a substantial illicit market that aims to filldemand, also brings police into close contact withplayers in the illicit drug trade. Recent investigationsinto police corruption in Australia have uncoveredexamples of cannabis-related police corruption thatinvolves large amounts of cannabis and money.215

There is a compelling argument that despite thehigh financial and social costs incurred by alegislative model where cannabis is illegal, theillegality of cannabis does very little to reduce itsuse. Thus, the overall harms associated with acriminal model are reasoned to be higher than theoverall harms associated with a decriminalisationapproach to cannabis legislation.

The ‘traditional’ model of cannabis legislation, inboth Australia and overseas, has been prohibitionwith criminal sanctions against the use of cannabis.Various differing methods of legislation on cannabisuse have been trialled in different locations across

the world. The one most favoured in Australia hasbeen termed ‘prohibition with civil penalties’whereby use and possession is illegal but there areno criminal consequences if fines or other civilpenalties are paid prior to a due date; large scalecultivation and supply of cannabis remain subject tocriminal sanctions.215

A comprehensive review of the different models oflegislation introduced to regulate cannabis use, bothin Australia and overseas, has demonstrated thatthere is little evidence that changing the legislativeenvironment has had any significant effect on levelsof cannabis use.215 This was also the conclusion in across-national analysis of cannabis use and attitudes,in the Netherlands, the United States and Australia;which found similar patterns and rates of use,suggesting that there was no link between rates ofuse and social policy.1030 A comprehensiveevaluation of the effect of introducing cannabis lawreform in South Australia (SA) in 1987 (the longestrunning example), found no evidence that changingthe law led to an increase in ‘population prevalenceof recent cannabis use’ compared to AustralianStates that had not changed their laws.1004 On theother hand, research on the SA scheme has foundevidence of significant ‘net-widening’, whichoccurs as more people are caught up in the legalsystem than might otherwise be the case, becausenotices are easy to issue.215 Another problem withthe South Australian Cannabis Expiation NoticeScheme was that there was evidence from lawenforcement sources of syndicates of growers inSouth Australia involved in the export of cannabis toother Australian States.200 However, these problemsappear more to do with the way the SA scheme wasimplemented rather than a problem with civilpenalty schemes as such.

In Victoria, a Cannabis Cautioning Program Pilot(CCPP) was conducted by the police in one district,for the last six months of 1997. Under this scheme,police were able to issue a caution to adults detectedin possession of or using a small quantity ofcannabis.1031 The evaluation of the pilotrecommended it be continued and extended toother police districts throughout the State (VictoriaPolice, 1998). This took effect in September 1998.An early conference paper noted that 5% of thosecautioned had again come to the attention of policewithin a one month period.1032 More recently, theVictorian cannabis cautioning scheme has beenevaluated as part of the Evaluation of the COAGinitiatives on Illicit Drugs. While this report was notavailable at the time of writing, it is believed thatconclusions based on client impacts are limited byvery small sample sizes.

��

�)# �� !��"�#��$��%��

A common argument against decriminalisingcannabis is that it will ‘send the wrong message’and will lead to increases in use. However, thisargument is not well supported by the evidence.215

Current evaluations of law reform, such as thatplanned for WA, will need to consider the issuesnoted above. So, too, will they need to ascertainwhether early initiation into cannabis use, whichappears to be a significant risk factor in laterengagement in antisocial and criminalbehaviour,1033 is encouraged or enhanced by theselegal changes.

ATS

During 2001/02, 240 clandestine laboratories werefound in Australia, which represents a steadyincrease from 1997/98. The majority were foundto be producing methylamphetamine with the basicingredient, pseudoephedrine, predominantlyextracted from cough and cold medications. Someconcern is expressed in the trend towards smalllaboratories, which will need greater resources fordetection.

Domestic seizures of ATS have increased, with 4861seizures during 1999/2000, which is an increase of657 on the previous year. The use of ATS continuesto increase and the AIDR maintains that, ‘despite thebest efforts of Australian law enforcement agencieslarge quantities of amphetamine are being producedin this country’ (p55).200

One approach to restricting the availability of ATS isthe control of precursor chemicals. Precursors arestarting compounds or ingredients which, whencombined with other essential chemicals andreagents, produce illicit drugs. The synthetic drugsproduced are predominantly ATS but there are alsosome opiate-like substances (such as ‘homebake’heroin).1034 All State and Territory police serviceshave now established chemical diversion desks tomonitor suspicious purchases of precursorchemicals. These chemical diversion desks also liaiseclosely with the chemical and pharmaceuticalindustries in their jurisdictions, in relation topharmaceutical theft and provision of adequatesecurity.200 A National Code of Conduct, adopted byall States and Territories, places voluntaryrestrictions on the sale of chemicals used formanufacturing methamphetamines and otherpsychostimulants and has been in place since 1994.However, only three states (NSW, Queensland andVictoria) have legislation that makes it an offence tobe in possession of a precursor with the intention tomanufacture a prohibited drug.200

The United Nations Drug Control Programme(UNDCP) notes that since the adoption of the 1988UN Convention against Illicit Traffic in NarcoticDrugs and Psychotropic Substances, the licit trade ofprecursor chemicals has been closely monitored butthat the illicit industry’s response has been to switchto alternative chemicals that do not fall under theregulations.1035 In Australia, methamphetaminemanufacturers are switching from pseudoephedrine,which has been the commonest precursor, toalternative pharmaceuticals obtained through legalpurchase and theft.168 The fact thatpseudoephedrine-based tablets are the mostcommon starting point for clandestine druglaboratories producing methamphetamine hasmeant that the manufacturers, distributors andpharmacy retailers of these products are now beingtargeted in every Australian jurisdiction.1034

Seizures of imported ephedra, ephedrine andpseudoephedrine have been made and, in May2000, a regime of criminal sanctions for the illegalimportation of precursor chemicals wasimplemented.1034 There are no current indicationsthat these processes have, as yet, reduced thedomestic manufacture of ATS.

Heroin

As has been noted previously, the situation withregards to heroin changed markedly between 1999/2000 and 2000/01. In the AIDR report on theearlier period, it was maintained that ‘despite thehigher number of heroin seizures of the Customsborder and in the various jurisdictions between1999 and 2000, there is no evidence that heroinavailability has diminished’ (p36).200 In the laterreport, they found that the heroin drought hadtaken hold in almost every State and Territory, withdiminished availability and purity and increasedprice. Concomitantly, heroin offences fell by 34.1%across the country.168 DUMA data for 2001,however, found that the declines in heroin use weresmaller outside of Sydney and the impact was ofshorter duration.202 In 2001/02, heroin availabilityand purity remained low, although some reportsindicated that the heroin drought was easing, in thefirst half of 2002 for Sydney, Canberra andMelbourne. Heroin offences again decreased fromthe previous year by 56.2%.165

Phenethylamines

Five phenethylamines producing laboratories weredetected in Australia during 1999/2000, two in2000/01 and none in 2001/02. The AIDRmaintains that there were few detected because ofthe difficulties in both obtaining the necessary

�)%��

ingredients and the production process.200 There isconcern that locally produced ‘ecstasy’ is morelikely to contain a variety of other substances thanphenethylamines, including ATS, heroin, ketamineand caffeine.

Cocaine

All jurisdictions other than the Northern Territoryrecorded seizures of cocaine between 1999 and2002 but the focal point for cocaine remains NSW.Arrests for use and supply in 1999/2000 wereabout 30% lower than in the previous year, butincreased by 50% in 2000/01. Cocaine offences fellby 6.5% from 2000/01 to 2001/02. Almost alloffences were in NSW, a finding supported byDUMA data.202 The AIDR considers that there is alarge undetected market for cocaine because, whilstsimilar amounts of cocaine and heroin are seizeddomestically, unlike heroin most of the cocaineseizures are made by AFP, suggesting that streettraders and dealers are not being apprehended.Cocaine arrests also represent a much smallerpercentage of total drug-related arrests than heroinarrests.168

Other drugs

There is little specific information available aboutthe number of arrests for consumer and provideroffences relating to LSD, mushrooms, GHB/GBLand ketamine. There are relatively few domesticdetections of hallucinogens, but there areindications that use of hallucinogens is generallyincreasing. LSD distribution has a relatively lowprofile in Australia, with most distributionoccurring through the postal system. Some policeservices report that OMCG are involved indistributing LSD in nightclubs and private parties.The availability and use of LSD are stable and low inmost jurisdictions, with the drug often takenconcurrently with other drugs. Mushrooms are notdistributed in Australia on a large scale and are notgenerally seen as a drug to be sold for profit.

Although there are no data for GHB-relatedoffences, border seizures of GHB have increasedsignificantly, from 3 in 2000/01 to 18 in 2001/02.Use is low and mostly confined to a subgroup ofnight club patrons. Ketamine hydrochloride is aveterinary anaesthetic and its use and availability isgenerally low. Ketamine powder from China wasseized at the border during 2001.168

&.'('3��--�� -��

Definition: All Australian jurisdictions have enactedlegislation dealing with the confiscation of theproceeds of crime. Confiscation legislation inAustralia is the subject of review, with some debatearound the question of conviction or non-conviction based confiscation.200


There is a debate around the issue of whetherlegislation requires that criminal activity be provedbeyond a reasonable doubt before assets areforfeited (conviction based) or whether proof to thecivil standard of the balance of probabilities that theperson has committed a serious offence (non-conviction) is sufficient. The WA Director of PublicProsecution, for example, maintains that with non-conviction based legislation, delays in courthearings do not hamper the effective confiscation ofcriminal property.1036

In relation to the Commonwealth Proceeds of CrimeAct 1987 (POC Act), the Australian Law ReformCommission (ALRC) concluded that a solelyconviction based regime failed to meet either theobjectives of the POC Act or public policyexpectations. As a consequence, it recommendedaugmenting the POC Act with a civil forfeitureregime enabling confiscation, upon proof to thecivil standard, of profits derived from engagementin prescribed unlawful conduct.1037

New South Wales has a non-conviction based civilforfeiture regime, as does Victoria, although thelatter applies only to offences of drug cultivation ortrafficking when the quantity of the drug involvedamounts to a commercial quantity.200 WesternAustralia has enacted a non-conviction Act underthe terms of which the confiscated assets will bespecifically directed to increasing the effort toameliorate the level and effect of crime in thecommunity.1036

Other initiatives in the financial investigation ofillicit drug offences include those implemented byThe Australian Transaction and Reports AnalysisCentre (AUSTRAC) to increase the capacity of lawenforcement agencies to investigate illicit drugssyndicates using financial intelligence. Some ofthese are funded under NIDS.200

It is difficult to determine the effectiveness oflegislation dealing with the confiscation of theproceeds of crime because there are no nationalreporting standards.200 In some jurisdictions it isdifficult to differentiate the confiscation informationcorresponding to crime types. It is argued that

��

�)' �� !��"�#��$��%��

uniform legislation would eliminate problemsassociated with different reporting standards, but ahighly committed coordinated national politicalresponse would be necessary if uniform legislationwere to be implemented. In terms of confiscatingthe proceeds of crime, an assets removal programwould work only if it was accompanied by greaterlaw enforcement capacity to follow complicatedmoney trails.200

In terms of the conviction or non-conviction basedconfiscation debate, as noted, Western Australia hasenacted a non-conviction Act. The WA Director ofPublic Prosecutions (DPP) reports that, although itis too early to know whether new legal provisionswill have a significant impact on the level of crimein the community, it is ‘already apparent that thenew law greatly facilitates the confiscation of crimeuse and crime derived property’ (p8) and it isforecast that the amount of crime-related propertyconfiscated will increase ‘dramatically’.1036

&.'(';�� +�� )-��+�� +)�� -��++��)��+��

There is strong international evidence, partly drawnfrom what is known in relation to other blackmarkets for commodities such as coffee andtobacco, that rendering drugs illegal greatlyincreases their price.176 Nevertheless, the ‘effectiveprice’ of a drug is much more than its retail price—it is a measure that incorporates dollar cost, the timeand effort required to purchase the drugs and therisk of being arrested or ‘ripped off’. This effectiveprice will vary from user to user and, arguably, willbe higher for novices than for experienced users.1038

Whilst the basic goals of supply reduction and druglaw enforcement are to minimise the supply ofdrugs to illicit markets, and increase the price andinconvenience of acquiring drugs, it is important torecognise that the effects of supply reduction arenot equally distributed among different drugs. Inthe US at the end of the 1980s, for example, acomparison of drug prices to consumption showedthat heroin had been successfully controlled in the1970s; cannabis had recently come under control;but the cocaine problem in the 1980s had not beencontrolled.1039

Critically, Moore demonstrated that the drug marketchanges over time, is geographically specific, andalso that law enforcement impacts on differentdrugs at different points in their production anddistribution cycle. For heroin (in the US), the majordifficulties seemed to lie in processing, exportingand distributing the drugs rather than getting them

across the border. For cocaine, the biggest difficultylay in distribution; whilst with cannabis, thegreatest difficulty lay in collecting and processingthe plants and importing them.

One of the reasons why it is difficult for interdictionto make further gains is the adaptability ofsmugglers. Interdiction efforts are relatively moreimportant in affecting the price of cannabis thanother drugs because interdiction efforts are moresuccessful against bulk shipments in non-commercial vessels. Heroin and cocaine—being lessbulky—can be more easily smuggled by air andcommercial shipping.1038 Smugglers, however,responded to this by carrying much smaller loadswhich meant more effort was required to interdictsimilar overall quantities.1040

In the early 21st century, the US epidemics ofcocaine and heroin addiction have essentially rundown. According to Reuter, there have been fewnew heroin addicts since the 1970s and few newcocaine addicts since the mid-1980s. Neithermethamphetamine nor cannabis abuse raises asmuch concern or has as many consequences asheroin or cocaine use. Heroin and cocaine priceshave continued to decline but the perceivedavailability (by high school students) of cocaine andheroin has not decreased over 10 years. Because ofadaptability in the market, there are no signs thatmore intense enforcement would yield furtherimprovements.1041 Once again, the differencesbetween these observations and patterns of drug usein Australia are obvious. Indeed, Weatherburnpoints out that all drug market modelling to date isbased on the US market for illicit drugs and on USlevels of investment in drug law enforcement. Hemaintains that it is unclear whether, and to whatextent, the results of US simulation studies on theeffects of supply-side drug law enforcement arerelevant in Australia given different patterns of illicitdrug use—more heroin, more amphetamines andless cocaine.397

In Australia, a number of early studies suggestedthat illicit drug law enforcement was not cost-effective and did not produce significant impacts ondrug markets.1042, 1043 A comprehensive two yearnational review of drug law enforcementthroughout Australia concluded that the impacts ofcriminal justice fell mainly on low-end distributorsand users rather than high-level operators, althoughtargeting the latter was the stated objective.1044

Other than current research funded by the NationalDrug Law Enforcement Research Fund,1028 thereappears to have been no major approach, since1996, to independently assess the effect of domestic

�)(��

policing on the supply of drugs (Wardlaw, Suttonand Makkai, personal communication). There is aclear need for evaluators to be able to assess theimpact of law enforcement activities on illicit drugsupply but as the ANAO notes, whilst traditionalmeasures such as price, purity and availability ofdrugs are valuable indicators in some respects, theyare neither comprehensive nor responsive tochanges in drug law enforcement activities.Preliminary research, both in Australia and overseas,suggests examples of indicators that could be usedto evaluate the output and outcomes of drug lawenforcement activities, but additional performanceindicators are required to evaluate the full range oflaw enforcement impacts on illicit drug markets.1034

Weatherburn has proposed a set of possibleperformance indicators for heroin law enforcementwhich provide a means of assessing policeperformance in minimising the harm associatedwith heroin and gauging what police are doing toachieve this objective.1045 In April 2002, the NSWPolice published performance indicators forassessing the effectiveness of drug law enforcement,which were developed as an outcome of the NSWDrug Summit in 1999.1046 A number of strategiesagainst which performance will be measured areoutlined, including monitoring and assessment. It isnot known whether other police services havedeveloped similar indicators but it is clearly animportant precursor of overall performanceevaluation.

��"�,��$��

The ANAO noted that the Customs and other lawenforcement agencies were ‘working in anenvironment where the size of the market isunknown’ and that:

… there is no national consensus or estimation of the amountof illicit drugs entering Australia or the national drug marketoverall. Because it is extremely difficult to quantify illicit drugactivity, policies and practices are based upon availableinformation and market intelligence. Seizure information usedby many organisations to estimate trends and supply is limitedto the activities undertaken by law enforcement agencies anddoes not provide sufficient data for a comprehensive analysis ofthe market. Information relating to drug users, changes in thepatterns of drug use and the demand placed on health caresystems must also be considered (p12).1021

The ANAO also recommended that multipleindicators of both supply and demand be used forevaluation of effectiveness.

The readily accessible quantitative supply side data needs to beanalysed in conjunction with quantitative and qualitativedemand side information. An in-depth statistical analysis needsto be undertaken to ensure that the information relates to thework that has been undertaken. With quantitative performancemeasures it would then be possible to compare achievementsacross periods of times and benchmark the activities involved(p117).1021

Additional performance indicators are required inorder to evaluate the full range of law enforcementimpacts, particularly the extent of disruption anddismantling of criminal enterprises.1034

Weatherburn notes that there is a lack of adequatedrug law enforcement performance indicators,which means that ‘we cannot judge the value of thepublic investment in DLE’ (p1) claiming that, unlikethe UK, little has been done to improve themeasurement of drug law enforcement inAustralia.1045

Others have also argued that contemporaryAustralian data sets are inadequate for evaluationand that what is needed is a re-examination ofcurrent sources of data to ensure that they areconsistent and of high quality, and to encouragepolicy-relevant research that utilises such data.1047

Data collected for administrative purposes, such asarrest or prison data, are inadequate for researchpurposes; although uniform crime statistics arecollected by the ABS, these are problematic in thatthey come from police records and can be eithervictim or incident-based and are dependent on whatis reported to, and recorded by, police.1047 Theseproblems make international comparisons difficult.Makkai argues that what is needed is an integratedmonitoring system based on national surveys, aswell as specialised collections such as DUMA andIDRS, which gather data about drug use andavailability from specific populations includingarrestees and injectors.1047

There are also difficulties with the reliability andvalidity of national surveys, such as the NDSHS, forestimating prevalence and trends of illicit drug use.The US Committee on Data and Research for Policyon Illegal Drugs reported that the usefulness ofpopulation-based surveys, such as householdsurveys, is limited if people refuse to take part, orgive inaccurate responses. These response problemsare particularly severe in the case of illegal activities,including illicit drug use. The Committeedetermined that response problems in the USHousehold Surveys cast doubt upon inferencesrelated to both levels of drug use and trends. Theyfound that non-response rates could be as high as25%, and that assumptions about the levels of use in

��

��) �� !��"�#��$��%��

the non-responding population could not always besustained. The problem with inaccurate respondingis of a different order because it is not known howlarge it is. The Committee recommended that a‘systematic and vigorous’ research program be setup to understand and monitor non-responding andinaccurate responses in national drug use surveys,and develop methods to reduce non-responding andresponse errors.176 In Australia, Makkai andMcAllister found that a sealed-booklet format forasking questions about drug use in householdsurveys produced a more accurate estimate of druguse than direct questions.1048

Large-scale data sets with detailed data on price andpurity, based on objective data, are not currentlyavailable in Australia or the US. The US Committeewas tasked, among other things, with assessingexisting data sources that support policy analysis,and identifying new data and research that mightenable the development of more effective means ofevaluating the consequences of alternative drugcontrol policies. They found that the nation’s abilityto evaluate its enforcement activities was severelyhampered by two major data deficiencies: theabsence of adequate data on drug consumption, andreliable data on drug prices.

The central problem is a woeful lack of investment in programsand data connection and empirical research that would enableevaluation of the nation’s investment in drug lawenforcement…. the nation is in no better position to evaluatethe effectiveness of enforcement than it was 20 years ago, whenthe recent intensification of enforcement began (p2–3).176

The Committee considered that current datacollections that involved self-report data of non-representative samples, or that were collected bylaw enforcement agencies, were not adequate. Itrecommended that an economic working group beestablished to develop, test and validate methodsand procedures. The results of these efforts shouldideally be widely reported in the professionalliterature and subject to careful review andanalysis.176

This advice in relation to US drug law enforcementis pertinent to Australia. Indeed, one view is thatAustralia is even further behind than the US in thedevelopment of adequate national indicator data formonitoring and evaluation purposes (Makkai,personal communication). This kind of researchcould be further facilitated by addressing thelegislative, ethical and other issues surroundingaccess by research agencies to the administrativedatabases held by various government agencies.1034,

1049

A shift towards more detailed evaluation ofAustralian drug law enforcement has been evidentin recent years but the difficulties in doing researchand evaluation in this area should not beunderestimated. On the one hand, an infrastructurefor research into drug-related law enforcement hasnot been developed in Australia in the same way ashas an infrastructure for prevention and treatmentresearch, and, on the other hand, the nature of lawenforcement is such that some operations and datahave to remain confidential. Recent initiativesinclude the commissioning of research that, as hasbeen noted above, has largely been undertaken byNDLERF.

Relevant research projects are currently underwayor out to tender.1028

� A study of the mechanics of cross-bordertrafficking of heroin.

� The Illicit Drug Reporting System (IDRS)expansion (Illicit Drug Users Survey), in 2000/01 and 2001/02.

� The IDRS expansion (amphetamine-typestimulants) in 2000/01 and 2001/02.

� Insertion of drug law enforcement questions inthe NDSHS.

� Heroin drought research.

� Cocaine use in NSW and Victoria.

� Benzodiazepine and pharmaceutical opiatemisuse and the relationship to crime.

� Development of methodologies to studyphenethylamines markets.

The results of these research activities shouldenhance the scope, depth and effectiveness of theresearch base that shapes Australian drug supplyreduction activities.

��#�* 5�� $��9

Much of the description in this section is based onthe Australian Illicit Drug Report (AIDR). The AIDRis an annual report compiled by the AustralianBureau of Criminal Intelligence (ABCI) (now theAustralian Crime Commission) and is the mostcomprehensive and authoritative description ofillicit drug use supply and law enforcement inAustralia. The report collates information fromState, Territory and Federal police services, otherlaw enforcement agencies, correctional services,forensic science laboratories, Directors of PublicProsecution, the Australian Customs Service, drug

��

and alcohol research institutions, and drug andalcohol treatment agencies. Much of theinformation is qualitative and obtained fromquestionnaires but also contains State, Territory andCommonwealth statistical collections. In 1996,Sutton and James described the forerunner, theAustralian Drug Intelligence Assessment (ADIA) tothe AIDR thus.

Written explicitly from a law enforcement perspective, theADIAs have apparently been well received by the enforcementcommunity … and we have been impressed by the increasingdepth and sophistication of the series over the last three years.Indeed, it seems that they are likely to become the standard‘public record’ of illicit drug supply and enforcement phenomenain Australia; already academic papers are beginning to citeADIA as one of their central sources of information …(p14).1044

Recent AIDRs, now in their eleventh edition, aremore comprehensive than they were at the time ofthat writing. This makes the most recent threeeditions appropriate sources of descriptive dataabout illicit drug supply and enforcement.165, 168, 200

The limitations to these data, however, should notbe overlooked. As Sutton and James point out, AIDRanalyses and conclusions are only as good as theconstituent data forwarded by the agencies and, inthe main, little attempt has been made to assess howgood these are.1044 Moreover, integration of suchdiverse sources of data inevitably results in somepatchiness. Notably, the AIDR relies heavily onreports from police services and there is little or noway to verify these. Makkai notes that police dataare based on organisation records and varyconsiderably across Australia. Offences reported tothe police are also a reflection of policing practices,and public actions and values, rather than absolutevalues.1047

The literature review by Weatherburn et al. has alsocontributed a great deal to this section.397 It does notclaim to be a systematic review and nomethodology for the selection of studies andcommentary articles for inclusion is given; there islittle detail about the specific studies reviewed.However, it is comprehensive, covering bothinternational and Australian literature, and theprimary author has a major Australian andinternational reputation as a criminologist and druglaw enforcement researcher.

Most of the descriptive and some of the evaluativematerial in this section has been drawn from publicdocuments, many accessed from web sites. Whereinternal evaluations are not subject to externalreview, it is not possible to make judgements abouttheir validity. It is noteworthy that many agenciesare now calling for evaluations to be published inpeer-reviewed journals, which can only add to theconfidence that can be placed in their findings.

��

�� !��"�#��$��%��

��%�� !!��

There is a significant link between alcohol and other drug use, and crime, particularly violenceand property crime. The cost to the community of such crime includes the cost of lawenforcement and legal and detention costs, which can be considerable. Outcomes such ascriminal records can have negative effects on individuals and their families, and recidivism is acontinuing problem.

Diversion and other judicial systems, such as drug courts, are designed to improve outcomes forboth the community and for offenders who commit drug-related crimes. Diversion involves agraduated series of interventions that are proportionate to the seriousness of the crime. It aims toprevent first offenders from entering the criminal justice system and to divert offenders withdrug problems into appropriate treatment.1050 Diversion programs are particularly slanted towardsearly stage drug-related offenders; late stage offenders may be offered an opportunity to havetheir offence heard in a drug court, which is a specific example of therapeutic jurisprudencedirected towards drug dependence and its outcomes.

The use of diversionary and judicial programs to reduce the demand for drugs is an extension ofstreet-level drug law enforcement. For those who have not been successfully diverted away fromdrug-related crime, there are a range of programs in prisons to restrict the supply of drugs intothe prison, reduce demand for drugs by prisoners, and minimise harmful drug use.

This chapter reviews international and Australian experiences with diversion, drug courts andprison programs. It is noted that many of these programs are relatively recent in Australia andalthough full evaluations are planned, most are not yet completed. Nevertheless, the evaluationsplanned, and in some cases implemented, for diversion programs and drug courts are exemplaryand should act as models for other law enforcement and judicial evaluations.

�� !�� $ �� 9�� !�� !��

�� )��$��

�� -�� 9�� 9��

�� ) �� '�� ) ��6��' �� 9�<'#� �� 9�<'#��

� �� )�� &�� G��"��9�� 9��

��

�� 7��G��<'#9��

��

�� ) �� (�� ))8�� /� �� 0 �� <'#9��<'#��

� �� " ��

�� %%%

��

�� !�� $ �� 9�� !�� !��

� �� <�� 9�� ' #��

�� "�� $� � ��

( �� , � �� 9��

�� <� �� 9��

8�� <��

�� 7 <� �� 9� ��#��

< �� !� �� 9��

>��7 *��

� �� $��

� #��

� ��

��

�� $�� $�� $��

Definition: Youths with a high number of risk factorsfor substance abuse are vulnerable to influences thatcan draw them into crime. Once engaged in crime,such youths have a higher likelihood of convictionand incarceration. Evaluation suggests thatincarceration is not effective at preventing eithercrime or drug abuse.


There has been some description of policeinvolvement with at-risk young people at acommunity level in an effort to reduce substanceabuse and/or involvement in a criminal gang.1051

These authors described a one week program oflocal police involvement aimed to expose at-riskyouths to more appropriate role models, broadentheir socioeconomic horizons, and provide themwith the experience of financial reward forcommunity-based work. The evaluation of theprogram suffers from a number of methodologicallimitations (most notably inadequate description ofthe sample, non-random selection of theparticipants, and an inadequate follow-up timeframe). The authors did, however, employ

triangulation in an effort to gather data from a rangeof different sources (e.g. direct observation, focusgroups of young people themselves, and surveys ofparents and staff). Although the feedback from allsources was generally positive, there was no actualpre- and post-testing of variables of interest. Someoutcomes from this short-duration interventionappear unusually positive, raising some doubts as tothe level of risk/complexity of the interventiongroup.

In a large, multi-site intervention aimed atmodifying the substance use behaviour ofinstitutionalised young people, Morehouse andTobler demonstrated at least short-term efficacy oftheir Residential Student Assistance Program (RSAP)across foster care, juvenile justice and psychiatricfacilities, and a broad age range of 13 to 19years.1052 These authors showed that RSAP waseffective in both reducing and preventing alcoholand other drug use across the settings studied. Theyalso established a dose-response relationship, withthose young people in the high dose treatmentgroup (i.e. receiving 12 to 30 hours of activeintervention) showing significantly betterreductions in alcohol and other drug use (asmeasured both by frequency and intensity of use).Notably, these effects were achieved without any ��

�� !��"�#��$��%��

involvement of the young people’s family;however, no long-term data are provided regardingthe sustainability of the gains, or the resourcesrequired to achieve this.

An investigation that did directly target familyfunctioning in a young offender population wasdescribed by Dembo et al.1053 The authors’ accountof the two year outcomes of a five year longitudinalstudy showed that their Family EmpowermentIntervention (FEI) was relatively more beneficial incases of serious versus non-serious offending, byreducing the frequency of emotional/psychologicalproblems and the amount of marijuana used.Benefits with respect to marijuana use were alsodescribed for non-serious offenders. Again, thelong-term sustainability of these gains is yet to bedetermined.

The US juvenile justice system has mandated entryto family intervention programs, such as FunctionalFamily Therapy and Behavioural Parent Education.Research has demonstrated that the longer-termoutcomes of these programs indicate reductions incrime and incarceration. Exploration of similarprograms would appear warranted in Australia.

��$�� !�� !!��

Definition: Diversion programs aim to coerce drugusers into treatment and to reduce their contact withthe justice system, as the latter may be harmful tolow-level offenders.397


This section relies heavily on a literature reviewprepared by the National Drug and AlcoholResearch Centre.1050 Although not a systematicreview, it is comprehensive and the major issuesand findings have been published in aninternational peer-reviewed journal.1054

In April 1999, the Council of AustralianGovernments (COAG) agreed to develop anationally consistent Diversion Initiative as an earlyintervention strategy to prevent a new generation ofdrug users emerging in Australia. The Initiativetargets illicit drug users early in their involvementwith the criminal justice system. Police, and insome cases courts, divert targeted offenders tocompulsory drug education or assessment; fromthere they are referred to a suitable drug educationor treatment program.1055

Diversion programs often target those whoseoffences are drug-related, as well as direct drug

offenders. As this encompasses a broad andheterogenous client group, a range of responses isrequired to take into account the nature of theoffender’s drug use and criminal behaviour.397

Diversion, as a strategy, is based on soundprinciples. A review of international literature onrecidivism found that criminal sanctions had littleeffect, whilst rehabilitation programs showed smallto moderate effect sizes indicating a small reductionin recidivism.1056 It was concluded that for criminalsanctions to reduce subsequent recidivism bothtreatment and rehabilitation components wereessential, and that programs delivered in thecommunity had better outcomes than thosedelivered in institutions. Diversion of Indigenouspeople is also consistent with Royal Commissionrecommendations that custodial sentences beavoided wherever possible.1050

Several issues arise from discussions of diversion,the foremost of which are the efficacy and ethics ofcoerced treatment. Coerced treatment describesthose situations where treatment is offered as analternative to incarceration or other legalsanctions.397 There are four types of rationale foroffering legally coerced treatment.397

� Reducing illicit drug use and drug-relatedcrime: there is only a small body of evidenceregarding legally coerced treatment and itseffects on consumption and crime. However,reviews have concluded that coerced treatmentoffers similar benefits to non-coerced treatment.Criminal justice clients typically remain intreatment longer than voluntary clients, and aslength of treatment has a beneficial impact ontreatment outcome, this may be a positivefactor.

� Reduced costs of drug-related crime and lawenforcement: treatment has been found to bemore cost-effective than incarceration.

� Lack of effectiveness of criminal sanctions:there is little evidence of effectiveness forcriminal sanctions in reducing recidivism, or ofacting as a general deterrent.

� Reducing the spread of BBV among prisoners:prisons are a high-risk environment for thetransmission of BBV.

Hall reported that there was no Australian evidenceon the efficacy of legally coerced treatment, but thatinternational literature indicated that legally coercedtreatment programs could be as effective inreducing both drug use and criminality as voluntarytreatment programs.1057 The WHO has concluded

��"��

that compulsory treatment is legally and ethicallyjustified only if the rights of the individual areprotected and if effective and humane treatment isprovided.1058 Coerced treatment should alwaysinvolve a choice between normal criminal and legaljustice procedures and treatment, and the offendershould have some choice in the type of treatmentthey receive.1054

The impact of non-custodial sentences on familieshas also to be considered. On the one hand,maintaining offenders in the local environment canbe counterproductive, either because it could bedeleterious for the family to have the offendertreated in the community or because families cancontribute to the offending behaviour; on the otherhand, positive family support from trainedprofessionals can facilitate treatment outcomes.1054

Net-widening occurs when a diversion initiativeincreases the number of people involved in thecriminal justice system, or when offenders receive amore serious sentence if the penalty for failing adiversion program is more serious that the penaltyfor the original offence. There is little empiricalresearch but anecdotal reports suggest that this canbe a problem for diversion programs unless specificsteps are taken to avoid it.1054

Evaluating the efficacy of treatment approaches canbe difficult,176 with selection bias and themotivation of offenders to succeed beingconfounding factors. Another concern is theappropriateness of outcome measures; manyresearch initiatives have considered either drug useor crime as the only relevant variable, overlookingthe importance of health, and psychosocialwellbeing.176 Suggested outcomes for Australiandiversion systems include: community safety andlaw enforcement, economics, health, ethics, andcommunity confidence.1050 General characteristics ofeffective diversion programs have been identified ina number of reviews.1050

&0'0'&�� ) ��+��)-��+�� :�� -�8 ��:9��++��"�)�"��-�� ,

It is anticipated the Australian Diversion Initiativewill result in:

� drug users being given early incentives toaddress their drug use problem, in many casesbefore incurring a criminal record,

� an increase in the number of illicit drug usersdiverted into drug education, assessment andtreatment, and

� a reduction in the number of people appearingbefore the courts for use or possession of smallquantities of illicit drugs.1055

The primary target group is individuals who havelittle or no past contact with the criminal justicesystem for drug offences and who are apprehendedfor use or possession of small quantities of any illicitdrug. Offenders diverted under the scheme haveaccess to appropriate drug education and/ortreatment. Family involvement in the program willbe encouraged, where appropriate. If possible,young offenders will have access to youth-specificservices and Indigenous offenders should be giventhe option of attending an appropriate Indigenousagency.1055 A report on diversion programs that arespecific to Indigenous people is in draft, but wasunavailable at the time of writing.

The Diversion Initiative is being implementedwithin each State and Territory over a four yearperiod (1999 to 2003) using national criteria foridentifying priority targets including: need;demand; presence of special groups, particularlyIndigenous Australians; feasibility; and appropriateexisting service infrastructure. Reference groups ineach jurisdiction will oversee and participate in theimplementation of the initiative. COAG providesfunding for expanded early intervention treatmentand rehabilitation placements and funding forassessment services is provided by both theCommonwealth and State/Territory Governments.States and Territories provide the law enforcementbasis for diversion. Commonwealth funding totalsapproximately $105 million over four years.1055

The range of programs varies between jurisdictionsand it is beyond the scope of this Monograph todescribe them, except to note that in all cases thereare various programs addressing different levels andtypes of offending. The suite of programs in eachjurisdiction has been developed in line withnational principles.

The provision for evaluation and monitoring isintegral to the Diversion Initiative. TheCommonwealth Department of Finance andAdministration (DOFA) has been assignedresponsibility for independent evaluation, withassistance from a Steering Committee that comprisesrepresentatives of the Australian National Councilon Drugs (ANCD), the IntergovernmentalCommittee on Drugs (IGCD) and keyCommonwealth departments. The evaluation isrequired to advise on the effectiveness of thepackage in contributing to arresting the growth inillicit drug use, preventing the uptake of illicit drugsby new users, and reducing the damage to ��

��# �� !��"�#��$��%��

individuals and the cost to the community of illicitdrug use.1059

An additional issue is detection of possible negativeconsequences of the initiative. These might includenet-widening, displacement of voluntary clientsfrom treatment services by diverted offenders,pressure to plead guilty when innocent,discriminatory application of programs amongsubgroups of the population, and effects ontreatment service staff coping with difficult, coercedclients.1059

Whilst comment on the effectiveness of thisinitiative must wait until the evaluation isconcluded, the literature suggests that it might wellbe effective because diversion is based on soundprinciples and coerced treatment can be both ethicaland effective.

��

Definition: Drug courts are special purpose courtswhose role is to administer cases of people guilty ofdrug-related criminal offences. The drug courtprocess incorporates an extensive treatment andrehabilitation program, undertaken with thesupervision and ongoing management of the court.

Summary: Evidence for implementation �

This section has been prepared on the basis ofreviews of a limited literature primarily related todrug courts in the US. Although there are Australiandrug courts in almost every jurisdiction, only oneset of evaluations (NSW) is complete.

Drug courts are one of many possible approaches todiverting drug users from the criminal justicesystem towards treatment, education andrehabilitation approaches.397 Drug courts havebecome more common around the world duringthe 1990s. There are various models of drug courtsin use around the world but they are likely to havethe following features in common:1060

� an integrated approach involving criminaljustice procedures, drug addiction treatment,and social welfare programs,

� ongoing involvement with the court,

� frequent substance abuse testing, with sanctionsfor failing the tests,

� frequent contacts with health and welfareservices,

� sanctions and rewards based on the offender’sbehaviour, and

� the offender has pleaded guilty and a custodialsentence is the likely alternative.1061

There are many possible aims of drug courts relatedto both criminal justice and the therapeutic/rehabilitation process.1060, 1061

These include:

� reducing levels of drug-related criminal activity,

� reducing imprisonment rates,

� reducing burdens on the judicial andcorrectional system,

� improving the health and psychosocialwellbeing of the participants, and

� reducing or eliminating drug use.

US drug courts were first established in the 1970s,with the first ‘modern’ drug court being establishedin 1989.1060 These courts arose in the context ofconcerns about rising crime and drug use problems,and evidence of continued re-offending despitetough sentencing regimes.1062 As with the Australianexperience, the actual form of the courts variesgreatly according to the jurisdiction involved.1062

US courts are typically confined to non-violentoffenders whose involvement in the criminal justicesystem is largely a result of their drug use, and assuch, often includes small-scale drug useoffences.1062 Many people involved would notnecessarily have been drug dependent.1061 This is avery different client base to Australian courts, whichhave tended to deal with clients from the moresevere end of the criminality spectrum.

There are two major meta-analyses of theeffectiveness of US drug courts1063, 1064 that representthe highest quality source of evidence available. Thefirst concluded that in general, drug courtsdemonstrated effectiveness in reducing drug useand criminal behaviour, for the duration of theprogram. The second study found that recidivismwas lower for participants in drug courts thancontrol groups. However, the second report alsodemonstrated that evaluations varied greatly in boththe nature of the programs evaluated and themethodological quality of the evaluations.1061

Australian drug courts have been implemented on apilot basis in WA, NSW, Victoria, SA, andQueensland. None are yet assured of continuedoperation1061 but political and community supportappears to be strong.1060 There are two youth drugcourts—in NSW and Western Australia. The WAyouth drug court is similar to the adult court, withmodifications in respect of length, type and

��%��

intensity of treatment.1061 The NSW youth drugcourt is established within the framework of theexisting Children’s Court and began operating, on31 July 2000, in two Children’s Courts in Westernand South Western Sydney. Its aim is to reduce druguse and offending behaviour among young peoplecharged with serious offences where alcohol orother drug use is a contributory factor.1065

The NSW youth drug court is designed as anintegrated system, which brings together elementsof the juvenile criminal justice system with a rangeof adolescent service providers. It aims to divertyoung offenders from custody by addressingholistically the health and welfare issues that areassociated with substance use and offending.1066

Interventions can include accommodation plans,case management, individual group or familycounselling, participation in education or vocationaltraining, health assessments and programs,recreational programming, urinalysis and courtattendance.1066

The evaluation of this court is being conducted bythe Social Policy Research Centre at the Universityof NSW. It consists of five studies includingstatistical monitoring, implementation review,outcomes study, program cost analysis and legalissues review. Only the second of these, theimplementation review, is currently available.1065 Itwas based on: interviews with 25 key stakeholdersand five participants; observations; and documentreview. It found that the program was operatingeffectively as a pilot, in that problems were beingidentified and addressed.

As mentioned above, there are now adult drugcourts in NSW, Queensland, WA, SA and Victoria,with the first court commencing in February 1999.Australian drug courts have quite markedly differentjurisdictional characteristics.1061

The typical features of Australian drug courts(largely based on the US model) include:

� integrated treatment and justice programs,

� collaborative approach between prosecution anddefence lawyers,

� early placement in treatment programs,

� drug testing,

� ongoing participation by the magistrate orjudge, and

� ongoing monitoring and evaluation.1060

Australian drug courts are aimed at heavy drug userswith a substantial criminal record who would

otherwise be facing custodial sentences.1061

According to Freiberg, all Australian drug courtshave a requirement that the defendant must pleadguilty to obtain the services of the court.1061

The inclusion criteria are fairly similar across allAustralian drug courts.1060 They typically include:

� being drug dependent, and that this contributesto offending,

� likely to be incarcerated,

� living in a certain prescribed area, and

� no history of sex offences, and in mostjurisdictions, no history of violent or seriousviolent offences.1060

The conditions of ongoing participation in theprograms are also similar across jurisdictions andinclude: reduced or eliminated illicit drug use,frequent drug testing, frequent contact with thecourt, various residential and offending restrictions,and participation in therapeutic programs.1061 Eachcourt has its own regime of rewards for appropriatebehaviour and sanctions for breaching ofconditions.1060

As noted above, there is little evaluative informationavailable to date on Australian drug courts, with theexception of evaluations of the NSW court. TheNSW drug court has a rigorous evaluation process,which includes a control group of referredoffenders who are randomly allocated to thetraditional court process.1060 The evaluation processcovers three main areas: (1) cost effectiveness (incontrast to other justice system interventions) andthe impact on re-offending, (2) health andwellbeing, and (3) a process evaluation of keyaspects of the drug courts functioning.1060

Of the 608 people accepted into the NSW programto March 2002, 446 exited. Of these, 19%successfully completed the program and received anon-custodial sentence; the other 81% wereremoved from the program and received a custodialsentence; 43% of all those who entered the programwere terminated either due to re-offending or othernon-compliance with program conditions.1067

The cost-effectiveness evaluation of the NSW drugcourt1068 appears to have been the first such study ofdrug courts anywhere in the world.1061 There aremany possible mechanisms to measure cost. At thebroadest range, costs might include health carecosts, justice and police costs related to thesubsequent rates of recidivism and costs of crime tothe victims.1061 A narrower view, such as that takenin the assessments of the NSW program, is to ��

��' �� !��"�#��$��%��

simply compare the cost of drug court participationwith the cost of serving the original sentence.1061

Even using this narrow criterion, which overlooksmany pertinent costs, evaluations of the NSWprogram found that the average cost for drug courtparticipants was $144 per day, whereas the cost forthe control group of equivalent offenders simplyserving out their sentence was $151 per day.1068

Thus, the cost of drug courts is essentiallycomparable to the cost of incarceration.

This evaluation also investigated recidivism rates fordrug and non-drug criminal offences, using acontrol group of eligible detainees who did notparticipate in the program.1068 The only statisticallysignificant differences between the two groups werethe time taken to re-offend (in relation to drugoffences only) and the rates of offending (for drugoffences). In relation to preventing further opiateuse, the drug court was significantly more cost-effective than the control group.

The health and wellbeing of participants was also asignificant key issue. Freeman reported on therelative health status of participants in the NSWprogram.1067, 1069 It was found that participants werein extremely poor health compared to the Australianpopulation when they commenced their program,but physical health showed significantimprovement, up to normal range within 12months. Social functioning and mental andemotional wellbeing improved, although the latterwere still lower than in the general population.Participants also reduced their expenditure on illegaldrugs while they were on the program.Importantly, over 60% of participants had theirprogram terminated in less than 12 months. Thelength of sentence was the only factor predictingtreatment retention, with longer sentences beingassociated with greater program retention.

A major process evaluation of the NSW drug courtwas undertaken in order to identify the strengthsand weaknesses, and outline how these weaknessesshould be addressed.1070 Overall, the NSWevaluation data indicates that the court iscontributing to a significant improvement in thehealth and wellbeing of its participants, and areduction in illicit drug use. The impact oncriminality and recidivism is less clear-cut, with thedata failing to demonstrate statistically significantreductions in recidivism and criminal re-offending(other than illicit drug use).

Both the Australian and US experiences suggest thatinitial implementation of a drug court program canbe a difficult time.1060, 1067 The American literaturealso demonstrates that evaluations that take place

too early in the process are likely to be unfairlydistorted by problems of initial implementation,and will not give a fair picture of the operation ofthe drug courts.1061

A central concern is the issue of the treatment ofoffenders whose offences involve alcohol ratherthan illicit drugs. NSW, SA and WA specify that theoffender must be, or appear to be, dependent onillicit drugs (excluding alcohol). Queensland merelystates that the person must be drug-dependent,without drawing a distinction between licit andillicit drugs. Victoria is the only jurisdiction that hasspecifically included alcohol-related offences. Thewidespread exclusion of alcohol is a serious concernin the eyes of some.1060, 1061 Many people presentingto the drug court are alcohol-dependent and alcoholhas a prominent role in public disorder and violentcrime.1060 The exclusion of alcohol from programsis of particular relevance to the Indigenouspopulation. Indigenous people are under-represented in drug courts, despite their over-representation in the criminal justice system.1061

Some authors have also raised concerns about thelength of involvement in drug courts; while mostcourts are based on a 12 month program, there isno apparent empirical justification for this timeframe.1060 Makkai has suggested that the courts mayneed to consider longer time frames, which willhave significant resource and policy implications.1060

All Australian drug courts are being evaluated and itremains to be seen whether courts in otherjurisdictions will have similar outcomes to thosefound in NSW. It would seem that the opportunityexists for comparative studies to assess whichelements, or combinations of elements, ofAustralian drug courts are the most cost-effective.

��"�� !��

A variety of strategies are used in prisons to controlor reduce the supply of, and demand for, drugs.Many of these include elements of harm reduction.Prison initiatives differ from jurisdiction tojurisdiction but generally include drug detectionprograms, treatment programs, education and peersupport programs and a variety of strategies tominimise harm. In general, little evaluation isavailable on the effectiveness of these programs.

��(��

&0'6'&�"�)�� -

Definition: There are drug detection programs in allprisons, typically associated with sanctions whendrug use is detected.200 Detection programs includedrug sniffer dogs, urine testing, and cell searches.


