TBM SYSTEMATIC REVIEW

A systematic review of single health behavior changeinterventions vs. multiple health behavior changeinterventions among older adults

Claudio R. Nigg, PhD,1,2 Camonia R. Long, PhD2

ABSTRACTMultiple behavior change is widely used to reducetargeted health behaviors; however, its effect onbehaviors such as physical activity, nutrition, andalcohol and tobacco use among older adults remainsinconclusive. The primary purpose of this systematicreview was to evaluate the effects of single healthbehavior change (SHBC) interventions vs. multiplehealth behavior change (MHBC) interventions amongolder age individuals. PubMed was searched forpublications on health behavior interventions from2006 to 2011. Twenty-one randomized clinical trialsassessed the effects of health behavior change in olderindividuals. Results were reviewed by a number ofhealth behaviors and effectiveness. Results revealedthat within SHBC interventions, physical activity orexercise behavior revealed that interventions were themost common and showed the most promise ininfluencing positive outcomes in physical activitybehavior among community-dwelling older adults.There were too few MHBC studies identified to allowconfident comparison to SHBC interventions. The MHBCfield is still at an early stage within the older adultliterature, and more attention is recommended toinvestigate if the benefits of MHBC apply to this agegroup.

KEYWORDS

Health behavior change, Older adults, Physicalactivity, Nutrition, Tobacco, Alcohol

According to the Center for Disease Control,National Center for Health Statistics (CDC), 80 %of older adults have one chronic condition, and50 % have at least two [4]. In 2007, the top fourmajor causes of death among older adults wereheart disease (28.2 %), cancer (22.2 %), stroke(6.6 %), and chronic lower respiratory disease(6.2 %). This demonstrates the need for morefocused public health research that identifies effectivechronic disease prevention efforts.Additionally, CDC projects that by 2030, the

number of U.S. adults aged 65 or older will morethan double to about 71 million adults [4]. Theseestimates have led the CDC to make several

recommendations for efforts that target to improvethe mental and physical health of all Americans intheir later years [4]. The National Report Card forthe State of Aging and Health in America depictsdata on 15 key indicators related to the health ofadults aged 65 years and older for the United States.Key indicators for health risk behaviors for olderadults include lack of activity, eating less than fivefruits and vegetables per day, obesity, and currentsmoking. In 2009, data for persons 65 and olderrevealed that 32.7 % of older adults engaged in noleisure time activity, 72.5 % were eating less thanfive fruits and vegetables daily, 23.8 % were obese,and 8.3 % were currently smoking [5]. The UnitedStates met five of the ten healthy people 2010targets for older adults ahead of schedule. Thetargets included oral health/complete tooth loss,people currently smoking, mammograms within thepast 2 years, colorectal cancer screenings, andcholesterol checked within the past 5 years; howev-er, only one target (people currently smoking) wasfrom the key indicators for health behavior as listedin [5]. This underscores the importance of identify-ing specific health intervention efforts that address

1Department of Public HealthSciences, John A. Burns School ofMedicine,University of Hawaii at Manoa,1960 East-West Road, Honolulu, HI96822, USA2University of Hawai‘i CancerCenter, Honolulu, HI, USACorrespondence to: C Niggcnigg@hawaii.edu

Cite this as: TBM 2012;2:163–179doi: 10.1007/s13142-012-0130-y

ImplicationsResearch: More MHBC studies are needed inolder adults to inform if such approaches areeffective, provide synergy, provide savings ofresources, are more cost effective; or if theyprovide too large of a burden, are too confusing;and ultimately if MHBC is a better investmentcompared to SHBC approaches.

Practice: Older adults respond well to behaviorchange interventions delivered one behavior at atime. Practical experience in MHBC needs to begained, evaluated, and documented to see ifMHBC interventions are feasible and acceptedby older adults.

Policy: Evidence-based single health behaviorchange programs exist for older adults; however,the evidence of multiple health behavior changedoes not yet exist.

TBM page 163 of 179

preventable health risk behaviors among olderadults.

SINGLE VS. MULTIPLE BEHAVIOR CHANGE APPROACHSingle health behavior change (SHBC) interventionsmay be more focused on specific content, lessconfusing, and able to more comprehensively ad-dress intervention components. This notwithstand-ing, intervening on single behaviors can be complexand challenging—treating multiple behaviors is evenmore so. Conventional wisdom has been that it isnot possible to treat multiple behaviors simulta-neously because it is too burdensome and places toomany demands on a person’s inherent ability tochange [21]. Multiple health behavior change(MHBC) interventions are defined as efforts to treattwo or more health behaviors either simultaneouslyor sequentially within a limited time period [22].However, MHBC interventions for a commonhealth objective, e.g., cancer prevention, diabetesself-management, or weight management, haveshown significant multiple behavior changes [13–15, 26, 27, 31]. For example, smokers treated fortwo or three behaviors were as effective in beingabstinent at long-term follow-up as those treated foronly smoking [28].The hypothesis is that treating multiple behav-

iors is as, or more, effective with each of the targetbehaviors without reducing the effectiveness oftreating one behavior at a time. MHBC interven-tions, if as effective or better, provide a better useof resources applicable to a variety of healthbehaviors [20, 23]. Moreover, since it has beennoted that multiple risks multiply the health careburden both in terms of medical consequencesand costs [7, 32], studying intervention effective-ness in this context is important. Goldstein et al.[10] concluded that large gaps remain in MHBCknowledge and efficacy.MHBC interventions also address covariation/

coaction concepts that individuals taking effectiveaction on one target behavior are much more likelyto take effective action on a second behavior andthat individuals are likely to take effective action onuntreated behaviors that are related to the treatedbehaviors [13]. Moreover, success in changing oneor more lifestyle behaviors also may increaseconfidence or self-efficacy to improve risk behaviorsin individuals who have a lower motivation tochange [22]. This is related to the concept of agateway behavior [21] where one specific healthbehavior change is thought to lead other healthbehavior changes. In addition to potentially greaterefficiency and impacts on health,MHBC interventionshave a high potential for greater real-world applicabil-ity and for providing information on effective treat-ments for behaviors that co-occur [23]. This isespecially true among older adults who are morelikely to have concurrent health conditions comparedto younger adults.

