modeling environmental burden of disease of asthma: p rotective factors and control options

NATIONAL INSTITUTE FOR HEALTH AND WELFARE

Modeling environmental burden of disease of asthma:Protective factors and control options as part of the TEKAISU project

Isabell RumrichNational Institute for Health and Welfare (THL)Kuopio, Finland

Master Thesis in the ToxEn programUniversity of Eastern Finland, Department of Environmental Science


Outline

• Introduction

• Background data

• Associated Factors

• Control Policies

• Discussion

2


Asthma Chronic inflammatory disease Prevalence as high as 9.4 % (2007) Currently only symptomatic treatment Pathology is characterized by miss-regulation of immune

responses

• Various factors have been proposed to be associated with onset or symptoms:

anthropogenic and natural environmental factors, lifestyle related stressors, pharmaceutical

stressors, internal factors, genetic susceptibility and co-morbidities

3


IHME estimates of BoD (YLDs) in 2010

http://viz.healthmetricsandevaluation.org/gbd-compare/

Asthma:

• 2% of total YLDs in 1990 and

2010

• Maximum for 5-9y old (2010)

13% of total YLDs

biggest contribution to

total YLD

4


Genes

Environmental Factors

Lifestyle

Co-morbidities

Exposure

Risk Ratio

Exposure can be changed by relatively easy measures- Already existing policies- Development of hypothetical policies Modelling of effect of exposure change

Can not be changed

Can be changed

Other factors

Reducible Fraction

From the Model to Control Policies

5

Asthma BoD

Attributable

Attributable


Selection of exposure factors

6

Literature Search

Review TableModel PoliciesLack of evidence;

Duplication of factors

Lack of data;Lack of significance

Impact on asthma burden

6

15 factors 6 factors35 factors235 articles

Databases: PubMed, Scopus, Web of Science – WoS (ISI), SpringerLink and Science Direct (Elsevier).

Search queries: asthma; asthma AND environment*; asthma AND risk; asthma AND environment* NOT atopy; asthma AND risk NOT atopy; asthma AND mechanism; asthma AND risk NOT occupation*; asthma AND environment* NOT occupation*


Outline• Introduction

• Background data



• Discussion

7

Reducible Fraction

Asthma BoD

Attributable

Attributable


8

Life Table (1986-2040) and Age Distribution

0

1,000,000

2,000,000

3,000,000

4,000,000

5,000,000

6,000,000Elderly 81-99yPensioner 66-80yWorking Age 26-65yYoung Adult 20-25yTeen 13-19yChild 7-12yPreschool Child 4-6yToddler 1-3yInfant 0yTotalObservedProjection

Year

Popu

latio

n


Incidence & Prevalence

• Incidence: number of new cases in a specific period of time

number of new individuals entitled to reimburse expenses for asthma medication during one year

• Prevalence: number of all cases at a specific time point

total number of individuals entitled to reimburse expenses for asthma medication at the end of a year

• Data provided by KELA statistics

9


Incidence and Prevalence – Total number of cases

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

2018

2019

2020

2021

2022

2023

2024

2025

2026

2027

2028

2029

2030

2031

2032

2033

2034

2035

2036

2037

2038

2039

2040

0

5,000

10,000

15,000

20,000

25,000

30,000

0

50,000

100,000

150,000

200,000

250,000

300,000

Start Estimation Incidence Prevalence

Year

Inci

denc

e (c

ases

)

Prev

alen

ce (c

ases

)

10


Background Rates at Baseline (2011)

11

Infant Toddler Preschool Child

Child Teen Young Adult

Working Age

Pensioner Elderly Total

Inci-dence rate ('000)

0.0499384092952026

6.62483137537616

5.7018279389569

3.68971903579171

2.34084494422293

1.55262728152047

2.30912061047853

3.7317215715274

2.03910383777603

2.74376768668239

1234567

Inci

denc

e ra

te

(per

100

0)



Working Age


Preva-lence rate ('000)