The AIDR summarises reports from correctiveservice agencies in every Australian jurisdiction onactivities designed to reduce the flow of illicit drugsinto correctional facilities.200 The range of strategiesincludes: the use of drug detector dogs, both activeand passive, to conduct cell and visitor searches;perimeter patrols; prison searches and barriercontrols; intelligence; search and seizure; urine drugtesting, both random and targeted; and penalties forprisoners caught trafficking. There are alsoidentified Drug User programs that create incentivesfor prisoners to reduce drug use by imposingpenalties such as loss of contact visits, whileencouraging participation in Drug-Free IncentivePrograms; drug detection sweat patches; and PrisonHealth Services prescription and storage protocolsfor recommended medications.200

There is very limited evidence on the effectivenessof these programs. The Queensland Department ofCorrective Services reports that urinalysis data foroffenders in custody indicate far less ongoing druguse within correctional centres, as the result ofpreventive action by staff, than in the community.The NSW Department of Corrective Servicesdescribes drug detector dog teams as ‘one of themost effective means of locating drugs’ but there isno quantification of this.1071 The South AustralianDepartment for Correctional Services reports on thenumber of drug searches conducted by the dogsquad during 1999/2000 and 2000/01 and theiroutcomes: recovery of drugs, pills, drug useimplements and home brew.1072 In Victoria thepercentage of prisoners testing positive for an illicitdrug diminished from 4.7% in 1998/99 to 4.4% in1999/2000.200

&0'6'.�"�� +�� +��-�� *�-)-�


The Victorian and South Australian prison systemsare conducting trials of differential penalties ofcannabis use, such that prisoners caught usingcannabis will receive less severe penalties than usersof more harmful drugs, in an attempt to limit theextent of switching from smokeable to injectabledrugs.1073, 1074

&0'6'0��-�� )��


All Australian prisons offer drug treatment, in someform. However, these appear to be mainly limitedto counselling services380 and there is great variationin the availability of access.1075 Treatment in thewider community is generally effective804 andprison offers an excellent opportunity to engagedrug users in drug treatment programs.380 However,little is known about the actual nature and quality ofprison-based treatment services and no evaluationsof their effectiveness are available.380

Methadone maintenance treatment (MMT) inprisons is surprisingly rare, given that it is the mosteffective heroin treatment approach in the generalcommunity381, 804 and that its effectiveness andrelative lack of risk as a treatment modality is nolonger seriously questioned.1076 Because of thesefacts, numerous expert groups, worldwide, havecalled for its widespread use in prisons.1076 A 2001research review identified only two evaluatedprograms, one in New York, and the other inNSW.1076

The NSW prison MMT program began in 1986.380

Evaluations noted that MMT reduced injecting riskbehaviour but only with an appropriately high doseof methadone, and only when the methadone wasprovided for the duration of the sentence.380 OtherAustralian States offer very limited methadoneprograms1073, 1074, 1077, 1078 and with a variety ofrestrictions on availability: to specific prisonergroups, such as those who were using methadone atthe time of admission, or those who are nearingrelease.1078 Some systems offer methadone on acompulsory dose-reduction model wherebyprisoners with sentences longer than six monthshave their methadone gradually reduced,1073, 1078

which is less effective than maintenanceprograms.804 Discontinuation of methadonetreatment while in prison has been shown to be arisk factor for unsafe injection practices.1076

Overall, the limited and often restricted availabilityof methadone programs in Australian prisons is aconcern, as is the provision of ineffective dose-reduction programs in preference to more effectivemaintenance-orientated programs. It should benoted that the Victorian prison system has recentlyundertaken to increase the number of prisonersbeing offered methadone or buprenorphinetreatment on a compulsory gradual withdrawalmodel.1073

��

��) �� !��"�#��$��%��

&0'6'(�"�)�5��) �-

Definition: Drug-free units provide support, includingcounselling, for abstinence from drug use. Variousincentive and punishment regimes are utilised,usually including a strict no-use criterion. Typicallyin return for continued abstinence, prisoners gainaccess to various extra privileges.200 Somejurisdictions have drug-free units and others areconsidering putting them into place.200, 1073


There are no Australian evaluations; an evaluation ofa US Therapeutic Community (TC) within a prison,which is a similar model, suggests that in prison TCalone may not be sufficient and is more effectivewhen followed with re-entry work release TC.1079

&0'6'6��-


There is a general evidence base suggesting thatmanagement of behaviour problems in prisons isbest accomplished by systems of incentive forpositive behaviour, rather than a reliance onpunitive responses for negative behaviours.1074 Inrecent times, some Australian prisons have adoptedthis strategy to include rewards and privileges as aresult of staying drug-free.1073 While there is nospecific evidence of the impact of this policy ondrug consumption, there is evidence that at ageneral level this approach is effective.1074

&0'6'7��)��


Education programs on the risks of drug use arefrequently offered within Australian prisons.200, 1073,

1074 In some jurisdictions, funding is provided forpeer support programs within the prison, in light ofthe success of such programs in the community. Noevidence on effectiveness is available.

&0'6'3��


Some strategies have proposed that much drug usein prison arises as a response to boredom, despair,separation, and loss of family, friends, and freedom.Policies that aim to normalise the prison experiencemay reduce stress and, therefore, drug use. Theseinclude constructive activities such as education andtraining, humane and courteous treatment ofprisoners, and efforts to humanise the physicalenvironment as much as possible.1074 There is noevidence of the effectiveness of such approaches.

&0'6';�� -�� +�-)�� +��-��

Definition: Transitional support upon release fromprison that might involve drug treatment agenciesand attention to post-release income support,education, employment and health.


Whilst there is no direct evidence for effectiveness,many programs are using these approaches in anattempt not only to improve the chances of releasedprisoners staying drug-free, but to generallyimprove the chances for them successfully adaptingto life in the community following a period ofincarceration.200, 1073, 1074 Concerns about the highlevel of heroin overdose in WA have led to theintroduction of a program involving extensivecollaboration with external agencies to supportprisoners leaving jail, including education on lowlevels of tolerance and risk of overdose, and toenhance the likelihood that they continue treatmenton release.200 No evaluation is available.

&0'6'<��-�� *+�� >�� 4)��


Bleach is available for sterilisation purposes in someAustralian prisons. These programs have beenreported to have the potential to reduce risk inregard to HIV/AIDS.380 A major concern, however,is that whilst bleaching injecting equipment may beeffective against HIV, it is not known whether it iseffective against hepatitis C.1080 A trial of theprovision of bleach has been discussed in WA.

&0'6'&=��+�� -,�� 2��


Needle provision in prisons has been advocated byvarious groups as a harm reduction initiative. Prisonofficers have generally been strongly opposed tosuch programs, threatening industrial action.380

Internationally, there are 19 prisons offeringsyringe exchange services.1076 Evaluations of theseprograms have consistently reported that levels ofdrug consumption either remained stable orreduced,381, 1076 syringe sharing was dramaticallyreduced, there have been no reports of syringesbeing used as a weapon and only one report ofneedlestick injury. There were no known incidencesof BBV transmission in any of these prisons.381, 1076

Implementation typically met with initial staffresistance but, in most cases, once the program wasestablished staff displayed favourable attitudestowards it.1076

��

&0'6'&&��+)�� -��-� ��-

Current evaluations of prison programs appear to beprimarily internal and as yet there do not appear tobe independent studies on the effectiveness of theseprograms in deterring or otherwise reducing druguse in Australian prisons. The Victorian Prison DrugStrategy is planning current and forthcomingresearch and evaluation which will be: scientificallyvalid, meet Department of Justice ethical guidelines,and be integrated with existing corrections researchand evaluation. Research results will be widelycommunicated, published for scrutiny in peerreview journals and presented at forums.1073 Oneinitiative is being evaluated by the University ofMelbourne’s Department of Criminology.

It follows that there is little evidence of benefit fromdrug testing programs in Australian prisons. Someinternational evaluation studies have reported thatthe introduction of testing programs made nodifference to consumption levels,381, but an Englishstudy reported that random mandatory drug testing,which plays a central role in reducing the supply ofdrugs in Australian prisons, promoted inmatesswitching from cannabis to heroin use, becausecannabis has a much longer urinary half-life thanheroin.1081 Such a switch, however, raises concernsabout injecting and the transmission of blood borneviruses. The switch from smokeable to injectabledrugs, as a consequence of random urine testing, isapparent in Queensland where the Illicit Drug UseAction Plan states:

An additional and disturbing aspect of illicit drug taking withinprisons is an emerging trend for further adverse effects fromprisoners tending to move from using ‘soft’ drugs which aremore easily detected than ‘hard’ drugs, including those takenintravenously, which are less easily detected (p4).1082

In Western Australia, an internal evaluation of apilot program was undertaken to test the operationaleffectiveness of using sweat patches in acommunity-based setting. Sweat patches are a wayof overcoming some of the problems associatedwith urinalysis. The pilot revealed that using sweatpatches could be an effective tool in managingclients under community supervision, but isconditional on resolution of administrative andoperational processes.1083

&0'6'&.��)��,

The two main foci in prison drug and alcoholprograms are: drug detection and deterrence, andtreatment provision. At the outset, it should benoted that prisons can be considered as ‘revolvingdoors’ for some drug users. Relationships in theDUMA dataset between arrest, imprisonment andopiate use demonstrate that there is a subset ofoffenders ‘enmeshed in a revolving door throughthe criminal justice system and heroinuse’(p1806).238 There is little evidence available onthe actual impact on inmate drug consumption ofmost prison policies, other than self-report from theprisons involved.200 Whilst there are notedexceptions, in general, methadone in Australianprisons seems to have limited availability despite itsdemonstrated efficacy in both the internationalliterature and in the NSW prison system. Deterrenceand detection programs are being modified in somejurisdictions in order to address the issue ofdifferential penalties for cannabis, a policy thatreceives general support from the literature.Deterrence programs appear to work best whenaccompanied by a corresponding system of rewardsfor appropriate behaviour.1074 Overall, there is noindependent evidence to suggest that deterrence-based programs are effective, with someinternational research suggesting that they make noimpact on consumption levels.381

Attitudes and beliefs of staff within the prisonsystem are traditionally described as ‘zerotolerance’, and in the past this has been an obstacleto provision of demand and harm reductionapproaches.1077 Harm minimisation is limited inmost jurisdictions to education and peer supportprograms, methadone maintenance in some prisons,and the provision of bleach for decontamination forwhich there is as yet no evaluation. The provision ofsterile needles to inmates in Australian prisons doesnot appear to be likely despite there beinginternational evidence that it is not associated withincreased drug use, reduces sharing of injectingequipment, and does not appear to give rise tohostile behaviour using needles as weapons. Giventhe high levels of BBVs, particularly hepatitis C inAustralia prisons, it seems that a trial should beconsidered.

��

�� !��"�#��$��%��

�� !!��

Harm reduction strategies are those that seek to minimise or limit the harms associated with druguse, without necessarily seeking to eliminate use. Such strategies have been part of Australiannational drug policy since the first policy was developed. Harm reduction strategies are oftenthought of mainly in terms of reducing the spread of blood-borne viruses among injecting drugusers, but there is a much wider range than this, with initiatives relating to the use of tobacco,alcohol, illicit drugs and pharmaceuticals.

This chapter reviews these strategies, including regulations to reduce passive smoking;approaches to reducing drink driving; thiamine supplementation to reduce alcohol-related braindisease; night patrols and sobering up shelters, particularly in Indigenous communities; thereduction of violence and the effect of violence in licensed drinking environments; reducingopiate overdose and blood-borne viruses, including safe injecting centres and retractablesyringes; and limiting harms associated with so-called ‘dance drugs’ (phenethylamines, LSD andketamine). Finally, approaches to reducing the use of benzodiazepines in drug assisted sexualassault are reviewed.

�� !�� $ �� 9�� !�� !��

�� *�� 7 �'�'�� , � �� ;�� 9

��

�� 8� � �� '� �� ;�� 9��

��

( �� 9 ��

� �� $��

*�� B@� �!��*� ��-�� '�� 2��> '� � �� 9��

� � ��

(�� '�� 9�� J��

�� 9�� 6��

!�� $��

� �� '�� 9�� '� �� '�� 9�� 9��

��

*�� 8�� L�� 9�� '�� 9�� $ ��

� �� <-�� #$L�� /� ��:3%@%10

�� %%%

��

�� !�� $ �� 9�� !�� !��

*�� +�� $ �� '�� 2.3�� 9��

� �� 6��

'�� *�� !�� /� � <�� 9 *�� 0��

!�� "�� 9��

�� L�� '� ��6�� $�� 9�� %

+�� 7� �� $�� $�� ,� ��

*�� <�� <�� 9�� '��%

�� '�� '� �� 9�� '��%

�� '��

�� <�� 9� ��

�� !�� 7�� 9�� /7�0

8� �� 9�(�8��

��7�9��

�� '�� ) ��$� ��

� 6��

< �� ( �� +!G9�� 9�� 9�� 9

�� 9��

��!��

'�� " �� '�� 9��

� ��

+ ��- �� '�� 9�� < �� " �� ( �� # �� 9�� !��

�� 9��

!�� '�� 9�� I�� J

�� %%%

��

�� !��"�#��$��%��

�� !�� $ �� 9�� !�� !��

�� ,�� +�� $� ��9� �� .��

�� '�� 9�� "�� 9�� 9��

�� 7 <� �� '� ��

�� 7 <� �� 9�� -H�� &��& �� /-H�0 ��

+��

>��7 *�� $��

� #��

� ��

��

communities.13 This separation of the terms harmreduction and harm minimisation was partly apolitical re-emphasis of the importance of supplyreduction and abstinence approaches in Australia’snational strategy.

Despite these definitional changes, it was evidentthat from the mid 1980s to the mid 1990s, themain focus of harm reduction or harmminimisation in Australia, and elsewhere, was onreducing the harm of drug use itself, particularlythat associated with injecting drug use and thetransmission of HIV/AIDS.1085 However, in theirreview of the National Drug Strategy, Single and Rohlsaw one of the key underlying principles of theharm reduction approach as that, in reducing drug-related harm, consideration should also be made ofthe ‘unintended harms which may result from thedrug control strategy itself’ (p47).1086 Recentintroduction of cautioning schemes for illicit drugsin all Australian states and territories (exceptcannabis in jurisdictions where civil penaltiesapply), is in part reflective, if not a consequence of,this recognition of the need to consider the harmscaused by drug control strategies themselves.

Australia’s pragmatic response to drug policy hasbeen internationally recognised, mainly for itssuccess in minimising the spread of HIV amonginjectors, their sexual partners and the generalcommunity.

��*�!�� 1��

Since the mid 1980s, the stated aim of Australia’snational responses to drug use, The NationalCampaign Against Drug Abuse and the NationalDrug Strategic Framework, has been one of ‘harmminimisation’. Until the late 1990s, this wasdefined as an approach that:

aims to reduce the adverse health, social and economicconsequences of alcohol and other drugs by minimising orlimiting the harms and hazards of drug use for both theindividual and the community, without necessarilyeliminating use (emphasis added) (p4).1084

In its review of the 1998 National Drug StrategicFramework, from 1998/99 to 2002/03, theAustralian Government restated its commitment to‘harm minimisation’ but couched this as policiesand procedures that aim to reduce drug-relatedharm and improve health, social and economicoutcomes by employing a range of measuresincluding: (1) supply reduction strategies aimed todisrupt the production and supply of illicit drugs;(2) demand reduction strategies designed to prevent theuptake of harmful drug use, including abstinence-oriented strategies to reduce use; and (3) a range oftargeted harm reduction strategies designed to reducedrug-related harm for particular individuals and

��"��

�� !��

The main goal of harm reduction is to reduce drug-related harm, rather than the use of drugs per se. ‘Usereduction may be a strategy to achieve harmreduction but when the goal of an interventionbecomes the reduction of use, then a program,policy or intervention should not be described asone of harm reduction’ (pp51–2).215

Harm reduction strategies assume that many peoplewill continue to use drugs and do not require,necessarily, that this use is ceased. Harm reductionapproaches concern the harms that are associatedwith continuing drug use, and attempt to reducethese harms.

Lenton and Single have noted that in recent timesthere have been at least three main uses of the termharm reduction in the drug research literature: (1)narrow definitions, which only apply the term topolicies and programs (such as needle exchange)that aim to reduce the harm for those whocontinued to use drugs; (2) broad definitions whichinclude any policy or program that aims to reducedrug-related harm; and (3) empirical definitions,which limit the use of the term harm reduction topolicies and programs where a reduction in harmcan be empirically demonstrated.1087 Havingarticulated the shortcomings in each of thesedefinitions, these authors offer an alternativesocioempirical definition.

A policy, programme or intervention should be called harmreduction if, and only if:

(1) The primary goal is the reduction of drug-related harmrather than drug use per se;

(2) where abstinence oriented strategies are included, strategiesare also included to reduce the harm for those who continue touse drugs; and,

(3) strategies are included which aim to demonstrate that, onthe balance of probabilities, it is likely to result in a net reductionin drug-related harm (p219).1087

�� 3��

There are few strictly harm reducing policiesrelating to tobacco, because the broad consensus isthat there is no safe level of use. The only harmreducing strategies in the literature relate to low taror ‘light’ cigarettes, supposedly safer alternatives tosmoking cigarettes. These include snuff and nicotinenasal sprays, and measures to reduce passivesmoking by non-smokers.

&('('&�%��-


The US Surgeon General has raised concerns aboutlow tar and so-called ‘light’ cigarettes.944 It has beenfound that smokers used to mid and high tarcigarettes tend to compensate by inhaling harder ifgiven a low tar cigarette to smoke. Terms such as‘light’ and ‘ultra light’ are misleading in that suchcigarettes are not necessarily any less harmful toconsumers. There appears an implied promise ofreduced toxicity in such promotions, and there havebeen proposals to regulate against misleadingpromotional terms, but no trace of the latter couldbe found.

&('('.��+�� +��,-,-��-


Some authors have proposed that other, lessharmful nicotine delivery systems could be aneffective harm reduction strategy for tobacco.1088

Given that a proportion of the harm from smokingis associated with smoke inhalation, some haveproposed that products may be developed in thefuture that allow delivery of nicotine to smokerswithout the need to inhale smoke. Australia has anational policy of not encouraging the developmentof safer alternatives to cigarettes for fear that thesemay encourage non-smokers to take up the habit, orex-smokers to relapse. By contrast, Sweden haspermitted the sale and promotion of a type oftobacco for nasal ingestion (‘snuff’’), which hastaken a significant market share away from tobaccocigarettes (Simon Chapman, personalcommunication). There is some concern that suchproducts may increase the risk of cancer of themouth but this is likely to be less of a serious healthissue than lung cancer.

&('('0��-�� -�� --��-��1� �

Definition: A diverse range of legislative andregulatory approaches, worldwide, that have theeffect of restricting smoking in public places inorder to prevent passive smoking.


Systematic reviews have demonstrated that,generally, these interventions appear to accomplishtheir aim of preventing smoke exposure in certainenvironments,944 and thereby act to reduce passive

��

��# �� !��"�#��$��%��

smoking in public places.468, 783 The protection ofnon-smokers from the effects of environmentaltobacco smoke, by restricting the places wheresmoking is permitted, is clearly a harm reductionmeasure.

Clean-air regulations contribute to a different socialclimate regarding the acceptability of smoking, andthis may indirectly influence the prevalence ofsmoking944 but there is no direct evidence that theseregulations reduce overall smoking prevalence.Public support for such legislative and regulatoryapproaches appears to be high, even amongsmokers.944 There is no evidence of adverseconsequences resulting from such policies andsupport for them is generally high.944 Enactment ofsuch policies is simple, effective and inexpensive.

��"�1�� !��

Drinking to intoxication is a major contributor toshort-term harm from alcohol, and accordingly,there are many policies aimed at reducing theharmful impact of intoxication. As outlinedpreviously, the harms related to alcohol can oftenstem from the acute consequences of one session ofintoxication, and this is an especially predominantproblem in the younger population. It has recentlybeen estimated, on the basis of the 1998 NDSHS,that 90% of all alcohol consumed by 18 to 24 yearold Australians was consumed in a manner thatplaced the drinker at risk of acute and/or chronicharm from alcohol.962 Reducing levels ofintoxication is discussed earlier in this document. Inthis section, we present the methods and approachesthat aim to reduce the levels of harm to thosealready intoxicated. In addition, some interventions,such as thiamine replacement, seek to reduce theimpact of long-term chronic alcohol consumption.

&('6'&�"�� 15�� +��,

There are a number of approaches that are effectivein reducing drink-driving mortality and morbidity.Four main strategies have been used.960

� Reducing overall levels of consumption: astrong effect of reducing population-level, orlocal-level, consumption of alcohol is areduction in alcohol-related crashes. Reducingoverall consumption has been discussed earlier.

� Separating drinking from driving: thisincludes advertising and promotionalcampaigns, increasing use of public transport,designated driver programs, and similarinitiatives. There is little direct evidence foreither the effectiveness, or the lack thereof, ofsuch approaches.

� Removing drunk drivers from the road: thisincludes a variety of approaches, includingsobriety tests, breath testing, training police inidentifying drunken drivers and similarprograms. In the Australian context, this mainlyinvolves random breath testing (RBT).

� Preventing recurrence: preventing drink-drivers from repeating their offences is anotherkey approach. This includes specific deterrenceeffects of fines and legal sanctions for offenders,the jailing of repeat offenders, and otherprograms such as interlock devices.

The diversion of drink-drivers into brief treatmentinterventions, usually based on educating theoffender about ways of avoiding drinking anddriving, is widely practiced in the US.1057 A USreview of such diversion programs found a modesteffect of an 8 to 9% reduction in repeated drink-driving offences,1089 although individual evaluationstend to be hampered by lack of, or inadequatelymatched, control groups. Nevertheless, a recentreview of the effectiveness of court proceduresapplying to drink-drivers found that diversion toeducative and treatment interventions andincapacitation were the most effective means ofreducing drink-driving recidivism, when comparedto punishment via incarceration and fines.1090

Lower BAC limits for young drivers

Definition: Because of the increased likelihood offatalities in teenage and younger drivers,960 manyplaces throughout the world have trialled andresearched lower maximum permitted BAC levelsfor young and probationary drivers.


The evidence base for the effectiveness is tentativebut it appears that having a lower permitted BAC foryoung drivers reduces risk of fatalities. A systematicreview identified six studies, of which only threereached significance.775 One of the studies reportingsignificance claimed that the effect was the mostpowerful for a BAC limit of 0, and effectiveness wasreduced for limits of 0.02 and 0.04 respectively.775

Random breath testing


Random breath testing has been shown to beeffective, in Australia and the US.775 It reducesfatalities, injuries and crashes. A key factor in thesuccess of RBT is the principle of generaldeterrence. For RBT to be effective in preventingpeople from driving drunk, it requires that there is a

��%��

public perception that there is a high chance ofbeing caught.186, 1091 In Australia, this has beenachieved by a combination of high-visibilitypolicing (roadblocks, ‘booze buses’) and frequentpublic advertising campaigns emphasising thelikelihood of drink-drivers being detected.

A Cochrane review concluded that randomscreening (a broader definition than RBT) waseffective in reducing fatalities and injuries fromroad crashes.783 The review also determined that thecommunity perception of being caught may be akey factor in the success of such campaigns.Australia’s screening program (RBT) was singledout for being particularly successful, in comparisonto other nations. The success of Australia’s RBTprogram is attributed to the way it is implemented,including the fact that all stopped drivers are tested,community perceptions of the chance of beingcaught is high, and high intensity ofimplementation.

Ignition interlocks

Definition: Ignition interlock devices require driversto provide a breath sample before starting theirvehicle. Interventions where drink-drivingoffenders are required to have such devices fitted totheir cars have been trialled in a variety ofjurisdictions, internationally.

Summary: Evidence for outcome effectiveness..................................................................... �� 2/2

Although the evidence base is relatively small, itappears that the fitting of ignition interlock devicesreduces the risk of re-offending, while they arefitted. A systematic review identified only onerandomised controlled trial on this issue, whichreported a relative risk of re-offending of 0.36 forthe group with interlocks.775 Re-offending is oftenan issue after the interlocks are removed.Furthermore, drivers will frequently only accepthaving an interlock fitted if the alternative is prisonor a heavy fine.960 However, during the period thatthe interlock is fitted, there is strong evidence thatre-offending is at extremely low levels.1092, 1093

Designated driver schemes


Various locations around the world have runpromotional and educational campaigns ondesignated driver schemes. The evidence base islimited, however, and a US review concluded thatsuch strategies are not particularly effective inproducing behaviour change.960 However, anAustralian evaluation of a designated driver

intervention for young adults, known as Pick-a-Skipper, was shown to at least achieve its basic aimof persuading a significant number of youngdrinkers who were intending to drive to and fromtheir location of drinking, to select non-drinkingdrivers as ‘skippers’, before they began consumingalcohol.851, 1094 This finding is also replicated in theinternational literature, which reports a rise in theprevalence of designated drivers, as measured byroadside surveys, from 5% to 25% between 1986and 1996.960 The US literature has also reported thatin a sample of 600 people convicted of drinkdriving, ‘many’ had identified a designated driver,and this driver reneged on the agreement and as aresult, ‘drunk-driving’ took place. Australian studieshave reported that 26% of respondents appointed asdesignated drivers had driven in this role whileunder the effects of alcohol.1095 Studies of USestablishments with designated driver programs(offering free soft drinks) found that participationwas low.960 The evidence base is weak and firmconclusions cannot be drawn. However, it is hard tosustain an argument that providing free soft drinksto drivers can increase the risk of drink-driving andit is likely they contribute, to a limited degree, inreducing risk. Since the cost of such schemes isborne by the retail alcohol industry, there is noopportunity cost of recommending such schemes.

General issues with licensing and licensed venues

Across a diverse range of contexts, it has typicallybeen shown that self-regulation of licensed drinkingvenues in the absence of traditional lawenforcement is ineffective.1096 Where practice isregulated by law (not serving under age patrons,not serving intoxicated patrons, restrictions ondiscounting), regulatory enforcement is generallyrequired to create compliance. The liquor market isfiercely competitive and it is often profitable toviolate the regulations. For example, it is difficultfor one hotel to comply with a no-discountingapproach while their competitors host crowded‘happy hours’.524

A generally recommended policy has beenpartnership approaches that include industryconsultation in program design, in conjunctionwith legal frameworks providing deterrence toviolation of regulations.524 Intelligence-drivenpolicing has also been recommended, based on therepeated findings that a small minority of licensedvenues is associated with the majority of incidencesof alcohol-related harm.973, 975

��

��' �� !��"�#��$��%��

&('6'.�� -)��+��

Definition: Wernicke-Korsakoff’s syndrome is a formof serious brain damage primarily caused by long-term heavy alcohol use. A substantial part of itscausation is a thiamine deficiency brought on bothas a direct consequence of heavy consumption ofalcohol over many years, and as an indirect result ofthe poor diet typically associated with heavy alcoholconsumption. Prevention is possible with thiaminesupplementation.


There are two main contenders for thiamineenrichment—bread-making flour and alcoholicbeverages. In 1991, Australia decided to supplementflour with thiamine. Thiamine fortification of beerand flagon wine rather than bread was originallyrecommended by the NHMRC, in 1987, butopposition from government and nutritionists led tothe compromise position of supplementation ofbread.1097 However, Wernicke-Korsakoff’ssyndrome occurs mainly in alcohol dependentdrinkers, and as beer and bulk wine (casks andflagons) is the preferred beverage of patients withWernicke-Korsakoff’s syndrome, beer and bulkwine supplementation is more likely to improvethiamine intake in heavy drinkers thansupplementation of bread.1097

A number of arguments for and against thesupplementation of different products have beenpresented, including the concern that alcoholicbeverages might be promoted and seen as ‘healthfoods’ if they were supplemented with thiamine,and this might impact adversely on overallconsumption.1098 Against that is the concern that thediet of very heavy drinkers is so poor that they areunlikely to obtain enough thiamine through foodalone.1099

Australian work reviewed by Ludbrook indicatesthat thiamine supplementation of beer would be acost-effective way of preventing the incidence ofKorsakoff’s syndrome.775 It was found that the mostcost-effective method was fortifying full-strengthbeer; the second most cost-effective involvessupplementation of beer and cask and flagon wines;whilst the least cost effective method was fortifyingbread-making flours with thiamine.1099 Thisresearch utilised dietary and drinking histories toestimate potential benefits from the three varioussupplementation approaches, in terms of reducingWernicke-Korsakoff’s. However, there is suggestiveevidence from studies of hospital admissions, thatthiamine supplementation of bread in Australia inthe late 1980s led to reduced prevalence ofWernicke-Korsakoff’s syndrome over time.1100

&('6'0��+�-� ��-��*��-� ��+)*-

Summary: Evidence is contra-indicative ..........� 0/1

As noted previously, there is strong Australianevidence that changes to trading hours cansignificantly impact on levels of violence in andaround licensed premises.972 Violence is oftenassociated with closing times of drinking venues inentertainment areas, believed to be associated withthe gathering of large numbers of intoxicatedpeople in the same physical space.525 For this reason,authors have proposed that staggered closing timesmay reduce violence by reducing the potential forlarge numbers of intoxicated people to gather.Other complementary measures proposed haveincluded ensuring efficient transport is available toremove intoxicated patrons, and more extensivepolice monitoring of locations with a great numberof licensed premises.525 A recent Scandinavianexperiment with staggering trading hours fornightclubs, by allowing all-night trading, did resultin a more even flow of problem incidents foremergency services to deal with but, unfortunately,at a much higher level than when clubs wererequired to close earlier.974 It is possible thatstaggered closing hours might be beneficial if theydo not result in an overall extension of trading time,which in practice may be hard to achieve.

&('6'(�%�� -��-�� )��)��+� ��


There is a substantial body of risk and protectivefactors associated with the risk of violence in andaround licensed premises. Many risk factors that arestatistically associated with increased or reduced riskof harm are modifiable. Therefore, the argumenthas been proposed that changing these risk factorsmay modify the chance of harm.

One mechanism for modifying drinkingenvironments with a view to reducing alcohol-related harm is through licensee codes of conduct,negotiated with police and other interested partiessuch as public health and local governmentpersonnel. These are usually referred to as AlcoholAccords. There are many possible components tosuch agreements such as responsible alcohol service,drink discounting bans, trained security personnel,provisions of food, use of safe glassware and alcoholcontainers, environmental modifications to reduceconflict and frustration and thereby risk of violence.

��(��

The Surfer’s Paradise Safety Action project was oneof the first examples of a comprehensiveintervention involving alcohol service venues. Itinvolved encouraging a code of conduct for barsand clubs in a popular entertainment area (on theGold Coast) that had developed a reputation forviolence. Interventions included regulating servingstaff, security staff, advertising, alcohol use andentertainment. Managers were regulated through arisk assessment system, and were ‘rehabilitated’ intothe local business community. Managers from othercities who had demonstrated a commitment toreform were involved. On top of this, a key factorwas observed to be the significant interpersonalskills of the project officer involved.524 This projectwas demonstrably successful, overall, in reducingalcohol-related harms including physical and verbalaggression. Consumption and drunkennessdeclined, and the physical environment improvedsubstantially. None of these improvements can beattributed to any one factor but as a package theprogram was successful. However, the benefitswere not maintained when the research teamdriving the project moved on, and conditionsdeteriorated after two years. There was also someevidence that improvements were, to a degree, aresult of displacement of problem patrons to otherlocations.524 For local level interventions, this mayonly be an issue for communities that do not havean ongoing active program.

Later work by the same team976 replicated theSurfers’ Paradise model, across two years, in threenorth Queensland towns. It was found that therewere striking reductions in physical violence in andaround licensed drinking environments, whichcould be causally linked to the intervention. Theauthors remain committed to the robustness of themodel but acknowledge that this was a preliminaryevaluation and that a wider range of data are neededbefore firmer conclusions can be drawn..976

There have been two other published evaluations ofAustralian licensing Accords.1101, 1102 The evaluationof the Geelong Accord found positive outcomes onsome evaluation criteria and not others, whilst theevaluation of the Fremantle Accord found noevidence of effectiveness against a number of hardoutcome criteria. One of the differences betweenthe two projects may have been the extent ofcredible enforcement of liquor laws thataccompanied the Geelong Accord.1096 There areclearly a number of implementation issues such asthis but there is no doubt that Accords can be aneffective vehicle for introducing some harmreducing practices into licensed drinking venues. Itis recommended that such voluntary regulation is

accompanied by effective liquor law enforcement,as discussed earlier.

&('6'6��+�-��+�--


The observation that some of the more severeinjuries from bar fights were linked to usingdrinking glasses as weapons has led to thesuggestion that using tempered or plastic glasswould reduce harm.524 There is no direct researchon success or failure of this intervention; however,many drinking venues have implemented thisstrategy and it is highly likely to be effective onpractical grounds alone. Similar arguments apply toseeking alternatives to glass beer bottles, which havealso been used as dangerous weapons.

&('6'7��-��

Definition: Encouraging eating with drinking has beenproposed on the basis that the service of food is aprotective factor.524


It is well established that drinking alcohol with afull meal can reduce BAC substantially—by up to60%.42 Provision of food in bars has been acomponent of some Alcohol Accords. In someAustralian jurisdictions, it is a requirement thatrestaurant licences restrict alcohol sales only topeople who are eating. There is Australian datashowing that these are low-risk licensed premises,both for violence and subsequent drinking anddriving.1103 However, the precise contribution tofood service, as opposed to type of clientele andother environmental factors, cannot be determined.There is also the contrary and commonly usedpractice of serving salty snacks, which increasealcohol consumption by increasing thirst.

&('6'3�#��)�� )-��) ��-

Night patrols

Definition: The most common harm reductionstrategy in Indigenous communities is the use ofnight patrols, which provide transport to safelocations for intoxicated persons.


There were 69 patrols operating in various locationsin 1999/2000.859 Most were designed to reducealcohol-related conflicts and harm. Mosey providedan overview and largely qualitative assessment ofthe operation and effectiveness of remote area night

��

��) �� !��"�#��$��%��

patrols in Central Australia.1104 Sputore and hercolleagues have conducted qualitative and limitedquantitative evaluations of patrols in Kununurra,Wyndham and Halls Creek, in WA.867, 868 While thequantitative measures were equivocal—because ofconfounding factors—people in the WAcommunities generally rated the patrols as effectivein reducing alcohol-related violence and gettingintoxicated people off the streets.

Sobering-up shelters

Definition: Sobering-up shelters provide a temporaryhaven for, and supervision of, intoxicated people atrisk of causing harm to themselves or others, anddivert intoxicated people from police custody.

Summary: Warrants further research .......................

In 1999/2000 there were 23 shelters.859 Daly andGvozdenovich conducted a qualitative evaluation ofshelters in three WA towns and found that theshelters were well-accepted by both clients andpolice.1105 Evaluation of a shelter in Kununurra,Western Australia, found that it was well acceptedand led to a significant reduction in policedetentions of intoxicated people.867

��#�,�� $��

Interventions to reduce mortality and morbidityassociated with heroin-related overdose are of twotypes: strategies to prevent the occurrence of heroinoverdoses, and strategies to improve themanagement of overdoses when they occur. Theformer include educating users about risk factors foroverdose (e.g. poly-drug use, using opiates whenalone, reduced tolerance with abstinence, etc.) andhow to minimise the risk factors, and expandingtreatment options, especially methadone. The latterinclude improving user response to overdose byencouragement to call an ambulance, and trainingin resuscitation; establishing protocols betweenpolice, ambulance services and user representatives(e.g. that police will not routinely attend ambulancecalls to overdose); setting up peer outreach, such asthat targeted through hospital emergencydepartments; and making naloxone hydrochloride(Narcan) available to users, peers, family membersand outreach workers. Since, in Australia andelsewhere, many of these interventions are stillbeing expanded, it is too early to see any impact onthe number of deaths.

&('7'&��)�� +�� -


There are many behavioural risk factors for opiateoverdose. These include injecting when alone,injecting while under the influence of alcohol and/or benzodiazepines, and injecting after a period ofabstinence or otherwise reduced tolerance.230, 1106

Accordingly, consideration has been given toeducating heroin users about overdose riskbehaviours. While such programs have beenconducted in many locations, actual evaluations oftheir impact are scarce. A South Australian studyinvolved a comprehensive education campaign, andliaison with police and ambulance.1106 Researchdemonstrated that the target audiences were wellaware of the key messages of the intervention andthere was tentative evidence that this was associatedwith an increased rate of ambulance attendance atoverdose, resulting in lower death rates.

&('7'.��+-�� ,�-��-�� -�

Definition: Concerns have been expressed that policeattendance at heroin overdoses will reduce thelikelihood that users will call an ambulance.1107


An inherent part of the South Australianintervention, referred to above, was developingambulance and police protocols for overdoseattendance, ensuring that police did not attendoverdoses unless required, and then distributinginformation about this to heroin users. The NationalHeroin Overdose Strategy includes two keyrecommendations.1108

� Develop protocols between police andambulance services that clarify thecircumstances under which ambulance servicesmay call on police to attend drug overdoses.

� Develop police protocols for attendance atoverdose (whether called by ambulance orother means), including use of discretion.

The National Strategy, therefore, reflects theprinciples that police policy should not act to reducethe likelihood that users will call an ambulance, andthat police need the flexibility to exercise discretionin order to avoid harm.

��

&('7'0��+�*�+�,��


The evidence shows that treatment, particularlymethadone maintenance treatment, substantiallyreduces the risk of heroin overdose.234 Heroin usersnot in treatment have a risk of mortality 13 timesthat of non-using aged peers, compared to 3.4times for those who are in methadone maintenancetreatment.408 Hall recommended that expansion ofthe treatment options available should be a majorstrategy for reducing opioid overdose mortality.This should include alternative pharmacotherapies,such as LAAM and buprenorphine, which may bemore attractive to older, long-term users.1109

&('7'(��-�� +�2� �� -��

Definition: Naloxone administration instantly reversesboth intoxication and overdose.


Treatment of heroin overdose by the use of heroinantagonists, such as naloxone (Narcan), is almostuniversally indicated in medical settings.230

Naloxone administration is safe and reliable. Oneinitiative that has been proposed to reduce the rateof opiate overdose is the distribution of naloxone tousers.230, 1110, 1111 To date, no Australian trial has beenconducted of such an intervention.

There is an international literature on naloxoneprovision to users which appears to support this useas a preventive measure. Naloxone has beenavailable over-the-counter to users in Italy since1995 and is also available in a variety of otherlocations. However, no comprehensive evaluationshave been performed on its effectiveness for thisuse.

Possible benefits of naloxone administration aregiven.

� Naloxone administration is a simple, easy-to-use method for immediately and effectivelytreating heroin overdose.

� Naloxone has no effect in the absence of heroin,and cannot be used as a recreational drug.

� Overdoses rarely occur alone, meaning thatthere is potential for others to provide aid.

� There is a minimal risk of adverse reactions tonaloxone

Possible complications include the following.

� Naloxone administration often requires first aidto maintain breathing until the naloxone takeseffect. Therefore, it is recommended thattraining in first aid take place in conjunctionwith naloxone distribution to users.

� There are legal complications surroundingmaking naloxone more available.

There are some criticisms and concerns about theproposition, as follows.

� It has been suggested that more hazardousopiate use may occur as a result of the securityof naloxone being available for peeradministration. Given that naloxoneadministration is quite unpleasant, this seemsunlikely.

� Naloxone will be little use to those injectingalone (40% of fatal overdoses).

� Naloxone follow-up care may be lacking, incontrast to situations where medicalprofessionals have been involved.

There have been no Australian trials and a well-controlled, well-designed and cautious trial to assessthe use of such a strategy in preventing overdosedeaths is warranted.1110, 1111

��%�,��!�� 5��@��

The major harms associated with injecting are thetransmission of blood-borne viruses (BBVs),overdose, and the reduction of public amenityrelating to visible injecting and discarded usedinjecting equipment. Accordingly, programs havebeen developed to address these concerns.

&('3'&��+�� -,�� -��*)�� -

Definition: Needle and syringe programs (NSP) havebeen the cornerstone of Australia’s response to BBVsin injecting drug users.1112 NSPs include not onlydedicated needle exchange programs, but alsopharmacy sales of injecting equipment.


The key aspect of successful BBV harm reduction forinjectors is that they should have easy access toclean injecting equipment—predominantly needlesand syringes but also swabs, spoons and sterilewater. Sharing non-needle elements of injectingequipment is believed to be implicated in thetransmission of hepatitis C.1113

��

�� !��"�#��$��%��

The initial impetus for the development of NSPs wasthe HIV/AIDS epidemic. While there has beenconsiderable concern in the US about the effect ofNSPs, the US Centres for Disease Control have failedto find any evidence that needle exchange programsaffect either frequency of injection among users orrates of recruitment to injecting drug use, whileconcluding that the programs significantly reduceneedle sharing.1114, 1115 A wide body of otherinternational research, including Cochrane reviews,confirms the finding that NSPs are effective inpreventing HIV infection and do not cause anyincrease in drug use.783, 804, 1116, 1117

The early and rapid establishment of needledistribution programs in Australia is one of theforemost reasons why the rate of HIV infectionamong Australian injectors has remained at less than3%. This is borne out in a report of the economiceffectiveness of NSPs in Australia. The Return onInvestments study investigated the effectiveness andcost-effectiveness of needle and syringe programs(NSPs) in Australia, from 1991 to 2000, and foundthat NSPs were effective in reducing rates of HIVinfection and hepatitis C. It was estimated that inAustralia, by the year 2010, 4500 HIV-relateddeaths and 90 HCV-related deaths, will have beenprevented, as well as 25 000 HIV infections and21 000 hepatitis C infections. In treatment costs,this would amount to savings of $2.4 to $7.7billion for an investment of almost $150 million. Inaddition, benefits for consumers in number of lifeyears gained and improved quality of life aredemonstrated.1118

The prevention of hepatitis C among injectors is agreater challenge than the prevention of HIV/AIDSbecause this epidemic was well established beforethe introduction of NSPs. It has been recommendedthat NSPs be expanded in sufficient quantity to meetdistribution targets of injectors always using sterileinjecting equipment. The demand for needles andsyringes has been rising exponentially, but fundinghas not always been available to meet thiscontinually increasing demand.1112

A common community concern about NSPprograms is that they increase the number ofneedles and syringes discarded in public places.However, ANCD research reviews have concludedthat, in general, NSP programs do not increase thenumbers of inappropriately discarded syringes.804

Other elements of Australia’s harm reductionapproach to BBVs in injectors are outreach and

education, and the provision of methadonemaintenance treatment (MMT). NSPs often offer avariety of ancillary services, including treatmentlinkage, education, information, screening andtesting services, and advocacy services.1116 Ingeneral, outreach programs have been shown topromote treatment entry and encourage somedegree of change in levels of risk behaviour.804 Ratesof needle sharing have decreased steadily over thisdecade, and it is plausible that these programs havecontributed to this behavioural change.1112 Some USstudies have indicated that outreach-based educationprograms were effective in reducing risk behavioursin injectors, including needle sharing and unsafesex.1119

Needle and syringe exchanges in Indigenouscommunities

The provision of free or cheap needles and injectingequipment is a key strategy in attempts to reducethe spread of BBVs among people who inject drugs.In 1999/2000, there were six needle exchangesspecifically for Indigenous people and an unknownnumber of Indigenous health services also providedclean needles and other injecting equipment.859

&('3'.�$�� ,8$$�9

It is clear that MMT assists in the reduction of HIV-risk behaviour and can reduce the incidence of HIVinfection. Participation in MMT is related to lessfrequent injecting, less frequent sharing, fewersharing partners and lower HIV seroprevalence.1120

In a review of literature addressing the impacts ofMMT on BBVs, Ward, Mattick and Hall concludedthat there was considerable evidence thatmethadone protects against HIV infection, becauseit reduces risk behaviours such as injecting andneedle sharing.1121 The same level of confidencecannot be applied to reductions in the risk ofhepatitis C—partly because many injectors will havecontracted hepatitis C before entering treatment. Tostate the obvious, if MMT reduces the prevalence ofinjecting, the occasions in which there is a risk ofHCV transmission are reduced. Since hepatitis Cprevalence is closely linked to duration of injecting,methadone and other treatments would be moreeffective against hepatitis C if they were undertakenearlier in the injector’s career.

��

&('3'0��)��-�� >�� -

Definition: Legally sanctioned environments wheredrug users can inject under some degree ofsupervision.


Other approaches to the reduction of harmsassociated with injecting include the introduction ofsupervised injecting centres. The evidence baserelated to the potential prevention benefits ofAustralian supervised injecting centres is sparse asthese centres are a relatively recent Australiandevelopment. There are a number of reviews ofinternational experiences; and six month and 12month progress reports of evaluations of Australia’sonly medically supervised injecting centre (MSIC),in Kings Cross.

Supervised injecting centres have been used in avariety of major city locations across the world in anattempt to reduce some of the harms associated withinjecting drug use.1124–1126 While there arenumerous variants, the basic notion is that a legallysanctioned environment is provided where drugusers can inject under some degree of supervision.Clean needles and syringes are usually available,surroundings are typically far more hygienic thannormal injecting environments, and staff areavailable to provide immediate resuscitation after anoverdose.

There are four main purported benefits ofsupervised injecting centres.1124

1. Reduction in harms to the general public(reductions in: discarded syringes, visibility ofstreet injection scenes, frequency of publicinjection and public intoxication).

2. Reduction in risk of overdose, both fatal andnon-fatal.

3. Reduction in risk of BBV transmission.

4. Improved access to treatment and other socialwelfare services.

There are three main criticisms of the generalconcept of injecting rooms.1124

1. Being seen to condone drug use, and thereby‘sending the wrong message’.

2. Causing a congregation of drug dealers aroundthe site—the ‘honey pot’ effect.

3. Concern that use of the centre will result indelays in entry to treatment.

The international (largely European) evidencesuggests that:

� the presence of injection rooms appears toreduce (but not eliminate) the visibility ofinjecting drug use,1124

� users see the rooms as a positive experience andwelfare and treatment access by users alsoappears to improve, with consequentimprovements in general health andfunctioning,1124–1126

� overdose rates and risks of fatality appear to beimproved in supervised injection centres. Todate, there has not been a single recorded deathat any injection centre worldwide.1124–1126 Giventhe sheer numbers of centres in operation (over45 in Europe alone), this is a significant feat. Italso seems that the rate of non-fatal overdoses islower than would normally be expected,

� whilst it is not possible to measure the impact ofsupervised injecting centres on rates of BBVtransmission, it appears that centres areassociated with substantially fewer riskyinjection practices.1124–1126

Despite these positive findings, the review literatureis clear that there have been few thorough impactevaluations on the European sites, and the majorityof the published literature does not appear inEnglish.1124 Evidence of outcome effectiveness is,therefore, not yet available.

The MSIC in Sydney’s Kings Cross was establishedon the recommendation of the NSW Wood RoyalCommission, and an 18 month trial is underway.Currently, only the 61122 and 121123 monthevaluation reports are available. The following is asummary of the key findings from both documents.

� Over the first 12 months, 2279 individualswere assessed and registered to use the centre,making a total of 31 675 visits.

� Most clients were male but female clients had ahigher average number of visits per client.

� For the first six months, cocaine was the mostcommonly injected drug (47%), followed byheroin (45%). At 12 months, heroin was themost common at 50% of visits, with cocaine at42% of visits.

� Clients made an average of 12 visits per 12months, although the range was wide at one to535 visits.

� After 12 months, one in 31 visits generated areferral to other health care services for further

��

�� !��"�#��$��%��

assistance. This is a total of 1007 referrals.Forty-four percent of these referrals were fordrug dependence treatment, 31% were forprimary health care services, and 25% were forwelfare services.

� One-third of client visits led to medical servicesother than supervision of injection, beingprovided to clients.

� Only a small percentage of visits (0.8%)resulted in drug-related clinical incidents. Theseincluded 184 heroin overdoses (resulting inzero fatalities), 50 cases of cocaine-relatedtoxicity, eight benzodiazepine overdoses, andeight non-heroin opioid overdoses.

� The six month report noted that all overdoseswere managed successfully, with no adverseconsequences.

� A total of 5958 sterile syringes were dispensedfor take-away use by clients at the six-monthpoint, in addition to the sterile equipmentprovided for each instance of injection.

The report that 184 heroin overdoses had occurredwith no fatalities attests to the efficacy of theapproach. Previous research has estimated thatunder normal Australian heroin usage conditions,one in 20 heroin overdoses is fatal.230 This wouldroughly translate to nine lives saved as a result ofthe supervised injecting centre.

In relation to the three concerns described above,the centre does not appear to have had anysignificant adverse impact on public order or publicamenity in the area. Evaluation of loitering andother criminal activity has been conducted by theNSW Bureau of Crime Statistics and Research. Theyreported that there has been no observable impacton crime rates, loitering rates, pedestrian trafficlevels, and there was no indication that the MSIChad any effect on either theft or violent acquisitiveoffences in Kings Cross Local Area Command basedon the COPS database from 1 May 2000 to 30December 2001.1127 It seems likely that MSIC hasenhanced access to treatment and welfare servicesby at least some its clients.

The final (18 month) report of the evaluation of theMSIC had not been completed at the time ofwriting, so that final judgements on its efficacycannot be made.