The purpose of this paper is to assess whetherSHBC interventions differ in effectiveness whencompared to MHBC interventions in older adultsby reviewing the literature. The primary healthbehaviors included physical activity, nutrition, alco-hol use, and tobacco use. To increase understandingin this topic, we identified conceptual issues poten-tially related to SHBC vs. MHBC interventions(Fig. 1).

METHODSEligibility criteria, data sources, selection of studies,data extraction, precision, and subgroup analyseswere identified a priori. The procedure was con-ducted in accordance with the Cochrane Reviewers’Handbook [12].

Eligibility criteriaFor this review, we included randomized controltrials (RCTs) with physical activity, nutrition, tobac-co, and alcohol behavior interventions either indi-vidually or in some combination on older adultspublished in the English language. The healthbehaviors included both adoption (e.g., physicalactivity) and cessation behaviors (e.g., smokingcessation). As the MHBC field is relatively young[24], the focus was on papers from 2006 to 2011. Weincluded studies with older adults (>55+) fromcommunity and clinical settings. We excluded trialswhere the intervention and the outcome wereincongruent, trials where the outcome variable wasnot a health behavior, and animal studies.

Information sources and searchA search of the PubMed database was conductedusing a systematic combination of key wordsdescribing health behavior: physical activity, nutri-tion, diet, alcohol, drinking, and smoking. Table 1presents the search terms used and the number ofarticles identified.

Data collection process and quality analysisFirst, the first author (CN) independently input allkey word search combinations into the PubMeddatabase. After inputting each search combination,the second author (CL) extracted all of the relevanttrials and interventions based on the title andabstract. Second, CN and CL then both indepen-dently evaluated all the applicable abstracts. Third,CL extracted all of the appropriate full-text articles.If there was a disagreement whether to include anabstract, both authors evaluated the article to cometo a joint consensus. For instance, several RCTswhere older adult participants were given vitamin Dsupplements were not included because the trialsinvolved physicians or other health care providersadministering the vitamin rather than the older adultperforming the behavior themselves. Moreover,

SYSTEMATIC REVIEW

TBMpage 164 of 179

other trials including exercise behavior but studyingfalls as the outcome were excluded because theintervention was not congruent with the outcomebehavior (Fig. 2).Titles and abstracts from each of the key word

combinations were reviewed, and after applyingexclusion criteria, 18 publications incorporating 21interventions were retained. Precision is defined as thenumber of relevant reports identified divided by thetotal number of reports identified [12], so for thisreview, precision was calculated through the followingequation:

Precision ¼ 21=300 ¼ :07 ¼ 7%:

Data extraction and codingThe following were extracted for each dataset:population, study recruitment setting, mean age,gender, ethnicity, target sample, health behavior,adoption vs. cessation behavior, results, conclusion,and effectiveness decision. Effectiveness decisionwas coded a “+” for significance in the rightdirection, a “0” for null findings, and a “−” forsignificance of findings in the opposite directionhypothesized. Since the data search yielded RCTswhich often combined single and multiple interven-tions, they were aggregated to see which category(single intervention or multiple intervention) wouldbe conceptually appropriate for data analysis. Forexample, an intervention may consist of one largestudy on physical activity in older adults and includetwo more trials. The trials could include assessmentsof physical activity and nutrition and/or physical

activity, nutrition, and smoking cessation behavior.Each of the interventions in the trial was then listedor coded as “single” if one health behavior wasbeing studied or “multiple” if more than one healthbehavior was being studied in the trial.

RESULTSThe literature search and review process (see Fig. 2)identified 18 RCTs that met the selection criteria andwere included. Sixteen of these studies were SHBCinterventions, and two were MHBC interventions inthe older adult population. The majority of theincluded interventions was also adoption interventions(16), while only two were cessation interventions. Themost frequently studied was physical activity orexercise behavior (14 studies). Of those, 12 wereSHBC interventions, and 2 were MHBC interven-tions. Four of the included interventions includednutrition or weight loss behaviors, two were SHBCand two were MHBC. Two included SHBC interven-tions addressed alcohol reduction. Other specificcharacteristics are displayed in Tables 2 and 3.Effectiveness ratings are also noted in Tables 2 and 3for each intervention selected.

SHBC intervention studiesOverall, there was a positive rating of effectiveness inthe SHBC interventions. Of the 16 SHBC interven-tions, the sample size ranged from 38 to 1,971 olderadults. Fourteen were conducted in a communitysetting whereas only two were conducted in a primary

Addictive vs. Non-addictive

behaviors

Theoretical Model/ Perspective

Adoption behaviors vs.

Cessationbehaviors

Older adults

Number of Behaviors

(e.g. 2, 3, 4, or one behavior as

a gateway to another

behavior)

Behavior Type

Population

SHBC vs. MHBC Interventions

Specific age range (young old, old old)

Setting (clinical vs. primary)

Special populations (disabilities, women, healthy)

Race, ethnicity

Within population studies (older adults with other populations vs. solely older adults)

TTM, SCT, HBM,

TRA/TPB

Fig 1 | Working conceptual map: issues related to SHBC vs. MHBC intervention. TTM transtheoretical model, SCT socialcognitive theory, HBM health belief model, TRA/TPB theory of reasoned action/theory of planned behavior, SHBC singlehealth behavior change, MHBC multiple health behavior change

SYSTEMATIC REVIEW

TBM page 165 of 179

care setting. The majority of the reviewed interven-tions included more females than males. The race/ethnicity of the selected interventions was mainly fromthe non-Hispanic White population.Fourteen of the SHBC studies examined adoption

behaviors and two studies assessed cessation behav-iors. Of the 12 SHBC studies evaluating physicalactivity or exercise among older adults, participantsgenerally improved their level of activity at follow-up (at 6–12 months). In the nutrition or weight lossbehavior studies, intervention participants did better

than the controls at follow-up (at least 6 months afterthe intervention), improving their intake of fruitsand vegetables and compliance to dietary recom-mendations. Both cessation interventions positivelyinfluenced the older adult participant’s intake ofalcoholic drinks, reducing the number beginning at2 weeks, up through 12 months. There were 16positive, 4 zero or neutral, and 2 negative effective-ness ratings for the SHBC interventions (Table 2).

