REVIEW

A systematic review and meta-analysis of social cognitivetheory-based physical activity and/or nutrition behavior changeinterventions for cancer survivors

Fiona G. Stacey & Erica L. James & Kathy Chapman &

Kerry S. Courneya & David R. Lubans

Received: 10 July 2014 /Accepted: 11 November 2014 /Published online: 29 November 2014# The Author(s) 2014. This article is published with open access at Springerlink.com

AbstractPurpose Little is known about how to improve and createsustainable lifestyle behaviors of cancer survivors. Interven-tions based on social cognitive theory (SCT) have shownpromise. This review examined the effect of SCT-based phys-ical activity and nutrition interventions that target cancersurvivors and identified factors associated with their efficacy.Methods A systematic search of seven databases identifiedrandomized controlled trials that (i) targeted adult cancersurvivors (any point from diagnosis); (ii) reported a primaryoutcome of physical activity, diet, or weight management; and(iii) included an SCT-based intervention targeting physicalactivity or diet. Qualitative synthesis and meta-analysis wereconducted. Theoretical constructs and intervention

characteristics were examined to identify factors associatedwith intervention efficacy.Results Eighteen studies (reported in 33 publications) metreview inclusion criteria. Meta-analysis (n=12) revealed asignificant intervention effect for physical activity (standard-ized mean difference (SMD)=0.33; P<0.01). Most studies(six out of eight) that targeted dietary change reported signif-icant improvements in at least one aspect of diet quality. NoSCT constructs were associated with intervention effects.There were no consistent trends relating to intervention deliv-ery method or whether the intervention targeted single ormultiple behaviors.Conclusions SCT-based interventions demonstrate promise inimproving physical activity and diet behavior in cancer survi-vors, using a range of intervention delivery modes. Furtherwork is required to understand how and why these interven-tions offer promise for improving behavior.Implications for Cancer Survivors SCT-based interventionstargeting diet or physical activity are safe and result in mean-ingful changes to diet and physical activity behavior that canresult in health improvements.

Keywords Cancer . Physical activity . Nutrition . Systematicreview . Social cognitive theory

Background

Due to population growth and aging and improved cancerdetection and treatment, the number of cancer survivors isincreasing [1, 2]. Many cancer survivors experience sideeffects from treatment and are at risk for secondary cancersand other chronic diseases such as cardiovascular disease anddiabetes [3]. Several systematic reviews and meta-analyseshave synthesized the findings of physical activity (PA) inter-ventions in cancer survivors [4–9]. These reviews concluded

F. G. Stacey (*) : E. L. JamesSchool ofMedicine and Public Health, University of Newcastle, W4,HMRI Building, Callaghan, NSW 2308, Australiae-mail: fiona.stacey@newcastle.edu.au

F. G. Stacey : E. L. JamesHunter Medical Research Institute, Callaghan, NSW, Australia

F. G. Stacey : E. L. JamesPriority Research Centre for Health Behavior, University ofNewcastle, Callaghan, NSW, Australia

F. G. Stacey : E. L. James :D. R. LubansPriority Research Centre in Physical Activity and Nutrition,University of Newcastle, Callaghan, NSW, Australia

K. ChapmanCancer Council New South Wales, Woolloomooloo, NSW, Australia

K. S. CourneyaFaculty of Physical Education and Recreation, University of Alberta,Edmonton, Alberta, Canada

D. R. LubansSchool of Education, University of Newcastle, Callaghan, NSW,Australia

J Cancer Surviv (2015) 9:305–338DOI 10.1007/s11764-014-0413-z

that being physically active improved fitness, strength, bodycomposition, fatigue, anxiety, depression, self-esteem, physicalfunction, bone health, and quality of life [5, 6, 8, 9] and reducedrisk of cancer recurrence and mortality among survivors ofbreast, bowel, prostate, and ovarian cancer [9, 10]. Dietaryinterventions report improved physical functioning [11, 12]and weight loss [13], lower levels of depression [12], and areduction in new cancer events [13]. Diet and PA also play arole in energy balance andweight management, an independentpredictor of cancer risk, and risk of recurrence and mortality[14] and contribute to the development of other chronic dis-eases, like diabetes and cardiovascular disease [15–17]. Forthese reasons, guidelines recommend PA (both aerobic andresistance), healthy diet, and healthy weight management forimproving the health and well-being [8, 9, 14, 18–21] of cancersurvivors across all phases of the cancer continuum [22]. Whileweight management is not a lifestyle behavior, it is the keytarget of lifestyle behavior strategies. Despite the potentialimpact of behavior on improved health outcomes, cancer sur-vivors’ compliance with health recommendations remains lessthan optimal and is similar to the general population [23–25].

Effective diet and PA interventions have the potential toimprove cancer survivors’ health, but little is known about whatinterventions work best. Interventions based on behavioraltheory are reported to be more effective than atheoretical ap-proaches [26, 27]. Theory-based research provides a frame-work for the development and evaluation of interventions[28] and facilitates understanding of the factors that mediatebehavior change and the reasons why the intervention workedor failed [29, 30]. Social cognitive theory (SCT) is one prom-ising theory for use in behavior change interventions [31],particularly as it provides a framework for understanding whypeople make andmaintain health behaviors. The key constructsof SCT include the following: (1) knowledge of health risks andbenefits; (2) perceived self-efficacy that a person can controltheir own health habits; (3) the expected costs and benefits oroutcome expectations; (4) health goals, both proximal anddistal intentions to engage in the behavior; (5) perceived facil-itators and social support; and (6) barriers to making changes[31]. In 2004, Bandura reported a framework that specified thekey determinants and the interplay between the key constructs(known as “reciprocal determinism”). Knowledge of healthrisks and benefits sets the scene for possible behavior change;however, it is not enough to prompt behavior change alone[31]. Self-efficacy influences outcome expectations and bar-riers/facilitators, and all constructs influence goals [31]. Allconstructs influence behavior and motivation and are influ-enced by the environment [31]. Self-efficacy is the centralconstruct in SCT because it influences behavior directly,through belief in their ability to apply skills effectively indifficult situations, and indirectly, through influence on goals,outcome expectations, and barriers and facilitators [31, 32].Self-efficacy is a major influence on motivation by determining

the goals people set for themselves, the strength of commitmentto them, and the outcomes they expect for their efforts [32].Self-efficacy allows the individual to gain knowledge anddevelop skills, and as self-efficacy increases, people expectpositive outcomes, overcome barriers, and show motivationand commitment to goals [32]. SCT constructs explain 40–71 % of the variance in PA behavior in adults [33–37] andhave been shown to explain dietary behavior in adults,explaining 14–35, 22–53, and 36–61 % of the variance in fat,fiber, and fruit and vegetable intake [37, 38]. SCT also offersprinciples on how to motivate people to make positive behaviorchange [31]. Previous meta-analysis of health outcomes trialswith cancer survivors concluded that SCT-based interventionsresulted in improvements in global affect, depression, socialoutcomes, objective physical outcomes, and specific quality oflife outcomes [39]. However, little is known about whetherinterventions based on SCT can positively impact on PA anddiet behaviors, and what constructs and intervention character-istics are associated with increased behavior change.