&('3'(�� -��-��


In Australia, the NHMRC recommendation ofuniversal vaccination to prevent hepatitis B has beenadopted but the uptake of vaccination by high-riskpopulations, such as injectors, is far from completefor a number of reasons. These include limitedknowledge about vaccine among injectors, and costto the recipient. Better means of making vaccineavailable to all injectors, including programs of freevaccine, are needed.1128

&('3'6��*+��-,�� -

Definition: Retractable syringes are intended to beused only once, after which the needle retracts intothe barrel of the syringe, preventing it from beingused for further injections, or, from causingneedlestick injury.

Summary: Evidence is contra-indicative .............. �

The Australian Government has recently announcedits intention to develop and introduce retractablesyringe technology, and has allocated 27.5 milliondollars in funding. The media release for the projectlists two main reasons for this proposal: publicconcern over needlestick injury from discardedsyringes in public places, and the risk of diseasetransmission through needlestick injuries in healthcare settings.1129

No syringe design can ever totally prevent sharingby injecting drug users. It is always possible to fullyload the syringe, and then administer partial dosesto two or more people.1130 This may actuallyincrease BBV risk, since conventional syringes can atleast be cleaned in between injections butretractable syringes cannot. A US research team hasconcluded that conventional needles will always besuperior to retractable needles in regard to reducingthe risk of BBV infection.1131 Retractable syringes arealso likely to be very expensive. There is a currentlysignificant lack of data in regard to feasibility ofusing retractable needles in Australia.

��'�A��.��!��0�/��0�-��!��

Harm reduction programs relating to the use of so-called ‘dance drugs’ have been of three main sorts:information campaigns aimed at users and in theform of guidelines for club owners and dance eventpromoters to help minimise environmental risks;employing trained outreach workers to provide

��"��

support and links to appropriate emergencyinterventions for those experiencing problems atdance events; and sites at events where users canhave their pills ‘tested’ or checked against a registerof previously tested pills and capsules.1132

&(';'&�� -


Information campaigns have focused on providingadvice about proper hydration, managing bodytemperature, and avoiding risky behaviours, such asunsafe sex or driving whilst intoxicated by dancedrugs. In Australia, these have included the RaveSafeproject, which has since become a generic name forprograms in this country aiming to promote safetyin the dance club or ‘rave’ environment.226

&(';'.�:)��+� �-�� -


In Australia, as internationally, various groups andorganisations have attempted to establish conductguidelines for dance parties or similar nightclubvenues.1133 These guidelines involve aspects such asadequate availability of cold drinking water, havingchill-out rooms, and hiring professional securitystaff. There is no direct evidence to say if theyreduce harm, or not. However, they are certainlyevidence-based in that acute drug-related harm inthis environment is often a function of overheating,and therefore ready availability of water and chill-out environments appears a sensible strategy.

&(';'0�� 5-��-� ��++-

Definition: Programs that aim to identify the contentsof ecstasy tablets and to provide this information tousers.


As discussed earlier, some of the harms associatedwith ecstasy use are caused by the fact that pillsoften contain a wide variety of psychoactivesubstances other than MDMA. Some of thesesubstances are inherently more harmful, such asPMA. Others, such as GHB and ketamine, require adifferent approach to manage and reduce risk.226 Inaddition, there is a distinct lack of quality Australianscientific data on the actual content of street-levelecstasy pills.1134 Various countries and organisationshave advocated or implemented various on-site pilltesting programs in an attempt to reduce the riskassociated with ecstasy use, and to inform theknowledge base on managing ecstasy-related risks.

On-site pill testing programs are used widelythroughout the European Union (EU).1135 Typically,these programs involve the provision of shortinformational sessions on safer use strategies whilepill testing takes place.1135

An argument often advanced against the provisionof timely pill testing data to users is that it gives animpression of safety to the consumption of MDMA,which it is held may lead to increasedconsumption.1134 There is no evidence to eithersupport or refute this statement.1135

The level of evaluation that has taken place in thevarious EU pill testing programs is generally verypoor, and little conclusive information is knownabout the effects of these programs.1135 There is aneed for more research and evaluation studies onthe entire range of effects of on-site pill testinginterventions.

The Australian Bureau of Criminal Intelligenceconducts some degree of off-site pill testing inAustralia, based on seized tablets,165, 168 but the dataare not made available to the research community.Adequate and timely access to these data wouldprovide further insight into the nature of ecstasymarkets in Australia and would allow far moreinformed research and assessment on the impact ofany future changes in legislation or supply controlapproaches regarding ecstasy.

There has been some media and community debatein Australia about the availability of home-testingkits for ecstasy pills. Typically, concerns have beenraised about the availability of such kits, eitherbeing perceived as condoning use or as possiblyincreasing use. There is no evidence availableallowing comment on the impact of availability oftesting kits on consumption levels.

��(�=��B ��B��

Harm reduction programs relating to the use ofbenzodiazepines have focused on the injecting oftemazepam gel capsules and the use offlunitrazepam (e.g. Rohypnol) in drug-assistedsexual assault. Temazepam gel capsules haverecently been moved to Schedule 8 of the NationalDrugs and Poisons Schedule, in order to reducetheir use by injectors.247 Flunitrazepam has alsobeen moved to Schedule 8 and there is evidencefrom the IDRS that its availability and usage byinjectors has decreased markedly since itsrescheduling, in 1998.1136

��

��# �� !��"�#��$��%��

&('<'&�",�-�� *� A��A�� -


In response to concerns about drinks being spikedwith drugs in nightclubs and bars, Rocheintroduced a blue dye to Rohypnol, which wasintended to make it more difficult to administer tosomeone’s drink without them noticing. However,Rohypnol has since been withdrawn from themarket, although it is believed that flunitrazepam isstill available under the brand name Hypnodorm(Malcolm Dobbins, Dept of Human Services,Victoria, personal communication). It is notpossible to comment on the effectiveness of addinga dye to the product, particularly in a poorly litvenue such as a nightclub. However, it is hard toimagine any disadvantage of adding such a dyeother than, perhaps, the extra cost to themanufacturer.

��)��

Until the phasing out of leaded petrol, harmreduction strategies also included the substitution ofunleaded for leaded petrol. When this wasintroduced in Maningrida, it led to a significantreduction in the number of hospital admissions dueto petrol sniffing.1137

��*�!�!��!��


Australia and Europe, in contrast to the US, havemore of a focus on harm minimisation strategies fordrug education. The systematic review by Whiteand Pitts did not include any evaluations of harmminimisation approaches (13 considered) becausestudies did not meet criteria for methodologicalrigour.654 The authors attribute this to approachesusually targeting hard-to-reach groups and,therefore, having problems with programimplementation and evaluation. However, a majorrecent Western Australian controlled study supportsthe view that school-based drug education based ona harm minimisation framework can be effective forreducing youth alcohol use, and related risk andharm associated with alcohol.155 In addition, Baerand others have demonstrated effectiveness of harmreduction approaches with US college age youth,753

and Newman and colleagues have shown that aharm reduction approach to drink-driving educationin year 9 was successful in increasing knowledgeand reducing the rate of riding with a drinkingdriver.1138 It did not, however, change alcoholconsumption. Many of the prevention programsfrom the US are based on an aim of abstinence andtherefore evaluate effectiveness in terms ofstatistically significant delay in onset of use. Furtherresearch is required, with long-term follow-up, toestablish whether programs can achieve a clinicallysignificant (as well as statistically significant)reduction in the harms associated with drug use.

��2 ��

Harm reduction is often thought of only as needleand syringe programs, but many more strategies areused and, in many cases, have been shown to beeffective. Programs with the strongest evidentialsupport are as follows:

� regulations to reduce passive smoking,

� random breath testing,

� needle and syringe distribution,

� treatment of opiate dependence to reduce risk ofoverdose and blood-borne viruses,

� hepatitis B vaccination.

By contrast, evidence does not support ‘light’cigarettes, staggered closing times and retractablesyringes, and these programs are contra-indicated.

��%��

��

��' �� !��"�#��$��%��

��(��

�� "��!!��

This chapter presents an overview of policy and future investment implications for theprevention of early developmental problems, early age drug use, risky use and harm. A Protectionand Risk Reduction Approach to Prevention is presented, acknowledging that broad socialdeterminants influence drug use. The approach incorporates the developmental risk andprotective factors conceptualisation of youth problem behaviours, and the brief intervention andharm reduction frameworks used with adults. An approach to prevention is outlined,emphasising the local community as an important setting for ensuring the coordinateddevelopment and delivery of the Protection and Risk Reduction Approach to Prevention acrossdifferent policy jurisdictions, and spanning appropriate periods of time. The emphasis on thelocal community flows from the requirement to tailor prevention strategies to varying localconditions, the emerging success of community approaches and the attraction of enhancingcommunity in order to address growing social disconnection.

An examination of patterns of drug use and harms reveals the importance of continuing efforts toreduce tobacco use. The need to increase investment relevant to alcohol is also identified, withescalating levels of youth alcohol use of particular concern. It is noted that illicit drug use isresponsible for disproportionate harms in some populations.

Continued and enhanced investment is recommended in the following four broad areas, withmaximum benefit obtained from the simultaneous delivery of complementary strategies.

Universal interventions to prevent tobacco use and risky alcohol use: legal drugs generate the great bulk ofhealth, economic and social drug problems in contemporary Australia. The bulk of problems arefound within mainstream society among persons with average levels of developmental risk. Earlyuse of legal drugs predicts later problematic alcohol and other drug use, as well as mental healthproblems. Parental and community role models encourage use among children and adolescents,suggesting the requirement for ‘whole population strategies’ to address overall levels of use andto break inter-generational patterns.

Universal interventions to reduce the supply of, and demand for, licit and illicit drugs: law enforcement strategiesare necessary to protect the community against the crime and social disorders that flow from theuse of prohibited drugs. Law enforcement plays a critical role in prevention by: reinforcingcommunity values against illicit drug use, controlling the supply of licit drugs, and divertingearly offenders to preventive interventions.

Targeted interventions to address vulnerable and disadvantaged groups, with particular attention to Indigenous Australians:interventions should provide evidence-based support to families at the key developmental stagesof infancy, pre-primary and primary school. The interventions have the potential to address thebulk of harms associated with the use of illicit drugs as well as a significant proportion ofproblems with legal drugs.

Treatment, brief intervention and harm reduction approaches for adolescents and adults with emerging, or developed, riskydrug use patterns: investment in proven treatment methods, whether abstinence-oriented or harm-reducing, reduce drug-related harm at the population level. Brief screening interventions have anuntapped potential for widespread application in primary health care and community settings.Family members, particularly children, need to be involved in treatment programs to help breakinter-generational patterns of substance use and related harm.

��

��) �� !��"�#��$��%��

�"��+��

In this final chapter, we describe what we have beenable to glean from our readings on drug-relatedharm. The overarching aim of this document is toprovide the evidence basis for a comprehensivenational agenda for the prevention of problemsrelating to drug use.

The evidence we have reviewed relates to:

� the nature and extent of the major drug-relatedharms,

� the underlying patterns of drug use,

� the social and developmental antecedents ofthese patterns, as well as for other problembehaviours, and

� the evidence base for interventions to reducerisk and increase protection, throughout the lifecourse.

In what follows, we summarise and synthesise whatis known in each of the above areas in order toframe conclusions and consider future directions. Anumber of unifying principles and concepts relevantto the Protection and Risk Reduction Approach toPrevention were outlined in the introduction ofChapter 1. The first implication flowing from ourreview is that there are opportunities for a morecoordinated national prevention agenda, throughthe adoption of an integrative policy framework.This approach integrates knowledge ofdevelopmental processes throughout the life course,with knowledge of broader macro-social andenvironmental influences on behaviour and healthoutcomes. The Protection and Risk ReductionApproach emphasises the importance of reducingthe known developmental risk factors that leadchildren and young people to become involvedwith risky drug use and harm, while also enhancingprotective factors. The framework acknowledgesthat targeted, early intervention strategies focusedon strengthening protective factors will be usefulfor children and youth with a high number ofdevelopmental risk factors. The approach shouldalso emphasise brief interventions and treatmentand harm reduction strategies, acknowledging thatsuch strategies can reduce drug-related harm fordrug users that have a high number of risk factors,while also improving developmental opportunitiesfor children. Effective regulation and lawenforcement should be considered as essentialelements of this approach to prevention, not just incontrolling the supply of drugs, but also ininfluencing community values about drug use,diverting early offenders and acting to protect thecommunity from crime and social disorder.

The favoured approaches to prevention are based onan examination of both early risk factors (e.g. socialand early developmental risks) and late risks relatingto patterns and circumstances of drug use. Withregard to late risk factors, we have identified somebasic concepts regarding the pharmacological andpsychological effects of different drugs, their dose-response effects, toxicity, associated behaviouralchanges and drug dependence. All of the harmsrelated to drug use can be attributed to: theinterplay between these psychobiological processes,individual behaviour, drug use settings and broadsocial influences. A national prevention agendaneeds to seek opportunities across many sectors toreduce both distal and proximal risks for drug-related harm.

The second implication of our overview leads us toemphasise a systems approach to drug preventionwhich acknowledges the:

� many levels of society in which there areinfluences on patterns of drug use and harm,

� many levels in which interventions may bedelivered, and

� importance of consistency across diverse levelsand sub-systems in terms of influences andinterventions.

There are many opportunities for synergies acrossdiverse areas of action. For example, investment intreatment and brief interventions is likely todiminish crime, road trauma, disease and otherserious harms associated with drug use. Lawenforcement practices can affect entry to treatmentfacilities. Judicial processes have the capacity todivert illicit drug users, at an early stage in theirdrug using career, into effective interventionprograms. There are also issues around attemptingto reconcile inconsistent influences on patterns ofdrug use, such as the powerful role modelling ofsmoking, drinking and sometimes drug useprovided by the media, as opposed to the lessfrequent mass media campaigns to deter risky druguse. It is important to review the manyopportunities for reducing risk and increasingprotection for individuals, families andcommunities across all levels and sub-systems ofmodern society.

To better organise the systems approach toprevention, we have emphasised the localcommunity as one of the primary levels forintegrating and coordinating planning within aProtection and Risk Reduction Approach toPrevention. An emphasis on the local communityoffers prospects for addressing some of the broad

��

social determinants related to both socialdisadvantage and disconnection that underlie aspectsof drug-related harm. A structured approach to localcommunity organisation also offers a promisingmethod for the coordinated application of evidence-based prevention strategies aimed at the reductionof developmental risk factors and enhancement ofprotective factors. By emphasising the communitylevel, the implications of the systems model for acoordinated approach to prevention across differentjurisdictions become clearer. Policy and researchopportunities flowing from the evidence reviewedwere summarised for each of several sub-systemsand jurisdictions within which interventions can beimplemented. It is critical that national and statepolicies, legislative models and regulations arebased on evidence about drug-related risk andharm, and enable and empower local communitiesto develop effective prevention strategies.

Prior to considering the range of opportunities forintervention and prevention, it is important to havea clear picture of: a) the overall patterns of drug-related harm as they currently impact onAustralians; and b) the overall patterns of risky druguse that underlie these harms.

�"�� !��1��2��

It is easier to quantify the health, safety andeconomic costs of drug use than the social and legalcosts. An overview was presented of key indices ofhealth and economic costs, by drug type and broadage groups, using most recent available data.

The key points revealed by this overview are asfollows.

� Tobacco is the leading cause of premature deathand hospitalisation among Australians. Most(77.8%) of tobacco-caused deaths involvepersons over 64 years of age. However,tobacco-caused deaths involving children andadults up to 64 years of age are still greater intotal than all deaths caused by alcohol and illicitdrugs combined for all age groups. Theeconomic costs of tobacco reflect this fact.

� Alcohol causes the deaths and hospitalisation ofslightly more children and young people thando all the illicit drugs combined and many morethan does tobacco. These deaths are almost allcaused by either intentional or unintentionalinjuries. While alcohol is also responsible forthe deaths of many more adults and elderlypeople than are the illicit drugs, there are amuch larger number of deaths believed to be

saved among older people as a result of, mainly,low-risk alcohol consumption, principallyamong older women.

� The estimates of economic costs are updated for1998 and demonstrate an importantcontribution to the costs of illicit drug use fromlaw enforcement and crime.

It is important to note that there are also likelyfuture health costs associated with current drug usethat are hard to estimate. However, it is inevitablethat given the wide prevalence of hepatitis C amonginjecting drug users in Australia, there will besubstantial mortality, morbidity and associatedeconomic costs as a result of higher incidence ofliver disease in this group, in future years. Inaddition, there are known to be social harmsimpacting on individual users of illicit drugs whoreceive criminal convictions.

�"�� -��1��2��

The patterns of risky drug use underlying the aboveharms can be classified into those caused by adverseinfluences on: the development of children andyouth (e.g. through role-modelling, exposure topassive smoking, impairment to parenting), earlyage use (resulting in developmental problems),regular use (e.g. liver disease and cancers),intoxication (e.g. injuries, overdoses andpoisonings), unsafe means of administration (e.g.BBVs from dirty needles, septicaemia), anddependence (e.g. withdrawal symptoms, treatmentcosts). Some types of harm straddle these categories,such as suicide and some strokes associated bothwith intoxication and regular use of alcohol, inparticular. Social harms, such as loss of employmentand impairment of personal relationships, can alsobe a function of the combined effects of a long-termpattern of heavy use, frequent episodes ofintoxication, and time and resources expendedacquiring drugs.

The main features of risky drug use patterns inAustralia are as follows.

� There has been a dramatic reduction in levels ofsmoking in Australia in recent decades, suchthat rates of daily adult smoking fell below 20%for the first time in the 2001 NDSHS.22 Smokingrates by young people, and young women inparticular, have been less resistant to change inrecent years and are a concern for future levelsof tobacco-caused mortality and morbidity.

��

�� !��"�#��$��%��

� Alcohol consumption in Australia has recentlyincreased slightly overall, and more markedlyamong young people. Two-thirds of PersonYears of Life Lost through risky alcohol use aredue, at least in part, to the short-term or acuteeffects from alcohol intoxication. A veryconservative estimate from the 1998 NDSHS isthat 67% of all alcohol consumed in Australia isdone so in a manner inconsistent with the latestNHMRC National Alcohol Guidelines.42 Foryoung adults that figure is 90%.

� Cannabis is the most widely used illicit drug inAustralia, though its use may have declined veryrecently. In most jurisdictions, any use at all ofcannabis poses the risk of receiving a criminalrecord. Around 10% of people become regularheavy users of cannabis and risk long-termhealth consequences and dependence.Combining cannabis with depressant drugs,such as alcohol, appears to interact to pose extrashort-term safety risks (e.g. before driving).Cannabis use during adolescence is associatedwith later mental health and conduct problems,though the causal processes remain unclear.

� Injection is the main risk behaviour in relationto health-related harms from other illicit drugs.The 2001 NDSHS indicates that less than 2% ofthe adult population report injecting illicit drugsat some time in the last 12 months.22 Injectionof opiates delivers a usually unknown quantityof the drug rapidly and poses a risk of overdose,especially if other CNS depressant drugs(alcohol and benzodiazepines) have also beenconsumed. Sharing of injecting equipment andassociated paraphernalia is a major risk factorfor the spread of BBVs.

� Heavy ‘binges’ on amphetamine-type drugs areassociated with reckless and aggressivebehaviour and, when sustained over days, mayprecipitate a psychosis. The risks of use of otherillicit drugs, such as phenethylamines, arepoorly understood.

� Finally, there are marked temporal anddevelopmental sequences concerning the agesof first use and the order of onset of use ofdrugs. It is apparent that early use of tobaccoand alcohol is predictive of later problems withtobacco dependence, alcohol and illicit drugs. Itis also clear from longitudinal research that useof alcohol and tobacco at an early age predictsprogression to heavier drug use, even afteradjusting for the influence of a range of knowndevelopmental risk factors. The mechanisms bywhich legal drugs serve as ‘gateways’ in some

sense for illegal drugs are not clear. Adolescentuse of cannabis significantly increases the risk oflater use of other illicit drugs, but nonetheless,only around 10% of cannabis users progress touse other illicit drugs.

�"�"�� !�� 3��!��2��"�

Patterns of drug use and related harms are notdistributed randomly across the population; thereare defined groups in contemporary Australia thatare over-represented in the statistics. Given thesubstantial contribution of tobacco, alcohol andillicit drug use to the total ‘burden of disease’ inAustralia, and worldwide, it is not surprising thatthese groups are usually also those that are over-represented in statistics on general ill-health. Firstly,across all drug types, being male and being youngare each independently highly predictive ofinvolvement in risky drug use and harm. Secondly,almost any measure of disadvantage will besimilarly associated with increased risk and harmfrom drugs, regardless of gender and age. Inrelation to legal drugs, however, there is evidenceof a U-shaped relationship such that, for example,both low and high income can be predictive ofgreater consumption and related harm. It is likely,however, that there are different underlyingpatterns, such as less frequent but higher intakedrinking associated with higher rates of acutealcohol-related harm among disadvantaged groups.This latter pattern of drinking and related harm ismost clearly expressed among Indigenous Australianpopulations who also have very high rates ofsmoking and a host of other health risk behaviours.The association of drug use and measures of socialdisadvantage is strongest for the illicit drugs versusthe licit, and also for more intensely problematicpatterns of drug use, including dependence.Addressing social disadvantage in all its forms hascome to be seen as a central issue for modern drugpolicy to address.

Related to findings of social disadvantage, there areindications that social disconnection is increasinglya modern driver underlying drug-related harm.Family breakdown, loss of community, increasingmobility and weakened religious institutions arestructural determinants undermining social stabilitythat have been identified as developmental riskfactors for drug-related harm. The emerging callswithin mental health promotion for a focus onsocial and community wellbeing are pertinent toefforts to address these more pervasive andinsidious determinants of drug-related harm.

��

�"�#�,��-�� $�� 2��#�

The social determinants perspective on health andhealth risk behaviours is one traditional approachinforming an important emerging perspective onthe genesis of youth drug use. Comprehensivereviews of longitudinal and other studies examiningsignificant influences on the drug use of youngpeople have identified factors and variables such asfamily functioning, school performance, peerinfluences, temperament and local drug availabilityas predictive of who will use drugs. This literaturewas discussed in depth in Chapter 6. The variableshighlighted above were combined to form overallsurvey measures of risk for, and protection against,a variety of problem behaviours, including druguse. Children with high scores on the risk scale andlow scores on the protection scale are more likely todrink in a risky fashion, smoke, use illicit drugs,experience mental health problems and exhibitconduct disorders. It is noteworthy that someelements of risk include adult modelling of drug useand the extent of community availability of licit andillicit drugs.

Evidence suggests that both early initiation andfrequency of youth drug use are most clearlypredicted by the cumulative number of elevated riskfactors, rather than by any specific risk factor. Foryouth on trajectories characterised by a highnumber of risk factors, there is evidence thatelevating the number of protective factors canreduce the likelihood of a range of adverseoutcomes. This knowledge emphasises theimportance of prevention activities and strategiesfocused on reducing risk factors and enhancingprotective factors within specific communities.When the examination of protective factors isextended beyond adolescence into adulthood, itbecomes apparent that many harm reductionstrategies are in concert with this communityemphasis on enhancing protective factors for thosewith a high number of risk factors.

�"�%�8�� 5��$�� $��!��9�,��-0�� $�� :��2��#�

The economic consideration of preventioninvestment leads to a search for methodology toassess the optimal way to combine approaches toprevention. To date, little economic evidence hasbeen forthcoming to ensure that an ‘efficient’ use ofsociety’s resources is being made across the drug

policy arena. Much evidence still rests on theindividual impact and effectiveness of programs,without specifying the resources needed to bringthis about. Most evaluations still do not focus on thepersonal and social resources required and thispartial approach leads to a potential waste of scarceresources. Within each section of this report, wehave presented data relevant to cost-benefits, wherethis was available.

The report has focused particularly on costing theeconomic benefits accruing through the impact ofprograms and strategies in reducing harmful druguse. It is important to reiterate that such costings arelikely to be conservative, particularly in the area ofdevelopmental prevention. This is because there areadditional social benefits beyond reduction ofharmful drug use that can be expected to flow fromprograms that more generally improve thedevelopmental opportunities of children and theircommunity social environments. For example, thereduction of a risk factor such as academic failure islikely to lead to greater completion of high school,increased attendance at college and greater jobopportunities, all of which can be costed as benefitsof early school-based prevention efforts. Likewise,pre- and postnatal home visits by public/community health nurses not only reduce maternalsubstance use and arrest rates, of the mother andeventually the child, but also reduce rates ofsubstantiated child abuse and neglect that representadditional cost savings of this approach. Acomprehensive costing of the benefits of apreventative approach is beyond the scope of thepresent report but more fully accounting benefitsshould be a longer-term goal in preventionplanning.

The present report has provided importantinformation that is relevant to the targeting ofprevention investment through further analysis of amajor Australian data set on risk, protection andadolescent problem behaviours. The analysisexamined the extent to which drug preventionpolicy should be universal in its application, or targetedto high-risk populations. It is clear that for regularsmoking and heavy alcohol use, most adolescentsexhibiting these behaviours by age 16 were eitheraverage or low on a cumulative index that summedrisk and protective factors. For weekly cannabis use,most youth reporting this behaviour were classifiedas high-risk, for all age groups. The trend in thedata was for the number of low and average-riskchildren to be weekly cannabis smokers to increasewith increasing age, suggesting that if age hadincreased further, they would have been in themajority. By contrast, for the other illicit drugs

��

�� !��"�#��$��%��

combined, there was a clear tendency for weeklyusers to be predominantly high-risk children, at allages. In summary, the prevention paradox holds forthe legal drugs, and may hold for older teenagers inrelation to cannabis, but does not hold for otherillicit drugs.

The above evidence suggests that whole-of-population, or universal strategies, are of particularimportance in relation to reducing the moreprevalent harms associated with tobacco and alcoholuse. However, strategies targeted to high-riskchildren and adolescents may be necessary inpreventing the harms associated with illicit druguse. As high-risk youth generally have high levels ofdrug use, the more targeted strategies will alsobenefit the prevention of harms associated withlegal drugs and cannabis. To maximise theireffectiveness, targeted strategies should be initiatedearly in the developmental pathway and aim toreduce risk factors, enhance protective factors andprevent or delay drug use. Where drug use isevident, interventions should continue through intoprotective strategies, aiming to contain theemergence of risky patterns of drug use or to reducethe likelihood of harm.

�"�'��$��3�� $�� 1,��

Evidence was summarised for the relativeeffectiveness of interventions and policies, frompre-conception through to prenatal care, antenatalcare, infancy, pre-primary, primary school,adolescence, young adulthood, adulthood and oldage. The quality of the research was highly variable.The types of outcomes examined ranged from:known risk factors for later drug use, age of onset ofuse of different drugs, intensity of drug use, todependence and experiences of problems relating todrug use. In some areas, it was possible torecommend wide implementation with confidence,in others there was theoretical support forrecommending that interventions be trialled, and inyet others there was no relevant literature uponwhich to draw information. In well-researchedareas, such as school-based drug education andcommunity action, it is possible to identifyprinciples to guide effective practice as well asdescriptions of model projects. In other areas, it waspossible to identify individual strategies that, ifimplemented, would definitely have preventivebenefit. The evidence base will now be brieflysummarised for each of the major areas examined.

&6';'&�� -�� +,�+��5-�� -�8 ��39

Information summarised in this report suggests arange of opportunities for encouraging healthychild development, and thereby preventingchildren’s drug use and progression to heavy andharmful use. Prior to birth, and also in childhood,the healthy development of children can beimpaired through parental tobacco, alcohol andillicit drug use. Further innovation investment willbe required to develop and evaluate health servicereorientation programs that can be effectivelyapplied to address these problems.

Programs of structured home visits to supportmothers, before and in the first two years afterbirth, have evidence supporting their effectivenesswhen targeted to vulnerable families. Theseprograms offer basic advice, practical assistance withnursing, and advocacy for access to services. Theyshow evidence of positive outcomes for maternaldrug use and infant health. Some programs targetedspecifically at vulnerable (e.g. drug using) mothershave shown benefits for later problem behaviours inchildhood and adolescence, including some modestreductions in drug use. A promising yet currentlyunder-utilised approach involves targeting parenteducation and support within drug treatmentpopulations. A coordinated investment to ensurehealthy child development in drug treatmentsettings is warranted.

There is evidence for positive outcomes (schooladjustment and academic attainment) and goodcost-benefit ratios from targeted programs toprepare children from high-risk families for primaryschool. There is also substantial evidence for thevalue of both universal and selective parentingprograms for pre-primary school age children onsimilar outcome variables.

Strong evidence exists for both universal andtargeted programs to support parents and strengthenfamilies of primary school children. Outcomes fromthese studies have been documented by followingchildren through to adolescence. Results have foundreduced smoking and alcohol use as well asreductions in other adolescent problem behaviours.

��"��

&6';'.�� -�� +�-�� -�8 ��;9

There is now an emerging evidence base foruniversal interventions focusing on adolescents’ useof alcohol and tobacco. A combination of well-designed and executed regulatory approachessupported by other components, such as school-based interventions, holds the most promise. Astrong research tradition relates to the evaluation ofclassroom-based school drug education programs.While once regarded as an area with little promiseand a weak evidence base, more recent programsdeveloped in Australia and the US have shownpositive outcomes, particularly in terms ofreductions in tobacco use. An Australian study hasdemonstrated reductions in risky alcohol use andalcohol-related harms. A clear set of principles hasbeen developed, based on the internationalliterature, for application in Australian schoolcurricula.

Available evidence revealed very strong support anda sound rationale for the enforcement of lawsprohibiting sales of tobacco and alcohol to personsunder legal purchasing age for these legal drugs.Similarly, there is evidence for the effectiveness ofmeasures that control the price of alcoholic drinksfavoured by young people.

There is evidence encouraging the furtherevaluation of parent education approaches and forfurther application and evaluation of familyinterventions to address illicit drug use. Someevidence provides support for the further evaluationof community mobilisation/community actionapproaches, social marketing and health servicereorientation; and, in particular, of briefintervention approaches for heavy alcohol users.There is more limited evidence, but strongtheoretical grounds, for further evaluating schoolorganisation and policy approaches to minimisedrug use and harm and encourage alternative drug-free recreational pursuits.

&6';'0�� -�� *��5*�-�� -�� +�*� ��-��--��)+��+��*+��-�8 ��<9

There are at least three main ways in whichcommon benefits can be obtained through broad-based preventive interventions addressing a widerange of health, social and criminal problembehaviours.

Firstly, there are benefits associated with universalprograms to reduce or eliminate the social and

developmental risk factors that predict thedevelopment of problem behaviours. For example,broad social and economic policies that seek toimprove conditions for the healthy development ofchildren and youth, reduce disadvantage, increaseequity, and strengthen community will have a rangeof benefits including lower rates of use of bothillicit and licit drugs and of the attendantconsequences. There is a special urgency in relationto addressing these issues as they affect IndigenousAustralians.

Secondly, benefits can be obtained throughprograms that target individuals and groups with ahigh number of developmental risk factors. Settingsinclude disadvantaged areas, family crisis, policeand court contacts, and mental health. Strategies toensure ready access to services and to reduce riskand enhance protective factors in children, familiesand communities, are warranted in these settings.As mentioned above, there is particular promisefrom the wider application of targeted programs toprepare disadvantaged children for school and toenable parents to manage behavioural problems inprimary school age children. As these programshave the potential to reduce risk factors that predictthe emergence of harmful drug use, investmentaimed at reducing drug-related harm should form acomponent in these investments.

Thirdly, benefits are available through programs foradolescents and adults that have high rates of drug-related harm. Programs for these populations focuson immediate protective strategies and maycoordinate to address physical and mental healthproblems, including levels of drug use. Examplesinclude broad-based health promotion interventionsdelivered by primary care health professionals suchas general practitioners, occupational healthworkers; community-wide health screening; andbrief intervention programs. The potential publichealth benefits for the broad application ofscreening and brief intervention programs,targeting a range of health risk behaviours, has notyet been realised in Australia. Programs such as theSNAP initiative are under way to encourage greaterdissemination.

Our investigation of broad-based preventionsuggested important commonalities between effortsto prevent drug-related harm and other preventionefforts relevant to crime, mental health and physicalhealth. Planning mechanisms that encouragecombined prevention investments fromjurisdictions concerned with mental health, crimeand health would appear warranted, supportingcommunity efforts to prevent youth drug use.

��

��# �� !��"�#��$��%��

&6';'(�� -�� ) ��-�+��)+��,� �� -��+��+��)�-��--��++��)�-�8 ��&&9

Regulation of the supply of both tobacco andalcohol products, supported by a range of publiceducation measures, is strongly supported in theresearch literature. Both tobacco and alcohol behavelike other products in the market place in that thelevel of demand is strongly influenced by price.Young people, as well as heavy drinkers andsmokers, are most affected by price increases. Atpresent there are consistent taxation policies in placeto maintain the high price of cigarettes in Australiathat should be sustained. For alcohol, from a publichealth point of view there are sound policies inplace in relation to beer and spirits, in that lowertaxes are paid on lower alcohol content drinks(through excise taxes). However, there is aweakness in current policy with the absence of analcohol content-based tax (excise or other) onwines. This results in the availability of very cheapbulk wines that are favoured by vulnerable groupsand problem drinkers.

The use of hypothecated taxes from tobacco andalcohol has a significant tradition in Australia, witha strong rationale and degree of supportingevidence of effectiveness. While a proportion ofbeer taxes collected in the year 2000/01 washypothecated to form the Alcohol Education andRehabilitation Foundation, there is no continuinghypothecation of either tobacco or alcohol taxes atthe national level; such taxes are no longerpermitted by law at the State/Territory level.

Restrictions of sales of both alcohol and tobacco tominors can be effectively enforced. There iscommunity support for strict enforcement of theselaws but also evidence, especially in relation toalcohol, that underage youth access is relativelyeasy. There is also evidence that enforcing laws onserving intoxicated customers can be achievedeffectively. In both cases, law enforcementapproaches, with or without education, are moreeffective than education alone (e.g. throughresponsible service training). Enforcement of liquorlaws is a prerequisite of effective management oflicensed establishments, with Accords andresponsible service training programs providing asupporting role.

Physical availability of alcohol, in terms of numbersof outlets and hours of sale, has increased inAustralia over the last decade. Australian andoverseas evidence clearly identifies late nighttrading for hotels and nightclubs as a source of

alcohol-related violence and road trauma. There isalso evidence to support controls on outlet density.No operational model for achieving this has beendeveloped and there are currently inadequateinformation systems in some licensing authorities toenable identification of high-risk premises, so as todetermine whether they should continue trading.

The development and enforcement of laws topunish and deter drink-driving in Australia havebeen major successes for public health and safety,with uniform laws in place across Australia.Continued implementation and monitoring forquality control is of great importance.

&6';'6�� -�� ) ��-�+� �� -��+��+��)�-��--��++��)�-�8 ��&&9

There is strong evidence that includes Cochranereviews, to show that public education campaignscan contribute to reductions in smoking and riskyalcohol use, especially if campaigns support otherpolicy measures such as tax increases and lawenforcement. Education about some aspects oftobacco and alcohol use can also be seen as aprerequisite for a range of other system-wideinterventions and, as such, are hard to evaluate asisolated strategies. National Alcohol Guidelines areone such example. Advice about low-risk drinkinglevels and various health risks are fundamental tothe delivery of educational messages: through themedia, by a range of health professionals inprimary, secondary and tertiary care, and in a rangeof health and community settings. Widedissemination, therefore, supports other evidence-based activities. Australia introduced standard drinklabelling as one dissemination strategy to supportdrinkers to apply national drinking guidelines,though the current labels need only be 1.5mmhigh. Labelling is an efficient means of deliveringhealth messages to smokers and drinkers, but it ishard to evaluate as an isolated strategy. The extent towhich labelling supports, and could support otherevidence-based interventions should beinvestigated. Similarly, media campaigns need to bedeveloped to facilitate other national strategies,regulatory interventions and core health messages.

There is a developing evidence base, mainly fromoverseas studies, that the community is an effectivelocation for organising and delivering preventionmeasures targeted at legal drugs, especially alcohol.This tradition has matured to the extent that a set ofguiding principles for sound process, optimalcontent and good outcomes can be distilled. Theweight of published evidence suggests that

��%��

community-based interventions that targetstructural policy change at the local level are moreeffective than approaches with the less focused aimof community mobilisation. Thus, communityaction to: restrict trading hours in high-riskcommunities, increase enforcement of drink-driving and liquor laws, and restrict local alcoholavailability, are reported to have achieved the mostpositive results. This is also one of the few areas ofdemonstrated benefit for interventions withinIndigenous communities. Programs such asCommunities That Care are being implemented inAustralia and combine elements of communitymobilisation and structured community action. Bysupporting local coalitions to tailor evidence-basedprevention strategies to local conditions, theseprograms hold the promise of encouraging a well-coordinated selection of prevention strategies.

&6';'7�� -�� ) ��-�+��)+��,��+��-�� ++��)�-�8 ��-&.�� &09

Laws shape community values and opinions aboutdrug use. On the one hand, laws express socialdisapproval that reinforces social norms againstillicit drug use and, on the other hand, they act as adeterrent against use. The role of law enforcementin prevention thus goes beyond apprehending drugusers and dealers. The current Diversion Initiativedemonstrates the importance of the lawenforcement role in the apprehension of early usersand referral to education, treatment or/and support.

The impact of laws prohibiting the sale, supply anduse of certain drugs is very hard to ascertain fromcurrent scientific evidence. These laws and styles ofenforcement have evolved in many countries,mostly over the past 100 years, in response to theincreasing availability of a range of psychotropicsubstances for recreational and non-medical use.When use of a substance is illegal, there aresubstantial barriers to overcome in order to obtainreliable and valid data about patterns of use.Furthermore, when prevalence of use is quite low,as is the case for opiate drugs, it becomes expensiveto obtain accurate prevalence estimates fromnational surveys. A recent high-level inquiry,conducted by the US Academy of Sciences176 intothe scientific status of US illicit drug policy, couldidentify very little rigorous research data on patternsof use, impact of prevention programs, drug puritylevels, drug-dealing and law enforcement. Thereport noted that most of what is ‘known’ about theprevention of illicit drug use and harm is from thebetter documented experiences with legal drugs.

Economic modelling from black markets in othercommodities (tea, coffee and tobacco) suggests thatrendering a substance illegal results in substantialincreases in its price. A major investment inimproved research is emphasised in the presentreport as a method of improving the futureevidence base for illicit drugs policy.

The available Australian research has not been ableto identify measurable impacts of law enforcementstrategies on levels of drug purity or use, but this isnot to say that these effects do not occur. There isevidence that contact with law enforcement canencourage participation in treatment but it may alsoincrease risky drug use practices.

While the published evidence regarding street-leveldrug law enforcement is weak, there is support forthe notion that drug laws may provide a generaldegree of deterrence to the population who are notengaged in drug use. However, as a consequence ofthe difficulty of preventing more than a smallminority of drug deals, there is little evidence forspecific deterrence (i.e. deterring drug users whocome into contact with the law from re-offending).

There is a small literature evaluating the impact ofchanging the precise legal status of cannabis,including some important Australian studies andreviews. These indicate that moving from criminalto civil penalties for use and possession of smallquantities of cannabis is not associated withsignificant increases in prevalence of its use. Alongwith other European countries, Australian Stateshave variously moved to ‘soften’ policing ofcannabis laws, in some instances introducingcautioning schemes for first time offenders (e.g.Victoria), in others by decriminalised possession ofsmall quantities (SA and ACT). WA proposes tointroduce decriminalisation, for which there will bea rigorous evaluation of intended and unintendedconsequences using baseline data already collected.

In relation to the effectiveness of community-basedand educational initiatives in the mass media and inschools regarding illicit drugs, as noted by therecent US Academy of Sciences report, most of whatis known has been applied from research focusingon legal drugs. To increase the chances of success,broad principles can be applied from the areas ofcommunity action, mass media campaigns andschool-based programs, and these have beenoutlined in earlier chapters. However, when itcomes to the issue of the component parts of theseprograms, there is a dearth of scientific data.

��

��' �� !��"�#��$��%��

&6';'3�� -�� *�� -��)+-��)��*+��-�8 ��&=9

There are important intersections between the aimof population-level prevention of drug-related harmand what has traditionally been considered to betreatment. Emerging evidence suggests thatinvestment in various forms of treatment will havebenefits in terms of community-level reductions incrime, road trauma, hospital admissions and otherserious drug-related harms.

There is an enormous potential to enhance thesesavings at the community level, in three main ways:

1. expanding brief intervention programs targetingsmoking and risky drinking, to a wide range ofprimary health care, workplace and othercommunity-based settings,

2. ensuring that treatment programs offeredinclude approaches with the strongest evidence-base and are made widely accessible, perhapsthrough a greater emphasis on delivery at thecommunity- rather than institutional-level,

3. incorporating interventions to support childrenin families with drug-using parents, in order toattempt to break inter-generational patterns oftransmission of problem drug use.

This last point is important for prevention.Smoking, risky drinking and drug use by parentsand other significant adults is associated withincreasing developmental risk factors for children.This means that there are important yet currentlyneglected opportunities to target preventionprograms in drug treatment settings (Chapter 8).

If drug treatment programs for adults and familiesare to realise their potential for preventative benefitsfor children, then they should be designed withprevention in mind. Treatment activities shouldinclude interventions to reduce developmental riskfactors and enhance protective factors, and a morecareful analysis of treatment impacts on childrenshould be encouraged in evaluations.

�"�(�+�� $��

There have been few formal evaluations ofIndigenous intervention projects—a facthighlighted in various reviews summarised earlier.The majority of available evaluations have examinedevidence for impacts on patterns of drug use andrelated harms, but not on fundamental social andenvironmental conditions that contribute to theseproblems. However, the evaluations that have beenundertaken have reached a number of similarconclusions about what makes an interventioneffective.233, 869, 988, 989, 1139 First and foremost amongthese is the support of, and control by, localcommunities. A prime example relates to theeffectiveness of community-initiated localrestrictions on alcohol availability. Given thediversity within the Indigenous population,interventions must be tailored to the needs ofparticular communities. Importantly, interventionsneed to be adequately resourced and supported.This entails not only funding for project activities,but providing appropriate staff training and support.

In the past 10 years or so, there have also been anumber of reports and submissions that includerecommendations to address substance misuseamong Indigenous Australians.292, 331, 1140, 1141 Therecommendations contain a wealth of detail that canusefully form the basis for any strategy to preventsubstance misuse and related harm amongIndigenous Australians. While manyrecommendations remain to be implemented,several key themes emerge. The first is the need toaddress the underlying social determinants ofIndigenous inequality. This includes the call forreal, but appropriate, economic development forIndigenous people. This call is not new but,recently, has been forcefully made by NoelPearson.294

The second theme emerging is the need forIndigenous people to be involved, as equal partners,at all stages in the development and implementationof strategies to address substance misuse. There isevidence from both Australia and overseas of theefficacy of Indigenous ownership and control ofinterventions addressing ill-health—this includesthe importance of giving preferential employmentin government agencies providing services forIndigenous peoples.1142

As important as Indigenous involvement is, it isinsufficient without adequate resourcing—the thirdtheme to emerge from recommendations to addresssubstance misuse. The Commonwealth Government

��(��

has asserted that there has been an increase in fundsexpended on Indigenous affairs over the past threedecades. However, a substantial amount of theexpenditure identified as specifically directed forIndigenous welfare should arguably have been theroutine responsibility of other Governmentdepartments. In other words, there has been amisleading impression of Indigenous peoplereceiving extra funding over and above thatreceived by other Australians.1143 Whatever thelevels of funding for Indigenous affairs over the pastthree decades, the expenditure has failed to meet thewell-documented needs or to remedy the social andeconomic inequalities that underlie and perpetuatethe high levels of substance misuse amongIndigenous Australians. This inadequacy of fundinghas also been identified with regard to funding forboth Indigenous health services1144, 1145 andIndigenous substance misuse services.859, 869

An important component of adequate resourcing isthe building of capacity to continue to provideadequate and appropriate services withincommunities and community organisations. Thisincludes infra-structural development, researchcapabilities, and staff development and support.With regard to the latter, despite several projectevaluations highlighting the shortage of staff withadequate training,869, 1146 in 1999/2000, less than3% of funds for projects directly targetingIndigenous substance misuse was allocated to staffdevelopment and support.859

Over the past three decades, governmentIndigenous health and substance misuse policy hasacknowledged a link between substance misuse andunderlying social issues. However, despite thisacknowledgement, substance misuse policy andservice planning has largely been developed inisolation from policies in other portfolio areas, suchas land, employment, education and housing.Furthermore, substance misuse services have beenimplemented inconsistently across different regionsof Australia, as attested by the mismatch betweenregional funding allocations and populationlevels.859 Concerns over such problems underlie thefourth theme to emerge from those reports makingrecommendations to address Indigenous substancemisuse. That is the need for a holistic andcoordinated approach that includes Indigenouscommunity-controlled organisations, all levels ofgovernment, and all sectors.

In looking to the future, experience withcommunity action frameworks provides apromising approach for integrating futureprevention investment within the Indigenous

population. The emphasis within the CommunitiesThat Care program on the empowerment of localcoalitions, and on evidence-based investmenttailored to local evidence of elevated risk factors anddepressed protective factors, appears to bringtogether many of the principles outlined above.

�"��)�2 ��

The present document has been prepared tosummarise the existing evidence base regarding‘what works’ in the prevention of substance useproblems, so that a comprehensive and nationalprevention agenda can be implemented withsynergistic actions across multiple governmentdepartments and sectors of society. This agendamust acknowledge the broad social and structuraldeterminants known to influence patterns of druguse and related harm.

It must also acknowledge the:

� various early developmental experiences andcircumstances that serve to create risk orprotection for the individual later in life,

� extent to which these are common acrossdiverse problem areas, such as mental health,crime, health and welfare,

� nature and extent of the harms associated withthe use of different drugs and the patterns ofuse, and

� circumstances of drug use that create the risk ofthese harms occurring.

The agenda will then need to identify opportunitiesfor remedying risk and increasing protection acrossall sectors of society and government, so as tocontribute to a national effort to reduce drug-relatedharm. The evidence for informing this process hasbeen provided in the foregoing pages.

Table 15.1 breaks down different objectives fordifferent policy jurisdictions and operationalsettings within a Protection and Risk ReductionApproach to Prevention. This table indicates that animportant goal for supply reduction strategies is theachievement of a well-coordinated set of operationsand activities at the community level. Byemphasising integration at the community level,synergies between supply control, demandreduction and harm reduction programs can beestablished. Program objectives for children includeuse reduction and may be different from theobjectives for adults, where harm reduction goalsbecome more prominent.

��

�") �� !��"�#��$��%��

��6��7�" �� "�� (��( ��

$�� $�� % �� !��

<�� '� �� % �� " �� !� ��

�� % �� / %�%��$�� /�� 0� !� �� 0�� 9�� /� �� 0�� % � �� % ��/�� 0��%

*�� # ��$�� " �� % �� $� �� %

�� %�( �� %

*�� *�� 8�� /�� %� �� % �� 0% '��$ �� %

��%�'�� %

7�" �� % ( �� %�� % �� %

7�" �� + �� % * �� % �� 6� ��

�� %

Figure 15.1 (p. 251) also provides a summary ofthe main opportunities for continued and enhancedinvestment to prevent drug-related harms. Thefigure depicts an integrated relationship betweenchildren’s drug use and the patterns of drug usemodelled more broadly by adults. Furthermore, itreflects the literature, which suggests that effectivelegal drug control influences illicit drug use andharm, and that a carefully coordinated mix ofinvestment, rather than any single service strategy,has the greatest chance of success. The programcomplexity once again suggests the importance oftailoring the mix of investment to the specific anddistinct needs of particular communities.

The Protection and Risk Reduction Approach toPrevention holds advantages, not simply forreducing the harm to Australian society from druguse, but also for broader social improvement goals.Investments in prevention should aim to maximisethe potential for early childhood development,while also acknowledging that development andsocialisation have ongoing threads in later years. Bysupporting communities to address both thedevelopmental needs of children and the broader

social conditions relevant to public safety and socialconnection, a combination of benefits can beachieved. These benefits range from the maximisingof human potential through to increasingproductivity and achievement, with ultimateoutcomes for improving both the wealth andwellbeing of the nation.