MHBC intervention studiesOverall, there was moderate rating of effectivenessin the MHBC interventions. Of the two MHBCinterventions, the sample sizes were 94 and 735older adults. Both were conducted in a communitysetting. Each of the interventions included morefemales than males, and the race/ethnicity wasmainly from the non-Hispanic White population.Both MHBC studies examined adoption behaviors,

namely, a physical activity behavior and a nutritionrelated behavior. In one study, the combination ofphysical activity and fruit and vegetable consumptionamong older adults improved only the nutritionbehavior and not the physical activity behavior atfollow-up. In fact, the physical activity behavior resultwas in the unexpected direction [3]. In the other study,participants improved in both the weight loss behaviorand the physical activity behavior (indicated by + signson Table 3). Thus, there were three positive, zeroneutral, and one negative effectiveness rating for theMHBC interventions (Table 3).

DISCUSSIONThe purpose of this paper was to systematicallyreview the literature and assess whether SHBCinterventions differ in effectiveness when compared

One database searched (PubMed)

Potential abstracts identified and screened (n=300)

Not RCTs excluded (n=52)

All Full RCT articles reviewed (n=248)

All Full articles 55+ reviewed (n=227)

Not 55+ excluded (n=21)

Not behavioral excluded (n=171)

All Full articles relevant behaviors included (n=56)

Intervention incongruent to outcome excluded (n=38)

All Full articles meet inclusion criteria (n=18)

Fig 2 | Flow chart of articles in the systematic review. RCTrandomized control trial

Table 1 | Search terms used and number of publicationsidentified

Search terms used Number ofresults

Older adults + random controltrial + physical activity

126

Older adults + random controltrial + nutrition

70

Older adults + random controltrial + diet

50

Older adults + random controltrial + alcohol

13

Older adults + random controltrial + drinking

13

Older adults + random controltrial + smoking

10

Older adults + random controltrial + multiple behavior

0

Older adults + random controltrial + multiple health behavior

0

Older adults + random controltrial + physical activity + nutrition

13

Older adults + random controltrial + physical activity + diet

5

Older adults + random controltrial + physical activity + smoking

0

Older adults + random controltrial + physical activity + alcohol

0

Older adults + random controltrial + physical activity + drinking

0

Older adults + random controltrial + nutrition + smoking

1

Older adults + random controltrial + nutrition + alcohol

1

Older adults + random controltrial + nutrition + drinking

2

Older adults + random controltrial + diet + smoking

1

Older adults + random controltrial + diet + alcohol

1

Older adults + random controltrial + diet + drinking

2

Older adults + random controltrial + smoking + alcohol

3

Older adults + random controltrial + smoking + drinking

2

SYSTEMATIC REVIEW

TBMpage 166 of 179

Table2|S

ystematic

review

ofSHBCinterven

tion

sforolde

rad

ults

Autho

rNum

ber

Recruitm

ent

setting

(com

mun

ity

vs.clinical)

Mea

nag

eGen

der

Ethn

icity

Beh

avior(s)

Ado

ption

vs.

cessation

beha

vior

Resu

ltsCon

clus

ion

Effectiven

ess

(−,0,

+)

Tanet

al.

2006

[33]

113

Com

mun

ity

6994

% Female

96%

Black

Physical

activity

(PA,as

mea

sured

bych

ange

inph

ysical

activity

levels)

Ado

ption

Resu

ltsrevealed

that

53%

ofthestud

ypa

rticipan

tswere

moreactive

than

theprevious

year

byself-repo

rt,as

compa

redto

23%

ofthecontrols

(P<

0.01

).Whe

nad

justed

forag

e,ge

nder,an

ded

ucation,

there

was

atren

dtoward

increa

sedph

ysical

activity

inthestud

ypa

rticipan

tsas

calculated

bya

kilocaloriepe

rwee

kincrea

seof

40vs.a16

%de

crea

sein

the

controls

(P=0.49

).Study

participan

tswho

repo

rted

“low

activity”at

baselin

eexpe

rien

cedan

averag

e11

0%

increa

sein

their

physical

activity

atfollo

w-up.

Amon

gthecontrols

who

werein

thelow

activity

grou

pat

baselin

e,therewas,

Resu

ltssu

ggestthat

ahigh

-intens

ity

voluntee

rprog

ram

that

isde

sign

edas

ahe

alth

prom

otion

interven

tion

can

lead

,in

thesh

ort

term

,to

sign

ificant

improvem

ents

inthelevelo

fph

ysical

activity

ofprevious

lyinactive

olde

rad

ult

voluntee

rs

+(PAat

follo

w-up)

SYSTEMATIC REVIEW

TBM page 167 of 179

onaverag

e,on

lya12

%increa

sein

physical

activity

(P=0.03

).Amon

gthosewho

wereprevious

lyactive,therewas

nosign

ificant

differen

ce(P=0.30

)Bak

eret

al.

2007

[1]

38Com

mun

ity

76.1

63.2

%Female

Not

repo

rted

Exercise

(EX,

asmea

suredby

6-min

walkan

dha

bitual

physical

activity

level)

Ado

ption

Therewereno

sign

ificant

chan

ges

in6-min

walk

performan

ce(1

%ch

ange

inexercise

grou

pvs.0.3%

chan

gein

controls,

P=0.81

).Sub

jects

withthelowest

baselin

e6-min

walkpe

rforman

cesh

owed

the

grea

test

improvem

entin

that

mea

sure

at10

wee

ks(r=−0.50

8,P=0.00

1)compa

redto

controls

Thefind

ings

ofthe

curren

tstud

ymay

beim

portan

twhe

ncons

idering

exercise

prescription

ina

practicalsetting

0(EX10

wee

ks)

Fielding

etal.20

07[8]

213

Com

mun

ity

76.5

68.5

%Female

24.9

%Non

-White

Physical

activity

(PA,

asmea

suredby

increa

sepa

rticipation

inmod

erate-

intens

itywalking

ataminim

umof

150min/w

eek,

inad

dition

tostreng

th

Ado

ption

Participationin

mod

erate-intens

ity

physical

activity

increa

sedfrom

baselin

eto

mon

ths

6an

d12

inPA

compa

redwith

health

education

controls

(P<0.001

).At12

mon

ths,

PA

Older

individu

alsat

risk

fordisability

canad

here

toa

regu

larprog

ram

ofph

ysical

activity

inalong

-term

rand

omized

trial

+(PAat

6mon

ths),+(PA

at12

mon

ths)

Table2|(continue

d)

SYSTEMATIC REVIEW

TBMpage 168 of 179

andba

lance

training

three

times

perwee

k)

participan

tswho

repo

rted

>150

min/

wk-1of

mod

erate

activity

demon

strateda

sign

ificantly

grea

ter

improvem

entin

theirSho

rtPh

ysical

Performan

ceBattery

score

compa

redwith

participan

tswho

repo

rted

<150

min/

wk-1of

mod

erate

activity

(P<0.017

)

Koltet

al.