There are currently no systematic reviews including multi-ple cancer types that synthesize both PA and diet behaviorchange interventions. While there is significant evidencesupporting the impact of diet and PA behavior on healthoutcomes, there is a need to move to interventions that testhow to motivate cancer survivors to make positive sustainablebehavior change. Current evidence suggests that cancer sur-vivors do not maintain PA behavior after participating in asupervised PA intervention [40, 41]. This review examined PAand diet behavior change interventions based on SCT incancer survivors with mixed diagnoses both during and aftercompletion of cancer treatment [42].

Aim

This systematic review and meta-analysis aimed to (1) syn-thesize randomized controlled trials (RCTs) evaluating theefficacy of SCT-based behavior change interventions on PAand/or diet behaviors for cancer survivors of mixed diagnosesand (2) identify successful strategies for behavior change thatcan be used to guide intervention development. Of importancefor a reviewwith this aim is the definition of what constitutes abehavior change trial. We have used the reference defined byCourneya [43], that is, trials where the primary outcome isbehavior (as opposed to a health outcomes trial where theprimary outcome is quality of life, fatigue, etc.).

Method

Search strategy

The review was guided by the PRISMA statement [44].Studies were identified by structured database search from

306 J Cancer Surviv (2015) 9:305–338

inception until September 2014, in PsycINFO, CINAHL,Cochrane Central Register of Controlled Trials, Embase,Medline, SportDiscus, and Web of Science using the follow-ing search strings:

1. (Cancer survivor) or (cancer patient) or cancer.2. Nutrition or diet or food or fruit or vegetable.3. (Physical activity) or exercise or weight or aerobic or

(strength training) or (resistance training) or walking.4. (Social cognitive theory) or (social cognitive) or (social

learning theory) or (behavio#r change theor*). Stringswere made up of 1+(2 or 3) +4.

A sample search strategy is listed in the Appendix 1.Searches were limited to English language articles and thosethat targeted humans. Study titles were screened for eligibilityby a single reviewer (FS). Full text of the remaining titles wasobtained and screened in hierarchical order with studies ex-cluded at the first reason for exclusion (FS).

1. Participants: adults aged 18 years or older, diagnosed withany cancer (at any point from diagnosis)

2. Outcomes: primary outcome of PA or diet or body weight(loss, or maintenance)

3. Intervention:

(a) Any intervention designed to influence any type ofPA or diet quality

(b) Based on Bandura’s SCT [31], or explicitly de-scribed and referenced any SCT component (suchas “self-efficacy”)

4. Comparator: any parallel control group5. Study design: RCTs

Data extraction

Data extraction was conducted by one author (FS), and theextracted data was independently checked by a second author(EJ, DL, or KC). Disagreements were resolved by consensus.Data extraction forms were developed, piloted with one trial,and amended (FS). The following data were extracted: studypopulation and eligibility, behavior change outcomes andfollow-up periods, intervention characteristics, and how thetheory constructs were operationalized and assessed. Whereauthors indicated a trial protocol number, the protocol wasretrieved, but no other attempt was made to obtain unpub-lished trial information.

Synthesis of results

Separate meta-analyses were planned for PA and dietary out-comes. However, due to heterogeneity in dietary outcomes,

meta-analyses were conducted only for studies that reportedthe effect of the intervention on total PA using RevManversion 5.1 [45]. As recommended by the Cochrane Collabo-ration, posttest means and their standard deviations were usedin the analysis. Intention-to-treat data was extracted frompapers. When studies compared multiple treatment groupswith a single control group (n=2), the sample size of thecontrol group was divided to avoid double counting. All datawere considered continuous, but as PA was measured usingvarious methods, we report the standardized mean difference(SMD) and their 95 % confidence intervals. Statistical hetero-geneity was examined using chi-squared and the I2 index tests.A guide to the interpretation of heterogeneity based on the I2

index is as follows: 0–40 % might not be important; 30–60 %may represent moderate heterogeneity; 50–90 % may repre-sent substantial heterogeneity; and 75–100 % considerableheterogeneity [46].

Subgroup analyses comparing the number of behaviorstargeted (multiple behaviors compared to one only), cancertype, and number of theoretical constructs operationalized(self-efficacy compared to multiple theoretical constructs)were planned. However, the limited number of studies andheterogeneity of included trials did not allow for subgroupanalyses.

Risk of bias

Risk of bias was assessed using the McMaster Quality As-sessment Tool [47], with a score of “strong,” “moderate,” or“weak” methodological quality assigned to each of six sec-tions (1. selection bias; 2. study design; 3. confounders; 4.blinding; 5. data collection methods; 6. withdrawals and drop-outs). A global rating was made based on the ratings fromeach of the six sections. As recommended, papers with no“weak” ratings were “strong” methodological quality; thosewith one “weak” rating were “moderate”; and those with twoor more “weak” ratings were “weak” [47]. Risk of bias wasundertaken by two independent reviewers (FS and EJ, or DL,or KC), with disagreements resolved by consensus (FS andEJ).

Results

Study selection

Figure 1 shows the flow of studies through the review processand the reasons for exclusion. Database searches resulted in2020 potentially relevant titles. The full text of 110 articleswas assessed for eligibility, and 18 studies (reported in 33publications) met inclusion for the review (Table 1). Studieswere grouped and reported by intervention topic: PA only (ten

J Cancer Surviv (2015) 9:305–338 307

trials), diet only (one trial), or multiple health behavior (PAand diet) (seven trials).

Risk of bias assessment

There was initially 75 % agreement between authors on thestudy assessment criteria and full consensus was achievedafter discussion. Risk of bias results are reported in Table 2.Of the ten PA-only studies, five were classified as strongmethodological quality [48–52], three as moderate [53–56],and two as weak [57, 58]. The diet-only trial was classified asmoderate [59]. In the sevenmultiple behavior studies, one wasclassified as strong [60], four were moderate [61–65], and two

weak [66, 67]. Three trials were excluded from the meta-analysis [57, 58, 67] due to being weak.

The most common areas with a high risk of bias wereselection bias, confounders, and blinding. Eleven studies[53, 54, 56–58, 61–67] were rated as weak in selection biascategory with less than 60 % of potentially eligible partici-pants recruited. Two studies [59, 66] were rated as weak as thecontrol of confounders was not described. No studies receiveda strong rating for blinding as (understandably given they arebehavior change trials), all participants were aware of theresearch question, and if the outcome assessor was also awareof the intervention status of participants, studies were rated asweak methodological quality [57, 58, 67].

Fig. 1 PRISMA flow diagram

308 J Cancer Surviv (2015) 9:305–338

Physical activity trials

There were ten trials that targeted PA alone [48–54, 56–58].