There are synergies between the prevention ofdrug-related use, risk and harm and the preventionof homelessness, crime, cardiovascular disease,cancer, injuries, mental illness and suicide, many ofwhich share underlying theoretical frameworks. TheProtection and Risk Reduction Approach toprevention has the potential to integrate thesevarying frameworks, providing an important basisfor improving coordination between differentprevention strategies. A closer link between researchand service delivery has the potential to strengthenprevention policy by better defining strategycombinations that are effective in reducing drug-related harm.

�"��

"�� 6��

��

�� !��)� �� $��

�� %

7��!� ��

( ��

8��

��

��

( ��

�� C�� C

��C��

� ��

�� !�� !��0�� 2-��

� ��

( ��

�"� �� !��"�#��$��%��

�"��

��

�"� �� !��"�#��$��%��

��&�� '�� Appendix A summarises a selection of behavioural follow-up research studies relevant to the consequencesof different patterns of adolescent tobacco use on subsequent development. Note, in interpreting this table,symbols are used to summarise the effect of earlier behaviour on later development. The symbol ‘Y’indicates a significant effect; ‘N’ a non-significant effect; the letter ‘a’ indicates analyses were adjusted forother factors (i.e. multivariate methods used); and ‘u’ indicates that analyses were unadjusted.

�� 6� ��

=�� :�� "�� @�� D�� !��

*�� :? �� <� ��2@ �� L�� < ��D��

:? �� 6� � <� 1: �� 6� �� < �� *�� /��$��0

:4 �� D� 1: �� )�, �� <H

:4%@ � � �� <� 13%@ �� N O�� )�� '� � ��

��:: �� D� :@ 8�� )�� $O�� <��

��:2 �� D� :> 8�� # � ��<��

:3%@ * �� D� :? 8�� G�+�

��

:3 �� D� :@ �� 6� �� G�+�:@ �� <� :>��:? �� G�+�

��:? �� <� ��2@ �� L�� < ��D��

:? �� 6� � D� 1: �� 6� �� < �� *�� /��$��0

:@ �� D� :4 �� )�, �� <H:4 �� D� 1: �� )�, �� <H

�� :3 �� D� :?%@ ��$�� ,�� -��

:3 �� 6� � <� 11 �� < �� *�� /��$��0

:> �� D�<� 21 + �� O�� :? �� 6� � <� 1: �� 6� �� $ < �� *��

/��$��0

��:? �� <� ��2@ !�� L�� < ��D��

��:? �� D� ��2@ � �� L�� < ��D��

:? �� 6� � <� 1@ !�� < �� *�� /��$��0

�""��

=�� :�� "�� @�� D�� !��

.�� :3 �� 6� � <� 1@%@ �� ,�� -��

/��$��0

:3 �� 6� � D� 1@%@ ( �� ,�� -��/��$��0

:3 �� 6� � D� 11 + �� < �� *�� /��$��0

&��:2��:> �� D� 1: �� * �� < ��D��

:3 �� 6� � <� 1@%@ ) �� ,�� -��/��$��0

:3 �� 6� � D� 11 �� N < �� *�� /��$��0 ��

:@ �� D� :4 ) �� )�, �� <H

:> �� 6� � D� 11 �� -�� < ��D��

:> �� 6� � <� 11 �� -�� < ��D��

:> �� 6� � D� 11 )�"�� -�� < ��D��

:4 �� D� 1: �� .� �� )�, �� <H

�� :3 �� 6� � D� 11 '�� * �� *��

/��$��0

:? �� 6� �� D� 11 �� E�� ,�� *��

�

D ��

< �� "�� /�% %�� 0

� �� "�� %

�"# �� !��"�#��$��%��

��&�� '�� Research has examined the consequences of adolescent alcohol use for the development of more extremealcohol use behaviours and also for tobacco use, illicit drug use and cannabis use. The other consequencesthat have been explored include mental health, health and social problems. Table B summarises the currentfindings.

�� -��$�� 6� ��

=�� *�� :�� "�� @�� D�� !��

*�� > �� D� :@ �� <H

��> �� D� :@ �� 6� �� <H ��> �� D� :@ �� <H

��5 �� <� :> �� # � ��<��

:= �� 6� � <� 1: �� .� � �� ,�� ' ��

:= �� 6� � <� 1: �� ,�� ' �� :: �� D� :@ �� )�� $O�� <��

:2E:4 -�� D� 1: �� .� � �� +�� ' ��

:2 �� D� :3��:@ +�� )�-�� '+�+(

��:3 �� D� :3 M�� 6� �� <H:3 �� 6� � D� 1@%@ ��$�� ,�� -��

/��$��0 ��

:3 M�� D� :> �� <H�� 6� ��

:3 �� 6� � D� 1: �� .� � �� ,�� ' �� :3 �� 6� � D� 1: �� ,�� ' ��

:> �� 6� � D� 1: �� .� � �� ,�� ' ��

:> �� 6� � D� 1: �� ,�� ' ��

:? �� 6� � D� 1@ ��N�� < �� *�� /��$��0 ��

:4 �� D� 1: �� )�, �� <H

:4 �� D� 1: +�� <H

1= M�� D� 1> '�� 6� �� <�� 6� ��

��5�::�:2 �� D� :=��:@ 8��5=�� # � ��<��

��5�::�:2 ( � ��/5=��0 D� :=��:@ 8��5=�� # � ��<��

��

��:? �� <� ��2@ 8�� L�� < ��D��

:? �� 6� � <� 1: �� 6� �� < �� *�� /��$��0

:4%@ �� D� 13%@ 8�� O�� )�� 'N��

��5�::�:2 �� D� :=��:@ ( � �� # � ��<��

/��5=��0 ��

�� %%%

�"%��

=�� *��:�� "�� @�� D�� !��

��5�::�:2 ( � ��/5=��0 D� :=��:@ ( � �� # � ��<��

/��5=��0 ��

:3 �� 6� � D� :@ �� 6� �� G�+�:3 �� D� :@ �� 6� �� G�+�

:3 �� D��/� �� 0 :>��:? �� G�+�

��:? �� D� ��2@ )��"�� L�� < ��D��

:? �� 6� � <� 1: �� 6� �� < �� *�� /��$��0

:4 �� D� 1: �� )�, �� <H

:4 �� D� 1: �� )�, �� <H

:4 � � �� D� 1: �� L�� <H� � ��

�� :3 �� 6� � <� 11 �� < �� *��

/��$��0:> �� D�<� 21 + �� O��

��:? �� D� ��2@ �� L�� < ��D��

��:? �� D� ��2@ !�� L�� < ��D��

:? �� 6� � <� 1: �� 6� �� < �� *�� /��$��0

:? �� 6� � <� 1@ !�� < �� *�� /��$��0

.�� :3 �� 6� � <� 1@%@ + �� ,�� -��

/��$��0

:3 �� 6� � <� 11 + �� < �� *�� /��$��0

&��5�::�:2 �� D� :=��:@ �� # � ��<��

��:3 �� D�/�� 0 < � �� )�,� � )��

/�� /��7��0�� 0 :3

��:3 �� <� < � !� �� )�,� � )�� /�� /� �� 0�� 0 :3

:3 �� 6� � <� 1@%@ ) �� ,�� -��/��$��0

:3 �� 6� � <� 11 ) �� < �� *�� /��$��0

:@ �� D�<� :4 ) �� )�, �� <H

:> + �� D� 11 �� -�� < ��D��

:> + �� <� 11 )�"�� -�� < ��D��

:> + �� D� 11 �� -�� < ��D��

:4 �� <� 1: �� .� �� )�, �� <H

�� %%%

�"' �� !��"�#��$��%��

=�� *��:�� "�� @�� D�� !��

&��:2��:> �� P 1: �� * �� < ��D��

:2��:4 �� 6� � <� :4��12 � �� &��:2��:4 �� 6� � <� :4��12 �� &��

:2E:4 -�� <� 1: � �� +�� ' ��

��:@ M�� D� :> �� <H

�� 6� ��

:@ M�� D� :> G�� <H�� 6� ��

:4 �� 6� �� D� 11 �� $�� < �� *��

:2E:4 -�� <� 1: �� +�� ' ��

$��::��:@ �� 6� � <� :�� <�� '�

::��:@ �� 6� � <� :�� <�� '�:2E:4 -�� D� 1: +�� +�� ' ��

$��:2 M�� D� :3��:5 ' �� 7��<��

�� 6� ��

:2 �� D� :3��:5 ' �� 7��<��

:3 �� 6� � <�B 11 '�� < �� *�� /��$��0

�

D �� < ��

� �� "�� /�% %�� 0

� �� "�� %

�"(��

�� $�� 6� ��

=�� :�� "�� @�� D�� !��

*��:@ �� D�<� :> �� <H:@E:> �� 6� �� <� :>E:4 �� . � �� <H

� � ��

��:? �� <� ��2@ �� L�� < ��D��

:? �� 6� � D� 1@ ��N�� < �� *�� /��$��0 ��

:? �� 6� � <� 1: �� 6� �� < �� *�� /��$��0

:4 �� D� 1: �� L�� <H��

��:@ �� D�<� :> 8�� <H

��:? �� <� ��2@ 8�� L�� < ��D��

:? �� 6� � <� 1: �� 6� �� < �� *�� /��$��0

��7��:: �� D� :@ �� )�� $O�� <��

��:2 �� <��/� �� 0 :> �� N �� # � ��<��D��/�� 0 � � �� 9� ��

� � ��

:3 # �� D� :>��:? �� G�+�

:@ �� 6� �� D� :>��:? �� G�+�:@ �� D� :> �� <H

:? �� 6� � D� 1@ ��N�� < �� *�� /��$��0 ��

:4 �� D� 1: �� L�� <H��

�� :3 �� 6� � D� 1@%@ �� ,�� -��

/��$��0:3 �� 6� � <� 11 �� < �� *��

/��$��0

:@ �� 6� �� D� :> !�� <H

:> �� D� 21 + �� O��

��:? �� D� ��2@ !�� L�� < ��D��:? �� D� ��2@ �� L�� < ��D��

��

:? �� 6� � <� 1: �� 6� �� < �� *�� /��$��0

�� %%%

��&��'�� Research has examined the consequences of adolescent cannabis use for the development of later drug usebehaviours, and also for mental health, health and social consequences. Table C summarises the currentfindings.

�#) �� !��"�#��$��%��

=�� :�� "�� @�� D�� !��

.�� :3 �� 6� � <� 1@%@ + �� $ ,�� -��

/��$��0 � ��

:3 �� 6� � D� 11 + �� < �� *�� /��$��0

&��.��:2��:5 �� D� 1: �� * �� < ��D��

:3 �� 6� � <� 1@%@ ) �� ,�� -��/��$��0

:3 �� 6� � <� 11 ) �� < �� *�� /��$��0

:@E:> �� 6� �� <� :>E:4 �� <H:@E:> �� 6� �� D� :?E:4 )�"�� <H

:@E:> �� 6� �� <� :>E:4 '�� <H

:@ �� D�<� :> ) �� $ � �� <H� ��

:@ �� <� :4 ) �� )�, �� <H:> + �� D� 11 �� -�� < ��D��

��

:> + �� <� 11 �� -�� < ��D��

:> + �� <� 11 )�"�� -�� < ��D��

:4 �� D� 1: ) �� )�, �� <H

$��:@ �� D�<� :> '�� <H

:@E:> �� 6� �� D� :>E:4 �� <H

:>�� D� :4 '�� -�� '��'

$��:3 �� 6� � D� 11 �� < �� *��

/��$��0:@E:> �� 6� �� D� :>E:4 �� <H

��:@ �� D�<� :> �� <H

:@E:> �� 6� �� D� :?E:4 G�� <H

:@E:> �� 6� �� D� :?E:4 �� <H

�

D �� < ��$��

� �� "�� /�% %�� 0

� �� "�� %

�#��

��

adulthood: Implications for substance abuseprevention. Psychological Bulletin 1992;112(1):64–105.

11. Kreitman N. Alcohol consumption and thepreventive paradox. British Journal of Addiction1986; 81(3):353–363.

12. McDonald D, Moore R, Norberry J,Wardlaw G, Ballenden N. Legislative Options forCannabis in Australia (Monograph No. 26).Canberra: Australian Government PrintingService, 1994.

13. Ministerial Council on Drug Strategy.National Drug Strategic Framework 1998-99 to2002-03; Building Partnerships. A Strategy to Reducethe Harm Caused by Drugs in our Community.Canberra: Commonwealth of Australia,November 1998.

14. Toumbourou J, Patton G, Sawyer S, OlssonC, Web-Pullmann J, Catalano R, et al.Evidence-Based Health Promotion: Resources forPlanning No. 2. Melbourne: AdolescentHealth, Health Development Section, PublicHealth Division, Department of HumanServices, 2000.

15. Cochrane Collaboration. Cochranebrochure, nd. Viewed 22 May 2002,<http://www.cochrane.org/cochrane/cc-broch.htm>.

16. White JM. Cochrane reviews and evidence-based practice. In Roche A, MacDonald J(eds) Systems, Settings, People: WorkforceDevelopment Challenges for the Alcohol and OtherDrugs Field. Adelaide: National Centre forEducation and Training on Addiction,2001;87–92.

17. Clarke M, Oxman AD. Cochrane Reviewers’Handbook 4.1.1. In The Cochrane Library, Issue4. Oxford: Update Software, 2000.

18. Rehm J. Review papers in substance abuseresearch. Addiction 1999; 94(2):173–176.

19. Oxman AD. Checklists for review articles.British Medical Journal 1994; 309:648–651.

20. Oxman AD, Guyatt GH. Validation of anindex of the quality of review articles. Journalof Clinical Epidemiology 1991; 44:1271–1278.

1. Collins DJ, Lapsley HM. Counting the Costs:Estimates of the Social Costs of Drug Abuse inAustralia in 1998-9. National Drug StrategyMonograph Series No. 49. Canberra:Commonwealth of Australia, 2002.

2. House of Representatives, StandingCommittee on Family and CommunityAffairs. Where to next? - A Discussion Paper. InquiryInto Substance Abuse in Australian Communities.Canberra: Commonwealth of Australia,September 2001.

3. Thorley A. Medical responses to problemdrinking. Medicine 1980; 35:1816–1822.

4. Epstein EE. Classification of alcohol-relatedproblems and dependence. In Heather N,Peters TJ, Stockwell T (eds) InternationalHandbook of Alcohol Dependence and Problems.Chichester: John Wiley and Sons, 2001;47–70.

5. Edwards G, Marshall EJ, Cook CCH. TheTreatment of Drinking Problems: A Guide for theHelping Professions. 3rd ed. Cambridge:Cambridge University Press; 1997.

6. Lenton S. The essence of prevention. InWilkinson C, Saunders B (eds) Perspectives onAddiction, Making Sense of the Issues. Perth:William Montgomery, 1996;73–80.

7. Holder HD. Prevention of alcohol-relatedproblems. Alcohol Health and Research World1989; 13(4):339–342.

8. Institute of Medicine, Committee onPrevention of Mental Disorders. Reducing Risksfor Mental Disorders: Frontiers for PreventiveIntervention Research. Mrazek PJ, Haggerty RJ(Eds). Washington, DC: National AcademyPress, 1994.

9. National Crime Prevention - Attorney-General’s Department. Pathways to Prevention -Developmental and Early Intervention Approaches toCrime in Australia - Full Report. Canberra:National Crime Prevention - Attorney-General’s Department, 1999.

10. Hawkins DJ, Catalano RF, Miller JY. Riskand protective factors for alcohol and otherdrug problems in adolescence and early

�#� �� !��"�#��$��%��

21. Higgins K, Cooper-Stanbury M, Williams P.Statistics on Drug Use in Australia 1998. DrugStatistics Series AIHW Catalogue No. PHE16. Canberra: Australian Institute of Healthand Welfare, 2000.

22. Australian Institute of Health and Welfare.2001 National Drug Strategy Household Survey. FirstResults. Canberra: Australian Institute ofHealth and Welfare, May 2002.

23. Kabesch M, von Mutius E. Adverse healtheffects of environmental tobacco smokeexposure in childhood. ACI International2000; 12:146–152.

24. Weitzman M, Byrd RS, Aligne A, Moss M.The effects of tobacco exposure onchildren’s behavioral and cognitivefunctioning: implications for clinical andpublic health policy and future research.Neurotoxicology and Teratology 2002; 24:397–406.

25. Kramer MS. Intrauterine growth andgestational duration determinants. Pediatrics1987; 80:502–511.

26. Kleinman JC, Madans JH. The effects ofmaternal smoking, physical stature andeducational-attainment on the incident oflow birthweight. American Journal ofEpidemiology 1985; 121:843–855.

27. MacArthur C, Knox EG. Smoking inpregnancy: effects of stopping at differentstages. British Journal of Obstetrics and Gynaecology1988; 95:551–555.

28. MacMahon B, Alpert M, Salber EJ. Infantweight and prenatal smoking habits.American Journal of Epidemiology 1965; 82:247–261.

29. Lumley J, Oliver S, Waters E. Smokingcessation programs implemented duringpregnancy. In The Cochrane Library. Oxford:Update Software, 2001.

30. Kline J, Stein M, Hutzler M. Cigarettes,alcohol and marijuana. Varying associationswith birthweight. International Journal ofEpidemiology 1987; 16:44–51.

31. Chang G, Wilkins-Haug L, Berman S, GoetzMA. Brief intervention for alcohol use inpregnancy: a randomized trial. Addiction1999; 94(10):1499–1508.

32. Dunn HG, McBurney AK, Ingram S, HunterCM. Maternal cigarette smoking during

pregnancy and the child’s subsequentdevelopment: physical growth to the age ofsix and a half years. Canadian Journal of PublicHealth 1976; 76:499–505.

33. Barrow FC, Huttly SRA, Victoria CG,Kirkwood B, Vaughan JP. Comparison ofthe causes and consequences of prematurityand intrauterine growth retardation: Alongitudinal study in southern Brazil.Pediatrics 1992; 90:238–244.

34. Breslau N, Chilcoat HD. Psychiatric sequelaeof low birthweight at 11 years of age.Biological Psychiatry 2000; 47:1005–1011.

35. Pharoah POD, Stevenson CJ, Cooke RWI,Stevenson RC. Prevalence of behaviordisorders in low birthweight infants. Archivesof Disease in Childhood 1994; 70:271–274.

36. Breslau N, Chilcoat HD, Johnson EO,Andreski VC, Lucia VC. Neurologic softsigns and low birthweight: their associationand neuropsychiatric implications. BiologicalPsychiatry 2000; 47:71–79.

37. Johnson EO, Breslau N. Increased risk oflearning disability in low birthweight boysat age 11 years. Biological Psychiatry 2000;47:490–500.

38. Olds D. Tobacco exposure and impaireddevelopment: a review of the evidence.Mental Retardation and Developmental DisabilitiesResearch Reviews 1997; 3:257–269.

39. Dunn HG, McBurney AK, Ingram S, HunterCM. Maternal cigarette smoking duringpregnancy and the child’s subsequentdevelopment:II. Neurological andintellectual maturation to the age of 6 1/2years. Canadian Journal of Public Health 1977;68:43–50.

40. Picone TA, Allen LH, Olsen PN, Ferris ME.Pregnancy outcome in North Americanwomen:II. Effects of diet, cigarette smoking,stress and weight gain on placentas and onneonatal physical and behavioralcharacteristics. American Journal of ClinicalNutrition 1982; 36:1214–1224.

41. Mattson S, Calarco KE, Chambers CD, LyonsJones K. Interaction of maternal smokingand other in-pregnancy exposures: analyticconsideration. Neurotoxicology and Teratology2002; 24:367.

42. National Health and Medical ResearchCouncil. Australian Alcohol Guidelines: Health Risks

�#��

and Benefits. Canberra: NHMRC, October2001.

43. Claren SK, Smith DW. The fetal alcoholsyndrome. New England Journal of Medicine1978; 298:1063–1067.

44. Habbick BF, Nanson JL, Snyder RE, CaseyRE. Mortality in foetal alcohol syndrome.Canadian Journal of Public Health 1997;88(3):181–183.

45. Aronson M, Kyllerman M, Sabel KG, SandinB, Olegard R. Children of alcoholicmothers: Developmental, perceptual, andbehavioral characteristics as compared tomatched controls. Acta Paediatrica Scandinavica1985; 74:27–35.

46. Coles CD, Smith IE, Lancaster JS, Falek A.Persistence over the first months ofneurobehavioral differences in infantsexposed to alcohol prenatally. Infant Behaviorand Development 1987; 10:23–37.

47. Gusella J, Fried P. Effects of maternal socialdrinking and smoking on offspring at 13months. Neurobehavioral Toxicology and Teratology1984; 6:13–17.

48. O’Connor MJ, Brill N, Sigman M. Alcoholuse in elderly primips: Relation to infantdevelopment. Pediatrics 1986; 78:444–450.

49. Streissguth AP, Aase JM, Clarren SK, RandelsSP, LaDue RA, Smith DF. Fetal alcoholsyndrome in adolescents and adults. Journal ofthe American Medical Association 1991;265:1961–1967.

50. Zuckerman B, Frank DA. Prenatal cocaineand marijuana exposure: Research andclinical implications. In Zagon IS, Slotkin TA(eds). Boston: Academic Press, 1992;125–154.

51. Harclerode J. The Effect of Marijuana onReproduction and Development. NIDA ResearchMonograph No. 31: National Institute ofDrug Abuse, 1980.

52. Fried PA. Marijuana use by pregnantwomen and effects on offspring: An update.Neurobehavioral Toxicology and Teratology 1982;4:451–454.

53. Zuckerman B, Frank DA, Hingson R, AmaroH, Levenson S, Kayne H. Effects of maternalmarijuana and cocaine use on fetal growth.New England Journal of Medicine 1989; 320:762–768.

54. Hingson R, Alpert JJ, Day N, Dooling E,Kayne H, Morelock S, et al. Effects ofmaternal drinking and marijuana use onfetal growth and development. Pediatrics1989; 70:539–542.

55. O’Connell CM, Fried PA. An investigation ofprenatal cannabis exposure and minorphysical anomalies in a low risk population.Neurobehavioral Toxicology and Teratology 1984;6:345–350.

56. Fried PA. Marijuana use by pregnantwomen: Neurobehavioral effects inneonates. Drug and Alcohol Dependence 1980;6:451–454.

57. Fried PA, Makin JE. Neonatal behaviorcorrelates of prenatal exposure to marijuana,cigarettes, and alcohol in a low-riskpopulation. Neurotoxicology and Teratology 1987;9:1–7.

58. Levy M, Koren G. Clinical toxicology of theneonate. Seminars in Perinatology 1992; 16:63–75.

59. Scher MS, Richardson GA, Coble PA, DayNL, Stoffer DS. The effects of prenatalalcohol and marijuana exposure:Disturbances in neonatal sleep cycling andarousal. Pediatric Research 1988; 24:101–105.

60. Day NL, Richardson GA. Prenatal marijuanause: Epidemiology, methodologic issues,and infant outcome. Clinics in Perinatology1991; 18:77.

61. Scragg RK, Mitchell EA, Ford RP, ThompsonJM, Taylor BJ, Stewart AW. Maternalcannabis use in the sudden death syndrome.Acta Paediatrica 2001; 90(1):57–60.

62. Lim G, Lovegrove M. The effects ofmarijuana use on pregnancy, the fetus andthe child. In; 1994.

63. Mayes LC, Granger RH, Bornstein MH,Zuckerman B. The problem of prenatalcocaine exposure. A rush to judgement.Journal of the American Medical Association 1992;267(3):406–408.

64. Mayes LC, Granger RH. Teratologic anddevelopmental effects of prenatal drugexposure: Alcohol, heroin, marijuana, andcocaine. In Lewis M (ed) A Comprehensive Text-book of Child and Adolescent Psychiatry.Philadelphia, PA: Williams and Wilkins,1995;374–382.

�#� �� !��"�#��$��%��

65. Mayes LC, Fahy T. Prenatal drug exposureand cognitive development. In Sternberg RJ,Grigorenko EL (eds) Environmental Effects onCognitive Abilities: Lawrence ErlbaumAssociates, 2001.

66. Chao CR. Cardiovascular effects of cocaineduring pregnancy. Seminars in Perinatology1996; 20(2):107–114.

67. Singer L, Farkas K, Kliegman R. Childhoodmedical and behavioral consequences ofmaternal cocaine use. Journal of PediatricPsychology 1992; 17(4):389–406.

68. Neuspiel DR. The problem of confoundingin research on prenatal cocaine effects onbehavior and development. In Lewis M,Bendersky M (eds) Mothers, Babies and Cocaine:The Role of Toxins in Development. New York:Lawrence Erlbaum Associates, 1995;95–109.

69. Desmond MM, Wilson GS. Neonatalabstinence syndrome: Recognition anddiagnosis. Addictive Diseases 1975; 2:113–121.

70. Finnegan LP. Clinical effect ofpharmacologic agents on pregnancy, thefetus, and the neonate. Annals of the New YorkAcademy of Sciences 1976; 281:74–89.

71. Hans SL. Maternal opioid use and childdevelopment. In Zagon IS, Slotkin TA (eds)Maternal Substance Abuse and The Developing NervousSystem. San Diego, CA: Academic Press,1992;177–214.

72. Kaltenbach K, Finnegan LP. Perinatal anddevelopmental outcome of infants exposedto methadone in-utero. Neurotoxicology andTeratology 1987; 9:311–313.

73. Zagon IS, McLaughlin P. An overview of theneurobehavioral sequelae of perinatalopioid exposure. Neurobehavioral Teratology1984:194–233.

74. Rosen TS, Johnson HL. Drug-addictedmothers, their infants, and SIDS. Annals of theNew York Academy of Sciences 1988; 533:533–595.

75. Jeremey RJ, Hans SL. Behavior of neonatesexposed in utero to methadone as assessedon the Brazelton scale. Infant Behavior andDevelopment 1985; 8:323–336.

76. Hans SL. Developmental consequences ofprenatal exposure to methadone. Annals of

New York Academy of Sciences 1989; 562:195–207.

77. Hans SL, Marcus J. Motor and attentionalbehavior in infants of methadonemaintained women. National Institute of DrugAbuse Research Monogram 1983; 43:287–293.

78. Hans SL, Marcus J, Jeremy RJ.Neurobehavioral development of childrenexposed in utero to opioid drugs. In Yanai J(ed) Neurobehavioral Teratology. New York:Elsevier, 1984;249–273.

79. Bauman P, Levine SA. The development ofchildren of drug addicts. International Journal ofAddiction 1986; 21:849–863.

80. Sowder BJ, Burt MR. Children of Heroin Addicts:An Assessment of Health, Learning, Behavioral, andAdjustment Problems. New York: Praeger; 1980.

81. Wilson GS. Clinical studies of infants andchildren exposed prenatally to heroin. Annalsof the New York Academy of Sciences 1989;562:183–194.

82. Perez-Reyes M, Wall ME. Presence of 89-tetrahydrocannabinol in human milk. NewEngland Journal of Medicine 1982; 307:819–820.

83. Astley SJ, Little RE. Maternal marijuana useduring lactation and infant development atone year. Neurotoxicology and Teratology 1990;12:161–168.

84. Mejta C, Lavin R. Facilitating healthyparenting among mothers with substanceabuse or dependence problems: Someconsiderations. Alcoholism Treatment Quarterly1996; 14(1):33–46.

85. CASA. No Safe Haven: Children of Substance AbusingParents. New York: National Center onAddiction and Substance Abuse at ColumbiaUniversity, 1998.

86. Kelleher K. Alcohol and drug disordersamong physically abusive and neglectfulparents in a community-based sample.American Journal of Public Health 1994;84(10):1586–1590.

87. Angus G, Hall G. Child Abuse and NeglectAustralia 1994-1995. Child Welfare Series No.16. Canberra: Australian Institute of Healthand Welfare, 1996.

88. Hans SL, Bernstein VJ, Henson LG. The roleof psychopathology in the parenting ofdrug-dependent women. Development andPsychopathology 1999; 11:957–977.

�#"��

89. Farkas K, Parran TV. Treatment of cocaineaddiction during pregnancy. ClinicalPerinatology 1993; 20:29–45.

90. Finkelstein N. Treatment of programmingfor alcohol and drug-dependent women.International Journal of Addiction 1993; 28:1275–1309.

91. Finnegan LP, Oehlberg SM, Regan DO,Rudrauff ME. Evaluation of parenting,depression and violence profiles inmethadone maintained women. Child Abuseand Neglect 1981; 5:267–273.

92. Baker B, Heller T. Preschool children withexternalizing behaviors: experience offathers and mothers. Journal of Abnormal ChildPsychology 1996; 24:513–532.

93. Bellinger DC. Interpreting the literature onlead and child development: The neglectedrole of the “experimental system”.Neurotoxicology and Teratology 1995; 17:201–212.

94. Hans SL. Studies of prenatal exposure todrugs. Focusing on parental care ofchildren. Neurotoxicology and Teratology 2002;24:329–337.

95. United States Environmental ProtectionAgency. Respiratory Health Effects of PassiveSmoking: Lung Cancer and Other Disorders. ReportNo. EPA/600/6-90/006F. Washington(DC): United States EnvironmentalProtection Agency, Office of Health andEnvironmental Assessment, Office ofResearch and Development, December1992.

96. Dybing E, Sanner T. Passive smoking,sudden infant death syndrome (SIDS) andchildhood infections. Human and ExperimentalToxicology 1999; 18(4):202–205.

97. Schenker MB, Samur JM, Speizer FE. Riskfactors for childhood respiratory disease: theeffect of host factors and homeenvironmental exposures. American Review ofRespiratory Disease 1983; 128:1038–1043.

98. Strachan D, Jarvis MJ, Feyerabend C. Therelationship of salivary cotinine torespiratory symptoms, spirometry andexercise induced bronchospasm in seven-year old children. American Review of RespiratoryDisease 1990; 142:147–151.

99. Rylander E, Pershagen G, Eriksson M,Bermann G. Parental smoking, urinary

continine and wheezing bronchitis inchildren. Epidemiology 1995; 6:289–293.

100. Nafstad P, Jaakkola JJK, Hagen JA.Breastfeeding, maternal smoking and lowerrespiratory tract infections. European RespiratoryJournal 1996; 9:2623–2629.

101. Strachan D, Jarvis MJ, Feyerabend C. Passivesmoking, salivary continine concentrationsand middle ear effusion in 7 year oldchildren. British Medical Journal 1989;298:1549–1552.

102. Klonoff-Cohen HS, Edelstein SL, LefkowitzES, Srinivasan AP, Kaegi D, Chang JC, et al.The effect of passive smoking and tobaccoexposure through breast milk on suddeninfant death syndrome. Journal of the AmericanMedical Association 1995; 273:795–798.

103. Hiley CMH, Morley CJ. Risk factors forsudden infant death syndrome: furtherchange in 1992-93. British Medical Journal1996; 312:1397–1398.

104. Tager IB, Weiss ST, Munoz A, et-al.Longitudinal study of the effects of maternalsmoking on pulmonary function inchildren. New England Journal of Medicine 1983;309:699–703.

105. Rona RJ, Chinn S, Du Ve Florey C. Exposureto cigarette smoking and children’s growth.International Journal of Epidemiology 1985;14:402–409.

106. Woodward A, Douglas RM, Graham NMH,Miles H. Acute respiratory illness inAdelaide children: breastfeeding modifiesthe effect of passive smoking. Journal ofEpidemiology and Community Health 1990;44:224–230.

107. Maran AGD, Wilson JA. Glue ear and speechdevelopment. British Medical Journal 1986;293:713–714.

108. Samet JM, Tager I, Speizer F. The relationbetween respiratory illness in childhood andchronic airflow obstruction in adulthood.American Review of Respiratory Disease 1983;127:523.

109. Shaheen SO, Barker JP, Shiell AW, et-al. Therelationship between pneumonia in earlychildhood and impaired lung function inlate adult life. American Journal of RespiratoryCritical Care Medicine 1994; 149:616–619.

�## �� !��"�#��$��%��

110. Kandel DB, Wu P, Davis M. Maternalsmoking during pregnancy and smoking byadolescent daughters. American Journal of PublicHealth 1994; 84:1407–1413.

111. Chassin L, Presson CC, Todd M, Sherman SJ,Rose J. Maternal socialization of adolescentsmoking: the intergenerational transmissionof parenting and smoking. DevelopmentalPsychology 1998; 34:1269–1273.

112. Jackson C, Henriksen L, Dickinson D,Messer L, Robertson S. A longitudinal studypredicting patterns of cigarette smoking inlate childhood. Health Education and Behaviour1998; 24:436–447.

113. Doll R. Effect of environmental tobaccosmoke on adults. In Stockwell T (ed) DrugTrials and Tribulations: Lessons for Australian DrugPolicy. Proceedings of an International Symposium.Perth: National Centre for Research into thePrevention of Drug Abuse, 1998; 1–14.

114. National Health and Medical ResearchCouncil. The Health Effects of Passive Smoking.Canberra: National Health and MedicalResearch Council, 1997.

115. White V. Australian Secondary Students’ Use ofOver-the-counter and Illicit Substances in 1999.National Drug Strategy Monograph SeriesNo. 46. Canberra: CommonwealthDepartment of Health and Aged Care, 2001.

116. White VM, Hill DJ, Letcher TR. Alcohol useamong Australian secondary students in1996. Drug and Alcohol Review 2000;19(4):371–379.

117. Johnson EO, Schutz CG, Anthony JC,Ensminger ME. Inhalants to heroin: aprospective analysis from adolescence toadulthood. Drug and Alcohol Dependence 1995;40:159–164.

118. Costello EJ, Erkanli A, Federman E, AngoldA. Development of psychiatric comorbiditywith substance abuse in adolescents: effectsof timing and sex. Journal of Clinical ChildPsychology 1999; 28(3):298–311.

119. Patton GC, Carlin JB, Coffey C, Wolfe R,Hibbert M, Bowes G. The course of earlysmoking: a population based cohort studyover three years. Addiction 1998; 93:1251–1260.

120. Jackson KM, Sher KJ, Wood PK. Trajectoriesof concurrent substance use disorders: a

developmental, typological approach tocomorbidity. Alcoholism: Clinical and ExperimentalResearch 2000; 24(6):902–913.

121. Kandel DB, Yamaguchi K, Chen K. Stages ofprogression in drug involvement fromadolescence to adulthood: further evidencefor the gateway theory. Journal of Studies onAlcohol 1992; (September):447–457.

122. McGee R, Williams S, Poulton R, Moffitt T.A longitudinal study of cannabis use andmental health from adolescence to earlyadulthood. Addiction 2000; 95(4):491–503.

123. Coffey C, Lynskey M, Wolfe R, Patton G.Initiation and progression of cannabis use ina population-based Australian adolescentlongitudinal study. Addiction 2000;95(11):1679–1690.

124. Newcomb MD, Felix-Ortiz M. Multipleprotective and risk factors for drug use andabuse: cross-sectional and prospectivefindings. Journal of Personality and SocialPsychology 1992; 63(2):280–296.

125. Fergusson DM, Lynskey MT, Horwood LJ.Childhood exposure to alcohol andadolescent drinking patterns. Addiction 1994;89:1007–1016.

126. Miller-Johnson S, Lochman JE, Coie JD,Terry R, Hyman C. Comorbidity of conductand depressive problems at sixth grade:substance use outcomes across adolescence.Journal of Abnormal Child Psychology 1998;26(3):221–232.

127. Guy SM, Smith GM, Bentler PM. Adolescentsocialization and use of licit and illicitsubstances: impact on adult health. Psychologyand Health 1993; 8:463–487.

128. Casswell S, Zhang J-F. Access to alcoholfrom licensed premises during adolescence:a longitudinal study. Addiction 1997;92(6):737–745.

129. Guo J, Hawkins JD, Hill KG, Abbott RD.Childhood and adolescent predictors ofalcohol abuse and dependence in youngadulthood. Journal of Studies on Alcohol 2001;62:754–762.

130. Hill KG, White HR, Chung I, Hawkins JD,Catalano RF. Early adult outcomes ofadolescent binge drinking: person andvariable centered analyses of binge drinkingtrajectories. Alcoholism: Clinical and ExperimentalResearch 2000; 23(6):892–901.

�#%��

131. Federman EB, Costello EJ, Angold A, FarmerEM, Erkanli A. Development of substanceuse and psychiatric comorbidity in anepidemiologic study of white and AmericanIndian young adolescents the Great SmokyMountains Study. Drug and Alcohol Dependence1997; 44(2–3):69–78.

132. Krueger RF. Personality traits in lateadolescence predict mental disorders inearly adulthood: a prospective-epidemiological study. Journal of Personality1999; 67(1):39–65.

133. Newcomb MD, Bentler PM. Impact ofadolescent drug use and social support onproblems of young adults: a longitudinalstudy. Journal of Abnormal Psychology 1988;97(1):64–75.

134. Williams P. Alcohol-related social disorder and ruralyouth: Part 2 - Perpetrators. Trends and Issues inCrime and Criminal Justice Series No. 149.Canberra: Australian Institute ofCriminology, March 2000.

135. Brook JS, Whiteman M, Finch SJ, Cohen P.Young adult drug use and delinquency:childhood antecedents and adolescentmediators. Journal of the American Academy ofChild & Adolescent Psychiatry 1996;35(12):1584–1592.

136. Brook JS, Whiteman M, Finch S, Cohen P.Mutual attachment, personality, and druguse: pathways from childhood to youngadulthood. Genetic, Social and General Psychology1998; 124(4):492.

137. Stacy AW, Newcomb MD. Adolescent druguse and adult drug problems in women:direct, interactive and mediational effects.Experimental and Clinical Psychopharmacology1999; 7(2):160–173.

138. Brook JS, Cohen P, Brook DW. Longitudinalstudy of co-occurring psychiatric disordersand substance use. Journal of the AmericanAcademy of Child and Adolescent Psychiatry 1998;37(3):322–330.

139. Bewley BR, Bland JM. Smoking andrespiratory symptoms in two groups ofschool children. Preventive Medicine 1976;5:63–69.

140. Adams L, Lonsdale D, Robinson M,Bawbone R, Guz A. Respiratory impairmentinduced by smoking in children insecondary schools. British Medical Journal1984; 288:891–895.

141. U.S. Department of Health and HumanServices. The Health Consequences of Smoking:Chronic Obstructive Lung Disease: A Report of theSurgeon General. Report No. PHS 84-50204.United States: Public Health Service. Officeon Smoking and Health, 1984.

142. Royal College of Physicians of London.Smoking and the Young. London: Royal Collegeof Physicians, 1992.

143. Townsend J, Wikes H, Haines A, Jarvis M.Adolescent smokers seen in general practice:health, lifestyle, physical measurements,and response to antismoking advice. BritishMedical Journal 1991; 303:947–950.

144. Lerner JV, Vicary JR. Difficult temperamentand drug use: analyses from the New Yorklongitudinal study. Journal of Drug Education1984; 14(1):1–8.

145. Newcomb MD, McGee L. Influence ofsensation seeking on general deviance andspecific problem behaviors fromadolescence to young adulthood. Journal ofPersonality and Social Psychology 1991;61(4):614–628.

146. Commonwealth Department of Health andAged Care. National Tobacco Strategy 1999 to2002-03. A Framework for Action. Canberra:Commonwealth Department of Health andAged Care, June 1999.

147. Chassin L, Pitts SC, DeLucia C, Todd M. Alongitudinal study of children of alcoholics:predicting young adult substance usedisorders, anxiety, and depression. Journal ofAbnormal Psychology 1999; 108(1):106–119.

148. Fergusson DM, Lynskey MT, Horwood LJ.Alcohol misuse and juvenile offending inadolescence. Addiction 1996; 91(4):483–494.

149. Andrews JA, Duncan SC. Examining thereciprocal relation between academicmotivation and substance use: effects offamily relationships, self-esteem, andgeneral deviance. Journal of Behavioral Medicine1997; 20(6):523–549.

150. Pedersen W, Skrondal A. Alcohol and sexualvictimization: a longitudinal study ofNorwegian girls. Addiction 1996; 91(4):565–581.

151. Heale P, Stockwell T, Dietze P, Chikritzhs T,Catalano P. Patterns of Alcohol Consumption inAustralia, 1998. National Alcohol Indicators

�#' �� !��"�#��$��%��

Project Bulletin No. 3. Perth: National DrugResearch Institute, Curtin University ofTechnology, 2000.

152. Lemmens P. Relationship of alcoholconsumption and alcohol problems at thepopulation level. In Heather N, Peters TJ,Stockwell T (eds) International Handbook ofAlcohol Dependence and Problems. Chichester: JohnWiley and Sons, 2001; 395–412.

153. White V. Australian Secondary Students’ Use ofAlcohol in 1999. National Drug StrategyMonograph No. 45. Canberra:Commonwealth of Australia, 2001.

154. Moon L, Meyer P, Grau J. Australia’s YoungPeople: Their Health and Wellbeing, the first report onthe health of young people aged 12-24 years.Catalogue No. PHE 19. Canberra: AustralianInstitute of Health and Welfare, 1999.

155. McBride N, Midford R, Farringdon F,Phillips M. Early results from a schoolalcohol harm minimization study: theSchool Health and Alcohol Harm ReductionProject. Addiction 2000; 95(7):1021–1042.

156. Fergusson DM, Horwood LJ. Early onsetcannabis use and psychosocial adjustment inyoung adults. Addiction 1997; 92(3):279–296.

157. Fergusson DM, Horwood LJ, Swain-Campbell N. Cannabis dependence andpsychotic symptoms in young people.Psychological Medicine 2003; 33(1):15–21.

158. Fergusson DM, Horwood LJ, Swain-Campbell N. Cannabis use and psychosocialadjustment in adolescence and youngadulthood. Addiction 2002; 97(9):1123–1135.

159. Fergusson DM, Lynskey MT, Horwood LJ.The short-term consequences of early onsetcannabis use. Journal of Abnormal Child Psychology1996; 24(4):499–512.

160. Bray JW, Zarkin GA, Ringwalt C, Junfeng Q.The relationship between marijuanainitiation and dropping out of high school.Health Economics 2000; 9:9–18.

161. Lynskey M, Hall W. The effects ofadolescent cannabis use on educationalattainment: a review. Addiction 2000;95(11):1621–1630.

162. Kandel DB, Simcha-Fagan O, Davies M. Riskfactors for delinquency and illicit drug use

from adolescence to young adulthood.Journal of Drug Issues 1986; 16(1):67–90.

163. Brook JS, Whiteman M, Cohen P, Shapiro J,Balka E. Longitudinally predicting lateadolescent and young adult drug use:childhood and adolescent precursors. Journalof the American Academy of Child & AdolescentPsychiatry 1995; 34(9):1230–1238.

164. Bachman JG, O’Malley PM, Johnston LD.Drug use among young adults: the impactsof role status and social environment. Journalof Personality and Social Psychology 1984;47(3):629–645.

165. Australian Crime Commission. AustralianIllicit Drug Report 2001-02. Canberra:Commonwealth of Australia, 2003.

166. Darke S, Ross J, Hando, J, Hall W,Degenhardt L. Illicit Drug Use in Australia:Epidemiology, Use Patterns and Associated Harm.National Drug Strategy Monograph SeriesNo. 43. Canberra: CommonwealthDepartment of Health and Aged Care, 2000.

167. Ridolfo B, Stevenson C. The Quantification ofDrug-caused Mortality and Morbidity in Australia,1998. Cat. No. PHE 29. Canberra: AustralianInstitute of Health and Welfare, 2001.

168. Australian Bureau of Criminal Intelligence.Australian Illicit Drug Report 2000-01. Canberra:Commonwealth of Australia, 2002.

169. Kaye S, Darke S, Topp L. An Examination ofCocaine Dependence Among Injecting and Non-Injecting Drug Users in Sydney. Technical ReportNo. 116. Sydney: National Drug andAlcohol Research Centre, 2001.

170. Kaye S, Darke S, McKetin R. The Prevalence,Patterns and Harms of Cocaine Use Among Injectingand Non-Injecting Drug Users in Sydney. TechnicalReport No. 99. Sydney: National Drug andAlcohol Research Centre, 2000.

171. Australian Bureau of Statistics. 1998: Causes ofDeath. Canberra: Australian Bureau ofStatistics, 1999.

172. Hall W, Degenhardt L, Lynskey M. Trends inOpioid Overdose and Suicide Mortality in Young Adultsin Australia 1964-1997. Technical Report No.67. Sydney: National Drug and AlcoholResearch Centre, 1999.

173. Lenton S, Humeniuk R, Heale P, Christie P.Infringement versus conviction: The socialimpact of a minor cannabis offence in South

�#(��

Australia and Western Australia. Drug andAlcohol Review 2000; 19:257–264.

174. Saggers S, Gray D. Dealing with Alcohol:Indigenous Usage in Australia, New Zealand andCanada. Melbourne: Cambridge UniversityPress; 1998.

175. McAllister I, Makkai T. The prevalence andcharacteristics of injecting drug users inAustralia. Drug and Alcohol Review 2001;20(1):29–36.

176. Manski CF, Pepper JV, Petrie CV. InformingAmerica’s Policy on Illegal Drugs: What We Don’tKnow Keeps Hurting Us. Washington, DC:National Academy Press, 2001.

177. Hillman SD, Silburn RS, Green A, Zubrick S.Youth Suicide in Western Australia Involving Cannabisand Other Drugs: Western Australian DrugAbuse Strategy Office, 2000.

178. Stockwell T, Bolderston H. Alcohol andphobias. British Journal of Addiction 1987;82:971–979.

179. Teeson M. Comorbidity in mental healthand substance use: causes, prevention andtreatment. Paper presented at NationalWorkshop on Comorbidity. Canberra, 6–7March 2000.

180. Centers for Disease Control and Prevention.National Tobacco Information OnlineSystem (NATIONS), 2001. Viewed July 192002, <http://apps.nccd.cdc.gov/nations/>.

181. Fitzsimmons G, Cooper-Stanbury M. 1998National Drug Strategy Household Survey; State andTerritory Results. Drug Statistics Series No. 5.Canberra: Australian Institute of Health andWelfare, 2000.

182. Australian Customs Service. Australian CustomsService Annual Report 2001-02. Canberra:Commonwealth of Australia, 2002.

183. Ryder D, Salmon A, Walker N. Drug Use andDrug-related Harm: A Delicate Balance. Melbourne:IP Communications; 2001.

184. Miller M, Draper G. Statistics on Drug Use inAustralia 2000. Drug Statistics Series No. 8.AIHW Cat. No. PHE 30. Canberra:Australian Institute of Health and Welfare,2001.

185. Peto R, Darby S, Deo H, Silcocks P, WhitleyE, Doll R. Smoking, smoking cessation, andlung cancer in the UK since 1950:

combination of national statistics with twocase-control studies. British Medical Journal2000; 321:323–329.

186. Stockwell T, Gruenewald P. Controls on thephysical availability of alcohol. In HeatherN, Peters TJ, Stockwell T (eds) InternationalHandbook of Alcohol Dependence and Problems.Chichester: John Wiley and Sons,2001;699–720.

187. Stockwell TR. Alcohol misuse and violence: anevaluation of the appropriateness and efficacy of liquorlicensing laws across Australia, report No. 5.Canberra: Australian GovernmentPublishing Service; 1995.

188. Stockwell T, Crosbie D. Supply and demandfor alcohol in Australia: Relationshipsbetween industry structures, regulation andthe marketplace. International Journal of DrugPolicy 2001; 12:139–152.

189. Catalano P, Chikritzhs T, Stockwell T, WebbM. Trends in Per Capita Alcohol Consumption inAustralia, 1990/91-1998/99. Monograph No. 4.Perth, Western Australia: National DrugResearch Institute, Curtin University ofTechnology; 2001.