2007

[16]

186

Clin

ical

74.1

62.4

%Female

100%

White

Physical

activity

(PA,

asmea

sured

using

theAucklan

dHea

rtStudy

Physical

Activity

Que

stionn

aire)

Ado

ption

Mod

erateleisure

physical

activity

increa

sedby

86.8

min/w

eekmorein

theinterven

tion

grou

pthan

inthe

controlgrou

p(P=

0.00

7).More

participan

tsin

the

interven

tion

grou

preache

d2.5hof

mod

erateor

vigo

rous

leisure

physical

activity

per

wee

kafter12

mon

ths(42vs.23

%,od

dsratio=2.9,

95%

confi

dence

interval=1.33

–

6.32

,P=0.00

7)

Teleph

one-ba

sed

physical

activity

coun

selin

gis

effectiveat

increa

sing

physical

activity

over

12mon

thsin

previous

lylow-

active

olde

rad

ults

+(PAat

12mon

ths)

Green

eet

al.20

08[11]

834

Com

mun

ity

74.7

72.9

%Female

79.5

%White,

13.2

%Hispa

nic-

Portug

uese,

7.3%

othe

r

Fruitan

dvege

table

cons

umption

(F&V,as

mea

suredby

the

NCIFruitan

d

Ado

ption

Theinterven

tion

grou

pincrea

sed

intake

by0.5to

1.0

servingmorethan

thecontrolgrou

p

Theinterven

tion

was

effectivein

increa

sing

the

intake

offruits

and

vege

tables

inolde

r

+(F&Vintake

at24

mon

ths)

SYSTEMATIC REVIEW

TBM page 169 of 179

Veg

etab

leScree

ner

andthe5ADay

Prog

ram

screen

er)

over

24mon

thsas

mea

suredby

theNCI

Fruitan

dVeg

etab

leScree

ner

andthe5ADay

Prog

ram

screen

er.

Themajorityof

the

participan

ts(58%)

perceivedthat

they

maintaine

dfive

ormoreservings

per

dayfor24

mon

ths

adults.Thosewho

maintaine

dtheir

levelof

perceived

intake

asfive

ormoreservings

per

daycons

umed

two

tofour

servings

per

daymorethan

thosewho

faile

dto

prog

ress

Resn

icket

al.20

08[30]

166

Com

mun

ity

7381

% Female

72%

Black

Physical

activity

(PA,

timesp

entin

exercise

and

overallph

ysical

activity)

Ado

ption

Atba

selin

e,the

participan

tsrepo

rted

that

they

enga

gedin

217min

ofexercise

per

wee

k,or

30min

daily.Therewere

statistically

sign

ificant

improvem

ents

inthetimesp

entin

exercise

(P=0.04

),in

theinterven

tion

grou

pthan

thosein

thecontrolgrou

p.Therewereno

sign

ificance

differen

cesin

overallph

ysical

activity

(F=.28,

P=

0.63

)be

twee

nthe

twogrou

ps

Thestud

yfind

ings

wereminim

ally

supp

ortedbu

tdo

sugg

estthat

future

exercise

interven

tion

sfor

olde

rad

ults

shou

ldinclud

eob

jective

mea

suresof

physical

activity

0(PAtimesp

ent

inexercise

and

overallp

hysical

activity

week)

Borschm

anet

al.

2010

[2]

121

Com

mun

ity

7063

% Female

100%

White

Walking

(WK,

asmea

suredby

ape

dometer,

coun

tedstep

s

Ado

ption

Nosign

ificant

differen

ceswere

foun

dbe

twee

nexpe

rimen

tal

This

stud

yha

shigh

lighted

metho

dological

cons

iderations

for

0(W

Kpe

rwee

k)

Table2|(continue

d)

SYSTEMATIC REVIEW

TBMpage 170 of 179

walke

dda

ilyfor1wee

k)grou

psin

step

spe

rda

yas

mea

sured

byape

dometer

PAhe

alth

prom

otionan

dresearch

with

cultu

rally

and

lingu

istically

diverse

olde

rad

ults.These

find

ings

supp

ort

thene

edfor

investigating

person

-cen

tered

approa

ches

toPA

prom

otion,

toovercome

individu

als'

person

alan

dcultu

ralba

rriers

Duruet

al.

2010

[6]

62Com

mun

ity

72.8

100%

Female

100%

Black

Walking

(WK,

asmea

suredby

pedo

meter)

Ado

ption

Interven

tion

participan

tsaverag

ed12

,727

step

spe

rwee

kat

baselin

e,compa

red

to13

,089

step

sam

ongcontrols

(P=

0.88

).At6mon

ths,

interven

tion

participan

tsha

dincrea

sedwee

kly

step

sby

9,88

3on

averag

ecompa

red

toan

increa

seof

2,42

6forcontrols

(P=0.01

6)

TheSisters

inMotion

interven

tion

ledto

anincrea

sein

walking

.This

isthe

firstRC

Tof

afaith-

basedph

ysical

activity

prog

ram

toincrea

seph

ysical

activityam

ongolde

rAfrican

-American

wom

enan

drepresen

tsan

attractive

approa

chto

stim

ulate

lifestyle

chan

gewithinthis

popu

lation

+(W

Kat

6mon

ths)

Linet

al.