Participants

Ten PA trials reported a total of 960 participants (range36–330). Six trials targeted breast cancer survivors [48,50–52, 54, 58, 68–72], one targeted colorectal cancersurvivors [53], one targeted both breast and bowel can-cer survivors [49], and two included cancers of mixeddiagnoses [56, 57]. Mean time since diagnosis was3.1 years (range 0.9–4.9 years) [50, 51, 53, 54,56–58]. Time since diagnosis was not reported in twostudies [49, 52]; however, one trial reported participantswere scheduled to begin chemotherapy [52]. All othertrial participants had completed active cancer treatment(excluding hormone treatment). In five studies, onlycancer survivors who were inactive or insufficientlyactive were eligible to participate [48–50, 57, 58]. Threetrials used a wait-list control group design [48, 51, 58],three had attention control groups [50, 56, 57], andthree had usual care control groups [49, 52, 54]. Onetrial used an attention control design, with the controlgroup offered a limited intervention (written materialsonly) at the end of the study [53].

Intervention characteristics

One intervention was delivered by email [48], one de-livered by mail [54, 55], and one delivered usingFacebook [56], and all others used a combination ofdelivery formats, including telephone [49, 50, 52, 53,57, 58], mail [50, 53], and face-to-face counseling[51–53, 57, 58]. The majority were home-based, withonly one intervention reporting supervised PA sessions[51]. Three were walking interventions [51, 52, 58], andfour had PA goals that were based on duration [49, 54,56, 57] and/or moderate intensity [50, 53, 54, 56, 57].One targeted resistance training [54].

Interventions were commonly 12 weeks in duration [48,50, 51, 53, 54, 56, 58] and ranged from 6 [52] to 18 weeks[57]. The average number of intervention contacts was 15, andranged from 1 [54] to 52 [56]. Intervention adherence washigh, ranging from 94 % compliance with home exercise logs[58] to 99 % of total contacts completed [51]. Telephonecounseling adherence was also high with a median of 9 (of11) calls completed [49] and a mean of 11 (of 12) callscompleted [50, 53]. The intervention delivered usingFacebook reported lower adherence, with 81% of interventionparticipants who reported receiving ten or more messagesfrom Facebook, and 49 % had made two or more Facebookposts [56].

Outcome assessment

Two trials used an objective measure (accelerometer) to assessPA behavior change [50, 51]. All others relied on self-reportmeasures [48, 49, 52–54, 56–58] or used an objective measurein a subsample only [58]. Effect sizes for PA behavior changewere reported in four studies (d=0.55–1.93) [51, 53, 54, 57].Three home-based walking interventions reported significantimprovements postintervention (6–12 weeks) to total PA (d=1.02; P=0.004) [51] and walking [52, 58]. Three moderateintensity interventions reported significant postinterventionincreases in PA (d=0.55; P<0.05) (d=1.93; P=0.02) [50,53, 57], and two reported nonsignificant increases to aerobicand moderate-vigorous PA [54, 56]. One trial that targetedresistance training reported significant improvements and thatthe odds of meeting the resistance training guidelines hadincreased by 3.38 in the tailored intervention group [54].

Of the four trials that reported follow-up assessments of6 months or longer [51, 53, 57, 68], only two reported behav-ior change 3 months after intervention completion [53, 68].One trial reported that accelerometer-assessed behavior wasmaintained [68], and one reported that there were significantpostintervention changes that were not maintained at 6 and12 month follow-ups [53]. Study retention was high, with amean retention rate of 86 % (range 71 % [57] to 95 % [50,53]). One trial [52] reported adverse events involving twoparticipants that experienced anemia, shortness of breath,and dizziness. Participants in this trial were undergoing activetreatment at the time of intervention.

Meta-analysis of SCT intervention effects on physical activity

Meta-analysis was conducted with 12 trials, which reportedPA outcomes [48–54, 56–58, 60, 63–65] in Fig. 2. Six trialswere not included in the meta-analysis [57–59, 61, 66, 67].Reasons for exclusion were as follows: diet-only study [59];did not report adequate information from the modified PArecall [61]; and did not report duration of PA [66]; or had ahigh risk of bias [57, 58, 67].

Two trials consisted of three study arms, which comparedtwo PA interventions to a standard recommendation control[54, 55], and a PA intervention to a PA and diet intervention,compared to an attention control group [60]. The results foreach intervention compared to the control group are reportedseparately in the meta-analysis. Results were pooled to estab-lish the effects of interventions on total PA at interventioncompletion. As there was moderate heterogeneity amonginterventions (χ2=22.71, df=13 [P=0.05]; I2=43 %), therandom effects models were used. The impact of interventionson PA immediately postintervention was significant (SMD=0.33 [0.23, 0.44], Z=6.34 [P<0.00001]) (Fig. 2). Sensitivityanalysis was undertaken that compared this analysis to a meta-analysis which included the three trials with high risk of bias

J Cancer Surviv (2015) 9:305–338 309

Tab

le1

Descriptio

nof

included

trials

Study

Participants;m

ean

age;cancer

type;tim

esincediagnosis

Interventio

n(type,intensity,

duratio

n)Studydesign

and

evaluatio

nOutcomes

Results

Retentio

n

PA-onlystudies

Shortetal.[54,55]

Country:A

ustralia

Participants:N

=330

Meanage:55

years

Cancertype:b

reast

Timesincediagnosis:

41monthssince

activ

etreatm

ent

(SD=39)

G1:

Standard

recommendatio

ncontrolg

roup

received

natio

nal

PAguidelines

brochure

G2:tailo

red-print:computer-tailo

red

A44-page

newsletters(3)

G3:

targeted-print

interventio

n:54-

page

A5booklet“Exercisefor

Health

”(1)

Type:aerobicPA

,atleastmoderate

intensity,for

30min

ormoreon

mostdaysof

theweek.In

G2and

G3,participantswerealso

encouraged

toperform

resistance

training

exercises1–3tim

esper

week

Intensity:G

2had3newslettersover

12weeks.G

3had1mailout

over

12weeks

Duration:

G2received

one

newslettereach

6weeks

Studydesign:3

arm

RCT

Follow-up:

4,10

months

Com

parisongroup:

usualcare

Outcomemeasures:Godin

leisure

timePA

a ,adherenceto

meetin

gaerobicandresistance

training

guidelines,m

eandaily

steps

(pedom

eter)

Primary:G2reported

statistically

significanteffecto

nself-reported

resistance

activ

ity(P<0.01)and

onmeetin

gtheresistance

training

guidelines

(P<0.01).G2andG3

reported

nonsignificant

improve-

mentsto

self-reportedaerobicac-

tivity.T

here

was

nosignificant

effectformeetin

gtheaerobicPA

guidelines

Other

behaviors:nonsignificant

increase

instep

countsforG2and

G3participants.G

1step

counts

decreased

90% (n=299)

Valleetal.[56]

Country:U

SA

Participants:N

=86

Meanage:31.7

years

Cancertype:3

1%

hematologic;2

0%

breast;1

5%

gynecologic

Timesincediagnosis:

58.2

months

(SD=44.0)

G1:

Facebook-based

self-helpcom-

parisongroup

G2:

Facebook-based

interventio

ngroupwith

weeklybehavioral

lesson

onPA

andbehavioral

strategies

(12)

(FITNET)

Type:150

minof

moderateintensity

PAperweek

Intensity:1

2weeklybehavioral

lessons(sentv

iaFacebook

message),discussion

questio

nsposted

onFacebook(16total),

resourcesandreminderseach

posted

once

perweek(24)

Duration:

minim

umof

4contacts

perweekover

12weeks

Studydesign:2

arm

RCT

Follow-up:

12weeks

Com

parisongroup:

self-directed

Facebook

group

Outcomemeasures:Godin

leisure

timeexercise

questio

nnairea ,

interventio

nadherenceand

acceptability

Primary:significantd

ifference

betweengroups

inestim

ated

change

inlig

htPA

minsperweek

over

12weeks.G

2participants

reported

increasesin

minsof

moderatetovigorous

PAandtotal

PA;h

owever,these

werenot

significant

77%

(n=66)

Rogersetal.