0.0499384092952026

11.6330509724248

22.8464419475655

27.582444402969

22.2055463464899

18.4908362329946

43.3166347842574

88.8904129195391

91.4031556110138

44.2017308067266

10

30

50

70

90

Prev

alen

ce r

ate

(p

er 1

000)


Burden of Disease - YLDYears Lived with Disability (YLD)

a) Incidence based:

b) Prevalence based:

YLDI = YLDP

P x DW = I x D x DW

D = I = Incidence; DW = Disability Weight; D = Duration; P = Prevalence

12


Years Lived with Disability – Total number of years

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

2018

2019

2020

2021

2022

2023

2024

2025

2026

2027

2028

2029

2030

2031

2032

2033

2034

2035

2036

2037

2038

2039

2040

0

2,000

4,000

6,000

8,000

10,000

12,000

Estimation YLD_I YLD_P

Year

Year

s Li

ved

with

Dis

abili

ty (Y

LD) (

Year

s)

13


Duration estimation

14

WHO Infant Toddler Preschool Child


Working Age


1986 15 1 1.38805970149254

2.48523206751055

3.9258883248731

6.31931464174455

7.688 7.34434753438443

5.87721198988805

9.06435643564356

6.3712833545108

3

8

13

18

Dur

ation

(Yea

rs)

WHO Infant Toddler Preschool Child


Working Age


2040 15 0.195616342359335

0.806177912580271

3.84778215154842

7.76801055459771

10.7554421016062

13.4208156826173

21.792208371086

27.7883762787015

56.1335667911903

19.1197291353824

5

15

25

35

45

55

Dur

ation

(Ye

ars)


Outline

• Introduction

• Background data

• Associated Factors– Risk Factors– Protective Factors


• Discussion

15

Reducible Fraction

Asthma BoD

Attributable

Attributable


Overview Risk Factors

Factor Exposed Population [%]

RR/OR Target Age [Years]

Dampness and Mold 15 1,34 0-99NO2 100 1,077 0-99Underweight 3 3,14 6PM2.5 100 1,16 0-99SHS (child) 4 1,32 0-13SHS (adult) 14 1,97 21-99Cat Allergy 7 1,67 7-8Dampness and Mold 15 1,37 0-99Dog Allergy 7 2,78 21-99Formaldehyde 2 1,02 0-2

16


Attributable incident cases and residual at baseline (2011)

17

Overview

Factor

0 2,000 4,000 6,000 8,000 10,000 12,000 14,000 16,000

Incidence (cases)

Residual; 53 %

Residual; 53 %

Attributable; 47 %

PM2.5; 12%

Allergen;11%

NO2; 10%

SHS;7%

Dampness& Mould;5%

Underweight;1%

Dog; 1%

Cat; 0% Formaldehyde; 0%

Smoking; 0%


18

Overview Protective Factors

Factor Exposed Population [%]

RR/OR Target Age [Years]

Cat 20 0,47 7-16Dog 24 0,57 7-16Breastfeeding 35 0,48 4-6Eurotium 4 0,57 6-12Penicillium 4 0,57 6-12


Prevented cases at baseline (2011) and background

19

Overview

Factor

0 2,000 4,000 6,000 8,000 10,000 12,000 14,000 16,000 18,000

Incidence (cases)

Prevented; 9%

BackgroundDog; 3%

Breastfeeding; 3% Cat; 2%

Eurotium; 1% Penicillium; 0%

Background


Outline

• Introduction

• Background data


• Control Policies– Tobacco Smoke– PM2.5

– Dampness and Mould– Pets

• Discussion

20

Reducible Fraction

Asthma BoD

Attributable

Attributable


21

Summary Risk & Protective Factors

-500

0

500

1,000

1,500

2,000Residual

PM2.5;1720

Allergen;1573

NO2;1560

SHS;985

D&M;719

Under-weight;

159 Dog;82

Smoking;62

Cat;26

Formalde-hyde;

0

Penicil-lium;-50

Euro-tium;-73

Cat;-344 Dog,

-491Breast-

feeding;-495

Attrib

utab

le In

cide

nce

base

line

(201

1) (c

ases

)