190. Distilled Spirits Industry Council of AustraliaInc. DSICA Submission to the Senate EconomicsLegislation Committee. Melbourne, Australia:DSICA, 2002.

191. Stockwell T, Heale P, Dietze P, Chikritzhs T,Catalano P. Patterns of Alcohol Consumption inAustralia. National Alcohol Indicators ProjectTechnical Report No. 3. Perth: NationalDrug Research Institute, Curtin Universityof Technology, 2001.

192. World Drink Trends. International BeverageAlcohol Consumption and Production Trends.Oxfordshire: World Advertising ResearchCenter; 2002.

193. National Drug Research Institute. CurrentInquiry: Substance Abuse in Australian Communities.Perth: National Drug Research Institute,Curtin University of Technology, 2000.

194. Chikritzhs T, Jonas HA, Stockwell TR, HealePF, Dietze PM. Mortality and life-years lostdue to alcohol: a comparison of acute andchronic causes. Medical Journal of Australia2001; 174:281–284.

195. Chikritzhs T, Jonas H, Heale P, Stockwell T,Dietze P, Hanlin K, et al. Alcohol-Caused Deathsand Hospitalisations in Australia, 1990-1997.

�%) �� !��"�#��$��%��

National Alcohol Indicators ProjectTechnical Report No. 1. Perth: NationalDrug Research Institute, 2000.

196. Degenhardt L, Hall W. The Association BetweenPsychosis and Problematic Drug Use Among AustralianAdults: Findings from the National Survey of MentalHealth and Well-Being. Technical Report No.93. Sydney: National Drug and AlcoholResearch Centre, 2000.

197. Gilman SE, Abraham HD. A longitudinalstudy of the order of onset of alcoholdependence and major depression. Drug andAlcohol Dependence 2001; 63:277–286.

198. Greenfield TK. Individual risk of alcohol-related disease and problems. In Heather N,Peters TJ, Stockwell T (eds) InternationalHandbook of Alcohol Dependence and Problems.Chichester: John Wiley and Sons,2001;413–438.

199. Chikritzhs T, Stockwell T, Jonas H,Stevenson C, Cooper-Stanbury M, Donath S,et al. Towards a standardised methodologyfor estimating alcohol-caused death, injuryand illness in Australia. Australian and NewZealand Journal of Public Health 2002;26(5):443–450.

200. Australian Bureau of Criminal Intelligence.Australian Illicit Drug Report 1999-2000.Canberra: Australian Bureau of CriminalIntelligence, 2001.

201. Topp L, Kaye S, Bruno R, Longo M,Williams P, O’Reilly B, et al. Australian DrugTrends 2001: Findings of the Illicit Drug ReportingSystem (IDRS). Monograph No. 48. Sydney:National Drug and Alcohol Research Centre,2002.

202. Makkai T, McGregor K. Drug UseMonitoring in Australia. 2001 AnnualReport on Drug Use Among PoliceDetainees. Research and Public Policy Series 2002;No. 41:1–73.

203. World Health Organization. Rational Use ofBenzodiazepines. Geneva: World HealthOrganization, 1996.

204. Berbatis C, Lockwood A. The Utilisation ofDrugs to Treat Anxiety, Insomnia and Depression inAustralia: Australian Statistics, Associated Morbidityand Mortality, Alternative Treatments andRecommendations for Improved Drug Use. Perth:National Centre for Research into thePrevention of Drug Abuse, 1988.

205. Dobbin M. Flunitrazepam (Rohypnol, Hypnodorm).Melbourne: Victorian Department ofHuman Services, 1997.

206. Goldberg L, MacKinnon DP, Elliot DL, MoeEL, Clarke G, Cheong J. The adolescentstraining and learning to avoid steroidsprogram. Archives of Pediatrics and AdolescentMedicine 2000; 154:332–338.

207. Maycock B, Beel A. Anabolic steroid abuseand violence. Crime and Justice Bulletin.NSW Bureau of Crime Statistics andResearch, 1997. Viewed September 2001,<http:www.attgendept.nsw.gov.au\boscar1.nsf\pages\index>.

208. Van de Ven P, Rawstorne P, Crawford J,Kippax S. Facts and Figures. 2000 Male Out Survey.Monograph No. 2. Sydney: National Centrein HIV Social Research, 2001.

209. Yesalis CE, Bahrke MS. Anabolic-androgenicsteroids. Current issues. Sports Medicine 1995;19(5):326–340.

210. Beel A, Maycock B, McLean N. Currentperspectives on anabolic steroids. Drug andAlcohol Review 1998; 17:87–103.

211. Delalande CL, Aitken CK, Mercuri P, StantonKA. Risky practices among people whoinject steroids. Medical Journal of Australia 1998;169:62.

212. Maycock B, Howat P. Factors contributingto anabolic androgenic steroid relatedviolence. Journal of Performance Enhancing Drugs1999; 2(3):4–15.

213. Topp L, Darke S, Bruno R, Fry C, HargreavesK, Humenuik R, et al. Australian Drug Trends2000: Findings from the Illicit Drug Reporting System(IDRS) Monograph No. 47. Sydney: NationalDrug and Alcohol Research Centre, 2001.

214. Swift W, Hall W, Teesson M. Cannabis usedisorders among Australian adults: Findings from theNational Survey of Mental Health and Wellbeing.Technical Report No. 78. Sydney: NationalDrug and Alcohol Research Centre, 1999.

215. Lenton S, Heale P, Erickson P, Single E, LangE, Hawks D. The Regulation of Cannabis Possession,Use and Supply. A Discussion Document Prepared forthe Drugs and Crime Prevention Committee of TheParliament of Victoria. Monograph No. 3. Perth:National Drug Research Institute, 2000.

216. Hall W. The public health significance ofcannabis use in Australia. Australian Journal ofPublic Health 1995; 19(3):235–242.

�%��

217. Hall W, Solowij N. Adverse effects ofcannabis. Lancet 1998; 352(9140):1611–1616.

218. Solowij N. Cannabis and Cognitive Functioning.United Kingdom: Cambridge UniversityPress; 1998.

219. Hall W, Teesson M, Lynskey M, DegenhardtL. The Prevalence in the Past Year of Substance Use andICD-10 Substance Use Disorders in Australian Adults:Findings from the National Survey of Mental Healthand Well-being. Sydney: National Drug andAlcohol Research Centre, 1998.

220. World Health Organization. Amphetamine-TypeStimulants: A report from the WHO Meeting onAmphetamines, MDMA and other Psychostimulants,Geneva, 12-15 November 1996. Geneva:Division of Mental Health and Prevention ofSubstance Abuse, World HealthOrganization (WHO), 1997.

221. Makkai T, McGregor K. Drug Use Monitoring inAustralia: 2000 Annual Report on Drug Use AmongPolice Detainees. Canberra: Australian Instituteof Criminology, 2001.

222. Degenhardt L, Darke S, Dillon P. GHB useamong Australians: characteristics, usepatterns and associated harm. Drug and AlcoholDependence 2002; 67:89–94.

223. Topp L. IDRS: Illicit Drug Reporting System April2001. Drug Trends Bulletin. Sydney:National Drug and Alcohol Research Centre,2001.

224. Curran HV. Is MDMA (‘Ecstasy’) neurotoxicin humans? An overview of evidence and ofmethodological problems in research.Neuropsychobiology 2000; 42:34–41.

225. Morgan MJ. Ecstasy (MDMA): A review ofits possible persistent psychological effects.Psychopharmacology 2000; 15:230–248.

226. Dillon P. 10 years of ecstasy and other partydrug use in Australia: what have we doneand what is there left to do? 2000. Viewed11 September 2001, <http://www.clubhealth.org.uk/pages/2000/p2_p4.htm>.

227. Turner JJD, Parrot AC. ‘Is MDMA a humanneurotoxin?’: Diverse views from thediscussants. Neuropsychobiology 2000; 42:42–48.

228. Darke S, Kaye S, Topp L. NSW Drug Trends2001. Findings from the Illicit Drug Reporting System

(IDRS). Technical Report No. 125. Sydney:National Drug and Alcohol Research Centre,2002.

229. Hall W, Lynskey M, Degenhardt L. Heroin Usein Australia: Its Impact on Public Health and PublicOrder. Monograph No. 42. Sydney: NationalDrug and Alcohol Research Centre, 1999.

230. Warner-Smith M, Lynskey M, Darke S, HallW. Heroin Overdose: Prevalence, Correlates,Consequences and Interventions. Monograph No.46. Sydney: National Drug and AlcoholResearch Centre, 2000.

231. Degenhardt L. Opioid Overdose Deaths in Australia.Sydney: National Drug and AlcoholResearch Centre, 2001.

232. Callaly T, Trauer T, Munro L, Whelan G.Prevalence of psychiatric disorder in amethadone maintenance population.Australian and New Zealand Journal of Psychiatry2001; 35:601–605.

233. d’Abbs P, MacLean S. Petrol Sniffing in AboriginalCommunities: A Review of Interventions:Cooperative Research Centre for Aboriginaland Tropical Health, 2000.

234. Darke S, Zador D. Fatal heroin ‘overdose’: areview. Addiction 1996; 91(12):1765–1772.

235. Gerostamoulous J, Staikos V, Drummer OH.Heroin-related deaths in Victoria: a reviewof cases for 1997 and 1998. Drug and AlcoholDependence 2000; 61:123–128.

236. Drug Policy Expert Committee. Drugs: Meetingthe Challenge. Melbourne: Drug Policy ExpertCommittee, 2000.

237. Darke S, Kaye S, Ross J. Transitions betweenthe injection of heroin and amphetamines.Addiction 1999; 94:1795–1804.

238. Makkai T. Patterns of recent drug useamong a sample of Australian detainees.Addiction 2001; 96(12):1799–1808.

239. National Centre in HIV Epidemiology andClinical Research. Annual Surveillance Report.HIV/AIDS, Viral Hepatitis and Sexually TransmissibleInfections in Australia. Sydney: University ofNew South Wales, 2001.

240. Darke S, Topp L, Hall W. IDRS: Illicit DrugReporting System April 2001. Drug TrendsBulletin, Supplement. Sydney: NationalDrug and Alcohol Research Centre, 2001.

�%� �� !��"�#��$��%��

241. Ross J, Cohen J, Darke S, Hando J, Hall J.Transitions Between Routes of Administration andCorrelates of Injecting Amongst Regular AmphetamineUsers in Sydney. Monograph No. 18. Sydney:National Drug and Alcohol Research Centre,1994.

242. Carruthers S, Loxley W. Attitudes of noviceheroin injectors towards non-injectingroutes of administration to prevent thetransmission of blood borne viruses.International Journal of Drug Policy 2002; 13:69–74.

243. National Centre in HIV Epidemiology andClinical Research. HIV/AIDS, Viral Hepatitis &Sexually Transmissible Infections in Australia. AnnualSurveillance Report. Sydney: National Centre inHIV Epidemiology and Clinical Research,2001.

244. Darke S, Kay S, Ross J. Geographicalinjecting locations among injecting drugusers in Sydney, Australia. Addiction 2001;96:241–246.

245. Hulse GK, Basso MR, Wodak A. Theinjecting drug user. In Hulse GK, White J,Cape G (eds) Management of Alcohol and OtherDrug Problems. Melbourne: Oxford UniversityPress, 2003.

246. Department of Human Services Victoria.Injecting Temazepam. Department ofHuman Services, Victoria, 2001. Viewed 5November 2001, <http://www.drugs.vic.gov.au/temazepam/facts.htm>.

247. Pearson D. Temazepam capsules - PBSRestrictions. In: Haines B, editor. Canberra:UPDATE; 2002.

248. Batey RG, Bollipo J. Hepatitis B. Drug andAlcohol Review 1996; 15:289–306.

249. Carruthers S, Loxley W, Phillips M, BevanJS. The Australian Study of HIV andInjecting Drug Use. Part II: predictingexposure to hepatitis C and hepatitis B. Drugand Alcohol Review 1997; 16:215–220.

250. Australian National Council on AIDSHepatitis C and Related Diseases Hepatitis CSub-Committee. Hepatitis C Virus ProjectionsWorking Group: Estimates and Projections of theHepatitis C Virus Epidemic in Australia 2002.Sydney: National Centre in HIVEpidemiology and Clinical Research, 2002.

251. Brown K, Crofts N. Health care costs of acontinuing epidemic of hepatitis C virusinfection among injecting drug users.Australian and New Zealand Journal of Public Health1998; 22:384–388.

252. Standing Committee on Social Issues.Hepatitis C: The Neglected Epidemic. Inquiry intoHepatitis C in New South Wales. Report No. 16.Sydney: Parliament of New South Wales,1998.

253. Burrows D, Bassett B. Meeting the Needs of Peoplein Australia Living with Hepatitis C: NationalHepatitis C Councils Education ReferenceGroup; 1996.

254. Unwin E, Codde J. Trends and patterns ofdrug-caused mortality in Australia andWestern Australia. Australian and New ZealandJournal of Public Health 1999; 23(4):352–357.

255. Corrao G, Bagnardi V, Zambon A, Arico S.Exploring dose-response relationshipbetween alcohol consumption and the riskof several alcohol-related conditions: a metaanalysis. Addiction 1999; 94:1551–1574.

256. Fingerhood M. Substance abuse in olderpeople. Journal of the American Geriatrics Society2000; 48(8):985–996.

257. Powell JE, McInnes E. Alcohol use amongolder hospital patients: Findings from anAustralian study. Drug and Alcohol Review 1994;13:5–12.

258. Schonfeld L, Dupree LW. Antecedents ofdrinking for early- and late-onset elderlyalcohol abusers. Journal of Studies on Alcohol1991; 52:587–592.

259. Olsen-Noll CG, Bosworth MF. Alcoholabuse in the elderly. American Family Physician1989; 39(4):173–180.

260. Dunne FJ. Misuse of alcohol or drugs byelderly people: May need specialmanagement. British Medical Journal 1994;308(6929):608–610.

261. Salazar Thomas V, Rockwood KJ. Alcoholabuse, cognitive impairment, and mortalityamong older people. Journal of the AmericanGeriatrics Society 2001; 49(4):415–421.

262. Johnson I. Alcohol problems in old age: Areview of recent epidemiological research.International Journal of Geriatric Psychiatry 2000;15:575–581.

�%��

263. Australian Bureau of Statistics. National HealthSurvey: Summary of Results. Canberra: ABS,1995.

264. Jorm AF, Grayson D, Creasey H, Waite L,Broe GA. Long-term benzodiazepine use byelderly people living in the community.Australian and New Zealand Journal of Public Health2000; 24:7–10.

265. Ray WA, Griffin MR, Downey W.Benzodiazepines of long and shortelimination half-life and the risk of hipfracture. Journal of the American Medical Association1989; 262:3303–3307.

266. Wang PS, Bohn RL, Glynn RJ, Mogun H,Avorn J. Hazardous benzodiazepineregimens in the elderly: effects of half-life,dosage, and duration on risk of hip fracture.American Journal of Psychiatry 2001; 158:892–898.

267. Hemmelgarn B, Suissa S, Huang A, Boivin J-F, Pinard G. Benzodiazepines use and therisk of motor vehicle crash in the elderly.Journal of the American Medical Association 1997;278:27–31.

268. Roughead EE, Gilbert AL, Primrose JG,Sansom LN. Drug-related hospitaladmissions: A review of Australian studiespublished 1988-1996. Medical Journal ofAustralia 1998; 168:405–408.

269. Commonwealth Department of HumanServices and Health. National Drug StrategyHousehold Survey: Urban Aboriginal and Torres StraitIslander Peoples Supplement 1994. Canberra:Australian Government Publishing Service,1996.

270. Kamien M. The Dark People of Bourke: A Study ofPlanned Social Change. Canberra: AustralianInstitute of Aboriginal Studies, 1978.

271. Stephenson J, Lenz P. Tobacco Smoking byAboriginal People in Wilcannia, Western NSW.Health Report No. 3. Dubbo: HealthPromotion Unit, NSW Department ofHealth, 1990.

272. Perkins J, Sanson-Fisher R, Blunden S,Lunnay D, Redman S, Hensley M. Theprevalence of drug use in urban Aboriginalcommunities. Addiction 1994; 89:1319–1331.

273. Guest C, O’Dea K, Carlin JB, Larkins RG.Smoking in Aborigines and persons of

European descent in southeastern Australia:prevalence and associations with foodhabits, body fat distribution and othercardiovascular risk factors. Australian Journal ofPublic Health 1992; 16(4):397–402.

274. Hoy WE, Norman RJ, Hayhurst BG, PugsleyDJ. A health profile of adults in a NorthernTerritory Aboriginal community, with anemphasis on preventable morbidities.Australian and New Zealand Journal of Public Health1997; 21(2):121–126.

275. Watson C, Fleming J, Alexander K. A Surveyof Drug Use Patterns in Northern Territory AboriginalCommunities: 1986-1987. Darwin: NorthernTerritory Department of Health andCommunity Services, 1988.

276. Forero R, Bauman A, Chen JXC, Flaherty B.Substance use and socio-demographicfactors among Aboriginal and Torres StraitIslander school students in New SouthWales. Australian and New Zealand Journal of PublicHealth 1999; 23:295–300.

277. Dunne MP, Yeo MA, Keane J, Elkins DB.Substance use by Indigenous and non-Indigenous primary school students.Australian and New Zealand Journal of Public Health2000; 24(5):546–549.

278. Gray D, Morfitt B, Ryan K, Williams S. Theuse of tobacco, alcohol and other drugs byyoung Aboriginal people in Albany.Australian and New Zealand Journal of Public Health1997; 21(1):71–76.

279. Gray D, Morfitt B, Williams S, Ryan K,Coyne L. Drug Use and Related Issues Among YoungAboriginal People in Albany. Perth: NationalCentre for Research into the Prevention ofDrug Abuse, Albany AboriginalCorporation, November 1996.

280. McLennan W, Madden R. The Health andWelfare of Australia’s Aboriginal and Torres StraitIslander Peoples. Catalogue No. 4704.0.Canberra: Australian Bureau of Statistics(ABS), 1999.

281. Kamien M. Aborigines and alcohol: intake,effects and social implications in a ruralcommunity in western New South Wales.Medical Journal of Australia 1975; 1:291–298.

282. Smith DI, Singh MN. Survey of Aboriginal AlcoholConsumption and Related Problems at Wiluna,Western Australia. Typescript. Perth: WesternAustralian Alcohol and Drug Authority,1987.

�%� �� !��"�#��$��%��

283. Hunter E, Hall W, Spargo R. Patterns ofalcohol consumption in the KimberleyAboriginal population. Medical Journal ofAustralia 1992; 156:764–768.

284. Gray D, Chikritzhs T. Regional variation inalcohol consumption in the NorthernTerritory. Australian and New Zealand Journal ofPublic Health 2000; 24(1):35–38.

285. Gray D, Saggers S, Atkinson D, Carter M,Loxley W, Hayward D. The Harm ReductionNeeds of Aboriginal People Who Inject Drugs. Perth:National Drug Research Institute, CurtinUniversity of Technology, September 2001.

286. Larson A, Shannon C, Eldridge C.Indigenous Australians who inject drugs:results from a Brisbane study. Drug and AlcoholReview 1999; 18:53–62.

287. Shoobridge J, Vincent N, Allsop S, Biven A.Using Rapid Assessment Methodology to ExamineInjecting Drug Use in an Aboriginal Community.Bedford Park: National Centre for Educationand Training on Addiction, 1998.

288. Brady M. Heavy Metal: The Social Meaning of PetrolSniffing in Australia. Canberra: AboriginalStudies Press; 1992.

289. Freeman P. Petrol sniffing in Amata, SouthAustralia. In Nganampa Health Council HealthReport 1985/86. Alice Springs: NganampaHealth Council, 1986;88–97.

290. Burns C, d’Abbs P, Currie BJ. Patterns ofpetrol sniffing and other drug use in youngmen from an Australian Aboriginalcommunity in Arnhem Land, NorthernTerritory. Drug and Alcohol Review 1995;14(2):159–169.

291. Halfpenny N. Update on the Prevalence of PetrolSniffing on the Anangu Pitjantjatjara Lands:Nganampa Health Council, 2000.

292. Royal Commission into Aboriginal Deathsin Custody, Johnston E, Commissioner. RoyalCommission into Aboriginal Deaths in Custody:National Report, 5 vol. Canberra: AustralianGovernment Publishing Service, 1991.

293. Langton M, Ah Mat L, Moss B, et al. Toomuch sorry business: the Submission of theNorthern Territory Aboriginal Issues Unit.In Commissioner JE (ed) Royal Commission intoAboriginal Deaths in Custody. Canberra:Australian Government Publishing Service,1991.

294. Pearson N. Our Right to Take Responsibility.Cairns, Qld: Noel Pearson and Associates,2000.

295. Measey MA, d’Espaignet ET, Cunningham J.Adult Morbidity and Mortality Due to TobaccoSmoking in the Northern Territory, 1986-1995.Darwin: Territory Health Services, 1998.

296. Kunitz SJ, Santow G, Streatfield R, de CraenT. The Health of Populations on North QueenslandAboriginal Communities: Change and Continuity.Working Paper No. 30. Canberra: NationalCentre for Epidemiology and PopulationHealth, Australia National University, 1992.

297. Peever M, Webster I. Deaths in an outbacktown in New South Wales. Aboriginal HealthProject Information Bulletin 1984; 5:15–20.

298. Smith L, Thomson N, Gray A. AboriginalMortality in New South Wales Country Regions,1980-1981. State Health Publication No.(IDS) 83-169. Sydney: New South WalesDepartment of Health, 1983.

299. Thomson N, Smith L. An analysis ofAboriginal mortality in NSW countryregions, 1980-1981. Medical Journal of Australia1985; 143(Suppl. 1):S49–S54.

300. Khalidi NA. Levels and trends of Aboriginalmortality in Central Australia. In Gray A(ed) A Matter of Life and Death: ContemporaryAboriginal Mortality. Canberra: AboriginalStudies Press, 1990;85–94.

301. Devanesen D, Furber M, Hampton D, et al.Health Indicators in the Northern Territory. Darwin:Northern Territory Department of Health,1986.

302. Dempsey KE, Condon CR. Mortality in theNorthern Territory 1979-97. Darwin: TerritoryHealth Services, 1999.

303. Hunter E. Changing Aboriginal mortalitypatterns in the Kimberley region of WesternAustralia, 1957-86: the impact of deathsfrom external causes. Aboriginal HealthInformation Bulletin 1989; 11:27–32.

304. Hicks DG. Aboriginal Mortality Rates in WesternAustralia. Perth: Health Department ofWestern Australia, 1985.

305. O’Neill D, ed. Aboriginal Morbidity and Mortalityin Western Australia. Perth: Health Departmentof Western Australia; 1985.

306. Unwin E, Thomson N, Gracey M. The Impactof Tobacco Smoking and Alcohol Consumption on

�%"��

Aboriginal Mortality and Hospitalisation in WesternAustralia: 1983-1991. Perth: HealthDepartment of Western Australia, 1994.

307. Brady M, Palmer K. Alcohol in the Outback: AStudy of Drinking in an Aboriginal Community.NARU Monograph. Darwin: AustralianNational University, North AustraliaResearch Unit, 1984.

308. Harris MF. Alcohol and physical disease inBourke Aborigines. Australian Drug and AlcoholReview 1989; 8:79–82.

309. Brady M. Where the Beer Truck Stopped: Drinking ina Northern Australian Town. NARU Monograph.Darwin: Australian National University,North Australia Research Unit, 1988.

310. Flanagan D, Reece G. Alcohol Related ProblemsAmong the Alice Springs Hospital Inpatient Population.Typescript. Sydney: University of Sydney,1982.

311. Harris MF, Sutherland D, Cutter G,Ballangarry L. Alcohol admissions in acountry town. Australian Drug and Alcohol Review1987; 6:195–198.

312. Hunt H. Alcoholism among Aboriginalpeople. Medical Journal of Australia 1981;1(Suppl.):1–3.

313. Patterson KM, Holman CDJ, English DR,Hulse GK, Unwin E. First-time hospitaladmissions with illicit drug problems inIndigenous and non-Indigenous WesternAustralians: an application of record linkageto public health surveillance. Australian andNew Zealand Journal of Public Health 1999;23(5):460–463.

314. Cawte JE. Parameters of kava used as achallenge to alcohol? Australian and New ZealandJournal of Psychiatry 1986; 20:70–76.

315. Mathews JD, Riley MD, Fejo L, Munoz E,Milns NR, Gardner ID, et al. Effects of theheavy usage of kava on physical health:Summary of a pilot survey in an Aboriginalcommunity. Medical Journal of Australia 1998;148(11):548–555.

316. Gray D, Saggers S. Submission to the Inquiry bythe Director of Liquor Licensing into ProposedRestrictions on the Sale of Packaged Liquor in PortHedland and South Hedland. Perth: National DrugResearch Institute, Curtin University ofTechnology, 2002.

317. Task Force on Aboriginal Social Justice,Daube M, (Chairperson). Report of the TaskForce on Aboriginal Social Justice. Perth:Government of Western Australia; 1994.

318. Harding R, Broadhurst R, Ferrante A, LohN. Aboriginal Contact With the Criminal JusticeSystem and the Impact of the Royal Commission intoAboriginal Deaths in Custody. Annandale, NSW:Hawkins Press, 1995.

319. Lyon P. What Everybody Knows About Alice: AReport on the Impact of Alcohol Abuse on the Town ofAlice Springs. Alice Springs: TangentyereCouncil, 1990.

320. Bolger A. Aboriginal Women and Violence: A Reportfor the Criminology Research Council and the NorthernTerritory Commissioner of Police. Darwin:Australian National University, NorthAustralia Research Unit, 1990.

321. The Aboriginal and Torres Strait IslanderWomen’s Task Force on Violence. TheAboriginal and Torres Strait Islander Women’s TaskForce on Violence Report. Brisbane: Departmentof Aboriginal and Torres Strait IslanderPolicy and Development, QueenslandGovernment, 2000.

322. Gordon S, Hallahan K, Henry D. Putting thePicture Together: Inquiry into Response by GovernmentAgencies to Complaints of Family Violence and ChildAbuse in Aboriginal Communities. Perth: State LawPublisher, 2002.

323. Kahn M, Hunter E, Heather N, Tebbutt J.Australian Aborigines and alcohol: a review.Drug and Alcohol Review 1990; 10:351–366.

324. Edmunds M. Doing Business: Socialisation, SocialRelations and social Control in Aboriginal Societies.Royal Commission into Aboriginal Deathsin Custody Discussion Paper No. 2.Canberra: Department of Prehistory andAnthropology, Australian NationalUniversity Press and Australian Institute ofAboriginal and Torres Strait Islander Studies,1990.

325. Hunter E. Aboriginal Health and History.Melbourne: Cambridge University Press;1993.

326. Brady M. Giving Away the Grog: AboriginalAccounts of Drinking and Not Drinking. Canberra:Commonwealth Department of Health andFamily Services, 1995.

�%# �� !��"�#��$��%��

327. Dalton V, Edwards R. Aboriginal Deaths in Prison1980 to 1998: National Overview. Trends andIssues in Crime and Criminal Justice.Canberra: Australian Institute ofCriminology, 1999.

328. Reser J. Aboriginal mental health. In Reid J,Trompf P (eds) The Health of AboriginalAustralia. Sydney: Harcourt Brace Jovanovich,1991.

329. Steenkamp M, Harrison J. Suicide andHospitalised Self-Harm in Australia. InjuryResearch and Statistics Series. AIHWCatalogue No. INJCAT 30. Adelaide:Australian Institute of Health and Welfare,2000.

330. Edwards RW, Madden R. The Health andWelfare of Australia’s Aboriginal and Torres StraitIslander Peoples 2001. Canberra: AustralianBureau of Statistics, August 2001.

331. National Aboriginal Community ControlledHealth Organisation. NACCHO Submission to theHouse of Representatives Standing Committee onFamily and Community Affairs Inquiry into SubstanceAbuse in Australian Communities. Canberra:National Aboriginal Community ControlledHealth Organisation, nd.

332. Swan P, Raphael B. Ways Forward: NationalAboriginal and Torres Strait Islander Mental HealthPolicy National Consultancy Report. Canberra:Australian Government Publishing Service,1995.

333. Department of Immigration andMulticultural Affairs. Population Flows:Immigration Aspects. Canberra: Department ofImmigration and Multicultural Affairs,2000.

334. Rissel C, McLellan L, Bauman A. Socialfactors associated with ethnic differences inalcohol and marijuana use by Vietnamese-,Arabic- and English-speaking youths inSydney, Australia. Journal of Paediatrics and ChildHealth 2000; 36(2):145–152.

335. Rissel C, McLellan L, Bauman A. Factorsassociated with delayed tobacco uptakeamong Vietnamese/Asian and Arabic youthin Sydney, NSW. Australian and New ZealandJournal of Public Health 2000; 24(1):22–28.

336. Maher L, Dixon D, Lynskey M, Hall W.Running the Risks: Heroin, Health and Harm in SouthWest Sydney. Monograph No. 38. Sydney:National Drug and Alcohol Research Centre,1998.

337. Louie R, Krouskos D, Gonzalez M, Crofts N.Vietnamese-speaking injecting drug users inMelbourne: the need for harm reductionprograms. Australian and New Zealand Journal ofPublic Health 1998; 22:481–484.

338. Jukic A, Pino N, Flaherty B. Alcohol and OtherDrug Use, Attitudes and Knowledge Amongst Italian-speakers in Sydney. Sydney: Drug and AlcoholDirectorate, NSW Health Department, 1997.

339. Jukic A, Pino N, Flaherty B. Alcohol and OtherDrug Use, Attitudes and Knowledge amongst Arabic-Speakers in Sydney. Sydney: New South WalesHealth Department., 1996.

340. Bertram S, Flaherty B. Alcohol and Other DrugUse, Attitudes and Knowledge amongst Spanish speakersin Sydney and Wollongong. Sydney, NSW: NSWHealth Department., 1993.

341. Everingham S, Martin A, Flaherty B. Alcoholand Other Drug Use, Attitudes and Knowledge amongstGreek-speakers in Sydney. Sydney, NSW: NSWHealth Department / Drug and AlcoholDirectorate, 1994.

342. Everingham S, Flaherty B. Alcohol and OtherDrug Use, Attitudes and Knowledge amongst Chinese-speakers in Sydney. Sydney, NSW: NSW HealthDepartment of Health / Drug and AlcoholDirectorate, 1995.

343. Thomas DB, Karagas MR. Migrant studies.In Schottenfeld D, Fraumeni JF (eds) CancerEpidemiology and Prevention. 2nd ed. New York:Oxford University Press, 1996.

344. Raftery J, Jones DR, Rosato M. The mortalityof first and second generation Irish migrantsin the UK. Social Science and Medicine 1990;31(5):577–584.

345. Rahav G, Hasin D, Paykin A. Drinkingpatterns of recent Russian immigrants andother Israelis: 1995 national survey results.American Journal of Public Health 1999;89(8):1212–1216.

346. Razum O, Zeeb H, Akgun HS, Yilmaz S.Low overall mortality of Turkish residentsin Germany persists and extpends into asecond generation: merely a healthymigrant effect? Tropical Medicine and InternationalHealth 1998; 3(4):297–303.

347. Gilbert MJ. Alcohol consumption patterns inimmigrant and later generation MexicanAmerican Women. Hispanic Journal ofBehavioural Science 1987; 9(3):299–313.

�%%��

348. Harding S. The incidence of cancers amongsecond generation Irish living in Englandand Wales. British Journal of Cancer 1998;78(7):958–961.

349. Harding S, Balajaran. R. Patterns of mortalityin second generation Irish living in Englandand Wales: longitudinal study. British MedicalJournal 1996; 312:1389–1392.

350. Harding S, Rosato M. Cancer incidenceamong first generation Scottish, Irish, WestIndian and Asian migrants living in Englandand Wales. Ethnicity and Health 1999; 4(1/2):83–92.

351. Young C. Changes in demographicbehaviour of migrants in Australia and thetransition between generations. PopulationStudies 1991; 45:67–89.

352. Scribner R. Paradox as paradigm - the healthoutcomes of Mexican-Americans. AmericanJournal of Public Health 1996; 86(303–305).

353. British Medical Journal. Style matters:ethnicity, race and culture. British MedicalJournal 1996; 312:1094.

354. Cavelaars AEJM, Kunst AE, Geurts JJM,Crialesi R, Helmert U, Lahelma E, et al.Educational differences in smoking:international comparison. British MedicalJournal 2000; 320:1102–1107.

355. Drever F, Whitehead M, Roden M. Currentpatterns and trends in male mortality bysocial class, based on occupation. PopulationTrends 1996; 86:15–20.

356. Glendinning A, Shucksmith J, Hendry L.Social class and adolescent smokingbehaviour. Social Science and Medicine 1994;38(10):1449–1660.

357. Lee AL, Crombie IK, Smith WCS, Tunstall-Pedoe HD. Cigarette smoking andemployment status. Social Science and Medicine1991; 33(11):1309–1312.

358. Romelsjo A, Lundberg M. Changes in thesocial class distribution of moderate andhigh alcohol consumption and of alcohol-related disabilities in Stockholm and inSweden. Addiction 1996; 91(9):1307–1323.

359. Rossow I, Amundsen. A. The disadvantageof being advantaged? On a social gradient inexcess mortality among alcohol abusers.Addiction 1996; 91(12):1821–1829.

360. Townsend J, Roderick P, Cooper J. Cigarettesmoking by socio-economic group, sex andage: effects of price, income and healthpublicity. British Medical Journal 1994;309:923–927.

361. Edwards G, Kyle E, Nicholls P, Taylor C.Alcoholism and correlates of mortality:implications for epidemiology. Journal ofStudies on Alcohol 1978; 39(9):1607–1617.

362. Harrison L, Gardiner E. Do the rich reallydie young? Alcohol-related mortality andsocial class in Great Britain, 1988 - 1994.Addiction 1999; 94(12):1871–1880.

363. Jarvis MJ, Wardle J. Social patterning ofindividual health behaviours: the case ofcigarette smoking. In Marmot M, WilkinsonRG (eds) Social Determinants of Health. NewYork: Oxford, 1999;240–255.

364. Mullen K. Religion and health: a review ofthe literature. International Journal of Sociology andSocial Policy 1990; 10(85–95).

365. Mullen K, Williams R, Hunt K. Irishdescent, religion and alcohol and tobaccouse. Addiction 1996; 91(2):243–254.

366. Abbotts J, Williams R, Ford G, Hunt K,West P. Morbidity and Irish-Catholicdescent in Britain: relating healthdisadvantage to behaviour. Ethnicity and Health1999; 4(4):221–230.

367. Francis LJ, Mullen K. Religiosity andattitudes towards drug use among 13 to 15year olds in England. Addiction 1993;88(5):665–672.

368. Klonoff EA, Landrine H. Acculturation andalcohol use among blacks: the benefits ofremaining culturally traditional. WesternJournal of Black Studies 1999; 23(4):211–216.

369. Caetano R, Katsunas L. Changes in drinkingpatterns among whites, blacks andHispanics, 1984 - 1992. Journal of Studies onAlcohol 1995; 56:558–565.

370. Mueser KT, Kavanagh D. Treatingcomorbidity of alcohol problems andpsychiatric disorder. In Heather N, PetersTJ, Stockwell T (eds) International Handbook ofAlcohol Dependence and Problems. Chichester: JohnWiley and Sons, 2001;627–648.

371. Khantzian EJ. The self-medicationhypothesis of addictive disorders: Focus onheroin and cocaine dependence. American

�%' �� !��"�#��$��%��

Journal of Psychiatry 1985; 142(11):1259–1264.

372. Kessler R, Nelson C, McGonagle K, EdlundM, Frank R, Leaf P. The epidemiology of co-occurring addictive and mental disorders:implications for prevention and serviceutilization. American Journal of Orthopsychiatry1996; 66:17–31.

373. Hall W, Johnston L, Donnelly N.Epidemiology of cannabis use and itsconsequences. In Kalant H, Corrigall WA,Hall W, Smart RG (eds) The Health Effects ofCannabis. Toronto: Centre for Addiction andMental Health, 1999;71–125.

374. Ashton H. Benzodiazepine withdrawal: anunfinished story. British Medical Journal 1984;288:1135–1140.

375. Rehm J, Room R, Monteiro M, Gmel G,Graham K, Rehn N, et al. Alcohol as a riskfactor for burden of disease. In WorldHealth Organization (ed) ComparativeQuantification of Health Risks: Global and RegionalBurden of Disease Due to Selected Major Risk Factors.Geneva: World Health Organization, 2004.

376. Kavanagh DJ. Comorbidity in mental healthand substance use: causes, prevention andtreatment. Paper presented at NationalWorkshop on Comorbidity. Canberra, 6–7March 2000.

377. Andrews G, Hall W, Teesson M, HendersonS. The Mental Health of Australians. Canberra:Commonwealth Department of Health andAged Care, April 1999.

378. Moeller FG, Dougherty DM. Antisocialpersonality disorder, alcohol, andaggression. Alcohol Research and Health 2001;25(1):5–11.

379. Williams P. Windows of opportunity forintervention and prevention in a diversionsenvironment: evidence from DUCO. Paperpresented at 2nd Australasian Conference onDrugs Strategy. Perth: Western AustralianPolice Service, 7–9 May 2002.

380. Dolan K, Crofts N. A review of riskbehaviours, transmission and prevention ofblood borne viral infections in Australianprisons. In Shewan D, Davies JB (eds) DrugUse and Prisons. An International Perspective.Amsterdam: Harwood Academic Publishers,2000;215–232.

381. Jurgens R. HIV/AIDS and drug use inprisons: moral and legal responsibilities ofprison. In Shewan D, Davies JB (eds) DrugUse and Prisons. An International Perspective.Amsterdam: Harwood Academic Publishers,2000;1–26.

382. Watson J. Hepatitis C. A Study of Prevalence in WAPrisons. Perth: Hepatitis Council WesternAustralia, 2002.

383. Brisbane City Council. The Lord Mayor’sIllicit Drug Taskforce Report 1999. BrisbaneCity Council, 1999. Viewed 30 October2001, <www.brisbane.qld.gov.au/council_at_work/health_safety/illicit_drugs/pdf/report1.pdf>.

384. Briscoe S, Donnelly N. Temporal and RegionalAspects of Alcohol-Related Violence and Disorder.Report No. 1. New South Wales: NationalDrug Research Institute, Curtin Universityof Technology and NSW Bureau of CrimeStatistics and Research, May 2001.

385. Graham K, West P. Alcohol and crime:examining the link. In Heather N, Peters TJ,Stockwell T (eds) International Handbook ofAlcohol Dependence and Problems. Chichester: JohnWiley and Sons, 2001;439–470.

386. Matthews S, Chikritzhs T, Catalano P,Stockwell T, Donath S. Trends in Alcohol-RelatedViolence in Australia, 1991/92-1999/00.National Alcohol Indicators Draft BulletinNo. 5. Perth: National Drug ResearchInstitute, Curtin University of Technology,January 2002.

387. Makkai T. Illicit drugs and crime. In GraycarA, Grabosky P (eds) Cambridge Handbook ofAustralian Criminology: Cambridge UniversityPress, 2002.

388. Makkai T. Drugs and property crime. InABCI Australian Illicit Drug Report, 1997-1998.Canberra: ABCI, 1999;105–113.

389. Makkai T. Drugs and crime: what does theevidence show? Paper presented at 2ndAustralasian Conference on Drugs Strategy.Perth: Western Australian Police Service, 7–9 May 2002.

390. Christie P. The Effects of Cannabis Legislation inSouth Australia on Levels of Cannabis Use. Adelaide:Drug and Alcohol Services Council, 1991.

391. Lenton S, Christie P, Humeniuk R, BrooksA, Bennet M, Heale P. Infringement VersusConviction: The Social Impact of a Minor Cannabis

�%(��

Offence Under a Civil Penalties System and StrictProhibition in Two Australian States. Canberra:Commonwealth Department of Health andAged Care, and the National Drug Strategy,Australian Government Publishing Service,1999.

392. Abbey A, Zawacki T, Buck PO, Clinton AM,McAuslan P. Alcohol and sexual assault.Alcohol Research and Health 2001; 25(1):43–51.

393. Cameron M. Young Men and Violence Prevention.Trends and Issues in Crime and CriminalJustice Series Report No. 154. Canberra:Australian Institute of Criminology, June2000.

394. Mazza D, Dennerstein L, Ryan V. Physical,sexual and emotional violence againstwomen: A general practice-basedprevalence study. Medical Journal of Australia1996; 164:14–17.

395. Bennett LW. Substance abuse and thedomestic assault of women. Journal ofAddictions Nursing 1998; 10(1):40–50.

396. Lang E, Stockwell T, Rydon P, Gamble C.Drinking Settings, Alcohol Related Harm, and Supportfor Prevention Policies. Technical Report. Perth:National Centre for Research into thePrevention of Drug Abuse, CurtinUniversity of Technology, 1992.

397. Weatherburn D, Topp L, Midford R, AllsopS. Drug Crime Prevention and Mitigation: A LiteratureReview and Research Agenda. Sydney: New SouthWales Bureau of Crime Statistics andResearch, 2000.

398. Johnson D. Age of Illicit Drug Initiation. Trendsand Issues in Crime and Criminal JusticeSeries Report No. 201. Canberra: AustralianInstitute of Criminology, March 2001.

399. Friedman AS, Glickman NW, Morrissey MR.What mothers know about theiradolescents’ alcohol/drug use and problemsand how mothers react to finding out aboutit. Journal of Drug Education 1988; 18:155–167.

400. Orford J. Control, confront or collude: howfamily and society respond to excessivedrinking. British Journal of Addiction 1992;87:1513–1526.

401. Velleman R, Bennett G, Miller T, Orford J,Rigby K, Tod A. The families of problemdrug users: a study of 50 close relatives.Addiction 1993; 88:1281–1290.

402. Orford J, Natera G, Davies J, Nava A, MoraJ, Rigby K, et al. Tolerate, engage orwithdraw: a study of the structure offamilies coping with alcohol and drugproblems in South West England andMexico City. Addiction 1998; 93:1799–1813.

403. Loxley W, Papadapoulos L. Forging Allianceswith Parent Organisations. A report to theResearch and Development Committee,Division of Health Sciences, CurtinUniversity of Technology. Perth: NationalCentre for Research into the Prevention ofDrug Abuse, February 1999.

404. Allsop S, Bush R, Phillips M, Midford R,Fowler G, Wooden M, et al. Alcohol and OtherDrugs in the Australian Workplace: A CriticalLiterature Review. Adelaide: National Centre forEducation and Training on Addiction, 1997.

405. Normand J, Lempert RO, O’Brien CP. Underthe Influence? Drugs and the American Work Force.Washington: National Academy Press;1994.

406. Hutcheson GD, Henderson MM, Davies JB.Alcohol in the Workplace: Cost and Responses.Research Series No. 59. Sheffield:Department for Education and Employment,August 1995.

407. Midford R. The nature and extent of drug-related harm in the community and theimplications for the workplace. In Allsop S,Phillips M, Calogero C (eds) Drugs and Work.Responding to Alcohol and Other Drug Problems in theAustralian Workplace. Melbourne: IPCommunications, 2001;42–55.

408. English D, Holman CDJ, Milne E, WinterMG, Hulse GK, Godde JP, et al. TheQuantification of Drug Caused Mortality in Australia.Canberra: Commonwealth Department ofHuman Services and Health, 1995.

409. Chikritzhs T, Stockwell T, Heale P, Dietze P,Webb M. Trends in Alcohol-Related Road Injury inAustralia, 1990-1997. National AlcoholIndicators Project Technical Report No. 2.Perth: National Drug Research Institute,Curtin University of Technology, 2000.

410. Chesher G. Cannabis and road safety. anoutline of research studies to examine theeffects of cannabis on driving skills andactual driving performance. In Parliament ofVictoria Road Safety Committee (ed) Inquiryinto the Effects of Drugs (other than alcohol) on Road

�') �� !��"�#��$��%��

Safety. Melbourne: Road Safety Committee,Parliament of Victoria, 1995;67–96.

411. Robbe HWJ. Influence of Marijuana on Driving.Maastricht: Institute for HumanPsychopharmacology, University ofLimburg; 1994.

412. Road Safety Committee. The Effects of Drugs(Other than Alcohol) on Road Safety. Report No.34. Melbourne: Parliament of Victoria, May1995.

413. Hunter C, Lokan R, Longo M, White J,White M. The Prevalence and Role of Alcohol,Cannabinoids, Benzodiazepines and Stimulants in Non-Fatal Road Crashes. Adelaide: Department ofClinical and Experimental Pharmacology,University of Adelaide, 1998.

414. Poyser C, Makkai T, Norman L, Mills L. DrugDriving Among Police Detainees in Three States ofAustralia: Final Report: Australian Institute ofCriminology and Queensland Police Service,2002.

415. Lynch J. Social epidemiology: someobservations about the past, present andfuture. Australasian Epidemiologist 2000; 7.3:7–15.

416. Blakely T. Counterfactual challenges tosocial epidemiology. Australasian Epidemiologist2000; 7(3):28–31.

417. Marmot M. Social determinants of health:from observation to policy. Medical Journal ofAustralia 2000; 172:379–382.

418. Marmot M, Wilkinson RG, eds. SocialDeterminants of Health. New York: Oxford;1999.

419. Spooner C, Hall W, Lynskey M. StructuralDeterminants of Youth Drug Use. Woden, ACT:Australian National Council on Drugs,2001.

420. Acheson D, Barker D, Chambers J, GrahamH, Marmot M, Whitehead M. IndependentInquiry into Inequalities in Health Report. UnitedKingdom: Secretary of State for Health,November 1998.

421. Wilkinson R. Commentary: liberty,fraternity, equality. International Journal ofEpidemiology 2002; 31:538–543.

422. Deaton A. Commentary: the convolutedstory of international studies of inequalityand health. International Journal of Epidemiology2002; 31:546–549.

423. Turrell G. Income inequality and health: insearch of fundamental causes. In EckersleyR, Dixon J, Douglas B (eds) The Social Originsof Health and Well-being. Cambridge:Cambridge University Press, 2001;83–104.

424. Mackenbach JP. Income inequality andhealth. British Medical Journal 2002; 324:1–2.

425. Deaton A. Health, Inequality, and EconomicDevelopment. CMH Working Paper Series.Paper No. WG1:3. Geneva: WHOCommission on Macroeconomics andHealth, 2001.

426. Brunner E, Marmot M. Social organization,stress, and health. In Marmot M, WilkinsonRG (eds) Social Determinants of Health. NewYork: Oxford, 1999;17–42.

427. Kawachi I. Why social epidemiology?Australasian Epidemiologist 2000; 7.3:5–6.

428. Najman JM. A general model of the socialorigins of health and well-being. InEckersley R, Dixon J, Douglas B (eds) TheSocial Origins of Health and Well-being.Cambridge: Cambridge University Press,2001;73–82.

429. Keating DP, Hertzman C. Modernity’sparadox. In Keating DP, Hertzman C (eds)Developmental Health and the Wealth of Nations.Social, Biological, and Educational Dynamics. NewYork: Guilford Press, 1999;1–17.

430. Daniels N, Kennedy BP, Kawachi I. Whyjustice is good for our health: The socialdeterminants of health inequalities. Daedalus1999; 128(4):215–251.

431. Wilkinson RG. Unhealthy Societies. The Afflictionsof Inequality. London: Routledge; 1996.

432. Mackenbach JP, Cavelaars E, Kunst AE,Groenhof F. Socioeconomic inequalities incardio-vascular disease mortality. EuropeanHeart Journal 2000; 21:1141–1151.

433. Spooner C, Mattick R, Noffs W. The Natureand Treatment of Adolescent Substance Abuse.Monograph No. 40. Sydney: National Drugand Alcohol Research Centre, 1999.