[51,68–70]

Country:n

otexplicitly

stated.A

uthorsbasedin

Illin

ois,USA

Participants:N

=41

Meanage:53

years

Cancertype:b

reast

G1:

wait-listcontrol

G2:

discussion

groupsessions

(6),

individual-supervisedexercise

(12),individualface-to-face

Studydesign:2

arm

RCT

Follow-up:

3,6months

Outcomemeasures:totalactivity

countsa ,steps,minsof

moderate-

vigorous

PA(accelerom

eter);

Godin

leisuretim

ePA

Primary:significantincreaseintotal

activ

itycounts(m

ean

difference=72,103).Effectsize

d=1.02

(P=0.004)

asmeasured

byaccelerometerat3monthsand

92%

(n=38)

310 J Cancer Surviv (2015) 9:305–338

Tab

le1

(contin

ued)

Study

Participants;m

ean

age;cancer

type;tim

esincediagnosis

Interventio

n(type,intensity,

duratio

n)Studydesign

and

evaluatio

nOutcomes

Results

Retentio

n

Timesincediagnosis:

34monthssince

surgery(0.7–134)

counselin

g(3),transitio

nto

home-basedprogram

Type:m

oderateintensity

with

the

aim

ofbuild

ingup

to150min

ofmoderatewalking

perweek

Intensity:21sessions

over3months

Duration:

multip

leexposure

(minim

umweekly)

Com

parisongroup:

wait-listcontrol

remainedsignificantat6

months

(meandifference=61,651;

P=0.06)

Other

behaviors:significantincrease

inmoderateandvigorous

minutes

(d=0.57;d

=0.54

(P=0.09)).

Nonsignificantincreasein

self-

reported

moderateandvigorous

activ

ity(d=0.16;P

=0.63)

Pinto

etal.[50,71];

Rabin

etal.[72]

Country:n

otexplicitly

stated.A

uthorsbasedin

Rhode

Island,U

SA

Participants:N

=86

Meanage:53

years

Cancertype:early

stagebreastcancer

Timesincediagnosis:

1.74

years(SD

1.49)

G1:

contactcontrol,w

eeklyphone

callandcancer

survivorship

writtensheets

G2:

weeklytelephonecounselin

g(12)

andcustom

ized

written

feedback

(4),homeexercise

logs,

pedometer

Type:m

oderateintensity

PA(55–

65%

ofmaxim

umheartrate)

walking

gradually

build

ingup

togoalof30

minofaccumulated

PAperdayon

atleast5

days

per

week

Intensity:16contactsover3months

Duration:

multip

leexposure

(minim

umweekly)

Studydesign:2

arm

RCT

Follow-up:

12weeks,

6months,

9months

Com

parisongroup:

attentioncontrol

Outcomemeasures:7dayPA

recalla,1

mile

walktesta ,

accelerometera ,percentage

ofbody

fat(skinfold

thickness)a ,

body

massindexa

Primary:significantincreaseinmins

ofPA

,and

1mile

walktest

(P<0.001).S

ignificant

between

groupdifferenceswereevident

fortotalm

insof

PAon

7dayPA

recall(P<0.001),higherweekly

minsof

moderateintensity

PA(P<0.001),highertotalenergy

expenditu

re(P<0.001)

at12

weeks

Nodifference

inbody

massindexor

percentage

ofbody

fat

Other

behaviors

95%

(n=82)

Bennettetal.[57]

Country:n

otexplicitly

stated.F

irstauthor

basedin

New

Zealand;

co-authorsbasedin

Portland,U

SA

Participants:N

=56

Meanage:

interventio

n56

years;control

60years

Cancertype:any

Timesincediagnosis:

4.8years(SD3.0)

interventio

n

G1:

contactcontrol

G2:

singleface-to-face

counselin

g(30min)with

3follo

w-uptele-

phonecalls

(20min

percall),pe-

dometer

Type:p

hysicalactivity

aim

toreach

30min

ofmoderateintensity

PAon

mostd

aysof

week

Intensity:4

contactsover

18weeks

Duration:multip

leexposure(atleast

2weeks

apart)

Studydesign:2

arm

RCT

Follow-up:

3,6months

Com

parisongroup:

attentioncontrol

Outcomemeasures:Com

munity

Health

yActivities

Model

Program

forSeniors(CHAMPS

)(caloricexpenditu

reperweekin

kilocalories

perweek)

a

Primary:significantincreasein

PAat6months(d=0.55;P

<0.05)

with

adifference

inPA

increase

over

timeof

1159

kcalperweek

betweenthetwogroups

Other

behaviors

Interventio

n.71.4

%(n=20);

control,

92.9

%(n=26)

Matthew

setal.2007

[58]

Country:U

SA

Participants:N

=36

Meanage:

interventio

n51

years;control

57years

G1:

wait-listcontrol

G2:

face-to-face

behavioral

counselin

g(1)andtelephone

counselin

g(5)

Type:w

alking

(moderateintensity

)build

ingfrom

20to

30min

per

session,3tim

esperweekto

30–

Studydesign:2

arm

RCT

Follow-up:

6,12

weeks

Com

parisongroup:

wait-listcontrol

Outcomemeasures:Com

munity

Health

yActivities

Model

Program

forSeniors(CHAMPS

)(energyexpenditu

reMET-hper

week)

a ,accelerometer(subsample

only),21-item

diethabits

Primary:significantincreaseinself-

reported

walking

(P=0.01),MET-

hperweek(P=0.01)with

differ-

ence

of10.2

MET-hperweekof

walking

at12

weeks

betweenthe

groups

Not

reported

J Cancer Surviv (2015) 9:305–338 311

Tab

le1

(contin

ued)

Study

Participants;m

ean

age;cancer

type;tim

esincediagnosis

Interventio

n(type,intensity,

duratio

n)Studydesign

and

evaluatio

nOutcomes

Results

Retentio

n

Cancertype:

postmenopausal

breast

Timesincediagnosis:

0.9years(range

0.7–1)

interventio

n

40min

persession,5tim

esper

week

Intensity:6

sessions

over

12weeks

Duration:decreasing

frequencyover

12weeks.F

ace-to-facecounsel-

ing(30min).Telephonecounsel-

ing(10–15

min

percall)

questio

nnaire,19-item

fruitand

vegetablescreener

Other

behaviors:Nosignificant

changesto

fruitand

vegetable

consum

ptionandoveralld

ietary

habitsandno

significantchanges

werenoted,although

thedatawas

notshown

Ligibeletal.[49]