2010

[17]

239

Com

mun

ity

68.7

74.2

%Male

86.1

%Non

-Hispa

nic,

White

Alcoh

olredu

ction

(AR,

asmea

sured

byredu

ctions

inthenu

mbe

rof

alcoho

licdrinks)

Cessation

Therewere

statistically

sign

ificant

differen

cesin

baselin

ealcoho

l-relatedfactors

Con

cern

abou

trisks,

read

inged

ucationa

lmaterial,an

dpe

rcep

tion

ofph

ysicians

providingad

vice

to

+(AR,

perwee

k)

SYSTEMATIC REVIEW

TBM page 171 of 179

betwee

nthetwo

grou

ps.Thirty-nine

percen

tof

the

sampleha

dredu

ceddrinking

within2wee

ksof

receivingtheinitial

interven

tion

,(13.4+7.9,

P=0.00

1)compa

redto

the

controls

redu

cedrinking

wereassociated

withea

rly

redu

ctions

inalcoho

lus

ein

olde

rat-riskdrinke

rs.

Und

erstan

ding

thesefactorswill

enab

lethe

developm

entof

better

interven

tion

strategies

toredu

ceun

healthyalcoho

lus

e

McM

urdo

etal.20

10[18]

204

Com

mun

ity

77.3

100%

Female

100%

White

Physical

activity

(PA,

asmea

suredby

chan

gein

daily

activity

coun

tsmea

suredby

accelerometry)

Ado

ption

After

adjustmen

tfor

baselin

edifferen

ces,

accelerometry

coun

tsincrea

sedmore

sign

ificantly

inthe

beha

vior

chan

geinterven

tion

(BCI)

grou

pat

3mon

ths

than

thecontrol

grou

p(P=0.00

2)an

dthepe

dometer

plus

grou

p(P=0.04

).By6

mon

ths,accelerometry

coun

tsin

both

interven

tiongrou

psha

dfallentolevelsthat

wereno

longer

statisticallyan

dsignificantly

diffe

rent

from

baseline

TheBCI

was

effective

inob

jectively

increa

sing

physical

activity

insede

ntary

olde

rwom

enat

3mon

ths,

but

physical

activity

reverted

toba

selin

eafterwithd

rawal

from

the

interven

tion

.Provisionof

ape

dometer

yielde

dno

addition

albe

nefitin

physical

activity

butmay

have

motivated

participan

tsto

remainin

thetrial

+(PAat

3mon

ths),0(PA

at6mon

ths)

Moo

reet

al.

2010

[19]

631

Clin

ical

68.4

71%

Male

87%

White,

9%

Hispa

nic,

3%

othe

r

Redu

ctionin

alcoho

lcons

umption

(RAC,as

mea

sured

Cessation

At3mon

ths,

relative

tocontrols,fewer

interven

tion

grou

ppa

rticipan

tswere

Amultifaceted

interven

tion

amon

golde

rat-risk

drinke

rsin

prim

ary

+(RACat

3mon

ths),+

(RACat

12mon

ths)

Table2|(continue

d)

SYSTEMATIC REVIEW

TBMpage 172 of 179

bythenu

mbe

rof

drinks

inthepa

st7da

ys,he

avy

drinking

(fou

ror

moredrinks

ina

day)

inthepa

st7

days

andrisk

score)

at-riskdrinke

rs(odd

sratio(OR)

0.41

,95

%confi

denceinterval

(CI)0.22

–0.75

);they

repo

rted

drinking

fewer

drinks

inthepa

st7

days

(rateratio(RR)

0.79

,95

%CI0.70

–

0.90

),less

heavy

drinking

(OR0.46

,95

%CI0.22

–0.99

),an

dha

dlower

risk

scores

(RR0.77

,95

%CI0.60

–0.94

).At

12mon

ths,

only

the

differen

cein

the

numbe

rof

drinks

remaine

dstatistically

sign

ificant

(RR0.87

,95

%CI0.76

–

0.99

)

care

does

not

redu

cethe

prop

ortio

nsof

at-

risk

orhe

avy

drinke

rs,bu

tdo

esredu

ceam

ount

ofdrinking

at12

mon

ths

Troyer

etal.

2010

[35]

298

Com

mun

ity

6082

.9%

Female

61%

White

Dietary

App

roachto

StopHypertens

ion

(DASH)dieting

(DD,as

mea

sured

byusing24

-hfood

recalls

atba

selin

e,6an

d12

mon

ths)

Ado

ption

Participan

tswho

received

mea

lswere20

%(P=

0.00

1)morelik

ely

toreach

interm

ediate

DASH

at6mon

thsan

dwere18

%(P=0.00

7)more

likelyto

mee

tsaturatedfatat

12mon

thsthan

were

thosewho

didno

treceivemea

ls.

Whe

nstratified

byrace

andincome,

gainswere

Deliveryof

seven

DASHmea

lspe

rwee

kwas

foun

dto

increa

secompliancewith

dietary

recommen

dation

sam

ong

noncom

pliant

olde

rad

ults

with

cardiovascular

disease

+(DDat

6mon

ths),+(DD

at12

mon

ths)

SYSTEMATIC REVIEW

TBM page 173 of 179

margina

llylarger

for

Whitesan

dhigh

erincomeindividu

als

vanStralen

etal.

2010

[37]

1,97

1Com

mun

ity

6457

% Female

100%

White

Physical

activity

(PA,

asmea

suredby

wee

klyminutes

oftotalPA

beha

vior

andwee

kly

minutes

offive

leisureactivities)

Ado

ption

Atba

selin

e,pa

rticipan

tswere

physically

active

for

onaverag

e63

4.9

min/w

eek(SD=

451.9).Pa

rticipan

tsin

the

environm

entally

tailo

redinterven

tion

increa

sedtheirPA

beha

vior

byalmost

50min/w

eekmore

than

participan

tsin

theba

sic

interven

tion

(environ

men

tvs.

basic=48

.5,95

%CI=

−6.3–

103.3;

P=0.08

)

Theresu

ltsfoun

dthat

providing

environm

ental

inform

ationis

aneffective

interven

tion

strategy

for

increa

sing

PAbe

havior

amon

golde

rad

ults,

espe

cially

amon

gcertainat-risk

subg

roup

s,su

chas

lower

educated

,overweigh

t,or

insu

fficien

tlyactive

participan

ts

−(PAat

baselin

e),

−(PAat

12mon

ths)

Teriet

al.