434. Kawachi I, Kennedy BP, Glass R. Socialcapital and self-rated health: A contextualanalysis. American Journal of Public Health 1999;89(8):1187–1192.

435. Broom D. Gender and health. In Germov J(ed) Second Opinion: An Introduction to Health

�'��

Sociology. Melbourne: Oxford UniversityPress, 1999;39–56.

436. Shaw M, Dorling D, Davey Smith G.Poverty, social exclusion, and minorities. InMarmot M, Wilkinson RG (eds) SocialDeterminants of Health. Oxford: OxfordUniversity Press, 1999;211–239.

437. Spooner C, Mattick R, Howard J. The Natureand Treatment of Adolescent Substance Abuse. FinalReport of the Adolescent TreatmentResearch Project. Monograph No. 26.Sydney: National Drug and AlcoholResearch Centre, 1996.

438. Eckersley R. Culture, health and well-being.In Eckersley R, Dixon J, Douglas B (eds) TheSocial Origins of Health and Wellbeing. Cambridge:Cambridge University Press, 2001;51–70.

439. Cashmore J. Family, early development andthe life course: common risk and protectivefactors in pathways to prevention. InEckersley R, Dixon J, Douglas B (eds) TheSocial Origins of Health and Well-being.Cambridge: Cambridge University Press,2001;216–224.

440. Leeder S. Social epidemiology and healthinequalities in Australia (and beyond).Australasian Epidemiologist 2000; 7(3):2–4.

441. Eckersley R, Dixon J, Douglas B, eds. TheSocial Origins of Health and Well-being.Cambridge: Cambridge University Press;2001.

442. Najman JM, Smith GD. The embodiment ofclass-related and health inequalities:Australian policies. Australian and New ZealandJournal of Public Health 2000; 24(1):3–4.

443. Najman JM. Class inequalities in health andlifestyle. In Waddell C, Petersen AR (eds)Just Health. Inequality in Illness, Care and Prevention.Melbourne: Churchill Livingstone,1994;27–46.

444. Turrell G, Mathers CD. Socioeconomicstatus and health in Australia. Medical Journal ofAustralia 2000; 172:434–438.

445. Kawachi I, Kennedy BP, Wilkinson RG.Crime: social disorganization and relativedeprivation. Social Science and Medicine 1999;48(6):719–731.

446. Saggers S, Gray D. Aboriginal Health and Society:The Traditional and Contemporary Struggle for BetterHealth. Sydney: Allen and Unwin; 1991.

447. House of Representatives, StandingCommittee on Family and CommunityAffairs. Health is Life: Report on the Inquiry intoIndigenous Health. Canberra: House ofRepresentatives, Standing Committee ofFamily and Community Affairs, 2000.

448. McLennan W, (Australian Statistician). Censusof Population and Housing: Aboriginal and TorresStrait Islander People, Australia 1996,. CatalogueNo. 2034.0. Canberra: Australian Bureau ofStatistics, 1998.

449. Julian R. Ethnicity, health andmulticulturalism. In Germov J (ed) Secondopinion. An Introduction to Health Sociology.Melbourne: Oxford University Press,1999;77–95.

450. Wilkinson RG. Putting the picture together:prosperity, redistribution, health, andwelfare. In Marmot M, Wilkinson RG (eds)Social Determinants of Health. New York:Oxford, 1999;256–274.

451. Corin E. The cultural frame: context andmeaning in the construction of health. InAmick BC, Levine S, Tarlov AR, ChapmanWalsh D (eds) Society and Health. New York:Oxford University Press, 1995.

452. Bartley M, Ferrie J, Montgomery SM. Livingin a high-unemployment economy:understanding the health consequences. InMarmot M, Wilkinson RG (eds) SocialDeterminants of Health. New York: Oxford,1999;81–104.

453. Stockwell T, Dyskin E, O’Connor J. TheImpact of the Tobacco Control Act (WA) 1990 onPerth School Children 1991-1994. Perth:National Centre for Research into thePrevention of Drug Abuse, CurtinUniversity of Technology, 1995.

454. Stockwell T, Daly A, Phillips A, Masters L,Midford R, Gahegan M, et al. Total versushazardous per capita alcohol consumptionas predictors of acute and chronic alcoholrelated harm. Contemporary Drug Problems 1996;23(3):441–464.

455. Heale P, Chikritzhs T, Jonas H, Stockwell T,Dietze P. Estimated alcohol-caused deaths inAustralia, 1990-97. Drug and Alcohol Review2002; 21(2):121–129.

456. Chikritzhs T, Stockwell T, Heale P, Dietze P,Webb M. Trends in Alcohol-Related Road Injury inAustralia, 1990-1997. National Alcohol

�'� �� !��"�#��$��%��

Indicators Project Bulletin No. 2. Perth:National Drug Research Institute, CurtinUniversity of Technology, May 2000.

457. Homel R, Elias G, Hay I. Developmentalprevention in a disadvantaged community.In Eckersley R, Dixon J, Douglas B (eds) TheSocial Origins of Health and Wellbeing. Cambridge:Cambridge University Press, 2001;269–279.

458. Corin E. The social and cultural matrix ofhealth and disease. In Evans RG, Barer ML,Marmour TR (eds) Why are Some People Healthyand Others Not? The Determinants of Health ofPopulations. New York: Aldine de Gruyter,1994.

459. Toumbourou J. Implementing Communities thatCare in Australia: A Community MobilisationApproach to Crime Prevention. Report No. 122.Canberra: Australian Institute ofCriminology, July 1999.

460. Bond L, Carlin JB, Thomas L, Rubin K,Patton G. Does bullying cause emotionalproblems? A prospective study of youngteenagers. British Medical Journal 2001;323:480–484.

461. Williams B, Sanson A, Toumbourou J, SmartD. Patterns and Predictors of Teenagers use of Licit andIllicit Substances in the Australian TemperamentProject Cohort2000.

462. Fergusson DM, Horwood LJ, Lynskey MT.The effects of unemployment on psychiatricillness during young adulthood. PsychologicalMedicine 1997; 27(2):371–381.

463. Public Health Division Victorian Departmentof Human Services. Drugs in a MulticulturalCommunity: An Assessment of Involvement.Melbourne: Public Health Division VictorianDepartment of Human Services, 2000.

464. Groves R. The lost families: refugees andInternational Year of the Family. DevelopmentBulletin 1993; 29:8–9.

465. Kumpfer KL, Olds DL, Alexander JF, ZuckerRA, Gary LE. Family etiology of youthproblems. In Ashery RS, Robertson EB,Kumpfer KL (eds) Drug abuse prevention throughfamily interventions, NIDA Research Monograph 177.Rockville, MD: National Institute on DrugAbuse, 1998;42–77.

466. Madden R, (Acting Australian Statistician).National Aboriginal and Torres Strait Islander Survey

1994: Detailed Findings. Catalogue No. 4190.0.Canberra: Australian Bureau of Statistics,1995.

467. Cunningham J. Occasional Paper: CigaretteSmoking Among Indigenous Australians, 1994.Catalogue No. 4701.0. Canberra: AustralianBureau of Statistics, 1997.

468. Ivers R. Indigenous Australians and Tobacco: ALiterature Review. Darwin: Menzies School ofHealth Research, 2001.

469. Hogg RS. Variability in behavioural riskfactors for heart disease in an AustralianAboriginal community. Journal of BiosocialScience 1994; 19(2):159–173.

470. Strauss A, Corbin J. Basics of Qualitative Research:Grounded Theory Procedures and Techniques.Newbury Park, CA: Sage Publications; 1990.

471. National Mental Health Strategy. Promotion,Prevention and Early Intervention for Mental Health.Canberra: Department of Health and AgedCare, 1999.

472. National Crime Prevention - Attorney-General’s Department. Pathways to Prevention:Developmental and Early Intervention Approaches toCrime in Australia. Appendices. Canberra:National Crime Prevention - Attorney-General’s Department, 1999.

473. Coie JD, Watt NF, West SG, Hawkins JD,Asarnow JR, Markman HJ, et al. The scienceof prevention: a conceptual framework andsome directions for a national researchprogram. American Psychologist 1993;48:1013–1022.

474. Kellam SG, Rebok GW. Buildingdevelopmental and etiological theorythrough epidemiologically based preventiveintervention trials. In McCord J, TremblayRE (eds) Preventing Antisocial Behavior:Interventions from Birth through Adolescence. NewYork: Guilford Press, 1992;162–195.

475. Stockwell T. Harm reduction, drinkingpatterns and the NHMRC DrinkingGuidelines. Drug and Alcohol Review 2001;20:121–129.

476. Newcomb MD, Maddahian E, Bentler PM.Risk factors for drug use amongadolescents: concurrent and longitudinalanalyses. American Journal of Public Health 1986;76:525–531.

�'��

477. Glantz MD, Leshner AI. Drug abuse anddevelopmental psychopathology. Developmentand Psychopathology 2000; 12:795–814.

478. Mitchell P, Sanson A, Spooner C, CopelandJ, Vimpani G, Toumbourou JW, et al. TheRole of Families in the Development, Identification,Prevention and Treatment of Illicit Drug Problems. Aliterature review prepared for the NationalIllicit Drug Strategy Working Committeeand the National Health and MedicalResearch Council. Canberra:Commonwealth of Australia, March 2001.

479. Catalano RF, Hawkins JD. The SocialDevelopment Model: a theory of antisocialbehavior. In Hawkins JD (ed) Delinquency andCrime: Current Theories. New York: CambridgeUniversity Press, 1996;149–197.

480. Lynskey MT, Fergusson DM. Childhoodconduct problems, attention deficitbehaviors, and adolescent alcohol, tobacco,and illicit drug use. Journal of Abnormal ChildPsychology 1995; 23(3):281–303.

481. Fergusson DM, Horwood LJ, Lynskey M.The childhoods of multiple problemadolescents: a 15-year longitudinal study.Journal of Child Psychology and Psychiatry 1994;35(6):1123–1140.

482. McCain MN, Mustard JF. Early Years Report:Reversing the Real Brain Drain. Final Report.Ontario: The Canadian Institute forAdvanced Research, 1999.

483. Silburn SR, Zubrick SR, Garton A, Gurrin L,Burton P, Dalby R, et al. Western AustralianChild Health Survey: Family and Community Health.Perth: Australian Bureau of Statistics andTVW Telethon Institute for Child HealthResearch, 1996.

484. Fergusson DM, Horwood LJ, Lynskey MT.The prevalence and risk factors associatedwith abusive or hazardous alcoholconsumption in 16-year-olds. Addiction1995; 90:935–946.

485. Toumbourou JW, Gregg ME. Impact of anempowerment-based parent educationprogram on the reduction of youth suiciderisk factors. Journal of Adolescent Health 2002;31(3):279–287.

486. Weinberg NZ. Adolescent substance abuse:a review of the past ten years. Journal of theAmerican Academy of Child and Adolescent Psychiatry1998; 37(3):252–261.

487. Pape H, Hammer T. How does youngpeople’s alcohol consumption changeduring the transition to early adulthood? Alongitudinal study of changes at aggregateand individual level. Addiction 1996;91(9):1345–1357.

488. Holdcroft LC, Iacono WG. Cohort effects ongender differences in alcohol dependence.Addiction 2002; 97:1025–1036.

489. Pulkkinen L, Pitkanen T. A prospective studyof the precursors to problem drinking inyoung adulthood. Journal of Studies on Alcohol1994; 55:578–587.

490. McGue M, Iacono WG, Legrand LN, ElkinsI. Origins and consequences of age at firstdrink. II. Familial risk and heritabilty.Alcoholism: Clinical and Experimental Research2001; 25(8):1166–1173.

491. Moscato BS, Russell M, Zielezny M, BrometE, Egri G, Mudar P, et al. Gender differencesin the relation between depressivesymptoms and alcohol problems: alongitudinal perspective. American Journal ofEpidemiology 1997; 146(11):966–974.

492. Fergusson DM, Horwood LJ. Does cannabisuse encourage other forms of illicit druguse? Addiction 2000; 95(4):505–520.

493. Prior M, Sanson A, Smart D, Oberklaid F.Pathways from Infancy to Adolescence: AustralianTemperament Project 1983-2000. Melbourne:Australian Institute of Family Studies, 2000.

494. Werner E, Smith R. Overcoming the Odds: High-Risk Children from Birth to Adulthood. New York:Cornell University; 1992.

495. Rutter M. Psychosocial resilience andprotective mechanisms. American Journal ofOrthopsychiatry 1987; 57(3):316–331.

496. Henry B, Feehan M, McGee R, Stanton W.The importance of conduct problems anddepressive symptoms in predictingadolescent substance use. Journal of AbnormalChild Psychology 1993; 21(5):469–480.

497. Huba GJ, Bentler PM. The role of peer andadult models for drug taking at differentstages in adolescence. Journal of Youth andAdolescence 1980; 9:449–465.

498. Hansen WB, Graham JW. Preventivealcohol, marijuana and cigarette use amongadolescents: peer pressure resistancetraining versus establishing conservative

�'� �� !��"�#��$��%��

norms. Preventive Medicine 1991; 20(3):414–430.

499. Vinson T. Unequal in Life: The Distribution of SocialDisadvantage in Victoria and New South Wales.Richmond, Vic: Jesuit Social Services, 1999.

500. Smart R, Adlaf E, Walsh G. Neighbourhoodsocio-economic factors in relation tostudent drug use and programs. Journal ofChild and Adolescent Substance Abuse 1994; 3:37–46.

501. Bobashev GV, Anthony JC. Clusters ofmarijuana use in the United States. AmericanJournal of Epidemiology 1998; 148:1168–1174.

502. Petronis KR, Anthony JC. Paths of“transmission” of cocaine use and perceivedrisk of cocaine use within U.S.neighbourhoods. American Journal ofEpidemiology 1999; 149:S43.

503. Ward Hinds M. Impact of local ordinancebanning tobacco sale to minors. Public HealthReports 1992; 107:355–358.

504. Forster JL, Murray DM, Wolfson M, BlaineTM, Wagenaar AC, Hennrikus DJ. Theeffects of community policies to reduceyouth access to tobacco. American Journal ofPublic Health 1998; 88:1193–1198.

505. Jason LA, Ji PY, Anes MD, Birkhead SH.Active enforcement of cigarette control lawsin the prevention of cigarette sales tominors. Journal of the American Medical Association1991; 266:3159–3161.

506. Perry CL, Williams CL, Forster JL, WolfsonM, Wagenaar AC, Finnegan JR, et al.Background, conceptualization and designof a community-wide research program onadolescent alcohol use: Project Northland.Health Education Research Theory and Practice 1993;8(125–136).

507. Flynn BS, Worden JK, Secker-Walker RH,Pirie PL, Badger GJ, Carpenter JH. Long-term responses of higher and lower riskyouths to smoking preventioninterventions. Preventive Medicine 1997;26:389–394.

508. Brody GH, Forehand R. Prospectiveassociations among family form, familyprocesses, and adolescents’ alcohol and druguse. Behaviour Research Therapy 1993;31(6):587–593.

509. Hawkins JD, Lishner D, Catalano RF.Childhood predictors and the prevention ofadolescent substance abuse. In Jones CL,Battjes RJ (eds) Etiology of Drug Abuse:Implications for Prevention. NIDA Research MonographNo. 56. Washington, DC: U.S. GovernmentPrinting Office, 1985;75–126.

510. Loeber R, Dishion TJ. Early predictors ofmale delinquency: A review. PsychologicalBulletin 1983; 94:68–99.

511. Spoth R, Redmond C, Hockaday C, Yoo S.Protective factors and young adolescenttendency to abstain from alcohol use: amodel using two waves of interventionstudy data. American Journal of CommunityPsychology 1996; 24:749–770.

512. Hussong AM, Chassin L. Substance useinitiation among adolescent children ofalcoholics: testing protective factors. Journalfor Studies in Alcohol 1997; 58:272–279.

513. Wills TA, DuHamel K, Viccaro D. Activitiyand mood temperament as predictors ofadolescent substance use: test of a self-regulation mediational model. Journal ofPersonality and Social Psychology 1995;68(5):901–916.

514. Campo AT, Rohner RP. Relationshipsbetween perceived parental acceptance-rejection. Psychological adjustment, andsubstance abuse among young adults. ChildAbuse and Neglect 1992; 16:429–440.

515. Dishion TJ, Capaldi D, Spracklen KM, Li F.Peer ecology of male adolescent drug use.Development and Psychopathology 1995; 7:803–824.

516. Blackson CT. Temperament and IQ mediatethe effects of family history of substanceabuse and family dysfunction on academicachievement. Journal of Clinical Psychology1995; 51(1):113–122.

517. Steele RG, Forehand R, Armistead L, BrodyG. Predicting alcohol and drug use in earlyadulthood: the role of internalizing andexternalizing behavior problems in earlyadolescence. American Journal of Orthopsychiatry1995; 65(3):380–388.

518. Simons RL, Whitbeck LB, Conger RD,Melby JN. The effect of social skills, values,peers and depression on adolescentsubstance use. Journal of Early Adolescence 1991;11:466–481.

�'"��

519. Kandel DB, Andrews K. Processes ofadolescent socialization by parents andpeers. International Journal of the Addictions 1987;22:319–342.

520. Dielman TE. School-based research on theprevention of adolescent alcohol use andmisuse: methodological issues andadvances. Journal of Research on Adolescence 1994;4(2):271–293.

521. Orford J. Excessive Appetites. A Psychological Viewof Addictions. 2 ed. Chichester: John Wileyand Sons; 2001.

522. Aitken SS, DeSantis J, Harford TC, Fe CacesM. Marijuana use among adults. Alongitudinal study of current and formerusers. Journal of Substance Abuse 2000; 12:213–226.

523. White HR, Johnson V, Buyske S. Parentalmodeling and parenting behavior effects onoffspring alcohol and cigarette use. Agrowth curve analysis. Journal of Substance Abuse2000; 12:287–310.

524. Homel R, McIlwain G, Carvolth R. Creatingsafer drinking environments. In Heather N,Peters TJ, Stockwell T (eds) InternationalHandbook of Alcohol Dependence and Problems.Chichester: John Wiley and Sons,2001;721–740.

525. Bryant M, Williams P. Alcohol and Other Drug-related Violence and Non-reporting. Trends andIssues in Crime and Criminal Justice SeriesReport No. 171. Canberra: AustralianInstitute of Criminology, 2000.

526. Bond L, Thomas L, Toumbourou J, PattonG, Catalano R. Improving the Lives of YoungVictorians in our Community: A Summary Report.Melbourne: Community Care Division,Victorian Department of Human Sevices,2000.

527. Hawkins JD, Arthur MW, Catalano RF.Preventing substance abuse. In FarringtonD, Tonry M (eds) Crime and Justice: A Review ofResearch: Vol. 18. Crime Prevention. Chicago:University of Chicago Press, 1995.

528. Loeber R, Stouthamer-Loeber MS, VanKammen W, Farrington DP. Initiation,escalation, and desistance in juvenileoffending and their correlates. Journal ofCriminal Law and Criminology 1991; 82:36–82.

529. Rose G. Sick individuals and sickpopulations. International Journal of Epidemiology1985; 14:32–38.

530. Stockwell T, Hawks D, Lang E, Rydon P.Unravelling the preventive paradox foracute alcohol problems. Drug and Alcohol Review1996; 15(1):7–15.

531. Gmel G, Klingemann S, Muller R, BrennerD. Revising the preventive paradox: TheSwiss case. Addiction 2001; 96(2):273–284.

532. Chabon SS, Lee-Wilkerson D, Green TJ.Drug-exposed infants and children: livingwith a lethal legacy. Clinics in CommunicationDisorders 1992; 2(2):32–51.

533. Ettlinger T. In harm’s way: recognizing andaddressing alcohol risk for ruraldisadvantaged pregnant mothers. PublicHealth Nursing 2000; 17(3):207–210.

534. Olds DL, Henderson CR, Kitzman HJ,Eckenrode JJ, Cole RE, Tatelbaum RC.Prenatal and infancy home visitation bynurses: recent findings. The Future of Children1999; 9(1):44–65.

535. Olds DL, Eckenrode JJ, Henderson CR.Long-term effects of home visitation onmaternal life course and child abuse andneglect. Fifteen-year follow-up of arandomized trial. Journal of American MedicalAssociation 1997; 278:637–643.

536. Olds DL, Henderson CR, Cole RE. Long-term effects of nurse home visitation onchildren’s criminal and antisocial behavior.Fifteen-year follow-up of a randomizedtrial. Journal of American Medical Association 1998;280:1238–1244.

537. World Health Organization. Tobacco FreeInitiative. International Consultation on EnvironmentalTobacco Smoke (ETS) and Child Health. Geneva,Switzerland: World Health Organization,11–14 January 1999.

538. Rikard-Bell G, Waters E, Ward J. Smokingrelated issues and paediatricians: asystematic review of the literature. 2001.

539. Waters E, Campbell R, Webster P, SpencerNJ. The effects of smoking cessation andprevention programs for families and carerson child health outcomes (Protocol). In TheCochrane Library. Oxford: Update Software,1999.

�'# �� !��"�#��$��%��

540. Copeland J, Hall W. A comparison ofpredictors of treatment drop-out of womenseeking drug and alcohol treatment in aspecialist women’s and two traditionalmixed-sex treatment services. British Journal ofAddiction 1992; 87(6):883–890.

541. Nakamura K, Takano T. Family involvementfor improving abstinence rates in therehabilitation process of female alcoholics.International Journal of the Addictions 1991;26(10):1055–1064.

542. Armstrong KL, Fraser JA, Dadds MR, MorrisJ. A randomised controlled trial of nursehome visiting to vulnerable families withnewborns. Journal of Paediatrics and Child Health2002; 35:237–244.

543. Meager I, Milgrom J. Group treatment forpostpartum depression: a pilot study.Australian and New Zealand Journal of Psychiatry1996; 30:852–860.

544. Feldman MA, Sparks B, Case L. Effectivenessof home-based early intervention on thelanguage development of children ofmothers with mental retardation. Research inDevelopmental Disabilities 1993; 14:387–408.

545. Serketich WJ, Dumas JE. The effectivenessof behavioral parent training to modifyantisocial behavior in children: a meta-analysis. Behavior Therapy 1996; 27:171–186.

546. Patterson GR, Reid JB, Dishion TJ. AntisocialBoys. Eugene, OR: Castalia Publishing; 1992.

547. Dadds MR, Schwartz S, Sanders MR. Maritaldiscord and treatment outcome inbehavioral treatment of child conductdisorders. Journal of Consulting and ClinicalPsychology 1987; 55(3):396–403.

548. Webster-Stratton C. Predictors of treatmentoutcome in parent training for conductdisordered children. Behavior Therapy 1985;16:223–243.

549. Webster-Stratton C. Stress: a potentialdisruptor of parent perceptions and familyinteractions. Special Issue: the stresses ofparenting. Journal of Clinical Child Psychology1990; 19(4):302–312.

550. Sanders M, Markie-Dadds C, Tully L, Bor W.The Triple-P-Positive Parenting Program: Acomparison of Enhanced, Standard, andSelf-directed behavioural familyintervention for parents of children withearly onset conduct problems. Journal of

Consulting and Clinical Psychology 2000;68(4):624–640.

551. Webster-Stratton C, Hammond M. Treatingchildren with early-onset problems: acomparison of child and parent traininginterventions. Journal of Consulting and ClinicalPsychology 1997; 65:93–109.

552. Webster-Stratton C. Preventing conductproblems in Head Start children:strengthening parenting competencies.Journal of Consulting and Clinical Psychology 1998;66(5):715–730.

553. Forehand R, Long N. Outpatient treatmentof the acting out child: procedures, longterm follow-up data, and clinical problems.Advances in Behavior Research and Therapy 1988;10:129–177.

554. Long P, Forehand R, Wierson M, Morgan A.Does parent training with youngnoncompliant children have long-termeffects? Behavior Research and Therapy 1994;32(1):101–107.

555. Schuhmann EM, Foote RC, Eyberg SM,Boggs SR, Algina J. Efficacy of parent-childinteraction therapy: interim report of arandomized trial with short-termmaintenance. Journal of Clinical Child Psychology1998; 27(1):34–45.

556. Eyberg S, Funderburk B, Hembree-Kigin T,McNeil C, Querido J, Hood K. Parent-childinteraction therapy with behaviour problemchildren: one and two year maintenance oftreatment effects in the family. Child andFamily Behavior Therapy 2001; 23:1–20.

557. Funderburk B, Eyberg S, Newcomb K,McNeil C, Hembree-Kigin T, Capage L.Parent-child interaction therapy withbehaviour problem children: maintenanceof treatment effects in the school setting.Child and Family Behavior Therapy 1998; 20:17–38.

558. McNeil C, Eyberg S, Eisenstadt T, NewcombK, Funderburk B. Parent-child interactiontherapy with behavior problem children:generalization of treatment effects to theschool setting. Journal of Clinical Child Psychology1991; 20(2):140–151.

559. Eisenstadt T, Eyberg S, McNeil C, NewcombK, Funderburk B. Parent-child interactiontherapy with behaviour problem children:relative effectiveness of two stages and

�'%��

overall treatment outcome. Journal of ClinicalChild Psychology 1993; 22:42–51.

560. Kazdin AE. Treatment of antisocialbehaviour in children: current status andfuture directions. Psychological Bulletin 1987;102:187–203.

561. Kazdin AE. Treatment of conduct disorder:progress and directions in psychotherapyresearch. Special Issue: toward adevelopmental perspective on conductdisorder. Development and Psychopathology 1993;5(1–2):277–310.

562. Christensen A, Johnson SM, Phillips S,Glasgow RE. Cost effectiveness in behavioralfamily therapy. Behavior Therapy 1980;11:208–226.

563. Greenwood PW, Model KE, Rydell P, ChiesaJ. Diverting Children from a Life of Crime: MeasuringCosts and Benefits. Santa Monica, CA: RANDCorporation; 1996.

564. Campbell F, Ramey C. Effects of earlyintervention on intellectual and academicachievement: a follow-up study of childrenfrom low-income families. Child Development1994; 65:684–698.

565. Andersson BE. Effects of day-care oncognitive and socioemotional competenceof thirteen-year-old Swedish schoolchildren. Child Development 1992; 63:20–36.

566. Schweinhart LJ, Barnes HV, Wiekart DP.Significant Benefits: The High/Scope Perry Pre-SchoolStudy Through Age 27. Monograph of the HighScope Educational Research Foundation No.10. Ypsilanti, MI: High Scope Press, 1993.

567. Osburn AF, Milbank JE. The Effects of EarlyEducation: A Report from the Child Health andEducation Study. Oxford: Clarendon Press;1987.

568. Lee V, Brooks-Gunn J, Schnur E, Liaw F. AreHead Start effects sustained? A longitudinalfollow-up comparison of disadvantagedchildren attending Head Start, no preschool,and other preschool programs. ChildDevelopment 1990; 61:495–507.

569. US Department of Health and HumanServices. Early Head Start research: Making adifference in the lives of infants andtoddlers and their families: the impacts ofEarly Head Start. Child Outcomes Researchand Evaluation, 2002. Viewed 29/10/02

2002, <http://www.acf.dhhs.gov/programs/core/ongoing_research/ehs/ehs_reports.html>.

570. Schweinhart LJ, Weikart DP. The High/Scope Preschool Curriculum ComparisonStudy through age 23. Early Childhood Research1997; 12:117–143.

571. Weikart DP, Schweinhart LJ. High/ScopePreschool Program outcomes. In McCord J,Tremblay RE (eds) Preventing Antisocial Behavior:Interventions from Birth to Adolescence. New York:Guilford Press, 1992;67–86.

572. Barnett WS. Benefit-cost analysis ofpreschool education: findings from a 25-year follow-up. American Journal ofOrthopsychiatry 1993; 63(4):500–508.

573. Barnett WS. Lives in the Balance: Age-27 Benefit-Cost Analysis of the High/Scope Perry PreschoolProgram. Monograph of the High/ScopeEducation Research Foundation No. 8.Ypsilanti, MI: High/Scope Press, 1996.

574. Karoly LA, Greenwood PW, Everingham SS,Hoube J, Kilburn MR, Sanders M, et al.Investing in our Children: What We Know and Don’tKnow About the Costs and Benefits of Early ChildhoodInterventions. Santa Monica: RANDCorporation, 1998.

575. Karoly LA, Kilburn MR, Bigelow JH,Caulkins JP, Cannon JS, Chiesa JR. AssessingCosts and Benefits of Early Childhood InterventionPrograms: Overview and Application to the StartingEarly Starting Smart Program. MR-1336. SantaMonica, CA: RAND Corporation, 2001.

576. McDonald L, Coe-Braddish D, Billingham S.Families and Schools Together: aninnovative substance abuse preventionprogram. Social Work and Education 1991;13:118–129.

577. Greenberg MT, Domitrovich C, BumbargerB. The prevention of mental disorders inschool-aged children: current state of thefield. Prevention and Treatment [Onlinejournal], Vol 4, March, 2001. Viewed<http://journals.apa.org/prevention/volume4/pre0040001a.html>.

578. DeWitt DJ, Steep B, Silverman G, Stevens-Lavigne A, Ellis K, Smythe C, et al.Evaluating an in-school drug preventionprogram for at-risk youth. Alberta Journal ofEducation Research 2000; XLVI(2):117–133.

�'' �� !��"�#��$��%��

579. Perry C, Williams CL, Veblen-Mortenson S,Toomey T, Komro K, Anstine P, et al.Project Northland: outcomes of acommunitywide alcohol use preventionprogram during early adolescence. AmericanJournal of Public Health 1996; 86:956–965.

580. MacKinnon D, Anderson Johnson C, PentzMA, Dwyer JH. Mediating mechanisms in aschool based drug prevention program:first-year effects of the MidwesternPrevention Project. Health Psychology 1991;10:164–172.

581. Spoth R, Redmond C, Shin C. Direct andindirect latent-variable parenting outcomesof two universal family-focused preventiveinterventions: extending a public health-oriented research base. Journal of Consulting andClinical Psychology 1998; 66(2):385–399.

582. Spoth RL, Redmond C, Shin C. Reducingadolescents’ aggressive and hostilebehaviors: randomized trial effects of a brieffamily intervention 4 years past baseline.Archives of Pediatrics and Adolescent Medicine 2000;154:1248–1257.

583. Spoth RL, Guyll M, Day SX. Universalfamily-focused interventions in alcohol-usedisorder prevention: cost-effectiveness andcost-benefit analyses of two interventions.Journal of Studies on Alcohol 2002; 63:219–228.

584. Catalano RF, Gainey RR, Fleming CB,Haggerty KP, Johnson NO. An experimentalintervention with families of substanceabusers: one-year follow-up of the focus onfamilies project. Addiction 1999; 94:241–254.

585. DeMarsh JK, Kumpfer KL. Familyenvironmental and genetic influences onchildren’s future chemical dependency.Journal of Child in Contemporary Society: AdvancedTheory of applied Research 1985; 18:117–152.

586. Lloyd C, Joyce R, Hurry J, Ashton M. Theeffectiveness of primary school drugeducation. Drugs: Education, Prevention and Policy2000; 7(2):109–126.

587. Wragg J. The longitudinal evaluation of aprimary school drug education program:did it work? Drug Education Journal of Australia1990; 4(1):33–44.

588. Wragg J. An Evaluation of a Model of DrugEducation. National Campaign Against DrugAbuse Monograph Series No. 22. Canberra:

Australian Government Printing Service,1992.

589. Hurry J, Lloyd C. A Follow-up Evaluation ofProject Charlie: A Lifeskills Drug EducationProgramme for Primary Schools. Drug PreventionInitiative Paper No. 16. London: HomeOffice, 1997.

590. Stevens MM, Freeman DH, Mott L, YouellsF. Three-year results of preventionprograms on marijuana use: the NewHampshire study. Journal of Drug Education1996; 26:257–273.

591. Peterson AV, Kealy KA, Mann SL, Marek PM,Sarason IG. Hutchinson smoking preventionproject: Long-term randomized trial inschool-based tobacco use prevention-resultson smoking. Journal of the National CancerInstitute 2000; 92(24):1979–1991.

592. Sussman S, Hansen WB, Flay BR, Botvin G.Correspondence re: Hutchison smokingprevention project: long-term randomziedtrial in school-based tobacco use - results onsmoking. Journal of the National Cancer Institute2001; 93:1267.

593. Johnson CA, Pentz MA, Weber MD, DwyerJH, Baer N, MacKinnon DP, et al. Relativeeffectiveness of comprehensive communityprogramming for drug abuse preventionwith high-risk and low-risk adolescents.Journal of Consulting and Clinical Psychology 1990;58:447–456.

594. Shope JT, Copeland LA, Maharg R, DielmanTE. Effectiveness of a high school alcoholmisuse prevention program. Alcoholism:Clinical and Experimental Research 1996; 20:791–798.

595. Elder J, Wildey MB, de Moore C, Sallis J,Eckhardt L, Edwards C, et al. The long-termprevention of tobacco use among juniorhigh school students: Classroom andtelephone interventions. American Journal ofPublic Health 1993; 83:1239–1244.

596. Prokhorov AV, Alexandrov AA. Preventionof adolescent smoking in Russia: Generalsituation and results from a Moscowrandomized trial. European Journal of PublicHealth 1994; 4:169–174.

597. Abernathy TJ, Bertrand LD. Preventingcigarette smoking among children: Resultsof a four-year evaluation of the PALprogram. Canadian Journal of Public Health 1992;83:226–229.

�'(��

598. Snow D, Tebes JK, Arthur MW, Tapasak R.Two-year follow-up of a social-cognitiveintervention to prevent substance use. Journalof Drug Education 1992; 22:101–114.

599. Klepp KI, Oygard L, Tell GS, Vellar OD.Twelve year follow-up of a school-basedhealth education program: the Oslo youthstudy. European Journal of Public Health 1994;4:195–200.

600. Perry CI, Kelder SH, Klepp K. Community-–wide cardiovascular disease prevention inyoung people: long-term outcomes of theClass of 1989 study. European Journal of PublicHealth 1994; 4:188–194.

601. Hawthorne G. The social impact of LifeEducation: Estimating drug use prevalenceamong Victorian primary school studentsand the statewide effect of the LifeEducation programme. Addiction 1996;91(8):1151–1159.

602. Snati SM, Cargo M, Brown S, Best JA,Cameron R. Dispositional risk factors forsmoking-stage transitions: A socialinfluences program as an effect modifier.Addictive Behaviors 1994; 19:269–285.

603. Shope JT, Kloska D, Dielman TE, Maharg R.Longitudinal evaluation of an enhancedAlcohol Misuse Prevention Study (AMPS)curriculum for grades six to eight. Journal ofSchool Health 1994; 64(4):160–166.

604. Hawkins JD, Guo J, Hill K, Battin-Pearson S,Abbott R. Long term effects of the SeattleSocial Development intervention on schoolbonding trajectories. Applied DevelopmentalScience: Special issue: Prevention as Altering the Courseof Development 2001; 5(4):225–236.

605. O’Donnell J, Hawkins JD, Catalano RF,Abbott RD, Day LE. Preventing schoolfailure, drug use, and delinquency amonglow-income children: long-termintervention in elementary schools. AmericanJournal of Orthopsychiatry 1995; 65(1):87–100.

606. Lonczak HS, Abbott RD, Hawkins JD,Kosterman R, Catalano RF. Effects of theSeattle social development project on sexualbehavior, pregnancy, birth, and sexuallytransmitted disease outcomes by age 21years. Archives of Pediatrics and Adolescent Medicine2002; 156(5):438–447.

607. Eddy JM, Reid JB, Fetrow RA. Anelementary school-based prevention

program targeting modifiable antecedentsof youth delinquency and violence: Linkingthe Interests of Families and Teachers(LIFT). Journal of Emotional and BehaviouralDisorders 2000; 8:165–176.

608. Kellam SG, Anthony JC. Targeting earlyantecedents to prevent tobacco smoking:findings from an epidemiologically basedrandomzied field trial. American Journal of PublicHealth 1998; 88:1490–1495.

609. Battistich V, Schaps E, Watson M, SolomonD. Prevention effects of the childdevelopment project: early findings from anongoing multisite demonstration trial.Journal of Adolescent Research 1996; 11(1):12–35.

610. Greenberg MT, Kusche CA, Cook ET,Quamma JP. Promoting emotionalcompetence in school-aged children: theeffects of the PATHS curriculum. Developmentand Psychopathology 1995; 7(1):117–136.

611. Bierman KL. Process of change during socialskills training with preadolescents and itsrelation to treatment outcomes. ChildDevelopment 1986; 57:230–240.

612. Bierman KL, Miller CL, Stabb SD. Improvingthe social behavior and peer acceptance ofrejected boys: effects of social skills trainingwith instructions and prohibition. Journal ofConsulting & Clinical Psychology 1987;55(2):194–200.

613. Coie JD, Krehbiel G. Effects of academictutoring on the social status of low-achieving, socially rejected children. ChildDevelopment 1984; 55(4):1465–1478.

614. Conduct Problems Prevention ResearchGroup. Initial impact of the Fast TrackPrevention Trial for Conduct Problems: II.Classroom effects. Journal of Consulting andClinical Psychology 1999; 67(5):648–657.

615. Conduct Problems Prevention ResearchGroup. Initial impact of the Fast TrackPrevention Trial for Conduct Problems: I.The high-risk sample. Journal of Consulting andClinical Psychology 1999; 67(5):631–647.

616. Conduct Problems Prevention ResearchGroup. Evaluation of the first three years ofthe Fast Track prevention trial with childrenat high-risk for adolescent conductproblems. Journal of Abnormal Child Psychology2002; 30(1):19–35.

�() �� !��"�#��$��%��

617. Delquadri J, Greenwood C, Whorton D,Carta J, Hall R. Classwide Peer Tutoring.Exceptional Children 1986; 52(535–542).

618. Greenwood C, Delquadri J, Hall R.Longitudinal effects of Classwide PeerTutoring. Journal of Educational Psychology 1989;81(371–384).

619. Heron T, Heward W, Cooke N, Hill D.Evaluation of Classwide Peer TutoringSystem: First graders teach each other sightwords. Education and Treatment of Children 1983;6(2):137–152.

620. Kohler F, Greenwood C. Effects of collateralpeer supportive behaviours within theClasswide Peer Tutoring Program. Journal ofApplied Behavior Analysis 1990; 23(3):307–322.

621. Sideridis G, Utley C, Greenwood C,Delquadri J, Dawson H, Palmer PR, S.Classwide Peer Tutoring: Effects on thespelling performance and social interactionsof students with mild disabilities and theirtypical peers in an integrated instructionalsetting. Journal of Behavior Education 1997;7(4):203–212.

622. Toumbourou JW, Gregg ME. Working withfamilies to promote healthy adolescentdevelopment. Family Matters 2001; 59:54–60.

623. Kacir C, Gordon DA. Interactive videodiskparent training for parents of difficult pre-teens [Unpublished Manuscript]: Universityof Ohio; 1997.

624. Spoth RL, Redmond C, Kahn JH, Shin C. Aprospective validation study of inclination,belief and context predictors of family-focused prevention involvement. FamilyProcess 1997; 36:403–429.

625. Jenkin C, Bretherton D. PACE ParentingAdolescents: A Creative Experience. Camberwell,Victoria: Australian Council for EducationalResearch, 1994.

626. Cohen D, Rice J. A parent-targetedintervention for adolescent substance useprevention. Lessons learned. Evaluation Review1995; 19:159–180.

627. Dishion TJ, Kavanagh K. A multilevelapproach to family-centred prevention inschools: process and outcome. AddictiveBehaviors 2000; 25:899–911.

628. Toumbourou JW, Gregg ME, Jenkin C,Holmes S. Symposia paper - Parentinginterventions to prevent youth drug abuse.Paper presented at 2nd International Drugsand Young People Conference. Melbourne,5 April 2001.

629. Bauman KE, Foshee VA, Ennett ST,Pemberton M, Hicks KA, King TS, et al. Theinfluence of a family program on adolescenttobacco and alcohol use. American Journal ofPublic Health 2001; 91(4):604–610.

630. Rohrbach LA, Hodgson CS, Flay BR, HansenWB, Pentz MA. Parental participation indrug abuse prevention: results from theMidwestern Prevention Project. Journal ofResearch on Adolescence 1994; 4:295–317.

631. Chou C, Montgomery S, Pentz MA,Rohrbach LA, Johnson CA, Flay BR, et al.Effects of a community-based preventionprogram on decreasing drug use in high-risk adolescents. American Journal of Public Health1998; 88:944–948.

632. Bank L, Marlowe JH, Reid JB, Patterson GR,Weinrott MR. A comparative evaluation ofparent-training interventions for families ofchronic delinquents. Journal of Abnormal ChildPsychology 1991; 19:15–33.

633. Blyth A, Bamberg J, Toumbourou J. BehaviourExchange Systems Training: A Program for ParentsStressed by Adolescent Substance Abuse.Camberwell, Australia: Australian Councilfor Educational Research, 2000.

634. Toumbourou JW, Blyth A, Bamberg J, ForerD. Early impact of the BEST intervention forparents stressed by adolescent substanceabuse. Journal of Community & Applied SocialPsychology 2001; 11:291–304.

635. Bamberg J, Toumbourou JW, Blyth A, ForerD. Profiling changes for families in the BESTprogram, for parents coping with youthsubstance abuse. Australian & New ZealandJournal of Family Therapy 2001; 22:189–198.

636. Dishion TJ, Andrews DW. Preventingescalation in problem behaviours with high-risk young adolescents: immediate and 1-year outcomes. Journal of Consulting and ClinicalPsychology 1995; 63:538–548.

637. Bry BH, Catalano RF, Kumpfer KL, LochmanJE, J. S. Scientific findings from familyprevention intervention research. In AsheryRS, Robertson EB, Kumpfer KL (eds) Drug

�(��

Abuse Prevention through Family Interventions: NIDAResearch Monograph 177. Rockville, MD: USDepartment of Human Services, NationalInstitute on Drug Abuse, 1998;103–129.

638. Szapocznik J, Kurtines WM, Perez-Vidal A,Brickman A, H. FF, Santisteban D, et al.Engaging adolescent drug abusers and theirfamilies into treatment: strategic structuralsystems approach. Journal of Consulting andClinical Psychology 1988; 56:552–557.

639. Liddle HA, Dakof GA. Efficacy of familytherapy for drug abuse: promising but notdefinitive. Journal of Marital and Family Therapy1995; 21:511–543.

640. Santisteban DA, Szapocznik J. Bridgingtheory, research and practice to moresuccessfully engage substance abusing youthand their families into therapy. Journal of Childand Adolescent Substance Abuse 1994; 3:9–24.

641. Stanton MD, Todd TC. The Family Therapy ofDrug Abuse and Addiction. New York: GuilfordPress; 1982.

642. Alexander JF, Parsons BV. Functional FamilyTherapy. Monterey, CA: Brooks and Cole;2002.

643. Klein NC, Alexander JF, Parsons BV. Impactof family systems intervention onrecidivism and sibling delinquency: a modelof primary prevention and programevaluation. Journal of Consulting and ClinicalPsychology 1977; 45:469–474.

644. Aos S, Barnoski R, Lieb R. Preventiveprograms for young offenders effective andcost-effective. Overcrowded Times 1998; 9:7–11.

645. Robbins MS, Alexander JF, Newell RM,Turner CW. The immediate effect ofreframing on client attitude in familytherapy. Journal of Family Psychology 1996;10:28–34.

646. Borduin CM, Mann BJ, Cone LT, HenggelerSW, Fucci BR, Blaske DM, et al.Multisystemic treatment of serious juvenileoffenders: Long-term prevention ofcriminality and violence. Journal of Consultingand Clinical Psychology 1995; 63:569–578.

647. Henggeler SW, Rodick JD, Borduin CM,Hanson CL, Watson SM, Urey JR.Multisystemic treatment of juvenileoffenders: effects on adolescent behaviour

and family interaction. Developmental Psychology1986; 22:132–141.

648. Henggeler SW, Melton GB, Smith LA.Family preservation using Multisystemictherapy as an effective alternative toincarcerating serious juvenile offenders.Journal of Consulting and Clinical Psychology 1992;60:953–961.

649. Cunningham PB, Henggeler SW. Engagingmultiproblem families in treatment: lessonslearned throughout the development ofMultisystemic Therapy. Family Process 1999;38:265–281.

650. Midford R, Lenton S, Hancock L. A CriticalReview and Analysis: Cannabis Education in Schools.Sydney: NSW Department of Education andTraining, 2000.

651. Midford R, Snow P, Lenton S. School-basedIllicit Drug Education Programs: A Critical Review andAnalysis. Literature review prepared for the Departmentof Employment, Training and Youth Affairs.Melbourne: Centre for Youth Drug Studies,Australian Drug Foundation, 2001.

652. Botvin GJ, Eli B, Dusenbury L, Botvin EM,Diaz T. Long-term follow-up results of arandomized drug abuse prevention trial in awhite middle-class population. Journal of theAmerican Medical Association 1995; 273:1106–1112.

653. Hansen WB. School-based substance abuseprevention: a review of the state of the artin curriculum, 1980 - 1990. Health EducationResearch 1992; 7:403–430.

654. White D, Pitts M. Educating young peopleabout drugs: a systematic review. Addiction1998; 93:1475–1487.

655. Dusenbury L, Botvin GJ, James-Ortiz S. Theprimary prevention of adolescent substanceabuse through the promotion of personaland social competence. Prevention in HumanServices 1990; 7:201–224.

656. Ballard R, Gillespie A, Irwin R. Principles forDrug Education in Schools. Belconnen, ACT:University of Canberra, 1994.

657. Coggans N, Watson A. Drug education:approaches, effectiveness and delivery.Drugs: Education, Prevention and Policy 1995;2(3):211–224.

658. Dusenbury L, Falco M. Eleven componentsof effective drug abuse prevention curricula.

�(� �� !��"�#��$��%��

Journal of School Health 1995; 65(10):420–425.

659. Kelder SH, Perry CL, Klepp KI, Lytle LL.Longitudinal tracking of adolescentsmoking, physical activity and food choicebehaviour. American Journal of Public Health1994; 87(7):1121–1126.

660. Tobler NS, Lessard T, Marshall D, OchshornP, Roona M. Effectiveness of school-baseddrug prevention programs for marijuanause. School Psychology International 1999;20(1):105–137.

661. Tobler NS, Stratton HH. Effectiveness ofschool-based drug prevention programs: ameta-analysis of the research. Journal ofPrimary Prevention 1997; 18(1):71–128.

662. Bandura A. Social Learning Theory. EnglewoodCliffs, N.J.: Prentice Hall; 1977.

663. McGuire WJ. Constructing Social Psychology:Creative and Critical Processes. New York:Cambridge University Press; 1976/1999.

664. Botvin GJ, Baker E, Filazzola AD, Botvin EM.A cognitive-behavioural approach tosubstance abuse prevention: one yearfollow-up. Addictive Behaviors 1990; 15:47–63.

665. Dusenbury L, Falco M, Lake A. A review ofthe evaluation of 47 drug abuse preventioncurricula available nationally. Journal of SchoolHealth 1997; 67(4):127–132.

666. Botvin GJ, Griffin KW, Diaz T, Scheier LM,Williams C, Epstein JA. Preventing illicitdrug use in adolescents: long-term follow-up data from a randomized control trial of aschool population. Addictive Behaviors 2000;25(5):769–774.

667. Gorman DM. Do school-based social skillstraining programs prevent alcohol useamong young people? Addiction Research 1996;4(2):191–210.

668. Hansen WB, McNeal RB, Jr. Drug educationpractice: results of an observational study.Health Education Research 1999; 14(1):85–97.

669. Klepp KI, Perry C, Kelder S. Alcohol andmarijuana use among adolescents: long-term outcomes of the Class of 1989 Study.Annals of Behavioural Medicine 1995; 17(1):19–24.

670. Arkin RM, Roemhild HF, Johnson CA,Luepker RV, Murray DM. The Minnesotasmoking prevention program: a seventh-grade health curriculum supplement. Journalof School Health 1981; Nov:611–616.