Country:U

SA

Participants:N

=121

Meanage:54

years

Cancertype:b

reast,

colonor

rectal

cancer

Timesincediagnosis:

notreported

G1:

usualcare,offeredone

consultatio

nwith

exercise

trainer

G2:

semistructuredtelephone

counselin

gandparticipant

workbook

Type:p

hysicalactivity

Intensity:1

0–11

semistructured

telephonecalls

(30–45

min

each)

over

16weeks

Duration:decreasing

frequencyover

16weeks

Studydesign:2

arm

RCT

Follow-up:

16weeks

Com

parisongroup:

usualcare

Outcomemeasures:7dayPA

recall

interview(changein

minutes

ofweeklyPA

)a

Primary:nonsignificant

increase

inphysicalactiv

ityminutes

per

weekby

40min

(P=0.13),and

MET-hperw

eekby

2h(P=0.23)

Other

behaviors

Interventio

n,79

%(n=48);

control,

85%

(n=51)

Wangetal.[52]

Country:T

aiwan

Participants:N

=72

Meanage:50

years

Cancertype:n

ewly

diagnosedbreast,

scheduledto

start

chem

otherapy

Timesincediagnosis:

Firstmeetin

gis

24hpriorto

participantsurgery

G1:

usualcare

G2:

weeklytelephonecalland

weeklyindividualface-to-face

meetin

gs(heartratemonito

r,pe-

dometer,exercisediary,androle

modelstory)

Type:h

ome-basedwalking

program

oflowto

moderateintensity

from

40to60

%,3

to5tim

esperw

eek,

atleast3

0min

persessionor

the

accumulationof

30min

perses-

sion

Intensity:1

2telephoneandface-to-

face

contactsover

6weeks

Duration:

2sessions

perweek

Studydesign:2

arm

RCT

Follow-up:

24h

priorto

day1of

chem

otherapy,

thedayof

chem

otherapy

mid-cycle,

6weeks

Com

parisongroup:

usualcare

Outcomemeasures:exercise

behavior

(Godin

leisuretim

equestio

nnaire)a

Primary:significantincreases

inphysicalactiv

ityatallfollow-up

timepointsfortheinterventio

ngroup(allP<0.05).The

differ-

ence

betweenthegroups

was

62.7,58.2,and38.8

min

atthe

second,third,and

fourth

follo

w-

up(P<0.001)

Other

behaviors

Interventio

n,86

%(n=30);

control,

86%

(n=32)

Pinto

etal.[53]

Country:U

SA

Participants:N

=46

Meanage:57

years

Cancertype:colon

orrectalcancer

Timesincediagnosis:

3.1years(SD1.6)

(intervention)

G1:

contactcontrol

groupoffered

writtenmaterialatstudy

completion

G2:

oneface-to-face

appointm

ent,

oneweeklytelephonecall(12),

weeklyPA

andcancer

survivor-

shiptip

sheet(12),feedback

letter

summarizingprogress

(4),pe-

dometer

Studydesign:2

arm

RCT

Follow-up:

3,6,

12months

Com

parisongroup:

contactcontrol,

offeredwritten

materialsat

study

completion

Outcomemeasures:7dayPA

recall

(self-reported

PA)a,C

ommunity

Health

yActivities

Model

Program

forSeniors(CHAMPS

),accelerometer

Primary:Significant

increase

inPA

at3monthsin

interventio

n(d=1.93)(P=0.02),butincreases

werenotm

aintainedat6and

12month

follo

w-ups.T

here

was

asignificantd

ifferencebetween

groups

at3monthsby

117min/

week(P<0.05)butn

otat6or

12months

Other

behaviors

Interventio

n,95

%(n=19);

control,

88%

(n=23)

312 J Cancer Surviv (2015) 9:305–338

Tab

le1

(contin

ued)

Study

Participants;m

ean

age;cancer

type;tim

esincediagnosis

Interventio

n(type,intensity,

duratio

n)Studydesign

and

evaluatio

nOutcomes

Results

Retentio

n

Type:m

oderateintensity

home-

basedPA

with

goalto

perform

moderateintensity

activ

ityaero-

bicactiv

ities

at64–76%

ofesti-

mated

maxim

umheartrate

Intensity:29contactsover12

weeks

Duration:

2contactsperweek

minim

umHatchettetal.[48]

Country:n

otexplicitly

reported.L

eadauthor:

Mississippi,U

SA

Participants:N

=85

Meanage:not

reported

Cancertype:b

reast

cancer

Timesincediagnosis:

44%

between0

and20

months;

27%

between21

and40

months;

19%

between41

and70

months

G1:

wait-listcontrol

G2:em

ailm

essages(8),accesstoan

e-counselor(experienced

exercise

physiologist)

Type:p

hysicalactivity

(emails

focusedon

changing

SCT

constructs)

Intensity:total8em

ailm

essages,

weeklyfor5weeks,then

fortnightly

fornext

6weeks

Duration:

once

perweek(5

weeks),

then

fortnightly

for6weeks

fora

totalof8

messagesover12

weeks

Studydesign:2

arm

RCT

Follow-up:

6,12

weeks

Com

parisongroup:

wait-listcontrol

Outcomemeasures:7dayPA

recalla

Primary:At12weeks,for

totaldays

ofexercise,therewas

asignificant

difference

betweenthegroups

(P<0.001)

with

theinterventio

nreporting2.05

moredays

ofexercise

comparedto

thecontrol

group(P<0.001)

Other

behaviors

Interventio

n,88.4

%(n=38);

control,

85.7

%(n=36)

Dieto

nly

Parsonsetal.[59]

Country:U

SA

Participants:N

=43

Meanage:64

years

Cancertype:p

rostate

Timesincediagnosis:

notreported—

receivingonly

activ

esurveillance

astreatm

ent

G1:

standard

care

control

G2:

telephonecounselin

gType:d

iet(7servings/day

vegeta-

bles;2

servings/day

wholegrains,

1serving/daybeans/legumes)

Intensity:totalof

13sessions

Duration:

13structured

telephone

counselin

gsessions

over

6months.Calld

urationof

25–

50min

Studydesign:2

arm

RCT

Follow-up:

6months

Com

parisongroup:

usualcare

Outcomemeasures:24

hdietary

recalla,blood

samples

(plasm

acarotenoid

concentration)

a

Primary:To

talvegetableandtomato

productintakesignificantly

increasedin

theinterventio

n(P<0.05).Nosignificantchanges

infruit,wholegrain,beans,gram

sof

fiberperday,or

fatintake

Other

behaviors

96.7

%(n=42)

Multip

lebehavior

studies

Dem

ark-Wahnefried

etal.S

TRENGTH

trial[60]

Country:U

SA

Participants:n

=90

Meanage:

41.8

years

Cancertype:

prem

enopausal

breast

Timesincediagnosis:

notreported,

interventionoccurs

duringchem

otherapy

G1:

attentioncontrol(calcium-rich

diet)

G2:

calcium-richdietandexercise

(telephone

counselin

gcontacts

(14),exerciseequipm

ent,heart

ratemonito

r,workbook,video-

tape)

G3:

calcium-richdietandexercise

andhigh

fruitand

vegetable,low-

Studydesign:3

arm

RCT

Follow-up:

3,6months

Com

parisongroup:

attentioncontrol

Outcomemeasures:%

body

fat

(wholebody

DXAscans)a ,144-

item

diethistoryquestio

nnaire,

LongitudinalS

tudy

PhysicalAc-

tivity

Questionnaire

(MET-h/

week),accelerom

eter

(kcal/d

ay)

Primary:Consistentincreases

forall

measuresof

adiposity

were

observed

overtim

eandam

ongall

groups.G

3hadsignificantly

lower

scores

for%

ofbody

fat

(minus

thetrunk)

(P<0.05)

Other

behaviors:no

significant

changesin

physicalactiv

ityover

timeor

betweenstudyarms

91.2

%(n=82)

J Cancer Surviv (2015) 9:305–338 313

Tab

le1

(contin

ued)

Study

Participants;m

ean

age;cancer

type;tim

esincediagnosis

Interventio

n(type,intensity,

duratio

n)Studydesign

and

evaluatio

nOutcomes

Results

Retentio

n

fatd

iet(resourcesprovided

toG2

+encouraged

tomaintainhigh

fruitand

vegetable,low-fatdietto

reduce

energy

density

ofthediet).

Goallevelsof

<20

%of

energy

from

fat,and>5servings

fruitand

vegetables

perday

Type:d

iet(high

fruitand

vegetable,

lowfat)andexercise

(aerobicand

strength

training)(aerobic

exercise

>3tim

es/week,

strength

training

everyotherday)

Intensity:m

ultip

lecontacts(atleast

fortnightly

)Duration:

14telephonecounselin

gcontacts(10–30

min)over

6months

There

wereno

differencesin

energy

intake

amongstudyarms.

How

ever,G

3exhibitedhigher

fruitand

vegetableintakes(by1.7

serves)andlower

fatintakes

(reductio

nof

5.2%

calories

from

fat)at6months

Cam

pbelletal.[61];

Reedy

etal.[111];

Koetal.[112]

Country:U

SA

Participants:N

=922

(n=266colorectal

cancer

survivors)

Meanage:66.5

years

Cancertype:

colorectalcancer

Timesincediagnosis:

7.6%:lessthan

1year

ago;

29%:

1–2yearsago;

57%:2

–5years

ago

G1:

generichealth

education(2

mailin

gs)andtailo

red-print

newsletters(4)afterfollo

w-up

completed

G2:

4tailo

red-printn

ewsletters

G3:

4telephonecalls

(20min

duratio

n)G4:

4tailo

red-printn

ewsletters+4

telephonecalls

(20min

duratio

n)Type:d

iet,physicalactiv

ity,

colorectalcancer

screening

Intensity:m

ultip

leexposure

(less

than

monthly)

Duration:

1year

Studydesign:4

arm

RCT.

2×2

design—

stratifiedby

colorectalcancer

and

noncolorectal

cancer

status

Follow-up:

6,12

months

Com

parisongroup:

usualcarewith

tailo

red

newslettersat

study

completion

Outcomemeasures:modifiedBlock

Food

FrequencyQuestionnaire

(36item)a,2-item

fruitand

vege-

tablescreeningquestio

nsa ,modi-

fied

7dayPA

recall(m

oderateto

vigorous

PAscore)

Primary:There

wereno

significant

changesin

fruitand

vegetable

consum

ptionin

colorectalcancer

survivors,usingtheFo

odFrequencyQuestionnaire.T

here

was

anonsignificant

increase

inG2interventio

nby

ameanof

1.0

serves/day

Using

the2-item

screeningques-

tions,all3interventio

ngroups

show

edstatistically

significant

increasesam

ongcolorectalcancer

survivors

Other

behaviors:Nosignificant

change

onphysicalactiv

ity,and

participantsin

all4

groups

were

less

activ

eatfollo

w-upcompared

tobaselin

e

79.7

%(n=735)

from

total

sample

Von

Gruenigen

etal.

[65,73]

Country:U

SA

Participants:N

=45

Meanage:55

years

Cancertype:

endometrialcancer

Timesincediagnosis:

20.6

months

median

(intervention)

G1:

standard

care

G2:

face-to-face

groupsessions,

telephone,or

writtennewsletters

Type:w

eightloss,PA

,eating

behaviors

Intensity:w

eeklycontact

Duration:

totalo

f21

sessions

offace-to-face

(11),telephone

(5),

Studydesign:2

arm

RCT

Follow-up:

3,6,

12months

Com

parisongroup:

usualcare

Outcomemeasures:weightchange

(kilo

gram

s)—measureda,PA

usingLeisure

ScoreIndexof

the

Godin

leisuretim

eexercise

questio

nnaire

(frequency

per

weekon

Leisure

Score

Indexfor

mild

,moderate,strenuousPA

),3dayfood

record

(vitamin

Cand

Primary:The

meandifference

inweightchangebetweenthetwo

groups

was

−4.9

kg(P=0.018)

at12

months.The

controlgroup

did

notd

emonstrateanysignificant

changesin

weightfrom

baselin

e.Meanweightchangeexpressedas

apercentage

from

baselin

eto

Interventio

n,78

%(n=18);

control,

90%

(n=20)

314 J Cancer Surviv (2015) 9:305–338

Tab

le1

(contin

ued)

Study

Participants;m

ean

age;cancer

type;tim

esincediagnosis

Interventio

n(type,intensity,

duratio

n)Studydesign

and

evaluatio

nOutcomes

Results

Retentio

n

writtennewsletters(5)over

6months

folateas

markerof

fruitand

vegetableintake,kilo

calories)

12monthswas

−3.1

%in

the

interventio

ncomparedto

1.0%

inthecontrolg

roup

(mean

difference

−4.1

%,P

=0.020)

Other

behaviors:At1

2months,

therewas

asignificantd

ifference

inLeisure

Score

Indexbetween

groups

(meangroupdifference

17.8,P

=0.002)

Therewereno

significantchanges

indiet.T

heinterventio

ngrouphada

lower

energy

intake

(kilo

calories)

butw

asnotstatistically

signifi-

cant

from

thecontrolg

roup

Von

Gruenigen

etal.

[67]

Country:U

SA

Participants:N

=75

Meanage:58.0

years

Cancertype:

endometrialcancer

Timesincediagnosis:

20.7

months

G1:

standard

care

G2:

face-to-face

groupsessions,in-

dividualphysiciancounselin

g,newsletters,telephone,and

email

contactw

ithregistered

dietician.

Receivedpedometer,heartrate

monito

r,hand

andankleweights

Type:w

eightloss,PA

,resistance

exercises,dietquality

Intensity:w

eekly(10),then

biweekly(6)groupsessions.

Physiciancounselin

gat3,6,and

12months

Duration:

minim

umof

19contacts

for12

months

Studydesign:2

arm

RCT

Follow-up:

3,6,

12months

Com

parisongroup:

standard

care

received

one

inform

ation

brochure

Outcomemeasures:measured

weightaandheight,w

aist

circum

ference,hipcircum

ference,

Godin

leisuretim

eexercise

questio

nnaire,2

×24

hdietary

recalls,pedom

eter

step

counts

Primary:significantd

ifferences

for

weightchangefrom

baselin

eto3,

6,and12

months(P<0.001).