Risk Factors

8 000

Protective Factors

Residual;7922


Control Policies

Policy Factor Reference

Tobacco SHS Kutvonen (2014); Savuton Suomi 2040

Smoking

PM2.5 PM2.5 Kutvonen (2014)

Dampness and Mould

Dampness and Mould HealthVent study

Pets Cat

Dog

22


Control Policies – TobaccoPolicy Exposure in

2013Change in Exposure

Explanation

BanSHS:4% Children9% Adults

Smoking:15% (15-24y)19% (25-44y)29% (45-64y)8% (65-84y)

Total ban 100% reduction

From 2015 onwards no exposure at all

50% Reduction

50% Reduction In 2015 50% reduction and then constant exposure

10% Reduction

10% Reduction From Exposure 2014 annually 10% reduction

23


24

Tobacco Exposure trends

19861988

19901992

19941996

19982000

20022004

20062008

20102012

20142016

20182020

20222024

20262028

20302032

20342036

20382040

0%

5%

10%

15%

20%

25%Smoking BaU Smoking 50% Reduction

Smoking 10% Reduction SHS BaU

SHS 50% Reduction SHS 10% Reduction

Year

Frac

tion

of P

opul

ation

bei

ng e

xpos

ed to

Tob

acco

Sm

oke


25

Impact of Tobacco Control Policy

BaU Ban 50% Reduction annual 10% Reduction0

5,000

10,000

15,000

20,000

25y

cum

ulati

ve In

cide

nce

(cas

es)

SHS SHS SHS

Smoking

Smoking

Smoking


Control Policies – PM2.5Policy Exposure in 2013 Change in

ExposureExplanation

Ban of Small Scale Wood Combustion (SSWC) in Urban Areas

Total: 8mg/m3 and 0,6 mg/m3 due to SSWC

Total ban 100% Reduction

Annually fraction due to SSWC is deleted from total exposure

Reduction of Small Scale Wood Combustion (SSWC) in Urban Areas

Total: 8mg/m3 and 0,6 mg/m3 due to SSWC

50% Reduction

Annually 50% of fraction due to SSWC is deleted from total exposure

Speed Limit of 35km/h in Urban Areas

Total: 8mg/m3 and 0,7 mg/m3 due to resuspension

40% Reduction

Annually 40% of fraction due to resuspension is deleted

26


27

Impact of PM2.5 Control Policy

BaU Ban 50% Reduction Speed Limit0

10,000

20,000

30,000

25y

cum

ulati

ve a

ttrib

utab

le In

cide

nce

(cas

es)

Small Scale Wood Combustion


Control Policies – Dampness and Mould

Policy Exposure in 2013

Change of Exposure Explanation

D&M 15% of total population

50% Reduction In 2015 50% reduction to 7,5% and then constant exposure

28


29

Impact of Dampness and Mold Control Policy

BaU 50% Reduction0

5,000

10,000

15,000

20,000

25y

cum

ulati

ve a

ttrib

utab

le In

cide

nce

(cas

es)


Control Policies – PetsPolicy Exposure in 2013 Change in

ExposureExplanation

Cat Risk20% of total Population

7% atopic 1,5%

50% increase

Increase in 2015, after that constant at 3,5%

Cat Protection

93% non-atopic 18,5%

Increase in 2015, after that constant at 46,5 %

Dog Risk24% of total Population

7% atopic 1,8%


Dog Protection

93% non-atopic 22,2%


30


31

Impact of Pet Control Policy

BaU 50% Increase

-10,000

0

10,000

20,000

30,000

40,000

50,000

25y

cum

ulati

ve a

ttrib

utab

le In

cide

nce

(cas

es)