671. Kishchuk N, O’Loughlin J, Paradis S,Masson P, Sacks-Silver G. Illuminatingnegative results in evaluation of smokingprevention programs. Journal of School Health1990; 60(448–451).

672. Vartiainen E, Fallonen U, McAlister AL,Puska P. Eight-year follow-up results of anadolescent smoking prevention program:the North Karelia youth project. AmericanJournal of Public Health 1990; 80:78–79.

673. NHS Centre for Reviews and Dissemination.Preventing the uptake of smoking in youngpeople. Effective Health Care 1999; 5:1–12.

674. Thomas R, Bushby K. School basedprogrammes for preventing smoking(Protocol). In The Cochrane Library. Oxford:Update Software, 1999.

675. Bruvold WH. A meta-analysis of theCalifornia school-based risk reductionprogram. Journal of Drug Education 1990;20:139–152.

676. Bruvold WH. A meta-analysis of adolescentsmoking prevention programs. AmericanJournal of Public Health 1993; 83:872–880.

677. Ennett ST, Tobler NS, Ringwalt CI,Flewelling RL. How effective is drug abuseresistance education: a meta-analysis ofProjectg CARE outcome evaluations. AmericanJournal of Public Health 1994; 84:1394–1401.

678. Rooney BL, Murray DM. A meta-analysis ofsmoking prevention programs afteradjustment for errors in the unit of analysis.Health Education Quarterly 1996; 23:48–64.

679. Tobler NS, Roona MR, Ochshorn P,Marshall DG, Streke AV, Stackpole KM.School-based adolescent drug preventionprograms: 1998 meta-analysis. Journal ofPrimary Prevention 2000; 20:275–336.

680. Wang LY, Crossett LS, Lowry R, Dent CW,Sussman S. Cost-effectiveness of a school-based tobacco-use prevention program.Archives of Pediatrics and Adolescent Medicine 2001;155:1043–1050.

681. Nutbeam D, Macaskill P, Smith C, SimpsonJM, Catford J. Evaluation of two school

�(��

smoking education programs under normalclassroom conditions. British Medical Journal1993; 306:102–107.

682. Murray D, Perry C, Griffin G, Harry K,Jacobs DR, Scmid L, et al. Results from astatewide approach to adolescent tobaccouse prevention. Preventive Medicine 1992;21:449–472.

683. Caplan M, Weissberg RP, Grober JS, Sivo PJ,Grady K, Jacoby C. Social competencepromotion with inner-city and suburbanyoung adolescents: effects on socialadjustment and alcohol use. Journal ofConsulting and Clinical Psychology 1992; 60:56–63.

684. Maggs JL, Schulenberg J. Reasons to drinkand not to drink: altering trajectories ofdrinking through an alcohol misuseprevention program. Applied DevelopmentalScience 1998; 2(1):48–60.

685. Bell RM, Ellickson PL, Harrison ER. Do drugprevention effects persist into high school?How project ALERT did with ninth graders.Preventive Medicine 1993; 22:463–483.

686. Graham JW, Anderson Johnson C, HansenWB, B.R. F, Gee M. Drug use preventionprograms, gender, and ethnicity: evaluationof three seventh-grade project SMARTcohorts. Preventive Medicine 1990; 19:305–313.

687. McBride N. School Health and AlcoholHarm Reduction Project: Reducing AlcoholRelated Harms in Young People [PhDthesis]. Perth: Curtin University ofTechnology; 2002.

688. Clayton RR, Cattarello AM, Johnstone BM.The effectiveness of drug abuse resistanceeducation (Project DARE): 5-year follow-upresults. Preventive Medicine 1996; 25:307–318.

689. Caulkins JP, Rydell CP, Sohler EveringhamSM, Chiesa JR, Bushway S. An Ounce ofPrevention, A Pound of Uncertainty: The Cost-Effectiveness of School-Based Drug PreventionPrograms. Santa Monica: RAND Corporation,1999.

690. Kim S, McLeod J, Shantzis C. An outcomeevaluation of Here’s Looking at You 2000.Journal of Drug Education 1993; 23(1):67–81.

691. Perry CL, Grant M, Ernberg G, FlorenzanoRU, Langdon MC, Myeni AD, et al. WHO

Collaborative Study On Alcohol Educationand Young People: Outcomes of a four-country pilot study. International Journal of theAddictions 1989; 24:1145–1171.

692. Olweus D. Bully/victim problems amongschool children: basic facts and effects of aschool based intervention program. InPepler DJ, Rubins KH (eds) The Developmentand Treatment of Childhood Aggression. Hillsdale,NJ: Lawrence Erlbaum Associates,1991;411–448.

693. Glover S, Burns J, Butler H, Patton GP.Social environments and the emotionalwell-being of young people. Family Matters1998; 49:11–16.

694. Eggert L, Thompson E, Herting J, NicholasL, Dicker B. Preventing adolescent drugabuse and high school dropout through anintensive school-based social networkdevelopment program. American Journal ofHealth Promotion 1994; 8(3):202–215.

695. Hastie PA, Sharpe T. Effects of a sporteducation curriculum on the positive socialbehavior of at-risk rural adolescent boys.Journal of Education for Students Placed at Risk1999; 4(8):417–430.

696. Wiist WH, Snider G. Peer education infriendship cliques: prevention of adolescentsmoking. Health Education Research 1991;6:101–108.

697. Macri I, Tsiantis J. Effects of a peer ledprogram on smoking prevention: a caseexample in Greece. International Quarterly ofCommunity Health Education 1997; 17:297–308.

698. Dijkstra M, Mester I, de Vries H, vanBruekelen G, Parcel GS. Effectiveness of asocial influence approach and boosters tosmoking prevention. Health Education Research1999; 14(791–802).

699. Bloor M, Frankland J, Langdon NP,Robinson M, Allerton S, Catherine A. Acontrolled evaluation of an intensive peer-led, schools-based, anti-smokingprogramme. Health Education Journal 1999;58:17–25.

700. Koumi I, Tsiantis J. Smoking trends inadolescence: report on a Greek school-basedpeer-led intervention aimed at prevention.Health Promotion International 2001; 16(65–72).

�(� �� !��"�#��$��%��

701. Tobler NS. Meta-analysis of 143 adolescentdrug prevention programs: Quantitativeoutcome results of program participantscompared to a control or comparisongroup. Journal of Drug Issues 1986; 16:537–567.

702. Wilhelmsen W, Laberg J, Klepp KI.Evaluation of two student and teacherinvolved alcohol prevention programs.Addiction 1994; 89:1157–1165.

703. van Teijlingen ER, Friend JAR, Twine F.Evaluation of Grampian Smokebusters: asmoking prevention initiative aimed atyoung teenagers. Journal of Public HealthMedicine 1996; 18:13–18.

704. Snow P, Munro G. Alcohol consumption inamateur Australian football clubs: Evidencefrom a rural region. Health Promotion Journal ofAustralia 2000; 10(3):237–243.

705. Schinke SP, Orlandi MA, Cole KC. Boys andgirls clubs in public housing developments:prevention services for youth at risk. Journalof Community Psychology 1992; OSAP SpecialIssue:118–128.

706. Tierney JP, Grossman JB, Resch NL. Making aDifference. An Impact Study of Big Brothers/BigSisters. Philadelphia, PA: Public/PrivateVentures; 1995.

707. Fors SW, Jarvis S. Evaluation of a peer-leddrug abuse risk reduction project forrunaway/homeless youths. Journal of DrugEducation 1995; 25:321–333.

708. Palinkas LA, Atkins CJ, Miller C, Ferreira D.Social skills training for drug prevention inhigh-risk female adolescents. PreventiveMedicine 1996; 25:692–701.

709. Hebert EE. Doing something about childrenat risk. National Institute of Justice Journal 1993;Nov:4–9.

710. Bellhouse B, Johnston G. Good PracticeDescriptive Case Studies. Early Intervention for YoungPeople with Drug Problems. Report prepared bythe Australian Drug Foundation. NorthMelbourne: Department of Human Services,2002.

711. Furlong MJ, Casa JM, Corral C, Gordon M.Changes in substance use patterns associatedwith the development of a communitypartnership project. Evaluation and ProgramPlanning 1997; 20:299–305.

712. Cheadle A, Pearson D, Wagner E, Psaty B,Diehr P, Koepsell T. A community basedapproach to preventing alcohol use amongadolescents on an American Indianreservation. Public Health Reports 1995;110:439–447.

713. Schinke SP, Tepavac L, Cole KC. Preventingsubstance use among Native Americanyouth: three-year results. Addictive Behaviors2000; 25:387–397.

714. Sowden A, Arblaster L. Communityinterventions for preventing smoking inyoung people. In The Cochrane Library, Issue 1.Oxford: Update Software, 2001.

715. Biglan A, Ary DV, Smolkowski K, Duncan T,Black C. A randomised controlled trial of acommunity intervention to preventadolescent tobacco use. Tobacco Control 2000;9:24–32.

716. Baxter AP, Milner PC, Hawkins S, Leaf M,Simpson C, Wilson KV. The impact of hearthealth promotion on coronary heart diseaselifestyle risk factors in school children:lessons learnt from a community-basedproject. Public Health 1997; 111:231–237.

717. Davidson L. Wensleydale Smokebusters ProjectReport. Northallerton: Northalleton HealthAuthority, 1992.

718. Murray DM, Prokhorov AV, Harty KC.Effects of a statewide antismoking campaignon mass media messages and smokingbeliefs. Preventive Medicine 1994; 23:54–60.

719. Piper DL, Moberg DP, King MJ. The Healthyfor Life Project: behavioural outcomes.Journal of Primary Prevention 2000; 21(1):47–73.

720. St Piette TL, Kaltreider DL, Mark MM, AikinKJ. Drug prevention in a communitysetting: a longitudinal study of the relativeeffectiveness of a 3-year primary preventionprogram in boys and girls clubs across thenation. American Journal of Community Psychology1992; 20:673–706.

721. Sussman S, Dent CW, Stacy AW, Craig S.One-year outcomes of Project Towards NoDrug Abuse. Preventive Medicine 1998:632–642.

722. Vartiainen E, Paavola M, McAlister A, PuskaP. Fifteen-year follow-up of smokingprevention effects in the North Karelia

�("��

Youth Project. American Journal of Public Health1998; 88:81–85.

723. Winkleby MA, Fortmann SP, Rockhill B.Cigarette smoking trends in adolescents andyoung adults: the Stanford Five-City Project.Preventive Medicine 1993; 22:325–334.

724. Gordon I, Whitear B, Guthrie D. Stoppingthen starting: evaluation of a community-based project to discourage teenagesmoking in Cardiff. Health Education Journal1997; 56:42–50.

725. Kaufman JS, Jason LA, Sawlski LM, HalpertJA. A comprehensive multi-media programto prevent smoking among Black students.Journal of Drug Education 1994; 24:95–108.

726. Pentz MA, Dwyer JH, MacKinnon DP, FlayBR, Hansen WB, Wang EY, et al. Amulticommunity trial for primaryprevention of adolescent drug abuse. Effectson drug use prevalence. Journal of the AmericanMedical Association 1989; 261:3259–3266.

727. Holder HD, Saltz RF, Grube JW, Voas RB,Gruenewald PJ, Treno A. A communityprevention trial to reduce alcohol-involvedaccidental injury and death: overview.Addiction 1997; 92(Suppl. 2):S155–S171.

728. Stevens MM, Freeman DH, Mott LA, YouellsFE, Linsey SC. Smokeless tobacco use amongchildren: the New Hampshire study.American Journal of Preventive Medicine 1993;9:160–167.

729. Halpern-Felsher BL, Ozer EM, Millstein SG,Wibbelsmann CJ, Fuster DC, Elster AB, et al.Preventive services in a health maintenanceorganization: how well do pediatriciansscreen and educate adolescent parents?Archives of Pediatrics & Adolescent Medicine 2000;154:173=179.

730. Franzgrote M, Ellen JM, Millstein SG, IrwinCE, Jr. Screening for adolescent smokingamong primary care physicians inCalifornia. American Journal of Public Health1997; 87(8):1341–1345.

731. Schorling JB, Klas PT, Willems JP, EverettAS. Addressing alcohol use among primarycare patients: differences between familymedicine and internal medicine residents.Journal of General Internal Medicine 1994;9(5):248–254.

732. Ellen JM, Franzgrote M, Irwin CE, Jr,Millstein SG. Primary care physicians’

screening of adolescent patients: a survey ofCalifornia physicians. Journal of Adolescent Health1998; 22(6):433–438.

733. Joffe A, Radius S, Gall M. Health counselingfor adolescents: what they want, what theyget, and who gives it. Pediatrics 1988;82(3):481–485.

734. Epner JEG, Levenberg PB, Schoeny ME.Primary care providers’ responsiveness tohealth-risk behaviours reported byadolescent patients. Archives of Pediatrics &Adolescent Medicine 1998; 152(8):774–780.

735. Blum R, Beuhring T, Wunderlich M,Resnick MD. Don’t ask, they won’t tell: thequality of adolescent health screening infive practice settings. American Journal of PublicHealth 1996; 86:1767–1772.

736. Klein JD, McNulty M, Flatau CN.Adolescents’ access to care: teenagers’ self-reported use of services and perceivedaccess to confidential care. Archives of Pediatricsand Adolescent Medicine 1998; 152(7):676–682.

737. Britto MT, Rosenthol SL, Taylor J, PassoMH. Improving rheumatologists’ screeningfor alcohol use and sexual activity. Archives ofPediatrics & Adolescent Medicine 2000; 154:478–483.

738. Warren CW, Kann L, Small ML, Santelli JS,Collins JL, Kolbe LJ. Age of initiatingselected health-risk behaviors among highschool students in the United States. Journal ofAdolescent Health 1997; 21:225–231.

739. Millstein SG, Irwin CE, Jr, Adler NE, CohnLD, Kegeles SM, Dolcini MM. Health-riskbehaviors and health concerns among yongadolescents. Pediatrics 1992; 3:422–428.

740. Ward JE, Gordon J, Sanson-Fisher RW.Strategies to increase preventive care ingeneral practice. Medical Journal of Australia1991; 154(523–531).

741. Thom B, Tellez C. Detecting and managingalcohol problems in general practice. BritishJournal of Addiction 1986; 81:405–418.

742. Sanci LA, Coffey CMM, Veit FCM, Carr-Gregg M, Patton GC, Day N, et al.Evaluation of the effectiveness of aneducational intervention for generalpractitioners in adolescent health care:randomised controlled trial. British MedicalJournal 2000; 320:224–230.

�(# �� !��"�#��$��%��

743. Paperny DMN, Hedberg VA. Computer-assisted health counselor visits: a low-costmodel for comprehensive adolescentpreventive services. Archives of Pediatrics &Adolescent Medicine 1999; 153(1):63–67.

744. Adger H, Jr. Problems of alcohol and otherdrug use and abuse in adolescents [Review].Journal of Adolescent Health 1991; 12(8):606–613.

745. Svoen N, Schei E. Adolescent smokingprevention - primary health care incooperation with local schools. A controlledintervention study. Scandinavian Journal ofPrimary Health Care 1999; 17:54–58.

746. Kottke TE, Battista RN, DeFriese GH, BrekkeML. Attributes of successful smokingcessation interventions in medical practice.A meta-analysis of 39 controlled trials.Journal of the American Medical Association 1988;259:2883–2889.

747. Russell M, Wilson C, Taylor C, Baker C.Effect of general practitioners’ advice againstsmoking. British Medical Journal 1979; 2:31–35.

748. Klein JD, Portilla M, Goldstein A, LeiningerL. Training pediatric residents to preventtobacco use. Pediatrics 1995; 96:326–330.

749. Hovell MF, Slymen DJ, Jones JA, HofstetterCR, Burkham-Kreitner S, Conway TL, et al.An adolescent tobacco-use prevention trialin orthodontic offices. American Journal of PublicHealth 1996; 86:1760–1766.

750. Lionis C, Kafatos A, Flachonikolis J, VakakiM, Tzortzi M, Petraki A. The effects of ahealth education program among Cretanadolescents. Preventive Medicine 1991; 20:685–699.

751. Marlatt AG, Baer JS, Kivlahan DR, DimeffLA, Larimer ME, Quigley LA, et al.Screening and brief intervention for high-risk college student drinkers: results from a2-year follow-up assessment. Journal ofConsulting and Clinical Psychology 1998;66(4):604–615.

752. Baer JS, Kivlahan DR, Blume AW, McKnightP, Marlatt AG. Brief intervention for heavy-drinking college students: 4-year follow-upand natural history. American Journal of PublicHealth 2001; 91(8):1310–1316.

753. Baer JS, Marlatt GA, Kivlahan DR, FrommeK, Larimer ME, Williams E. An experimental

test of three methods of alcohol riskreduction with young adults. Journal ofConsulting and Clinical Psychology 1992;60(6):974–979.

754. Rickert VI, Graham CJ, Fisher R, Gottlieb A,Trosclair A, Jay MS. A comparison ofmethods for alcohol and marijuanaanticipatory guidance with adolescence.Journal of Adolescent Health 1993; 14:533–539.

755. Oliansky D, Wildenhaus K, Manlove K,Arnold T, Schoener E. Effectiveness of briefinterventions in reducing substance useamong at-risk primary care patients in threecommunity-based clinics. Substance Abuse1997; 18:95–103.

756. Werch C, Anzalone D, Brokiewicz L, FelkerJ, Carlson J, Castellon-Vogel A. Anintervention for preventing alcohol useamong inner city middle-school students.Archives of Family Medicine 1996; 5(3):146–152.

757. Catron T, Weiss B. The Vanderbilt school-based counseling program: an interagency,primary-care model of mental healthservices. Journal of Emotional and BehaviouralDisorders 1994; 2(4):247–253.

758. Sowden AJ, Arblaster L. Mass mediainterventions for preventing smoking inyoung people. In The Cochrane Library, Issue 4.Oxford: Update Software, 2001.

759. Hafstad A. Provocative appeals in anti-smoking mass-media campaigns targetingadolescents - the accumulated effect ofmultiple exposures. Health Education Research1997; 12:227–236.

760. Flay BR, Miller TQ, Hedeker D, Siddiqui O,Britton CF, Brannon BR, et al. Thetelevision, school, and family smokingprevention and cessation project VIII.Student outcomes and mediating variables.Preventive Medicine 1995; 24:29–40.

761. Flay BR, Hansen WB, Johnson CA, CollinsLM, Dent CW, Dwyer KM, et al.Implementation effectiveness trial of a socialinfluences smoking prevention programusing schools and television. Health EducationResearch 1987; 2:385–400.

762. Bauman KE, LaPrelle J, Brown JD, Koch GG,Padgett CA. The influence of three massmedia campaigns in variables related toadolescent cigarette smoking: results of a

�(%��

field experiment. American Journal of PublicHealth 1991; 81:597–604.

763. Flynn BS, Worden JK, Secker-Walker RH,Badger GJ, Geller BM. Cigarette smokingprevention effects of mass media and schoolinterventions targeted to gender and agegroups. Journal of Health Education 1995;26:45–51.

764. Flynn BS, Worden JK, Secker-Walker RH,Badger GJ, Geller BM, Costanza MC.Prevention and cigarette smoking throughmass media intervention and schoolprograms. American Journal of Public Health1992; 82:827–834.

765. Flynn BS, Worden JK, Secker-Walker RH,Pirie PL, Badger GJ, Carpenter JH, et al.Mass media and school interventions forcigarette smoking prevention: effects 2years after completion. American Journal ofPublic Health 1994; 84:1148–1150.

766. Worden JK, Flynn BS, Solomon IJ, Secker-Walker RH, Badger GJ, Carpenter JH. Usingmass media to prevent cigarette smokingamong adolescent girls. Health EducationQuarterly 1996; 23:453–468.

767. Goldman LK, Glantz SA. Evaluation ofantismoking advertising campaigns. Journal ofthe American Medical Association 1998; 279:772–777.

768. Department of Community Services andHealth. The First Twelve Months of the DrugOffensive Campaign against Adolescent Alcohol Abuse.Unpublished Report. Canberra: Departmentof Community Services and Health, 1990.

769. Carroll T. The Role of Social Marketing CampaignsWithin Australia’s National Drug Strategy. ASubmission to the National Drug StrategyEvaluation 1996. Canberra: CommonwealthDepartment of Health and Family Services,November 1996.

770. Carroll T, Lum M, Taylor J, Travia J.Evaluation of the Launch Phase of the National AlcoholCampaign. Sydney: Government Departmentof Health and Aged Care, September 2000.

771. Donohew L, Lorch E, Palmgreen P.Sensation seeking and targeting of televisedanti-drug PSAs. In Donohew L, Sypher HE,Bukoski WJ (eds) Persuasive Communication andDrug Abuse Prevention. New Jersey: LawrenceErlbaum Associates, 1991;209–226.

772. Carroll T, Taylor J, Lum M. Evaluation of theDrug Offensive “Speed Catches Up With You”Amphetamines Campaign - 1993-1995. Sydney:Commonwealth Department of Health andFamily Services, 1996.

773. Bertram S, Worsley J, Carroll T. Evaluationof the launch phase of the National IllicitDrug Campaign. Research and MarketingGroup, Population Health Division,Department of Health and Ageing, 2002.Viewed 25/10/02 2002, <http://www.health.gov.au/pubhlth/publicat/document/reports/nidc_eval_1.pdf>.

774. Carruthers S, McDonald C. The availabilityof cigarettes to minors in Perth, WesternAustralia. Tobacco Control 1995; 4:49–52.

775. Ludbrook A, Godfrey C, Wyness L, Parrot S,Haw S, Napper M, et al. Effective and Cost-Effective Measures to Reduce Alcohol Misuse inScotland: A Literature Review. Aberdeen:University of York, 2002.

776. Grube JW. Preventing sales of alcohol tominors: results from a community trial.Addiction 1997; 92(Suppl. 2):S251–S260.

777. Howell JC, Krisberg B, Hawkins JD, Wilson.JJ, eds. A Sourcebook: Serious, Violent, and ChronicJuvenile Offenders. New York: SagePublications; 1995.

778. Commonwealth Department of EducationScience and Training. National Framework forProtocols for Managing the Possession, Use and/orDistribution of Illicit & Other Unsanctioned Drugs inSchoolsJune 2000.

779. Withers G, Russell J. Education for Resilience:Prevention & Intervention Strategies for Young PeopleAt Risk. Camberwell, Victoria: AustralianCouncil for Educational Research Press,2001.

780. Alcohol and other Drugs Council ofAustralia. Working towards a nationalhomelessness strategy, 2001. Viewed 20November 2001, <www.adca.org.au/submissions/homeless%20strategy.doc>.

781. Bond L, Thomas L, Toumbourou J, PattonG, Catalano RF. A Profile of Youth in ourCommunity. Report prepared for theDepartment of Human Services. Melbourne:Centre for Adolescent Health, 2000.

782. National Crime Prevention - Attorney-General’s Department. Pathways to Prevention -

�(' �� !��"�#��$��%��

Developmental and Early Intervention Approaches toCrime in Australia - Summary Volume. Canberra:National Crime Prevention - Attorney-General’s Department, 1999.

783. Contributors to the Cochrane Collaborationand the Campbell Collaboration. Evidence fromSystematic Reviews of Research Relevant toImplementing the ‘Wider Public Health’ Agenda.York: NHS Centre for Reviews andDissemination, 2000.

784. Commonwealth Department of Health andAged Care, Australian Institute of Healthand Welfare. National Health Priority Areas Report:Cardiovascular Health. Canberra: AustralianInstitute of Health and Welfare, 1999.

785. Commonwealth Department of Health andAged Care. Cardiovascular Health: A Report onHeart, Stroke and Vascular Disease. National HealthPriority Areas Report Summary. AIHW CatalogueNo. PHE 12. Canberra: Australian Instituteof Health and Welfare, 1998.

786. Australian Institute of Health and Welfare.Heart, Stroke and Vascular Diseases. National Centrefor Monitoring Cardiovascular Disease. CatalogueNo. CVD 13. Canberra: Australian Instituteof Health and Welfare, 2001.

787. Joint Advisory Group on General Practiceand Population Health. Smoking, Nutrition,Alcohol and Physical Activitiy (SNAP) Framework forGeneral Practice. Canberra: CommonwealthDepartment of Health and Aged Care, June2001.

788. Commonwealth Department of Health andAged Care. Australia’s National TobaccoCampaign. Every Cigarette is Doing YouDamage, 2000. Viewed 13 December 2001,<http://www.health.gov.au/pubhlth/publicat/document/tobaccamp_2.pdf>.

789. Commonwealth Department of Health andAged Care. Directions in Injury Prevention. Report2: Injury Prevention Interventions-Good Buys for theNext Decade. Canberra: Commonwealth ofAustralia, April 1999.

790. Dinh-Zarr T, Diguiseppi C, Heitman E,Roberts I. Preventing injuries throughinterventions for problem drinking: Asystematic review of randomized controlledtrials. Alcohol and Alcoholism 1999; 34(4):609–621.

791. Gepkens A, Gunning-Schepers LJ.Interventions to reduce socioeconomic

health differences. A review of theinternational literature. European Journal ofPublic Health 1996; 6:218–226.

792. Herrman H. The need for mental healthpromotion. Australian and New Zealand Journal ofPsychiatry 2001; 35:709–715.

793. Davis C, Martin G, Kosky R, O’Hanlon A.Early Intervention in the Mental Health of YoungPeople: A Literature Review: Department ofHealth and Aged Care, 2000.

794. Commonwealth Department of Health andAged Care. National Action Plan for Promotion,Prevention and Early Intervention for Mental Health.Canberra: Commonwealth Department ofHealth and Aged Care, 2000.

795. Commonwealth Department of Health andAged Care. Promotion, Prevention and EarlyIntervention for Mental Health - A Monograph.Canberra: Commonwealth Department ofHealth and Aged Care, 2000.

796. Dadds MR. The comorbidity of substanceuse and emotional disorders: potential forinnovative prevention strategies. Paperpresented at National Comorbidity ProjectWorkshop, 6–7 March 2000.

797. Hall W, Lynskey M, Teesson M. What iscomorbidity and why does it matter? In.Sydney: National Drug and AlcoholResearch Centre; 1999.

798. Beyond Blue: the national depressioninitiative, 2001. Viewed 2 July 2002,<http://www.beyondblue.org.au/site/organisation/annualreport_2001.pdf>.

799. Mitchell P. Valuing Young Lives: Evaluation of theNational Youth Suicide Prevention Strategy.Melbourne: Australian Institute of FamilyStudies, 2000.

800. Canterbury Suicide Project. Bulletin No 9August 1996. Review of papers from thestudy, 1996. Viewed 9 April 2002, <http://www.chmeds.ac.nz/research/suicide/suicide.htm>.

801. Beautrais AL. Risk factors for suicide andattempted suicide among young people.Australian and New Zealand Journal of Psychiatry2000; 34:420–436.

802. Neeleman J, Farrell M. Suicide andsubstance misuse. British Journal of Psychiatry1997; 171:303–304.

�((��

803. Mitchell P. Primary Prevention and EarlyIntervention. Technical Report No. 2.Melbourne: Australian Institute of FamilyStudies, 2000.

804. Gowing LR, Proudfoot H, Henry-EdwardsSM, Teesson M. Evidence Supporting Treatment.The Effectiveness of Interventions for Illicit Drug Use.Woden: Australian National Council onDrugs, 2001.

805. Commonwealth Government Department ofFamily and Community Services. StrongerFamilies and Communities Strategy. Canberra:Australian Government Printing Service,2000.

806. Brown JM. The effectiveness of treatment.In Heather N, Peters TJ, Stockwell T (eds)International Handbook of Alcohol Dependence andProblems. Chichester: John Wiley and Sons,2001;497–508.

807. Proudfoot H, Teesson M. Investing in Drug andAlcohol Treatment. Technical Report No. 91.Sydney: National Drug and AlcoholResearch Centre, 2000.

808. Gowing LR, Ali R, White J. The Management ofOpioid Withdrawal: An Overview of ResearchLiterature. Parkside: Drug and AlcoholServices Council, August 2000.

809. Department of Human Services Victoria.Developing Best Practice Drug and Alcohol TreatmentService and Support Models for Young People ofCambodian, Lao and Vietnamese Origin. Victoria:Drug Treatment Services, Aged, Communityand Mental Health Division, Department ofHuman Services, May 1998.

810. Alati R. The Role of Alcohol and Drug Rehabilitationand Treatment amongst Aboriginal People. Canberra:Commonwealth Department of HumanServices and Health, 1996.

811. O’Brien CP, McLellan AT. Myths about thetreatment of addiction. Lancet 1996;347:237–240.

812. Stockwell T, Thiagarajan S, McGrath D, LangE. The measurement of alcohol dependenceand impaired control in communitysamples. Addiction 1994; 89:167–174.

813. Silagy C, Mant D, Folwer G, Lancaser T.Nicotine replacement therapy for smokingcessation. In The Cochrane Library, Issue 3.Oxford: Update Software, 2000.

814. Heather N. Pleasures and pains of ourfavourite drug. In Heather N, Peters TJ,Stockwell T (eds) International Handbook ofAlcohol Dependence and Problems. Chichester: JohnWiley and Sons, 2001;5–14.

815. Glaser FB, Heather N, Drummond DC,Finney JW, Lindstrom L, Sutton S, et al.Comments on Project MATCH: matchingalcohol treatments to client heterogeneity.Addiction 1999; 94(1):31–69.

816. McLellan AT, Hagan TA, Levine M, Gould F,Meyers K, Bencivengo M, et al.Supplemental social services improveoutcomes in public addiction treatment.Addiction 1998; 93(10):1489–1499.

817. Gossop M, Marsden J, Stewart D. NTORS AtOne Year: Changes in Substance Use, Health andCriminal Behaviour One Year After Intake. London:Department of Health, 1998.

818. Shearer J, Wodak A, Mattick RP, van Beek I,Lewis J, Hall W, et al. A Randomised ControlledTrial of the Feasibility of Monitoring ControlledPrescribing of Dexamphetamine. Technical ReportNo. 75. Sydney: National Drug and AlcoholResearch Centre, 1999.

819. Swift W, Copeland J, Lenton S. Cannabisand harm reduction. Drug and Alcohol Review2000; 19(1):101–112.

820. Hannifin J. The ownership debate: cannabisand the concept of cannabism. In ChesherG, Consroe P, Musty R (eds) Marijuana: AnInternational Research Report, National CampaignAgainst Drug Abuse, Monograph Series 7. Canberra:Australian Government Publishing Service,1988.

821. Miller NS, Gold M. The diagnosis ofmarijuana (cannabis) dependence. Journal ofSubstance Abuse Treatment 1989; 6:183–192.

822. Miller NS, Gold MS, Pottash AC. A 12-steptreatment approach for marijuanadependence. Journal of Substance Abuse Treatment1989; 6:241–250.

823. Grenyer B, Solowij N, Peters R. A Guide toQuitting Marijuana. Sydney: National Drug andAlcohol Research Centre; nd.

824. Hancock L, Williams S, Shand D, Quigley A.Focus: A Guide to Cutting Back on Cannabis. Perth:Next Step Specialist Drug and AlcoholServices, In preparation 2002.

�)) �� !��"�#��$��%��

825. Bleeker A, Malcolm A. Mulling it Over: HealthInformation for People Who Use Cannabis. Sydney:Manly Drug Education and CounsellingCentre; 1998.

826. Hancock L, Quigley A, Shand D.Development of a brief intervention foradult cannabis users: eliminating guessworkthrough a consultative process. Paperpresented at Annual Conference of theAustralian Professional Society on Alcoholand other Drugs (APSAD). Canberra, 8–10November 1999.

827. Hulse GK, Tait RJ. Six-month outcomesassociated with a brief alcohol interventionfor adult in-patients with psychiatricdisorders. Drug and Alcohol Review 2002;21:105–112.

828. Ward J, Mattick R, Hall W. Key Issues inMethadone Maintenance Treatment. Sydney:National Drug and Alcohol Research Centre;1992.

829. Mattick RP, Oliphant D, Ward J, Hall W,White J. The Swiss Scientific Studies of MedicallyPrescribed Narcotics. The Effectiveness of Other OpioidReplacement Therapies: LAAM, Heroin, Buprenorphineand Naltrexone. Technical Report No. 43.Sydney: National Drug and AlcoholResearch Centre, 1997.

830. Smart RG, Mann RE. The impact ofprograms for high-risk drinkers onpopulation levels of alcohol problems.Addiction 2000; 95(1):37–52.

831. Kristenson H, Ohein H, Hulten-Nosslin MB,Trell E, Hood B. Identification andintervention of heavy drinking in middle-aged men: results and follow-up of 24-60months of long-term study withrandomized controls. Alcoholism: Clinical andExperimental Research 1983; 7(2):203–209.

832. Teesson M, Proudfoot H. Interventions foralcohol dependence, abuse and excessivedrinking. Paper presented at NationalWorkshop on Alcohol Research Priorities.Adelaide, March 2001.

833. Weatherburn D, Lind B. Street-level druglaw enforcement and entry into methadonemaintenance treatment. Addiction 2001;96:577–587.

834. Gerstein DR, Harwood H, Suter N, MalloyK. Evaluating Recovery Services: The California Drugand Alcohol Treatment Assessment (CALDATA).

Sacramento: Department of Alcohol andDrug Programs, 1994.

835. Harwood HJ, Hubbard RL, Collins JJ, RachalJV. The costs of crime and the benefits ofdrug treatment: a cost-benefit analysis usingTOPS data. In Leukefed CG, Tim FM (eds)Compulsory Treatment of Drug Abuse - Research andClinical Practice. Washington DC: NationalInstitute on Drug Abuse, 1988;209–235.

836. Gossop M, Marsden J, Stewart D, EdwardsC, Lehmann P, Wilson A, et al. The nationaltreatment outcome research study in theUnited Kingdom: six month follow-upoutcomes. Psychology of Addictive Behaviors 1997;11(4):324–337.

837. Swedish Council on Technology Assessmentin Health Care. Treatment of alcohol anddrug abuse: an evidence-based review.International Journal of Technology Assessment inHealth Care 2002; 18(1):145–154.

838. Dwyer T, Pierce JP, Hannam CD, Burke N.Evaluation of the Sydney ‘Quit for Life’ anti-smoking campaign. Part 2: Changes insmoking prevalence. Medical Journal of Australia1986; 144.

839. National Institute of Alcohol Abuse andAlcoholism. Community-based preventionapproaches. In 10th Special Report to the USCongress on Alcohol and Health. Bethesda, MD:National Institute of Alcohol Abuse andAlcoholism, 2000;397–411.

840. Casswell S, Gilmore L. An evaluatedcommunity action project on alcohol. Journalof Studies on Alcohol 1989; 50(4):339–346.

841. Rhode Island Department of Health. FinalReport of the Rhode Island Community Alcohol Abuseand Injury Prevention Project, Volume I: Descriptionand Process Evaluation. Providence, RI: RhodeIsland Department of Health, 1989.

842. Rhode Island Department of Health. FinalReport of the Rhode Island Community Alcohol Abuseand Injury Prevention Project, Volume II: StatisticalOutcome Evaluation. Providence, RI: RhodeIsland Department of Health, 1989.

843. Putnam SL, Rockett IRH, Campbell MK.Methodological issues in community-basedalcohol-related injury prevention project:attribution of program effects. In GreenfieldTK, Zimmerman R (eds) Experiences withCommunity Action Projects: New Research in thePrevention of Alcohol and Other Drug Problems (CSAP

�)��

Prevention Monograph 14). Washington, DC:U.S. Government Printing Office, 1993.

844. Stout RL. Prevention experiments in thecontext of ongoing community processes:opportunities or obstacles for research. InHolder HD, Howard JM (eds) CommunityPrevention Trials for Alcohol Problems. Westport,CT: Praeger, 1992.

845. Cooper M, Midford R, Jaeger J, Hall C.Partysafe Evaluation Report. Perth: National DrugResearch Institute, Curtin University ofTechnology, 2001.

846. Aguirre-Molina M, Gorman DM.Community-based approaches for theprevention of alcohol, tobacco, and otherdrug use. Annual Review of Public Health 1996;17:337–358.

847. Gorman DM, Speer PW. Preventing alcoholabuse and alcohol related problems throughcommunity interventions: A review ofevaluation studies. Psychology and Health 1996;11:95–131.

848. Wallin E, Norstrom T, Andreasson S.Alcohol prevention targeting licensedpremises: a study of effects on violence.Journal of Studies on Alcohol 2003; 64(2):270–277.

849. Holder HD, Saltz RF, Grube JW, Treno A,Reynolds RI, Jr, Voas RB, et al. Summingup: lessons from a comprehensivecommunity prevention trial. Addiction 1997;92(Suppl. 2):S293–S301.

850. Midford R, Boots K. COMPARI: insightsfrom a three year community based alcoholharm reduction project. Australian Journal ofPrimary Health - Interchange 1999; 5(46–58).

851. Boots K, Midford R. ‘Pick-a-Skipper’: anevaluation of a designated driver program toprevent alcohol related injury in a regionalAustralian city. Health Promotion International1999; 14(4):337–345.

852. Casswell S. A decade of community actionresearch. Substance Use and Misuse 2000;35(1&2):55–74.

853. Mundy J. Grassroots Action. Connexions 2000;August–October.

854. Murphy T. NSW Drugs Summit—a viewfrom the other side. Drug and Alcohol Review2000; 19:134–136.

855. Bolleter A, Loxley W. Prevention in action:evaluating the Community PartnershipsInitiative. Paper presented at 3rdInternational Conference on Drugs andYoung People. Sydney, 13–15 May 2002.

856. Tobler N. Drug prevention programs canwork: research findings. Journal of AddictiveDiseases 1992; 11(3):1–28.

857. Sciacca J. Peer health education: a powerfulyet inexpensive helping strategy. PeerFacilitator Quarterly 1987.

858. Holder H, Giesbrecht N. Concepts andissues for community based action toprevent alcohol and other drug relatedproblems. Paper presented at Symposium onExperiences with Community ActionProjects for the Prevention of Alcohol andOther Drug Problems. Toronto, 1989.

859. Gray D, Sputore B, Stearne A, Bourbon D,Strempel P. Indigenous Drug and Alcohol Projects:1999-2000. ANCD Research Paper No. 4.Canberra: Australian National Council onDrugs, 2002.

860. Brady M. Broadening the Base of Interventions forAboriginal People With Alcohol Problems. TechnicalReport No. 29. Sydney: National Drug andAlcohol Research Centre, 1995.

861. Brady M, Hunter E. Talking About AlcoholWith Indigenous Patients: A BriefIntervention For Health Professionals.Helping Aboriginal And Torres StraitIslander Patients At Risk Of Alcohol-RelatedProblems. In. Canberra: Office of Aboriginaland Torres Strait Islander Health,Department of Health and Aged Care; 2001.

862. Hunter E, Brady M, Hall W. NationalRecommendations for the Clinical Management ofAlcohol-Related Problems in Indigenous Primary CareSettings. Brisbane: University of Queensland,Australian Institute of Aboriginal and TorresStrait Islander Studies and National Drugand Alcohol Research Centre, 1999.

863. Couzos S, Murray R. Aboriginal Primary HealthCare: An Evidence-Based Approach. Melbourne:Oxford University Press; 1999.

864. O’Connor R, Associates Pty Ltd. Report on theAboriginal Alcohol Treatment/ RehabilitationProgrammes Review and Consultation. Typescript.Perth: Western Australian Alcohol and DrugAuthority, 1988.

�)� �� !��"�#��$��%��

865. d’Abbs P. Responding to Aboriginal SubstanceMisuse: A Review of Programs Conducted by theCouncil for Aboriginal Alcohol Program Services(CAAPS), Northern Territory. Darwin: NorthernTerritory Drug and Alcohol Bureau,Department of Health and CommunityServices & Aboriginal and Torres StraitIslander Commission, 1990.

866. Miller K, Rowse T. CAAAPU (Central AustralianAboriginal Alcohol Programs Unit): An Evaluation.Menzies Occasional Papers Issue No. 1/95.Darwin: Menzies School of Health Research,1995.

867. Sputore B, Gray D, Bourbon D, Baird K.Evaluation of Kununurra-Waringarri AboriginalCorporation and Ngnowar-Aerwah AboriginalCorporation’s Alcohol Projects. Perth: NationalCentre for Research into the Prevention ofDrug Abuse, Curtin University ofTechnology, 1998.

868. Sputore B, Gray D, Sampi C. Review of theServices Provided by Jungarni-Jutiya Alcohol ActionCouncil Aboriginal Corporation. Perth: NationalDrug Research Institute, Curtin Universityof Technology, 2000.

869. Gray D, Saggers S, Sputore B, Bourbon D.What works? A review of evaluated alcoholmisuse interventions among AboriginalAustralians. Addiction 2000; 95(1):11–22.

870. Edwards G, Frances R, Lehmann T.Community Report: Injecting Drug Use Project.Melbourne: Victorian Aboriginal HealthService Cooperative, 1998.

871. Brady M. The Grog Book: Strengthening IndigenousAction on Alcohol. Canberra: CommonwealthDepartment of Health and Family Services;1998.

872. Alati R. Evaluation of the Koori Alcohol and DrugPrevention Project: Final Report. Melbourne: KooriHealth Unit, Victorian Department of Healthand Community Services, 1993.

873. Milne C, Josif P, Lynn D. Raypirri: Evaluation ofthe Northern Territory Bush tour Project, April-June1993. Canberra: Public Affairs Branch,Commonwealth Department of Health,Housing, Local Government andCommunity Services, 1993.

874. Barber J, Walsh C, R. B. Alcohol Education forAboriginal Children. National Campaign AgainstDrug Abuse Monograph Series No. 10.Canberra: Government Publishing Service,1989.

875. Sheehan M, Schonfeld C, Hindson E, BallardR. Alcohol education in an indigenouscommunity school in Queensland, Australia.Drugs: Education, Prevention and Policy 1995;2:259–273.

876. Gray D, Sputore B, Walker J. Evaluation ofan Aboriginal health promotion program: acase study from Karalundi. Health PromotionJournal of Australia 1998; 8:24–28.

877. Walley C, James R. Producing culturallyappropriate health promotion resources forAboriginal people. Aboriginal and Islander HealthWorker Journal 1995; 19(1):6–9.

878. Dale G. Jabby don’t smoke: developingresources to address tobacco consumptionin remote Aboriginal communities. Paperpresented at 11th National HealthPromotion Conference: Building SocialCapital in the 21st Century. Perth, 1999.

879. Eades D, Ambrose M. Yamatji smokingprevention program. Paper presented at11th National Health PromotionConference: Building Social Capital in the21st Century. Perth, 1999.

880. Johnston F, Beecham R, Dalgleish P,Malpraburr T, Gamarania G. TheManingrida be smoke free project. HealthPromotion Journal of Australia 1998; 8(1):12–17.

881. Anonymous. Recommendations from theworkshop ‘Preventing smoking-relateddiseases among Aboriginals’. Aboriginal HealthInformation Bulletin 1990; 14:14–15.

882. O’Connor M, Parker E, Mieklejohn B,Oldenburg B, Alati R. Heart healtheducation resources for Aboriginal andTorres Strait Islander communities: thehealth worker’s view. Aboriginal and IslanderHealth Worker Journal 1999; 23(2):20–23.

883. Australian Medical Association, AustralianPharmaceutical Manufacturer’s Association.Indigenous Smoking: Issues and Responses. Canberra:Australian Medical Association, AustralianPharmaceutical Manufacturer’s Association,2000.

884. San Roque C, in association with, Lefay LB,Cook B, Cook E, et al. The Western Line Project:A Report to Living with Alcohol: Injartnama AboriginalCorporation Alcohol, Drugs and Volatile SubstancesMisuse Prevention Activities. Typescript. AliceSprings: Injartnama Aboriginal Corporation,1999.

�)��

885. Stojanovski AJ. Mt Theo Story - About PetrolSniffing: July 1993 Through September 1994.Typescript. Yuendumu1994.

886. Morice R, Swift H, Brady M. Petrol SniffingAmong Aboriginal Australians: A Resource Manual.Adelaide: Western Desert Project, FlindersUniversity of South Australia, 1981.

887. Simpson M. Atyempe: “The Strong Ironwood Tree”.The DETOUR Project: A New Learning Place forAboriginal Youth. Alice Springs: TangentyereCouncil, 1995.

888. Burns C, Currie BJ, Clough A, Wuridjal R.Evaluation of strategies used by a remoteAboriginal community to eliminate petrolsniffing. Medical Journal of Australia 1995;163(2):82–86.

889. Lowe HJ, Franks C, Japaltjarri AS. Rationaleand Operation of Petrol Sniffing Prevention Team.Typescript. Alice Springs1985.

890. Kirby M, Denihan A, Bruce I, Radic A,Coakley D, Lawlor BA. Benzodiazepine useamong the elderly in the community.International Journal of Geriatric Psychiatry 1999;14:280–284.

891. Philips M. Drug testing at work. How doesit work? Is it worthwhile? In Midford R,Heale P (eds) Under the Influence? Issues andPracticalities of Alcohol and Other Drug Testing in theWorkplace. Proceedings of a Forum held in Perth,Western Australia, 8 October 1996. Perth:National Centre for Research into thePrevention of Drug Abuse, 1997;43–60.

892. Bellamy A. Workplace drug testing - theemployer’s perspective. In Midford R, HealeP (eds) Under the Influence? Issues and Practicalitiesof Alcohol and Other Drug Testing in the Workplace.Proceedings of a Forum held in Perth, WesternAustralia, 8 October 1996. Perth: NationalCentre for Research into the Prevention ofDrug Abuse, 1997;65–68.

893. Nolan J. Employee drug testing: somerecent legal developments. In Allsop S,Phillips M, Calogero C (eds) Drugs and Work.Responding to Alcohol and Other Drug Problems in theAustralian Workplace. Melbourne: IPCommunications, 2001;60–73.

894. Cooke T. Workplace drug testing - theemployee’s perspective. In Midford R, HealeP (eds) Under the Influence? Issues and Practicalitiesof Alcohol and Other Drug Testing in the Workplace.Proceedings of a Forum held in Perth, Western

Australia, 8 October 1996. Perth: NationalCentre for Research into the Prevention ofDrug Abuse, 1997;61–64.

895. Ralph M. Alcohol and other drugs in theworkplace - why be concerned? In MidfordR, Heale P (eds) Under the Influence? Issues andPracticalities of Alcohol and Other Drug Testing in theWorkplace. Proceedings of a Forum held in Perth,Western Australia, 8 October 1996. Perth:National Centre for Research into thePrevention of Drug Abuse, 1997;31–42.

896. Holder H. Prevention of alcohol problemsin the workplace: a public policyperspective. In Roman PM (ed) AlcoholProblem Intervention in the Workplace: EmployeeAssistance Programs and Strategic Alternatives. NewYork: Quorum Books, 1990;361–368.

897. Roman PM, Blum TC. The workplace andalcohol problem prevention. Alcohol Researchand Health 2002; 26:49–57.

898. Midford R. Alcohol and other drugs. InWyatt A, Oxenburgh M (eds) ManagingOccupational Health and Safety. Sydney: CCHAustralia Ltd, 2003.

899. Ames G. Research and strategies for theprimary prevention of workplace alcoholproblems. Alcohol Health and Research World1993; 17(1):19–27.

900. Shahandeh B. Drug and alcohol abuse in theworkplace: consequences andcountermeasures. International Labour Review1985; 124(2):207–223.

901. Trice H. Work-related risk factors associatedwith alcohol abuse. Alcohol Health and ResearchWorld 1992; 16(2):106–111.

902. Ames G, Janes C. A cultural approach toconceptualising alcohol and the workplace.Alcohol Health and Research World 1992;16(2):112–119.

903. Corry A. Controls on drug use. In Allsop S,Phillips M, Calogero C (eds) Drugs and Work.Responding to Alcohol and Other Drug Problems in theAustralian Workplace. Melbourne: IPCommunications, 2001;105–144.