2011

[34]

273

Com

mun

ity

79.2

61.9

%Female

90.5

%White,

3.3%,A

sian

orPa

cific

island

er,4.8

%.B

lack

not

Hispa

nic,0.4

%Hispa

nic,

0.7%

native

American

,0.4%

multiracial

Physical

activity

(PA,

asmea

suredby

theSea

ttle

Protocol

for

Activity(SPA

)curriculum

for

home-ba

sed

physical

activity

includ

inga6-min

walktest)

Ado

ption

At3mon

ths,

participan

tsin

the

SPA

exercisedfor

moreminutes

(39.3,

CI=

0.2–

78.4,P=

0.04

9)an

dha

dsign

ificantly

better

self-repo

rted

streng

ththan

inthe

health

prom

otionor

routinemed

ical

care

control

cond

itiongrou

ps.

Over18

mon

ths,

theSPA

participan

tscontinue

dto

repo

rt

Older

adults

participatingin

low

levels

ofregu

lar

exercise

can

establishan

dmaintainaho

me-

basedexercise

prog

ram

that

yields

immed

iate

and

long

-term

physical

andaffective

bene

fits

+(PAas

mea

suredby

theSPA

at3

mon

ths),+(PA

asmea

sured

bytheSPA

at18

mon

ths)

Table2|(continue

d)

SYSTEMATIC REVIEW

TBMpage 174 of 179

morewee

kly

exercise

minutes

(31.4,

CI=

4.7–

58.1,

P=0.02

)than

non-

SPA

participan

ts

Rejeskiet

al.20

11[29]

83Com

mun

ity

66.9

62.7

%Female

79.5

%White,

18.1

%Black,2.4%

othe

r

Physical

activity

(PA,

asmea

suredby

participatingin

30+min

ofmod

erate-

intens

ityactivity

inmost,ifno

tall,

days

ofthewee

kforatotalwee

kly

accumulationof

150+

min)

Ado

ption

Levels

ofmod

eratePA

at6mon

thswere

sign

ificantly

high

erin

thePA

grou

p(P=0.00

28)than

the

successful

weigh

tloss

andph

ysical

activity

orag

ing

grou

ps(PA(m

ean

(SE)

=18

9.5(13.7)

min/w

eek,

WL+

PA=22

3.4(12.9)

min/w

eek,

and

SA=12

3.8(13.9)

min/w

eek

Add

itiona

lresearch

isrequ

ired

toexam

ine

themed

itationa

lprocessessu

chas

weigh

tloss

and

successful

aging

that

relatedto

physical

activity

inolde

rad

ults

+(PAat

6mon

ths)

Weins

tock

etal.

2011

[36]

1,65

0Com

mun

ity

70.8

63.6

%Female

48.3

%White,

15.2

%Black,35

.8%

Hispa

nic,

0.7%

Physical

activity

(PA,

asmea

suredby

pedo

meter

use)

Ado

ption

Inthetelemed

icine

grou

pcompa

red

withtheus

ualcare

grou

p,therate

ofde

clinein

PA(P=0.01

28)was

sign

ificantly

less

over

time.

Significant

mea

nen

dpoint

differen

ces

wereob

served

for

PA(P=0.00

3).

Pedo

meter

usewas

sign

ificantly

associated

withPA

(P=0.00

06)an

dPI

(P<0.000

1)

Thetelemed

icine

interven

tion

redu

cedratesof

declinein

PAin

olde

rad

ults

with

diab

etes.

Pedo

metersmay

beahe

lpful

inexpe

nsivead

junct

todiab

etes

initiativesde

livered

remotelywith

emerging

tech

nologies

+(PA)

SYSTEMATIC REVIEW

TBM page 175 of 179

Table3|S

ystematic

review

ofMHBCinterven

tion

sin

olde

rad

ults

Autho

rNum

ber

Recruitm

ent

setting

(com

mun

ity

vs.clinical)

Mea

nag

eGen

der

Ethn

icity

Beh

avior(s)

Ado

ption

vs.

cessation

beha

vior

Resu

ltsCon

clus

ion

Effectiven

ess

(−,0,

+)

Cam

pbell

etal.

2009

[3]

735

Com

mun

ity

66.5

49.4

%Female

64.6

%White,

35.4

%Black

Fruitan

dvege

table

cons

umption

(F&V,

mea

suredas

how

oftenfood

was

cons

umed

inthelast

mon

th),

physical

activity

(mea

suredas

minutes

per

wee

ksp

entin

very

hard,

hard,an

dmod

erate

intens

eactivity)

Ado

ption

Asign

ificant

increa

sein

F&Vcons

umption

was

foun

dforthe

combine

dinterven

tion

grou

pin

theen

tire

sample(P<

0.05

).Anincrea

seof

1.0servingin

the

combine

dinterven

tion

grou

p(baseline:

mea

n5.4,

2.8SD;follo

w-up:

mea

n6.4,

2.8SD),

compa

redto

a0.5

servingforea

chinde

pend

ent

interven

tion

and

little

chan

gefor

controls

(P<0.01).

Forph

ysical

activity,

none

ofthe

interven

tion

sprod

uced

statistically

sign

ificant

improvem

ents.At

baselin

e,averag

eself-repo

rted

MVPA

was

approxim

ately

290minutes

per

wee

k,mee

ting

the

Cen

ters

forDisea

seCon

trol

recommen

dation

ofat

least

This

stud

yindicates

that

combining

tailo

ring

and

motivationa

linterviewing

may

bean

effective

and

cost-effective

metho

dfor

prom

oting

dietary

beha

vior

chan

geam

ongolde

rhe

althy

adults

+(F&V

cons

umption),

−(PAat

follo

w-

up)

SYSTEMATIC REVIEW

TBMpage 176 of 179

150min

per

wee

kof

mod

erate

andvigo

rous

physical

activity.Atfollo

w-up,

participan

tsin

all

stud

ycond

itions

repo

rted

less

physical

activity

compa

redto

baselin

e,an

dthere

wereno

differen

ces

betwee

ngrou

psRe

jeskiet

al.20

11[29]

94Com

mun

ity

66.8

67% Female

86.2

%White,

13.8

%Black

Weigh

tloss

(WL,

asmea

suredby

redu

cing

caloric

intake

toprod

ucea

weigh

tloss

ofap

proxim

ately

0.3kg

/wee

kfor

thefirst6

mon

thsof

trea

tmen

tforatotalof

weigh

tloss

of7–

10%;

physical

activity

(PA,mea

sured

bypa

rticipating

in30

+min

ofmod

erate-

intens

ity

activity

inmost,ifno

tall,da

ysof

the

wee

kforatotal

wee

kly

accumulation

of15

0+min)

Ado

ption

Therewas

statistical

sign

ificancerelatedto

thenu

mbe

rof

days

individu

alsmee

ttheir

calorie(r=0.25

,P=0.02

)an

dsaturatedfat(r=0.29

,P<0.01)

goals.