Dog Protection

Cat Protection

Dog Risk

Cat Risk


32

Reduction Potential of Control Policies

Ban 50% Reduction

annual 10%

Reduction

Ban 50% Reduction

Speed Limit

50% Reduction

50% In-

crease Pets

-5,000

0

5,000

10,000

15,000

20,000

SHS16 549

Smoking1 555

SHS4 544

Smoking581

SHS8 774

Smoking847

2 496 1 246 1 345

8 733Cat

10 188

Cat-812

Dog9 382

Dog-1 759

25y

cum

ulati

ve In

cide

nce

(red

uced

cas

es)

Tobacco Wood CombustionDampness


Impact of combined Control Policies

33


34

Reducible Fraction of the total 25y cumulative Incidence



Working Age

Pensioner Elderly

-10%

0%

10%

20%

30%

40% Dog_PDog_RCat_PCat_RD&MPMSmokingSHSNet

Age group

Redu

cibl

e Fr

actio

n of

to

tal 2

5y c

umul

ative

In-

cide

nce



Working Age

Pensioner Elderly

-10.0 %

0.0 %

10.0 %

20.0 %

30.0 %

40.0 %Dog_PDog_RCat_PCat_RD&MPMSmokingSHSNet

Age group

Redu

cibl

e Fr

actio

n of

tota

l 25

y cu

mul

ative

inci

denc

e

More realistic

Most efficient


35

Efficiency Control Scenarios - Incidence

20152017

20192021

20232025

20272029

20312033

20352037

203910,000

11,000

12,000

13,000

14,000

15,000BaU Most Efficient More Realistic

Year

Inci

denc

e (c

ases

)


36

Efficiency Control Scenarios - Prevalence

20152017

20192021

20232025

20272029

20312033

20352037

2039150,000

170,000

190,000

210,000

230,000

250,000

270,000 BaU Most Efficient More Realistic

Year

Prev

alen

ce (c

ases

)


37

Efficiency Control Scenarios – combined Incidence & Prevalence

20152017

20192021

20232025

20272029

20312033

20352037

2039150,000

175,000

200,000

225,000

250,000

275,000BaU Most efficient More realistic

Year

Prev

alen

ce (c

ases

)


Outline

• Introduction

• Background data



• Discussion

38


Limitations• Population Life Table

– Neglecting of (Im-)migration

• Use of YLD instead of DALY– Each year a very low number of death due to asthma

neglected

• Discounting– Discounting decreases estimates for future years compared to

non-discounted estimates

39


Uncertainties

• Trend estimations– Uncertainty about the future trends in asthma and exposures

• Evidence– Overall very weak (association with atopy)– PM source has impact on toxicological profile

• Duration– Duration has impact on incidence based YLD estimate longer

duration increases YLD estimate

40


Conclusion

• Accumulation of prevalent cases in older age groups

• Asthma duration is increasing and age dependent

• About half of the total BoD can be theoretically explained

• BoD can be reduced (up to 20%) by reducing exposure to risk factors

41


42

Thank you for your attention!


Life Table (1986 – 2040)

43

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

2018

2019

2020

2021

2022

2023

2024

2025

2026

2027

2028

2029

2030

2031

2032

2033

2034

2035

2036

2037

2038

2039

2040

4,800,000.0

5,000,000.0

5,200,000.0

5,400,000.0

5,600,000.0

5,800,000.0

6,000,000.0

Observed Life Table Pop Projection

Year

Popu

lati

on (i

n M

io)