Mean(95%

CI)difference

betweengroups

at6monthswas

−4.4

kg[−5.3,−3

.5],P<0.001

andat12

monthswas

−4.6

kg[−5.8,−3

.5],P<0.001.Mean

percentw

eightchangein

the

interventio

nwas

−4.1

%as

comparedto−0

.8%

incontrolsat

6monthsand−3

.0%

and+

1.4%

at12

months

Other

behaviors:mean(95%

CI)

difference

inchange

at6months

was

100min

perweek[6,94],

P=0.038andat12

monthswas

89min

perweek[14,163],

P=0.020.Meanchange

inpedometer

step

countsfrom

baselin

eto

6monthswas

2353

intheinterventio

ngroupversus

−9.4stepsperdayin

theusual

care

group(differenceof

[95%

CI]of

2362

(494,4230);

P=0.015)

Meandifference

inchange

intotal

fruitand

vegetableintake

was

0.91

servings

perday(P<0.001)

78.7

%(n=59).

Interven-

tion,

85.4

%(n=35);

control,

70.6

%(n=24)

J Cancer Surviv (2015) 9:305–338 315

Tab

le1

(contin

ued)

Study

Participants;m

ean

age;cancer

type;tim

esincediagnosis

Interventio

n(type,intensity,

duratio

n)Studydesign

and

evaluatio

nOutcomes

Results

Retentio

n

at6monthsand0.92

(P<0.001)

at12

months.Meandifference

inchange

inkilocalories

between

groups

was

−228.8,−

217.8,and

−187.2

kcalat3,6,and

12months(P<0.001)

Dem

ark-Wahnefried

etal.[62];Dem

ark-

Wahnefriedetal.[63];

Mosheretal.[78];

Wilk

insonetal.[113];

Christy

etal.[74].—

FRESHST

ART

Country:U

SA

Participants:N

=543

Meanage:57

years

Cancertype:b

reast,

prostate

Timesincediagnosis:

3.83

months(SD

2.74)

G1:

attentioncontrol

G2:

tailo

red-printn

ewslettersand

workbook

Type:d

ietand

physicalactiv

ityIntensity:initialw

orkbookand(6)

tailo

rednewslettersevery7–

9weeks

for10

months

Duration:

totalo

f7contactsfor

10months

Studydesign:2

arm

RCT

Follow-up:

1year,

2years

Com

parisongroup:

attentioncontrol

Outcomemeasures:numberof

goal

behaviorspracticed

(percentage

adoptinggoalbehavior

inatleast

2areas)a ,7dayPA

recall,

diet

historyquestio

nnaire,D

iet

QualityIndexmeanscore

Primary:both

armssignificantly

improved

theirlifestylebehaviors

(P<0.05).Significant

difference

betweengroups

inpracticeof

2or

moregoalbehaviors(P<0.0001)

(16%

greaterin

interventio

nparticipants)

Other

behaviors:significant

differencesbetweengroups

inexercise

minutes

perweek

(P=0.02)(+20

min/week

interventio

n),fruitandvegetables

perday(P=0.01)(+0.5servings

interventio

n),totalfat(P<0.0001)

(−2.3%

interventio

n),saturated

fat(P<0.0001)(−1.0%)

Interventio

n,93.4

%(n=253);

control,

97.8

%(n=266)

Djuricetal.[64]

Country:U

SA

Participants:N

=40

Meanage:52

years

Cancertype:b

reast

Timesincediagnosis:

notreported

although

either

scheduledforor

startin

gchem

otherapy

inthenext

2weeks

G1:

controlg

roup

received

written

dietandphysicalactiv

itymaterialsandpedometer(sam

eas

G2),and

bimonthly

study

newsletters

G2:writtendietandphysicalactiv

itymaterials,pedom

eter,telephone

counselin

g(byadieticiantrained

inmotivationalinterview

ing)

Type:h

ighfruitand

vegetable,low-

fatd

iet,weightcontrol,30min

perdayof

moderate-to-vigorous

PAIntensity:m

ultip

lecontacts(atleast

monthly)

Duration:

totalo

f19

calls,w

ritten

materials,and

pedometer,over

12months

Studydesign:2

arm

RCT

Follow-up:

6,12

months

Com

parisongroup:

attentioncontrol

with

written

materialsand

pedometer

(sam

eas

the

interventio

ngroup)

Outcomemeasures:measured

weightand

body

fata,19-item

fruitand

vegetablescreener,17-

item

percentage

ofenergy

from

fat,24

hdietrecall,

Wom

en’s

Health

Initiativevalid

ated

PAquestio

nnaire

Primary:thepercentage

ofbody

fat

increasedby

1.2%

inthecontrol

groupanddecreasedby

0.07

%in

theinterventio

ngroup.Weight

decreasedby

0.8kg

at12

months

Other

behaviors:To

talp

hysical

activ

ityincreasedto

ameanof

364min

perweekandmoderate/

vigorous

activ

ityincreasedto

ameanof

315min

perweekat

12months,slightly

belowthe

targetof

350min

perweekof

moderate/vigorous

activ

ity.F

orfruitand

vegetableintakesfrom

unannouncedrecalls,the

number

ofservings/day

increasedonly

inthetelephonearm,and

themean

reported

intake

at12

monthswas

justabovetheminim

uminter-

ventiongoalof7servings

perd

ay,

notcountingpotatoes.T

here

was

Interventio

n.65

%(n=13);

control.

85%

(n=17)

316 J Cancer Surviv (2015) 9:305–338

Tab

le1

(contin

ued)

Study

Participants;m

ean

age;cancer

type;tim

esincediagnosis

Interventio

n(type,intensity,

duratio

n)Studydesign

and

evaluatio

nOutcomes

Results

Retentio

n

asignificantincreasein

fruitand

vegetableservings

by3.1serves

from

baselin

eto

12months

Djuricetal.[66]

Country:U

SA

Participants:N

=48

Meanage:mean:

51.7

years

Cancertype:b

reast

Timesincediagnosis:

notreported—

although

needed

tohave

been

diagnosedwith

inthepast4years

G1:

standard

care

G2:

weightw

atchers(freecoupons

toattend

each

week)

(52)

G3:

telephonecounselin

gby

dietician(24calls),andmailed

writtenmaterial(12)

G4:

weightw

atchersfree

coupons

(52),dietician-deliv

ered

tele-

phonecounselin

g(24),m

ailed

writtenmaterial(12)

Type:w

eightlossgoal(10%

baselin

eweighto

ver6months)

bydecreasing

energy

andfat

intake,and

30–45min

moderate

activ

itymostd

aysof

theweek

Intensity:m

ultip

lecontacts

(minim

ummonthly)

Duration:To

talcontactsvaried

from

36(G

3),52(G

2),88(G

4),in

12month

interventio

n

Studydesign:4

arm

RCT

Follow-up:

3,6,

12months

Com

parisongroup:

usualcare

Outcomemeasures:weighta,3

day

food

record,physicalactivity

logs

(self-reported

intentional

exercise)

Primary:significantd

ifferencein

weightlossat12

monthsfor

participantsin

G3(m

ean8kg

loss)andG4(m

ean9.4kg

loss).