Inad

dition

,thosein

the

WL+PA

lost

8.6%

(17.8kg

)of

their

body

weigh

tin

the

first6mon

ths,

whe

reas

weigh

tloss

was

approxim

ately

1%

inbo

thPA

(2.7

kg)an

dSA(2.3

kg).

Levels

ofmod

eratePA

at6mon

thswere

sign

ificantly

high

erin

WL+PA

(P<0.001

)as

compa

redwithSA:PA

(mea

n(SE)=18

9.5

(13.7)

min/w

eek,

WL

+PA

=22

3.4(12.9)

min/w

eek,

and

SA=12

3.8(13.9)

min/w

eek

Resu

ltsillus

trate

that

WL+PA

canbe

effectivein

improving

olde

rad

ults'ea

ting

beha

vior

and

that

thesech

ange

sare

prospe

ctively

relatedto

the

amou

ntof

weigh

tloss

+(W

Lby

caloric

intake

),+(PA

bymod

erate-

intens

ity

activity)

SYSTEMATIC REVIEW

TBM page 177 of 179

to MHBC interventions in older adults. The primaryhealth behaviors included were physical activity,nutrition, alcohol use, and tobacco use. To ourknowledge, this is the first review to attempt tocompare SHBC interventions to MHBC interven-tions among older adults. The idea of identifyingwhich type of intervention (SHBC vs. MHBC) ismost effective among older adult populations hasbeen alluded to in several studies [9, 22, 25]. Ourpreliminary findings show that there were substan-tially more (>4 times) positive SHBC interventionbehavior changes than MHBC interventions behav-ior changes identified, while the number of negativeSHBC interventions behavior changes was higherthan the number of negative MHBC interventions(2:1 ratio). However, there were also substantiallymore SHBC interventions found overall thanMHBC interventions (8:1 ratio). We strongly cau-tion any comparative interpretations of SBHC vs.MHBC interventions at this time. In fact, there aretoo few MHBC interventions at this point to makeany confident comparative conclusions at this point.Overall, the SHBC interventions on physical

activity or exercise behavior revealed the mostpositive outcomes in physical activity behavioramong community-dwelling older adults. In addi-tion, two of the SHBC studies included positivenutrition/weight loss behaviors, and two includedpositive alcohol cessation behaviors. Single behaviorintervention targets one behavior, and interventionsmay be more focused on specific content, lessconfusing, and able to more comprehensively ad-dress intervention components compared to multi-ple behavior approaches. The SHBC interventionfindings reflect that single behavior health interven-tions can be effective in physical activity, nutrition,and alcohol use. These findings are important andunderline to continue implementing effective effortsto promote the health behaviors of older adults.Conversely, it is too early to conclude about the

effectiveness of MHBC interventions in older adultsalthough there is some preliminary evidence pre-sented that these intervention types are feasible andcan produce positive results in this population [29]including underserved/minorities. Identifying ifMHBC interventions are effective with older adults,with underserved and minority older adults, and forwhat conditions, may improve the way healthbehavior change interventions are designed tomaximize benefits. This type of knowledge has thepotential to improve the efficiency and impacts thatinterventions have on health for greater real-worldapplicability [23].Limitations to be considered are that this review

was limited to peer-reviewed studies on older adultsaccessed through one database, PubMed, usingspecific key word choices. Due to the nature of thisprocess, some RCT may have been missed. Tomitigate this limitation, the reference list of eachstudy in this review was cross-checked for otherviable studies. Moreover, the limited number of

MHBC interventions limited any conclusions forthese types of studies at this time. Finally, althoughmany of these interventions reported significantresults, the majority of these studies were conductedwith community-dwelling older adults who were ofa similar background limiting the possibility ofgeneralizing these results to other populations.

Future directions/conclusionsMore MHBC studies are needed in older adults atthis time to inform if such approaches are effective,provide synergy and savings of resources or aremore cost effective, provide too large of a burden,are too confusing, and if this is a better investmentcompared to SHBC approaches.

Acknowledgments: Support for CRL was provided by R25 CA090956.

1. Baker MK, Kennedy DL, Boble PL, et al. Efficacy and feasibility ofa novel tri-modal robust exercise prescription in a retirementcommunity: a randomized, controlled trial. J Am Geriatr Soc.2007;55(1):1-10.

2. Borschman K, Moore K, Russell M, Ledgerwood K, Renehan E, LinX. Overcoming barriers to physical activity among culturally andlinguistically diverse older adults: a randomized controlled trial.Australas J Aging. 2009;29(2):77-80.

3. Campbell MK, Carr C, DeVellis B, et al. A randomized trial oftailoring and motivational interviewing to promote fruit andvegetable consumption for cancer prevention and control. AnnBehav Med. 2009;38(2):71-85.

4. Centers for Disease Control and Prevention Healthy AgingProgram State of Aging and Health in America Available at:http://www.cdc.gov/aging/data/stateofaging.htm. AccessedSeptember 20, 2011.

5. Centers for Disease Control and Prevention Chronic DiseasePrevention and Health Promotion Healthy Aging. Available at:http://www.cdc.gov/chronicdisease/resources/publications/AAG/aging.htm. Accessed September 20, 2011.

6. Duru OK, Sarkisian CA, Leng M, Mangione CM. Sisters in motion:a randomized controlled trial of a faith-based physical activityintervention. J Am Geriatr Soc. 2010;58(10):1863-1869.

7. Edington DW, Yen LT, Witting P. The financial impact of changesin personal health practices. J Occup Environ Med.1997;39:1037-1046.

8. Fielding RA, Katula J, Miller ME, et al. Activity adherence andphysical function in older adults with functional limitations. MedSci Sports Exerc. 2007;39:1977-2004.

9. Fisher EB, Fitzgibbon ML, Glasgow RE, et al. Behavior matters.Am J Prev Med. 2011;40(5):e15-e30.

10. Goldstein MG, Whitlock EP, Depue J. Multiple behavioral riskfactor interventions in primary care: summary of researchevidence. Am J Prev Med. 2007;27:61-79.