Age groups

44

Age Group Start End Absolute 1986

% 1986 Absolute 2011

% 2011 Absolute 2040

% 2040

Infant 0 0 67 221 1,3 60 074 1,1 72 314 1,3

Toddler 1 3 211 339 4,2 183 099 3,4 214 476 3,8

Preschool Child

4 6 209 640 4,1 178 890 3,3 210 889 3,7

Child 7 12 441 028 8,7 348 265 6,4 410 752 7,2

Teen 13 19 468 675 9,3 446 420 8,3 460 318 8,1

Young Adult 20 25 465 347 9,2 398 035 7,4 378 248 6,7

Working Age 26 65 2 591 580 51,2 2 885 081 53,4 2 485 483 43,6

Pensioner 66 80 480 214 9,5 670 736 12,4 891 563 15,6

Elderly 80 99 152 366 3,0 230 003 4,3 571 632 10,0

Total 0 99 5 058 012 99,9 5 400 603 99,9 5 695 675 99,9

Absolute 0 >100 5 058 119 5 401 267 5 700 200


45

Disability Weights

0

0,1 0,2 0,3 0,5 0,6 0,7 0,8 0,9

1

Asthma

Meningitis

PerfectHealth

Death

Dental caries

Acute mycardialinfarction

1st stroke ever

Liverneoplasm

Leukemia

Cretinism

SevereDepressiveEpisode

0,4


46

Estimation

19861989

19921995

19982001

20042007

20102013

20162019

20222025

20282031

20342037

20400.0

10.0

20.0

30.0

40.0

50.0

60.0

Infant Toddler Preschool Child Child TeenYoung Adult Working Age Pensioner Total Elderly

Year

Dura

tion

(Yea

rs)


47

0

1,000,000

2,000,000

3,000,000

4,000,000

5,000,000

6,000,000

0.00%

0.05%

0.10%

0.15%

0.20%

0.25%

0.30%

0.35%

0.40%

Total Years Lived

relative YLD_WHO

relative YLD_P/I

Age Group

Year

s Liv

ed

Rela

tive

Frac

tion

Year

s Liv

ed w

ith D

isabi

lity

(YLD

)

0

1,000,000

2,000,000

3,000,000

4,000,000

5,000,000

6,000,000

0.00%

0.05%

0.10%

0.15%

0.20%

0.25%

0.30%

0.35%

0.40%

0.45%

Total Years Livedrelative YLD_WHO

Age Group

Year

s Liv

ed

Rela

tive

Frac

tion

Year

s Liv

ed w

ith D

isabi

lity

(YLD

)

Comparison YLD_I and YLD_P

1986

2011


48

Infant

Toddler

Presch

ool Child

ChildTe

en

Young A

dult

Worki

ng Age

Pensio

ner

Elderl

yTo

tal0

1,000,000

2,000,000

3,000,000

4,000,000

5,000,000

6,000,000

0.00%

0.05%

0.10%

0.15%

0.20%

0.25%

0.30%

0.35%

0.40%Total Years Livedrelative YLD_WHOrelative YLD_P/I

Age Group

Year

s Liv

ed

Rela

tive

Frac

tion

Year

s Liv

ed w

ith D

isabi

lity

(YLD

)

0

1,000,000

2,000,000

3,000,000

4,000,000

5,000,000

6,000,000

0.00%

0.05%

0.10%

0.15%

0.20%

0.25%

0.30%

0.35%

Total Years Livedrelative YLD_WHOrelative YLD_P/I

Age Group

Year

s Liv

ed

Rela

tive

Frac

tion

Year

s Liv

ed w

ith D

isabi

lity

(YLD

)

2015

2040


Attributable YLD_I & attributable YLD_P – Comparison I

WHO P/I0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000

10,000

DogCatFormaldehydePM2.5D&MSmoking_aggregatedSHS_aggregatedResidual

Asthma Duration

Attrib

utab

le Y

ears

Live

d w

ith D

isabi

lity

(YLD

) (Ye

ars)

Baseline (2011)

49


Attributable YLD_I & attributable YLD_P – Comparison

WHO P/I0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000Dog

Cat

Allergen

Formaldehyde

PM2.5

Underweight

NO2

D&M

Smoking_aggregated

SHS_aggregated

Residual

Asthma Duration

Attrib

utab

le Y

ears

Liv

ed w

ith D

isabi

lity

(YLD

) (Ye

ars)

WHO P/I0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000

Asthma Duration

Attrib

utab

le Y

ears

Liv

ed w

ith D

isabi

lity

(YLD

) (Ye

ars)