904. Makkai T. Drug Use Monitoring in Australia(DUMA): Drug Detection Testing. AIC Researchand Public Policy Series No. 25. Canberra:Australian Institute of Criminology, 2000.

905. Lewis J. Drug Detection and its Role in LawEnforcement. Trends and Issues in Crime and

�)� �� !��"�#��$��%��

Criminal Justice Series Report No. 205.Canberra: Australian Institue ofCriminology, May 2001.

906. Moskowitz H, Robinson CD. Effects of LowDoses of Alcohol on Driving Related Skills: A Review ofthe Evidence. Washington, DC: NationalHighway Traffic Safety Administration,1988.

907. Baker T. The plutonic gold experience. InMidford R, Heale P (eds) Under the Influence?Issues and Practicalities of Alcohol and Other DrugTesting in the Workplace. Proceedings of a Forum heldin Perth, Western Australia, 8 October 1996. Perth:National Centre for Research into thePrevention of Drug Abuse, 1997;69–76.

908. Allsop S, Phillips M. An overview of drugtesting in the workplace - key note address.In Midford R, Heale P (eds) Under theInfluence? Issues and Practicalities of Alcohol and OtherDrug Testing in the Workplace. Proceedings of a Forumheld in Perth, Western Australia, 8 October 1996.Perth: National Centre for Research into thePrevention of Drug Abuse, 1997;1–18.

909. Tinelli G, Hamlett L, Smithson M. Fit forduty... or is it? The Voice 1999:20–21, 27–28, 30.

910. Kinder B, Pape N, Walfish S. Drug andalcohol education: a review of outcomestudies. International Journal of the Addictions1980; 15:1035–1054.

911. Shain M. Alternatives to drug testing:employee assistance and health promotionprograms. In McDonald S, Roman PM (eds)Research Advances in Alcohol and Drug Problems,Volume 11, Drug Testing in the Workplace. NewYork: Plenum Press, 1994;257–277.

912. McVey D, Stapleton J. Can anti-smokingtelevision advertising affect smokingbehaviour? Controlled trial of the HeathEducation Authority for England’s anti-smoking TV campaign. Tobacco Control 2000;9:273–282.

913. Boots K, Midford R. Mass media marketingand advocacy to reduce alcohol-relatedharm. In Heather N, Peters TJ, Stockwell T(eds) International Handbook of Alcohol Dependenceand Problems. Chichester: John Wiley andSons, 2001;805–822.

914. Hastings G, Haywood A. Social marketingand communication in health promotion.Health Promotion International 1991; 6(2):135–145.

915. Egger G, Donovan R, Spark R. Health and theMedia: Principles and Practices for Health Promotion.Sydney: McGraw Hill; 1993.

916. Backer TE, Rodgers EM, Sopory P. DesigningHealth Communication Programs: What Works?Newbury Park, CA: Sage Publications; 1992.

917. Montagne M, Scott DM. Prevention ofsubstance use problems: models, factors andprocesses. International Journal of the Addictions1993; 28:1177–1208.

918. Pierce JP, Dwyer T, Frape G, Chapman S,Chamberlain A, Burke N. Evaluation of theSydney “Quit for Life” anti-smokingcampaign. Part I. Achievement ofintermediate goals. Medical Journal of Australia1986; 144:341–344.

919. Pierce JP, Macaskill P, Hill D. Long-termeffectiveness of mass media led anti-smoking campaigns in Australia. AmericanJournal of Public Health 1990; 80:565–569.

920. Hassard K. Australia’s National Tobacco Campaign:Evaluation Report Volume One. Canberra:Commonwealth Department of Health andAged Care, 2000.

921. Hassard K. Australia’s National Tobacco Campaign:Evaluation Report Volume Two. Canberra:Commonwealth Department of Health andAged Care, 2000.

922. Holder HD, Treno AJ. Media advocacy incommunity prevention: news as a means toadvance policy change. Addiction 1997;92(2):S189–S199.

923. Strunge H. Danish experiences of nationalcampaigns on alcohol, 1990-1996. Drugs:Education, Prevention and Policy 1998; 5(1):73–79.

924. Carroll T. Australia’s National alcoholeducation and risk awareness campaign:Generating a focus and creatingopportunities. In. Brisbane: CommonwealthDepartment of Human Services and Health;1995.

925. Cameron M, Newstead S. Mass mediapublicity supporting police enforcementand its economic value. Paper presented atPublic Health Association of Australia 28thAnnual Conference Symposium on MassMedia Campaigns in Road Safety, 30September 1996.

�)"��

926. Plant A, Macaskill P, Lo SK, Pierce JP. Reportof the Evaluation of the Anit-heroin Campaign.Conducted by the National CampaignAgainst Drug Abuse. Sydney: Department ofPublic Health, University of Sydney, 1988.

927. National Drugs Campaign: Campaignoutline. Department of Health and Ageing,nd. Viewed 28/10/02 2002, <http://www.drugs.health.gov.au/media/campaign.htm>.

928. Bertram S, Worsley J, Carroll T. Evaluation ofthe Launch Phase of the National Illicit DrugsCampaign. Chapter 1: Overview. Research Report.Sydney: Government Department of Healthand Ageing, January 2002.

929. DeJong W. College students’ use ofdesignated drivers: the data are in. PreventionPipeline 1997; Nov/Dec(25–27).

930. Wallack L. Two approaches to healthpromotion in the mass media. World HealthForum 1990; 11:143–164.

931. Wallack L, Dorfman L, Jernigan D, ThembaM. Media Advocacy and Public Health—Power forPrevention. Newbury Park, CA: SagePublications; 1993.

932. Erickson A, Ricks A, McKenna J, Romano R.Past lessons and new uses of the media inreducing tobacco consumption. Public HealthReports 1990; 105:244–257.

933. Chapman S, Lupton D. The Fight for PublicHealth—Principles and Practice of Media Advocacy.London: British Medical Journal PublishingGroup; 1994.

934. Hawks D. Not Much to Ask for, Really! TheIntroduction to Standard Drink Labelling inAustralia. In; 1996.

935. Stockwell T, Single E. Standard unit labellingof alcohol containers. In Plant M, Single E,Stockwell T (eds) Alcohol. Minimising the Harm.What Works? London: Free Association Press,1997;85–104.

936. Holder HD. Mass communication as anessential aspect of community prevention toreduce alcohol-involved traffic crashes.Alcohol, Drugs and Driving 1994; 10:295–307.

937. National Health and Medical ResearchCouncil. Is there a Safe Level of Daily Consumptionof Alcohol for Men and Women? 1st edition.Canberra: Australian Goverment PrintingServices, 1987.

938. National Health and Medical ResearchCouncil. Is there a Safe Level of Daily Consumptionof Alcohol for Men and Women? 2nd edition.Canberra: Australian Government PrintingServices, 1992.

939. Collins DJ, Lapsley HM. The Social Costs of DrugAbuse in Australia in 1988 and 1992. Canberra:Commonwealth Department of HumanServices and Health, 1996.

940. Stockwell T, Beel A. Public Support for theIntroduction of Standard Drink Labels on AlcoholContainers. Perth: National Centre forResearch into the Prevention of Drug Abuse,Curtin University of Technology, 1994.

941. National Food Authority. A Guide to Meeting theNew Standard Drink Labelling Requirement forAlcoholic Beverages. Canberra: National FoodAuthority, 1995.

942. Australian Institute of Health and Welfare.2001 National Drug Strategy Household Survey: Stateand Territory Supplement. Drug Statistics Series 10.Canberra: Australian Institute of Health andWelfare, 2002.

943. Greenfield TK, Graves K, Kaskutas L.Alcohol warning labels for prevention:national survey findings. Alcohol Health andResearch World 1993; 17:67–75.

944. U.S. Department of Health and HumanServices. Reducing Tobacco Use: A Report of theSurgeon General. Atlanta, Georgia: U.S.Department of Health and Human Services,Centers for Disease Control and Prevention,National Center for Chronic DiseasePrevention and Health Promotion, Office onSmoking and Health, 2000.

945. Chaloupka FJ. Macro-social influences: Theeffects of prices and tobacco control policieson the demand for tobacco products. Nicotineand Tobacco Research 1999; 1(S1):S105–109.

946. Borland R, Hill D. Initial impact of the newAustralian tobacco health warnings onknowledge and beliefs. Tobacco Control 1997;6(4):317–325.

947. Wyllie A, Zhang JF, Casswell S. Positiveresponses to televised beer advertisementsassociated with drinking and problemsreported by 18 to 29 -year-olds. Addiction1998; 93:749–760.

948. Wyllie A, Zhang J-F, Casswell S. Responsesto televised alcohol advertisements

�)# �� !��"�#��$��%��

associated with drinking behaviour of 10-17-year-olds. Addiction 1998; 93:361–372.

949. Grube JW, Nygaard P. Adolescent drinkingand alcohol policy. Contemporary Drug Problems2001; 28:87–131.

950. World Health Organization. Report of TechnicalAdvisory Group on Alcohol Promotions and YoungPeople. Valencia, Spain7-10 May Inpreparation 2002.

951. Geller ES, Kalsher MJ, Clarke SW. Beerversus mixed drink consumption atfraternity parties: a time and place for low-alcohol alternatives. Journal of Studies on Alcohol1991; 52:197–203.

952. Plant M, Single E, Stockwell T, eds. Alcohol:Minimising the Harm. What Works? London;New York: Free Association Books Ltd;1997.

953. Solomon R, Payne J. Alcohol liability inCanada and Australia: sell, serve and besued. Tort Law Review 1996; 4(3):188–241.

954. Gruenewald P, Stockwell T. Availabilitypolicy and the reduction of alcohol-relatedharm. In: Alcohol and Public Policy Project,WHO. Berkeley; 2001.

955. Alcohol and other Drugs Council ofAustralia. Drug Policy 2000: A New Agenda forHarm Reduction: Alcohol and other DrugsCouncil of Australia, June 2000.

956. Stockwell T, Single E. Reducing harmfuldrinking. In Stanton P, Grant M (eds) Alcoholand Pleasure. A Health Perspective. Philadelphia:Taylor and Francis, 1999;357–373.

957. Saltz RF. Prevention where alcohol is soldand consumed: server intervention andresponsible beverage service. In Plant M,Single E, Stockwell T (eds) Alcohol: Minimisingthe Harm. What Works? New York: FreeAssociation Books Ltd, 1997;72–84.

958. Graham K, Homel R. Creating safer bars. InSingle E, Stockwell T (eds) Alcohol: Minimisingthe Harm. What Works? New York: FreeAssociation Books Ltd, 1997;171–192.

959. Osterberg E. Effects of price and taxation. InHeather N, Peters TJ, Stockwell T (eds)International Handbook of Alcohol Dependence andProblems. Chichester: John Wiley and Sons,2001;685–698.

960. McKnight AJ, Voas RB. Prevention ofalcohol-related road crashes. In Heather N,

Peters TJ, Stockwell T (eds) InternationalHandbook of Alcohol Dependence and Problems.Chichester: John Wiley and Sons,2001;741–770.

961. Treno AJ, Holder HD. Prevention at thelocal level. In Heather N, Peters TJ,Stockwell T (eds) International Handbook ofAlcohol Dependence and Problems. Chichester: JohnWiley and Sons, 2001;771–784.

962. Stockwell T, Dietze P, Chikritzhs T, CatalanoP, Heale P. How much alcohol is drunk inAustralia in excess of the new Australianalcohol guidelines? Letter to the Editor.Medical Journal of Australia 2002; 176(2):91–92.

963. Norstrom T, Skog O-J. Alcohol andmortality: methodological and analyticalissues in aggregate analyses. Addiction 2001;96 Suppl 1:S5–S17.

964. Chikritzhs T, Jonas H, Heale P, Dietze P,Hanlin K, Stockwell T. Alcohol-Caused Deathsand Hospitalisations in Australia, 1990-1997.National Alcohol Indicators Project BulletinNo. 1. Perth: National Drug ResearchInstitute, Curtin University of Technology,1999.

965. Catalano P, Chikritzhs T, Stockwell T, WebbM, Rohlin C-J, Dietze P. Trends in Per CapitaAlcohol Consumption in Australia, 1990/91 -1998/99. National Alcohol Indicators ProjectBulletin No. 4. Perth: National DrugResearch Institute, Curtin University ofTechnology, May 2001.

966. Stockwell T. Alcohol policy, harm reductionand the prevention paradox. Journal of HealthPromotion for Northern Ireland 2001; 15:22–25.

967. Gruenewald PJ, Stockwell T, Beel A, DyskinE. Beverage sales and drinking and driving:the role of on-premise drinking places.Journal of Studies on Alcohol 1999; 60(1):47–53.

968. Stockwell T, Masters L, Philips M, Daly A,Gahegan M, Midford R, et al. Consumptionof different alcoholic beverages as predictorsof local rates of night-time assault and acutealcohol-related morbidity. Australian and NewZealand Journal of Public Health 1998;22(2):237–242.

969. Godfrey C. Can tax be used to minimiseharm? A health economist’s perspective. InPlant M, Single E, Stockwell T (eds) Alcohol.Minimising the Harm. What Works? London: FreeAssociation Books Ltd, 1997;29–42.

�)%��

970. Casswell S. Population level policies onalcohol: are they still appropriate given that“alcohol is good for the heart”? Addiction1997; 92(S1):S81–S90.

971. Beel A, Stockwell T. The Introduction of 0.05Legislation in Western Australia: A PreliminaryResearch Report. Perth, Western Australia:National Centre for Research into thePrevention of Drug Abuse, CurtinUniversity of Technology, 1993.

972. Chikritzhs T, Stockwell T. The impact oflater trading hours for Australian publichouses (hotels) on levels of violence. Journalof Studies on Alcohol 2002; 63(5):591–599.

973. Briscoe S, Donnelly N. Assaults on LicencedPremises in Inner-Urban Areas. Report No. 2.Sydney: NSW Bureau of Crime Statistics andResearch and the National Drug ResearchInstitute, October 2001.

974. Ragnarsdóttir T, Kjartansdóttir A,Davíðsdóttir S. Effect of extended alcoholserving hours in Reykjavík. In Room R (ed)The Effects of Nordic Alcohol Policies. Helsinki:Nordic Council for Alcohol and DrugResearch, 2002;145–154.

975. Catalano P, Stockwell TR. Driving After Drinkingin Western Australia. Perth: National DrugResearch Institute, Curtin University ofTechnology, 2002.

976. Hauritz M, Homel R, McIlwain G, BurrowsT, Townsley M. Reducing violence inlicensed venues through community safetyaction projects: the Queensland experience.Contemporary Drug Problems 1998; 25:511–551.

977. Braithwaite J. The new regulatory state andthe transformation of criminology. BritishJournal of Criminology 2000; 40:222–238.

978. Bourbon D, Saggers S, Gray D. IndigenousAustralians and Liquor Licensing Legislation. Perth:National Centre for Research into thePrevention of Drug Abuse, CurtinUniversity of Technology, 1999.

979. Gray D, Drandich M, Moore L, Wilkes T,Riley R, Davies S. Aboriginal well-being andliquor licensing legislation in WesternAustralia. Australian Journal of Public Health1995; 19(2):177–185.

980. d’Abbs P. Restricted areas and Aboriginaldrinking. In Vernon J (ed) Alcohol and Crime:Proceedings of a Conference Held 4-6 April 1989.

Canberra: Australian Institute ofCriminology, 1990.

981. Larkins KP, McDonald DN. Recent NorthernTerritory liquor control initiatives. AustralianAlcohol/Drug Review 1984; 3:263–271.

982. Douglas M. Restrictions of the hours of salein a small community: a beneficial impact.Australian and New Zealand Journal of Public Health1998; 22(6):714–719.

983. d’Abbs P, Togni S. The Derby Liquor LicensingTrial: A Report on the Impact of Restrictions onLicensing Conditions between 12 January 1997 and12 July 1997. Darwin: Menzies School ofHealth Research, 1997.

984. d’Abbs P, Togni S, Crundall I. The TennantCreek Liquor Licensing Trial, August 1995 - February1996: An Evaluation (Revised Edition). Darwin:Menzies School of Health Research and NTLiving With Alcohol Program, June 1996.

985. Gray D, Saggers S, Atkinson D, Sputore B,Bourbon D. Beating the grog: an evaluationof the Tennant Creek liquor licensingrestrictions. Australian and New Zealand Journal ofPublic Health 2000; 24(1):39–44.

986. d’Abbs P, Togni S, Stacey N, Fitz J. AlcoholRestrictions in Tennant Creek: A Review Prepared forthe Beat the Grog Committee, Tennant Creek, NorthernTerritory. Darwin: Menzies School of HealthResearch, 2000.

987. d’Abbs P, Togni S, Duquemin A. Evaluation ofRestrictions on the Sale of Alcohol from Curtin SpringsRoadside Inn, Northern Territory. A ReportPrepared for the Office for Aboriginal andTorres Strait Islander Health. Darwin:Menzies School of Health Research, 1998.

988. d’Abbs P, Togni S. Liquor licensing andcommunity action in regional and remoteAustralia: a review of recent initiatives.Australian and New Zealand Journal of Public Health2000; 24(1):45–53.

989. Gray D. Indigenous Australians and liquorlicensing restrictions (invited and refereededitorial). Addiction 2000; 95(10):1469–1472.

990. Brady M. Indigenous and governmentattempts to control alcohol use amongAustralian Aborigines. Contemporary DrugProblems 1990; 17(2):195–220.

991. Brady M, Palmer K. Alcohol in the Outback: TwoStudies in Drinking. Darwin: North Australia

�)' �� !��"�#��$��%��

Research Unit, Australian NationalUniversity, 1984.

992. Martin DF. The Supply of Alcohol in RemoteAboriginal Communities: Potential Policy Directionsfrom Cape York. Discussion paper No. 162/1998. Canberra: Centre for AboriginalEconomic Policy Research, AustralianNational University, 1998.

993. d’Abbs P. Out of sight, out of mind?Licensed clubs in remote Aboriginalcommunities. Australian and New Zealand Journalof Public Health 1998; 22(6):679–684.

994. Health Insurance Commission. About thePharmaceuticals Benefit Scheme (PBS) -About Doctor Shopping, 2001. Viewed 3June 2002, <http://www.hic.gov.au/yourhealth/our_services/apbs/doctorshop.htm>.

995. Dobbin M. The Victorian temazepaminjection prevention initiative. Health ofVictorians. The Chief Health Officer’sBulletin, 2002. Viewed 21 May 2002,<http://www.heatlh.vic.gov.au/chiefhealthofficer/downloads/vol2no1jan2002/temazepam.pdf>.

996. Dobbin M. Temazepam Injecting In Victoria 2001.Discussion Paper. Melbourne: Department ofHuman Services, Victoria, November 2001.

997. Drugs and Crime Prevention Committee.Inquiry into the Inhalation of Volatile Substances.Melbourne: Parliament of Victoria, January2002.

998. Grasmick HG, Bursik RJ. Conscience,significant others, and rational choice:Extending the deterrence model. Law andSociety Review 1990; 24(3):837–861.

999. Braithwaite V. The Community Hopes, Fears andActions Survey: Goals and Measures. WorkingPaper No. 2. Canberra: Centre for TaxSystem Integrity, Research School of SocialSciences, The Australian NationalUniversity, 2001.

1000. Makkai T, Braithwaite J. Reintegrativeshaming and compliance with regulatorystandards. Criminology 1994; 32(3):361–385.

1001. MacCoun RJ. Drugs and the law: Apsychological analysis of drug prohibition.Psychological Bulletin 1993; 113(3):487–512.

1002. Schneider AL, Ervin L. Specific deterrence,rational choice, and decision heuristics:Applications in juvenile justice. Social ScienceQuarterly 1990; 71(3):584–601.

1003. Weatherburn D, Jones C. Does Prohibition DeterCannabis Use? Contemporary Issues in Crimeand Justice Bulletin No. 58. Sydney: NSWBureau of Crime Statistics and Research,August 2001.

1004. Ali R, Christie P, Lenton S, Hawks D, SuttonA, Hall W, et al. The Social Impacts of theCannabis Expiation Notice Scheme in South Australia– Summary Report. Monograph No. 34.Canberra: Publications Productions Unit,Commonwealth Department of Health andAged Care, National Drug Strategy, 1999.

1005. Stafford MC, Warr M. A reconceptualizationof general and specific deterrence. Journal ofResearch in Crime and Delinquency 1993;30(2):125–135.

1006. Caulkins JP. Law Enforcement’s Role in a HarmReduction Regime. Contemporary Issues inCrime and Justice Bulletin No. 64. Sydney:New South Wales Bureau of Crime Statisticsand Research, January 2002.

1007. Dunworth T, Mills G. National Evaluation ofWeed and Seed. National Institute of Justice,USA, 1999. Viewed 6 June 2003, <http://www.ojp.usdoj.gov/nij/pubs-sum/176358.htm>.

1008. Mazerolle LG, Price JF, Roehl J. Civilremedies and drug control: a randomizedfield trial in Oakland, California. EvaluationReview: A Journal of Applied Social Research 2000;24(2):212–241.

1009. Jacobson J. Policing Drug Hot-Spots. PoliceResearch Series Paper 109. London: Policingand Reducing Crime Unit, Home Office,1999.

1010. Weisburd D, Green L. Policing drug hot-spots: findings from the Jersey City DMAExperiment. Justice Quarterly 1995;12(4):711–735.

1011. Williams P, White P, Teece M, Kitto R.Problem-oriented Policing: Operation Mantle - A CaseStudy. Trends and Issues in Crime andCriminal Justice Series Report No. 190.Canberra: Australian Institue ofCriminology, February 2001.

�)(��

1012. Graycar A, Nelson D, Palmer M. LawEnforcement and Illicit Drug Control. Trends andIssues in Crime and Criminal Justice SeriesReport No. 110. Canberra: AustralianInstitute of Criminology, April 1999.

1013. Maher L, Dixon D, Swift W, Nguyen T. Anhhai: Young Asian Background People’s Perceptions andExperiences of Policing. Sydney: University ofNew South Wales Faculty of Law, 1997.

1014. Grapendaal M, Leuw E, Nelen H. A World ofOpportunities: Life-Style and Economic Behavior ofHeroin Addicts in Amsterdam. Albany, NJ: StateUniversity of New York Press, 1995.

1015. Dixon D, Coffin P. Zero tolerance policingof illegal drug markets. Drug and Alcohol Review1999; 18:477–486.

1016. Braga A. The effects of hot spots policing oncrime. In Heston AW, Weiner NA,Farrington D, Welsh BC (eds) The Annals ofThe American Academy of Political and Social Science.What Works in Preventing Crime? Systematic Reviewsof Experimental and Quasi-experimental Research.Thousand Oaks: Sage Publications,2001;104–125.

1017. Ratcliffe J. Burglary Reduction and the Myth ofDisplacement. Trends and Issues in Crime andCriminal Justice Series No. 232. Canberra:Australian Institue of Criminology, 2002.

1018. Green L. Policing Places with Drug Problems.Thousand Oaks: Sage Publications; 1996.

1019. Commonwealth Department of Health andAgeing. National illicit drug strategy“Tough on drugs”. Population HealthDivision, 2001. Viewed 25 June 2002,<http://www.health.gov.au/pubhlth/strateg/drugs/illicit/index.htm>.

1020. Commonwealth Department of Health andAged Care. National Action Plan on Illicit Drugs2001 to 2002-03. Canberra: Commonwealthof Australia, July 2001.

1021. Australian National Audit Office. DrugDetection in Air and Containerised Sea Cargo andSmallcraft. Canberra: ANAO, 2002.

1022. Australian Federal Police. National Illicit DrugStrategy Initiatives November 1997-April 2001.Second Edition. Canberra: Australian FederalPolice, 2001.

1023. Australian Customs Service. Australian CustomsService Annual Report 2000-01. Canberra:Commonwealth of Australia, 2001.

1024. McFadden M, Mwesigya S-E. Estimating theeconomic value of AFP fraud and druginvestigations. Platypus Magazine, 2001.Viewed 25 June 2002, <http://www.afp.gov.au/raw/Publications/Platypus/Dec01/ecvalue.pdf>.

1025. Bush WM. The Australian heroin drought:the case for an inquiry into its causes andthe flood of methamphetamines, 2002.Viewed 3 February 2002, <http://www.ffdlr.org.au/commentary/HeroinDrought.PDF>.

1026. Weatherburn D, Jones C, Freeman K,Makkai T. The Australian Heroin Drought and itsImplications for Drug Policy No. 59. Sydney:NSW Bureau of Crime Statistics andResearch, October 2001.

1027. Topp L, Degenhardt L, Day C. The Causes,Effects and Implications of the Heroin Shortage in theACT, NSW and Victoria. Sydney: National Drugand Alcohol Research Centre, December2001.

1028. Commonwealth Department of Health andAgeing. National Drug Law Enforcement ResearchFund Grants List. April 2002.

1029. Brooks A, Stathard C, Moss J, Christie P, AliR. Costs Associated with the Operation of the CannabisExpiation Notice Scheme in South Australia.Adelaide: Drug and Alcohol ServicesCouncil, 1999.

1030. Makkai T, McAllister I. Marijuana in Australia:Patterns and attitudes. Canberra: AustralianGovernment Publishing Service, 1997.

1031. Victoria Police Strategic DevelopmentDepartment. Evaluation of the Cannabis CautioningProgram Pilot. 1998.

1032. Ditchburn P. Cannabis Cautioning Programand the Drug Diversion Pilot - VictoriaPolice. Paper presented at AustralasianConference on Drug Strategy. Adelaide, 27– 29 April 1999.

1033. McAllister I, Makkai T. Antisocial behaviouramong young Australians while under theinfluence of illicit drugs. Australian and NewZealand Journal of Criminology 2003; 36(2):211–222.

1034. Australian Police Ministers Council. NationalSupply Reduction Strategy for Illicit Drugs. June2002.

1035. United Nations International DrugProgramme. World Drug Report. Oxford:Oxford University Press; 1997.

��) �� !��"�#��$��%��

1036. Government of Western Australia. Office ofthe Director of Public Prosecutions Western Australia.Annual Report 2000/2001. Perth: Governmentof Western Australia, 2001.

1037. Australian Law Reform Commission.Confiscation that counts: a review of theProceeds of Crime Act 1987. AustralasianLegal Information Institute, 1999. Viewed25 June 2002, <http://www.austlii.edu.au/au/other/alrc/publications/reports/87/>.

1038. Moore MH. Policies to achievediscrimination on the effective price ofheroin. The American Economic Review 1973;63:270–277.

1039. Moore MH. Supply reduction and drug lawenforcement. In Tonry M, Wilson JQ (eds)Drugs and Crime. Crime and Justice: A Review of theResearch. Chicago: University of ChicagoPress, 1990;109–157.

1040. Reuter P. Can the borders be sealed? InWeisheit R (ed) Drugs, Crime and the CriminalJustice System. Cincinnati, OH: AndersonPublishing Co. and Academy of CriminalJustice Sciences, 1990;13–26.

1041. Reuter P. The Limits of Supply-Side Drug Control.Santa Monica: RAND Corporation, 2001.

1042. Green P, Purnell I. Measuring the Success of LawEnforcement Agencies in Australia in Targeting MajorOffenders Relative to Minor Offenders. Adelaide:National Police Research Unit, 1995.

1043. Weatherburn D, Lind B. Drug Law EnforcementPolicy and its Impact on the Heroin Market. Sydney:New South Wales Bureau of Crime Statisticsand Research, 1995.

1044. Sutton A, James S. Evaluation of Australian Drugand Anti-trafficking Law Enforcement: NationalPolice Research Unit, 1996.

1045. Weatherburn D. Performance Indicators for DrugLaw Enforcement. Crime and Justice BulletinNo. 48. Sydney: New South Wales Bureauof Crime Statistics and Research, February2000.

1046. New South Wales Police. NSW Illicit Drug LawEnforcement Performance Indicators. Sydney:Organisational Policy and Development,NSW Police, April 2002.

1047. Makkai T. Linking Drugs and Criminal Activity:Developing an Integrated Monitoring System. Trendsand Issues in Crime and Criminal Justice

Series Report No. 109. Canberra: AustralianInstitute of Criminology, April 1999.

1048. Makkai T, McAllister I. Measuring socialindicators in opinion surveys: a method toimprove accuracy on sensitive questions.Social Indicators Research 1992; 27:169–186.

1049. Makkai T. Statistical systems: Crime. InSmelser NJ, Baltes PB (eds) InternationalEncyclopedia of the Social and Behavioral Sciences:Elsevier Science Ltd, 2001;15053–15056.

1050. Spooner C, Hall W, Mattick R. A StrategicOverview of the Diversion of Drug-Related Offenders inNSW. Technical Report No. 96. Sydney:National Drug and Alcohol Research Centre,June 2000.

1051. Thurman QC, Giacomazzi A, Bogen P.Research note: cops, kids, and communitypolicing - an assessment of a communitypolicing demonstration project. Crime andDelinquency 1993; 39(4):554–564.

1052. Morehouse E, Tobler N. Preventing andreducing substance use amonginstitutionalised adolescents. Adolescence2000; 35(137):1–28.

1053. Dembo R, Shemwell M, Guida J, SchmeidlerJ, Pacheco K, Seeberger W. A longitudinalstudy of the impact of a familyempowerment intervention on juvenileoffender psychosocial functioning: a firstassessment. Journal of Child and AdolescentSubstance Abuse 1998; 8(1):15–54.

1054. Spooner C, Hall W, Mattick R. An overviewof diversion strategies for Australia drug-related offenders. Drug and Alcohol Review2001; 20:281–294.

1055. Ministerial Council on Drug Strategy. COAGIllicit Drug Diversion Initiative: Report to Heads ofGovernment2001.

1056. Howells K, Day A. The Rehabilitation of Offenders:International Perspectives Applied to AustralianCorrectional Systems. Trends and Issues inCrime and Criminal Justice No. 112.Canberra: Australian Institute ofCriminology, 1999.

1057. Hall W. The role of legal coercion in thetreatment of offenders with alcohol andheroin problems. Australian and New ZealandJournal of Criminology 1997; 30:103–120.

1058. Porter L, Arif A, Curran W. The Law and theTreatment of Drug and Alcohol Dependent Persons - A

��

Comparative Study of Existing Legislation. Geneva:World Health Organization, 1986.

1059. Health Outcomes International Pty Ltd forDepartment of Finance and Administration.Monitoring and Evaluation Framework for the COAGIllicit Drugs Initiatives. St Peters, SAMarch 2000.

1060. Makkai T. The emergence of drug treatmentcourts in Australia. Substance Use and Misuse2002; 37(12–13):1567–1594.

1061. Freiberg A. Australian drug court: aprogress report. In: 2nd AustralasianConference on Drugs Strategy. Perth:Western Australian Police Service; 2002.

1062. Makkai T. Drugs Courts: Issues and Prospects.Report No. 95. Canberra: AustralianInstitute of Criminology, September 1998.

1063. Belenko S. Research on Drug Courts: A CriticalReview. New York: National Center onAddiction and Substance Abuse at ColumbiaUniversity, 1998.

1064. Belenko S. Research on Drug Courts: A CriticalReview, 2001 Update. New York: NationalCenter on Addiction and Substance Abuse atColumbia University, 2001.

1065. Flick M, Eardley T. Evaluation of the NSW YouthDrug Court Pilot Program. First ImplementationReview. SPRC Report No. 5/01. Sydney:Social Policy Research Centre, June 2001.

1066. Dive R, Killen M, Cole D, Poder A. NSWyouth drug court trial. Paper presented atThe 2nd Australasian Conference on DrugsStrategy. Perth: Western Australian PoliceService, 7–9 May 2002.

1067. Freeman K. New South Wales Drug CourtEvaluation: Health, Well-Being and ParticipantSatisfaction. Sydney: New South Wales Bureauof Crime Statistics and Research, 2002.

1068. Lind B, Weatherburn D, Chen S, ShanahanM, Lancsar E, Haas M, et al. New South WalesDrug Court Evaluation: Cost-Effectiveness. Sydney:New South Wales Bureau of Crime Statisticsand Research and Centre for HealthEconomics Research and Evaluation, 2002.

1069. Freeman K. New South Wales Drug CourtEvaluation: Interim Report on Health and Well-Beingof Participants. Contemporary Issues in Crimeand Justice Bulletin No. 53. Sydney: NewSouth Wales Bureau of Crime Statistics andResearch, February 2001.

1070. Taplin S. The New South Wales Drug CourtEvaluation: A Process Evaluation. Sydney: NewSouth Wales Bureau of Crime Statistics andResearch, 2002.

1071. Department of Corrective Services. AnnualReport 2000/2001. Brisbane: QueenslandGovernment, 2001.

1072. Department for Correctional Services. AnnualReport 2000/2001. Adelaide: Government ofSouth Australia, 2001.

1073. Office of the Correctional ServicesCommissioner. Victorian Prison Drug Strategy2002. Melbourne: State of Victoria, 2002.

1074. Department for Correctional Services. Drugand Alcohol Strategy Project. FinalReportFebruary 1998.

1075. Ryder D, Kraszlan K, Walker N, Fowler H,Omaji P. Drug Court Pilot Feasibility Study. InterimReport. Perth: Edith Cowan University,November 1998.

1076. Rutter S, Dolan K, Wodak A, Heilpern H.Prison-based Syringe Exchange. A Review of theInternational Research and Program Development.Technical Report No. 112. Sydney: NationalDrug and Alcohol Research Centre, 2001.

1077. Ministry of Justice Western Australia.Discussion paper for the Ministry of Justice.The provision of bleach availabilityprograms for prisoners in WesternAustralia. In. Perth; 2001. p. 30.

1078. Dolan K. Surveillance and Prevention of Hepatitis CInfection in Australian Prisons: A Discussion Paper.Technical Report No. 95. Sydney: NationalDrug and Alcohol Research Centre, May2000.

1079. Inciardi JA, Martin SS, Butzin CA, HooperRM, Harrison LD. An effective model ofprison-based treatment for drug-involvedoffenders. Journal of Drug Issues 1997;27(2):261–278.

1080. McGeorge J, Crofts N, Burrows C.Disaffected or disinfected? The nationaldisinfection of used injecting equipmentproject. Paper presented at ‘Out There Too’Social Research and Practice Forum.Adelaide, 14–15 March 1996.

1081. Gore SM, Bird AG, Cassidy J. Prisoners’views about the drugs problem in prisons,and the new Prison Service drug strategy.Communicable Disease and Public Health 1999;2(3):196–197.

�� !��"�#��$��%��

1082. Department of Corrective Services. AddressingIllicit Drug Use By Offenders: Action Plan. Brisbane:Queensland Government, 2001.

1083. Close H. Pilot of Sweat Patch Usage Within TwoWA Community Corrections Centres. Final EvaluationReport. Perth: Department of Justice WesternAustralia, September 2001.

1084. Department of Health Housing LocalGovernment and Community Services.National Drug Strategic Plan 1993-97. Canberra:Australian Government Printing Service,1993.

1085. Hawks D, Lenton S. Harm reduction inAustralia: has it worked? A review. Drug andAlcohol Review 1995; 14(3):291–304.

1086. Single E, Rohl T. The National Drug Strategy:Mapping the Future. Report Commissioned bythe Ministerial Council on Drug Strategy.Canberra: Australian Government PrintingService, 1997.

1087. Lenton S, Single E. The definition of harmreduction. Drug and Alcohol Review 1998;17(2):213–220.

1088. Borland R, Scollo M. Opportunities forharm minimization in tobacco control. Drugand Alcohol Review 1999; 18:345–353.

1089. Wells-Parker E, Bangert-Drowns R,McMillen R, Williams M. Final results froma meta-analysis of remedial interventionswith drink/drive offenders. Addiction 1995;90:907–926.

1090. Voas RB, Fisher DA. Court procedures forhandling intoxicated drivers. Alcohol Researchand Health 2001; 25:32–42.

1091. Rydon P, Stockwell T. Local regulation andenforcement strategies for licensedpremises. In Single E, Stockwell T (eds)Alcohol: Minimising the Harm. What Works? NewYork: Free Association Books Ltd,1997;211–229.

1092. Voas RB, Marques PR, Tippetts AS, BeirnessDJ. The Alberta Interlock Program: theevaluation of a province-wide program onDUI recidivism. Addiction 1999; 94:1849–1860.

1093. Marques PR, Voas RB, Tippetts AS, BeirnessDJ. Behavioral monitoring of DUI offenderswith the Alcohol Ignition InterlockRecorder. Addiction 1999; 94:1861.

1094. Boots K, Midford R. Promoting Safe Parties:Evaluation of the 1993 Geraldton Partysafe Campaign.Perth: National Centre for Research into thePrevention of Drug Abuse, Division ofHealth Sciences, Curtin University ofTechnology, 1995.

1095. Stevenson M, Palamara P, Rooke M,Richardson K, Baker M, Baumwol J. Drinkand driving: what’s the skipper up to?Australian and New Zealand Journal of Public Health2001; 25(6):511–513.

1096. Stockwell T. Responsible alcohol service:lessons from evaluations of server trainingand policing initiatives. Drug and Alcohol Review2001; 20:257–265.

1097. Drew LR, Truswell AS. Wernicke’sencephalophathy and thiamine fortificationof food: time for a new direction? MedicalJournal of Australia 1998; 168(11):534–535.

1098. Wodak A, Richmond R, Wilson A. Thiaminfortification and alcohol. Medical Journal ofAustralia 1990; 152(2):97–99.

1099. Connelly L, Price J. Preventing theWernicke-Korsakoff syndrome in Australia:cost-effectiveness of thiamin-supplementation alternatives. Australian andNew Zealand Journal of Public Health 1996;20(2):181–187.

1100. Joing Ma J, Truswell AS. Wernicke-Korsakoff syndrome in Sydney hospitals:before and after thiamine enrichment offlour. Medical Journal of Australia 1995;163:531–534.

1101. Hawks D, Rydon P, Stockwell TR, White M,Chikritzhs T, Heale P. The Evaluation of theFremantle Police Licensee Accord: Impact on ServicingPractices, Harm and the Wider Community. Perth:National Drug Research Institute, CurtinUniversity of Technology, 1999.

1102. Rumbold G, Malpass A, Lang E, Cvekovski S,Kelly W. An Evaluation of the Geelong Local IndustryAccord: Final Report. Melbourne: TurningPoint, 1998.

1103. Stockwell T. Regulation of the licenseddrinking environment: a major opportunityfor crime prevention. Crime Prevention Studies1997; 7:7–34.

1104. Mosey A. Central Australian Night Patrols: AReview. Alice Springs: Drug and AlcoholServices Association, 1994.

��

1105. Daly A, Gvozdenovic A. Evaluation of theImpacts of Public Drunkenness DecriminalisationLegislation and the Establishment of Sobering UpFacilities on Aboriginal People in Northern WesternAustralia, 2 vols. Perth: Western AustralianAlcohol and Drug Authority, 1994.

1106. McGregor C, Hall K, Ali R, Christie P,Braithwaite R, Darke S. It’s Rarely Just the ‘h’:Addressing Heroin Overdose Among South AustralianHeroin Users Through a Process of IntersectoralCollaboration. Parkside: Drug and AlcoholServices Council, 1999.

1107. Ministerial Council on Drug Strategy.National Heroin Overdose Strategy. Canberra:Commonwealth of Australia, 2001.

1108. Ministerial Council on Drug Strategy.National Heroin Overdose Strategy. Summary.Canberra: Commonwealth Department ofHealth and Aged Care, July 2001.

1109. Hall W. Reducing the toll of opioidoverdose deaths in Australia. Drug and AlcoholReview 1995; 18:213–220.

1110. Lenton S, Hargreaves K. A trial of naloxonefor peer administration has merit, but willthe lawyers let it happen? Drug and AlcoholReview 2000; 19:365–369.

1111. Lenton S, Hargreaves K. Should we conducta trial of distributing naloxone to heroinusers for peer administration to prevent fataloverdose? Medical Journal of Australia 2000;173:260–263.

1112. Loxley W. Doing the possible: Harmreduction, injecting drug use and bloodborne viral infections in Australia.International Journal of Drug Policy 2000;11:407–416.

1113. Crofts N, Dore G, Locarnini S. Hepatitis C. AnAustralian Perspective. Melbourne: IPCommunications; 2001.

1114. Lurie P, Reingold AL, Bowser B, Chen D,Foley J, Guydish J, et al. The Public HealthImpact of Needle Exchange Programs in the UnitedStates and Abroad. San Francisco: University ofCalifornia; 1993.

1115. Paone D, Des Jarlais DC, Gangloff R,Milliken J, Friedman SR. Syringe exchange:HIV prevention, key findings and futuredirections. International Journal of the Addictions1995; 30(12):1603–1645.

1116. HIVDent. Evidence-Based Findings on theEfficacy of Syringe Exchange Programs: AnAnalysis from the Assistant Secretary forHealth and Surgeon General of the ScientificResearch Completed Since April 1998. HIVDent, 2001. Viewed 26 September 2001,<http:\\www.hivdent.org\publicp\ppebsotsr062000.htm>.

1117. Center for AIDS Prevention Studies. DoesNeedle Exchange Work? University ofCalifornia, San Francisco, 1998. Viewed 25October 2001, <http://www.ama-assn.org/special/hiv/preventn/prevent3.htm>.

1118. Commonwealth Department of Health andAgeing. Return on Investment in Needle and SyringePrograms in Australia. Summary Report. Canberra:Commonwealth Department of Health andAgeing, 2002.

1119. Coyle SL, Needle RH, Normand J. Outreach-based HIV prevention for injecting drugusers: A review of published outcome data.Public Health Reports 1998; 113(Supplement1):19–30.

1120. Darke S. Injecting drug users and theHuman Immunodeficiency Virus: What dowe know? Drug and Alcohol Review 1992;11:153–161.

1121. Ward J, Mattick RP, Hall W. MethadoneMaintenance Treatment and Other Opioid ReplacementTherapies. Amsterdam: Harwood AcademicPublishers, 1998.

1122. Kaldor JM, Lapsley HM, Mattick R,Weatherburn D. Six Month Process Report on theMedically Supervised Injecting Centre. Sydney:National Drug and Alcohol Research Centre,2002.

1123. Kaldor J, Lapsley H, Mattick RP,Weatherburn D, Kimber J, MacDonald M.Twelve-month Process Evaluation Report on theMedically Supervised Injecting Centre. Sydney:National Drug and Alcohol Research Centre,May 2002.

1124. Dolan K, Kimber J, Fry C, Fitzgerald J,McDonald D, Trautmann F. Drugconsumption facilities in Europe and theestablishment of supervised injectingcentres in Australia. Drug and Alcohol Review2000; 19:337–346.

�� !��"�#��$��%��

1125. Kimber J, Dolan K, Wodak A. InternationalSurvey of Supervised Injecting Centres (1999-2000).Technical Report No. 126. Sydney: NationalDrug and Alcohol Research Centre, 2001.

1126. Lindesmith Center. Safer Injecting Rooms. NewYork: Lindesmith Center, 1999.

1127. Dillon P. Key findings from the 12-monthreport of the Medically Supervised InjectingCentre, 2002. Viewed 24 May 2002,<http://ndarc.med.unsw.edu.au/ndarc.nsf/website/News.pressreleases.MSIC.2002>.

1128. Loxley W. Hepatitis C and B in people whoinject drugs: testing and vaccination - majoropportunities for prevention. Paperpresented at NDRI Symposium “InjectingDrugs: The Prevention of Blood BorneViruses and Overdose”. West Perth, 4August 2000.

1129. Communicable Diseases and HealthProtection Branch, Population HealthDivision. Retractable Needle and Syringe Technology.Media Release. Canberra: Department ofHealth and Aged Care, 2002.

1130. Des Jarlais DC. “Single-use” needles andsyringes for the prevention of HIV infectionamong injection drug users. Journal of AcquiredImmune Deficiency Syndromes and HumanRetrovirology 1998; 18(Suppl 1):S52–S56.

1131. Caulkins JP, Kaplan E, Lurie P, O’Connor T,Ahn S-H. Can difficult-to-reuse syringesreduce the spread of HIV among IDUs?Interfaces 1998; 28:23–33.

1132. Lenton S, Boys A, Norcross K. Raves, drugsand experience: Drug use by a sample ofpeople who attend raves in WesternAustralia. Addiction 1997; 92(10):1327–1337.

1133. New South Wales Government. Code ofPractice for Dance Parties. Sydney: New SouthWales Government, 1998.

1134. Winstock A, Vingoe L. Pill-testing: Harmreduction or just a bitter pill to swallow?National Addiction Centre - London, 2000.Viewed 11 August 2001, <http://www.clubhealth.org.uk/pages/2000/s5_p2.htm>.

1135. European Monitoring Centre for Drugs andDrug Addiction. On-Site Pill-Testing Interventionsin the European Union. Vienna: EuropeanMonitoring Centre for Drugs and DrugAddiction, 2001.

1136. Fry C, Bruno R. Trends in Benzodiazepine UseAmong Injecting Drug Users in Victoria and Tasmania:Findings from the Illicit Drug Reporting System. DrugTrends Bulletin. Canberra: CommonwealthDepartment of Health and Aged Care, June2001.

1137. Burns C, Currie B, Powers JR. Unleadedpetrol: a useful harm-reduction strategy forpetrol sniffers? In Brady M, Torzillo P (eds)Workshop on Lead And Hydrocarbon Toxicity FromChronic Petrol Inhalation: 5-6 May 1994, AliceSprings. Canberra: Australian GovernmentPublishing Service, 1995.

1138. Newman I, Anderson C, Farrell K. Rolerehearsal and efficacy: two 15 monthevaluations of a ninth-grade alcoholeducation program. Journal of Drug Education1992; 22(1):55–67.

1139. Gray D, Saggers S, Sputore B. Interventionsfor Indigenous Australians: the broadercontext. In Best Practice Interventions in Correctionsfor Indigenous People: Papers from a Conference inSydney 8-9 October 2001. Sydney: AustralianInstitute of Criminology in conjunctionwith the Department of Corrective ServicesNSW, 2002.

1140. National Aboriginal Health StrategyWorking Party. A National Aboriginal HealthStrategy. Canberra: Australian GovernmentPublishing Service, 1989.

1141. Lindorff K. Tobacco: Time for Action. Final Report.National Aboriginal and Torres StraitIslander Tobacco Control Project. Canberra:National Aboriginal Community ControlledHealth Organisation, 2002.

1142. Kunitz S. Disease and Social Diversity: The EuropeanImpact on the Health of Non-Europeans. New York:Oxford University Press; 1994.

1143. Gardiner-Garden J. Identifiable CommonwealthExpenditure on Aboriginal and Torres Strait IslanderAffairs. Current Issues Brief No. 18 1997–98.Canberra: Parliament of Australia,Department of the Parliamentary Library,2001.

��"��

1144. Deeble J, Mathers C, Smith L, Goss J, WebbR, Smith V. Expenditures on Health Services forAboriginal and Torres Strait Islander People.Canberra: Department of Health and FamilyServices, 1998.

1145. Mooney G. Australian health care: fairenough? or only an un-fair go? Paperpresented at Health Division Forum. Perth:Curtin University of Technology, 2002.

1146. Commonwealth Department of Health andAged Care. Review of the Commonwealth’sAboriginal and Torres Strait Islander Substance MisuseProgram. Final Report. Canberra:Commonwealth of Australia, 1999.

1147. Holder HD, Saltz RF, Grube JW. Acommunity prevention trial to reducealcohol-involved accidental injury anddeath: overview. Addiction 1997; 92(Suppl.2):S155–S171.

1148. Guy S, M., Smith G, M., Bentler P, M.Consequences of adolescent drug use andpersonality factors on adult drug use. Journalof Drug Education 1994; 24(2):109–132.

1149. Baumrind, D. The influence of parentingstyle on adolescent competence andsubstance use, Journal of Early Adolescence 1991:11; 56–95.

· isfp iowa strengthening family program jag joint advisory group kav knowledge, attitude and...

Documents