11. Greene GW, Fey-Yensan N, Padula C, Rossi S, Rossi JS, Clark PG.Change in Fruit and Vegetable intake over 24 months in olderadults: results of the SENIOR Project intervention. Gerontologist.2008;48(3):378-387.

12. Higgins JPT, Green S (editors). Cochrane Handbook for SystematicReviews of Interventions Version 5.1.0 [updated March 2011].The Cochrane Collaboration, 2011. Available from: www.co-chrane-handbook.org.

13. Johnson SS, Pavia AL, Cummins CO, et al. Transtheoreticalmodel-based multiple behavior intervention for weight manage-ment: effectiveness on a population basis. Prev Med. 2008;46(3):238-246.

14. Jones H, Edwards L, Greene G, et al. The efficacy of stages ofchange interventions in diabetes self-care and control. In: 61stAnnual Conference of the American Diabetes Association. Phila-delphia, PA; 2001.

15. Jones H, Edwards L, Vallis TM, et al. Changes in diabetes self-carebehaviors make a difference in glycemic control: the DiabetesStages of Change (DiSC) study.Diabetes Care. 2003;26(3):732-737.

16. Kolt GS, Schofield GM, Kerse N, Garrett N, Oliver M. Effect oftelephone counseling on physical activity for low-active olderpeople in primary care: a randomized, controlled trial. J AmGeriatr Soc. 2007;55:986-992.

SYSTEMATIC REVIEW

TBMpage 178 of 179

17. Lin JC, Karno MP, Barry KL, Blow FC, Davis JW, Tang L, Moore AA.Determinants of elderly reductions in drinking in older-at-riskdrinkers participating in the intervention arm of a trial to reduceat-risk drinking in primary care. J Am Geriatr Soc. 2010;58:227-233.

18. McMurdo ME, Sugden J, Argo I, et al. Do pedometers increasephysical activity in sedentary older women? A randomizedcontrolled trial. J Am Geriatr Soc. 2010;58:2099-2106.

19. Moore AA, Blow FC, HoffingM, et al. Primary care-based interventionto reduce at-risk drinking in older adults: a randomized control trial.Soc Study Addict. 2010;106:111-120.

20. Nigg CR, Allegrante JP, Ory M. Theory-comparison and multiple-behavior research: common themes advancing health behaviorresearch. Heal Educ Res. 1999;17:670-679.

21. Patterson R. Changing Patient Behavior: Improving Outcomes in Healthand Disease Management. San Francisco: Jossey-Bass; 2001:238.

22. Prochaska JO. Multiple health behavior research represents thefuture of preventive medicine. Prev Med. 2008;46:281-285.

23. Prochaska JJ, Nigg CR, Spring B, Velicer WF, Prochaska JO. Thebenefits and challenges of multiple health behavior change inresearch and in practice. Prev Med. 2010;50:26-29.

24. Prochaska JJ, Spring B, Nigg CR. Multiple health behavior changeresearch: an introduction and overview. Prev Med. 2008;46(3):181-188.

25. Prochaska JJ, Velicer WF, Nigg CR, Prochaska JO. Methods ofquantifying change in multiple risk factor interventions. PrevMed. 2008;46:260-265.

26. Prochaska JO, Velicer WF, Rossi JS. Multiple risk expert systemsinterventions: impact of simultaneous stage-matched expert systeminterventions for smoking, high-fat diet, and sun exposure in apopulation of parents. Heal Psychol. 2004;23(5):503-516.

27. Prochaska JO, Velicer WF, Redding C. Stage-based expertsystems to guide a population of primary care patients to quitsmoking, eat healthier, prevent skin cancer, and receive regularmammograms. Prev Med. 2005;41(2):406-416.

28. Prochaska JJ, Velicer WF, Prochaska JO, Delucchi K, Hall SM.Comparing intervention outcomes in smokers treated for singleversus multiple behavioral risks. Heal Psychol. 2006;25(3):380-388.

29. Rejeski WJ, Mihalko SL, Ambrosius WT, Bearon LB, McClel-land JW. Weight loss and self-regulatory eating efficacy inolder adults: the cooperative lifestyle intervention program.The J Gerontol, Ser B Psychol Sci Soc Sci. 2011;66(3):279-286.

30. Resnick B, Luisi D, Vogel A. Testing the Senior ExerciseSelf-efficacy Project (SESEP) for use with urban dwellingminority older adults. Public Health Nurs. 2008;25(3):221-234.

31. Rossi J (editor). The healthy changes trial biomedical results. InL. Ruggiero (Chair). Helping people change: the impact of stage-based behavior change interventions in diabetes care. In: 28thAnnual Conference of the American Association of DiabetesEducators. Louisville, KY; 2001.

32. Stewart AL, Verboncoeur CJ, McLellan BY, et al. Physical activityoutcomes of CHAMPS II: a physical activity promotion programfor older adults. The J Gerontol Ser A Biol Sci Med Sci. 2001;56(8):M465-M470.

33. Tan EJ, Qiau-Li X, Tao L, Carlson M, Fried L. Volunteering: aphysical activity intervention for older adults—the ExperienceCorps program in Baltimore. J Urban Health Bull NY Acad Med.2006;83(5):954-969.

34. Teri L, McCurry SM, Hogsdon RG, Gibbons LE, Buchner DM,Larson EB. A randomized controlled clinical trial of the Seattleprotocol for activity in older adults. J Am Geriatr Soc.2011;59:1188-1196.

35. Troyer JL, Racine EF, Ngugi GW, McAuley WJ. The effect of home-delivered Dietary Approach to Stop Hypertension (DASH) mealson the diets of older adults with cardiovascular disease. Am JClin Nutr. 2010;91:1204-1212.

36. Weinstock RS, Brooks G, Palmas W, et al. Lessened decline inphysical activity and impairment of older adults with diabeteswith telemedicine and pedometer use: results from the IDEATelstudy. Age Aging. 2011;40:98-105.

37. van Stralen MM, de Vries H, Bolman C, Mudde AN, Lechner L.Exploring the efficacy and moderators of two computer-tailoredphysical activity interventions for older adults: a randomizedcontrol trial. Ann Behav Med. 2010;39:139-150.

SYSTEMATIC REVIEW

TBM page 179 of 179