1986 2006


Comparison studies – Methods WHO IHME EBoDE SETURI HealthVent Thesis

Target year 2004 2010 2004 2006 2010 2011

YLD estimate YLD_I YLD_P YLD_I YLD_I YLD_I YLD_IYLD_P

Disability Weight

0,04 0,009-0,132

0,04 0,04 0,04 0,04

Duration 15 years - 15 years 15 years 15 years 15 years-

Discounting Yes No YesNo

No Yes No

Source Asthma Data

WHO ? WHO WHO WHO KELA statistics

51


Comparison with other studiesStudy Year Factor Estimate

(YLD)Thesis (YLD)

WHO 2002 Asthma 9 526 8 974WHO 2004 Asthma 9 000 8 191HealthVent 2010 Asthma a 2 023* 2 037HealthVent 2010 PM2.5 1,a 8 653* 1 049HealthVent 2010 SHS 2,a 278* 591HealthVent 2010 Dampness & Mould 3,a 340* 397EBoDE 2010 SHS 692 604EBoDE 2010 Formaldehyde 9 0

52

* In DALYsa Includes only attributable to poor indoor air quality1 includes asthma, lung cancer, CV-diseases, COPD2 includes lung cancer, ischemic heart disease, asthma3 includes respiratory infections, asthma


Risk estimates for stressors• Risk estimates for stressor were available for short

window of time linear regression used for extrapolation for longer

period of time

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 221

1.5

2

2.5

3

3.5

f(x) = − 0.142307692307692 x + 3.98846153846154

f(x) = 0.356666666666667 x + 1f(x) = − 0.145 x + 4.01

Age (years)

Rela

tive

Risk

53


Tobacco Statistics Finland

19791980

19811982

19831984

19851986

19871988

19891990

19911992

19931994

19951996

19971998

19992000

20012002

20032004

20052006

20072008

20092010

20110

5

10

15

20

25

30

35

40

45

f(x) = − 3.99251291881455 ln(x) + 39.6842744960782R² = 0.717417262429346

Total Logarithmic (Total) 15-24 25-44 45-64

Year

Smok

ing

Popu

latio

n (%

)

54


55

PM Exposure trends

20142015

20162017

20182019

20202021

20222023

20242025

20262027

20282029

20302031

20322033

20342035

20362037

20382039

20404.5

5

5.5

6

6.5

7

7.5

8

BaU Ban SSWC Reduction SSWCSpeet Limit

Year

Tota

l am

bien

t PM

con

cent

ratio

n (m

g/m

3)


56

2015 2016 2017 2018 2019 2020

BaU Inc

BaUPrev

Control 1Preventing Inc

-x% -x% -x% -x% -x%

+a1-x% +a2-x% +a3-x% +a4-x% +a5-x%

Year

Control Inc

Control 1 Prev

Control 2Preventing Prev

-y% -y% -y% -y% -y%

Control 2 Prev

+a1 +a2 +a3 +a4 +a5

From change in Incidence to change in Prevalence


Summary I

• How much of the burden of asthma can be explained by known environmental risk factors? Which are the ones with the most impact?25-50% with PM2.5 and SHS having the biggest

impact

• Are there any protection factors capable of preventing a significant fraction of onset or symptoms of asthma?

Yes, but very weak evidence

57


Summary II

• Are the two different modeling approaches comparable? Are differences in the burden of disease estimates due to changes in the incidence or prevalence rate?

Incidence based has bigger focus on younger age groups and prevalence based estimates have bigger focus on older age groups

• Does the reduction of environmental exposures lead theoretically to a reduction of burden of disease?

10% of total BoD and 30% of attributable BoD

58


Summary III • Which control policies approach has theoretically a

bigger impact on DALYs?

Ban of Tobacco (SHS) and Increase of Pets

• Can any causality between onset and aggravation regarding environmental factors be identified?

No

• Does it make a difference to use a constant duration of disease or an age-dependent estimate?

Yes (assumed that duration is equal to Prev/Inc)

59


Conclusion

”Essentially, all models are wrong, but some are useful” (George Box)

Many uncertainties, but nevertheless, the model gives an overview over the order of magnitude of impact of exposures on asthma

Results can be used as support for decision making in public health policies

60

modeling environmental burden of disease of asthma: p rotective factors and control options

Documents