Therewas

anonsignificantlossof

2.5kg

inG2,andan

increase

of0.85

kgin

G1(control

group)

Other

behaviors:nonsignificant

decreasesin

energy

intake

(kilo

calories

perday)

ineach

ofthethreeinterventio

ngroups

(by

447–616kcalperday),and

nonsignificant

decrease

infat

intake

(%of

energy

from

fat)in

the3interventio

ngroups

(by2–

11%)at12

months.The

control

groupenergy

intake

remainedthe

same(decreaseof

126kcalper

day),and

increasedfatintake(by

7%)

There

was

nodifference

inweight

loss

betweenwom

enwho

self-

reported

intentionalexercise(be-

yond

daily

activ

ities),andthose

who

reported

nointentionalac-

tivities

ineach

studygroup

81.3

%(n=39)

Ggroup

aDenotes

prim

aryoutcom

e

J Cancer Surviv (2015) 9:305–338 317

[57, 58, 67]. There was no change to the impact of interven-tions on total PA (SMD=0.34 [0.24, 0.44]). However, therewas an increase in heterogeneity (χ2=30.31, df=16 [P=0.02];I2=47 %).

Diet-only trial

One trial reported dietary outcomes only [59]. The trialtargeted men receiving active surveillance for prostate cancer,with a dietary counseling intervention delivered by telephoneover 6 months. Men completed self-report measures andprovided blood samples for objective assessment of caroten-oid intake at completion of the intervention. Retention ratewas 97 %, with significant increases to vegetable consump-tion. There was no change to fruit, whole grains, beans, or fatconsumption [59].

Multiple behavior trials

Seven studies focused on multiple behaviors (PA and diet)[60, 61, 63–66].

Participants

A total of 1107 participants were randomized (range 40–543).Three studies recruited newly diagnosed patients [60, 62–64],and patients were diagnosed with breast [60, 62–64, 66],prostate [62, 63], colorectal [61], or endometrial [65, 67, 73]cancer. Only three trials reported time since diagnosis, whichvaried from a mean of 3.8 months [63], to a mean of20.6 months [65, 67]. In two trials, participants were sched-uled to start chemotherapy [60, 64]. Three trials were aimedexclusively at overweight or obese breast [66] or endometrialcancer survivors [65, 67], with the aim of achieving weightloss through changing PA and diet behaviors. Two trialsreported PA [66] or PA and diet [64] outcomes for the purposeof intervention adherence.

Three trials had a usual care control group [65–67], and oneused a usual care comparison group with tailored newslettersat study completion [61]. Two had attention control groups[60, 63], and one used attention control with the same writtenmaterials and pedometer as the intervention group [64].

Intervention characteristics

All of the interventions were home-based and did not includeany supervised PA. All trials targeted both PA and diet behav-iors. Four of these targeted weight loss [60, 64–67], or pre-vention of weight gain [60, 64–66], through changing PA anddiet behaviors [60, 64–66]. Four trials were aimed at increas-ing fruit and vegetables and reducing fat [60, 61, 63, 65, 67],one targeted only energy and fat [66], two included bothaerobic and strength activity [60, 67], two targeted moderate

or moderate-to-vigorous PA [64, 66], and the remaining threetargeted PA [61, 63, 65].

Interventions were delivered using telephone counseling[60, 61, 64–67], written materials [60–62, 64, 65, 67], or faceto face [65–67]. Most trials used multiple delivery modes,with one that used tailored newsletters [63]. Interventionduration was 6 months [60, 65], 9 months [61], 10 months[62], or 12 months [64, 66, 67]. The average number ofcontacts was 27 (ranging from 4 [61] to 88 [66]). Studiesreported high adherence with all intervention componentsranging from 73 to 100 %, with no difference between deliv-ery modes.

Outcome assessment

Objectively assessed weight or body fat was reported as theprimary outcome in five studies [60, 64–67]. All trialsassessed diet using a range of self-reported measures [60,61, 63–67]. Only one trial used an objective measure of PA(accelerometer) [60], and the remaining trials assessed PA byself-report only [61, 62, 64–67].

Follow-up periods were 6 months in one study [60],12 months [61, 64–67], or 2 years [62]. However, only twostudies reported follow-up beyond postintervention time point[63, 65]. At 12 months, one trial reported significant differ-ences in PA levels, with no difference in diet [65]. After 2 yearsof follow-up, both study groups hadmaintained increased fruitand vegetable consumption, decreased saturated fat, and im-proved overall diet quality [74]. The mean retention rate was84 % (range 75 % [64] to 96 %[63]). No adverse events werereported in two studies [60, 61, 63, 64]. In two studies thatreported adverse events, 10–13% (n=4; n=74) [63, 64] of thetotal sample reported serious adverse events that led towithdrawal.

Five of the seven studies [60, 61, 63, 64, 67] reportedsignificant improvements in one or more aspects of diet qual-ity, as assessed by self-report, over the medium to long term(6 months to 2 years). The remaining two studies reportednonsignificant decreases in energy [65, 66] and fat intake inthe intervention groups [66]. Inconsistent improvements infruit and vegetable consumption were reported using a two-item screening question; however, these improvements werenot found when using the comprehensive Food FrequencyQuestionnaire results [61]. At 6 months, significant improve-ments were reported for vegetables [60], fruit [60], combinedfruit and vegetables [67], and decreased fat [60]; however,there was no change for energy [60]. At 12months, significantimprovements were reported for fruit and vegetables, by amean of 0.5 [63], 0.9 [67] to 3.1 serves per day [64]. Therewas also a significant reduction by 2.3 % in the proportion ofenergy from fat [63], which is supported by similar reductionsin other trials [64–66]. There were two trials that reported noeffect on PA at 3, 6, or 12 months [60, 61]. At 12 months, four

318 J Cancer Surviv (2015) 9:305–338

trials reported improvements in PA, ranging from 20 min perweek (P=0.02) [63] to 89 min per week [67] to 117 min perweek [64], and a difference of 17.8 (P=0.002) on the LeisureScore Index [65].

Theoretical framework

Recent development of behavior change taxonomies [28, 75,76] has encouraged consistent reporting of behavior changetechniques. We have mapped the SCT constructs to identifythe behavior change techniques that align with each construct,using the taxonomy designed to change PA and eating behav-iors [75] in Table 3.

Self-efficacy was the most commonly assessed construct[48–50, 52, 55–57], with four PA studies that assessed otherSCTconstructs [48, 51, 53, 55] (Table 4). Two studies [52, 57]reported that the study was based on Bandura’s self-efficacytheory, and the only construct operationalized was self-effica-cy. Four PA interventions that used telephone or face-to-facecounseling reported that the counseling principles were basedon SCT [49, 51, 53, 58]. Five studies reported that the studywas based on SCT and the transtheoretical model (TTM), orelements of TTM, such as stages of change [50, 53, 57, 61,

62]. Stage of change was assessed in four studies [50, 51, 53,57] with reference to both the TTM and social cognitivetheories. However, one trial assessed stage of change, despitenot providing any reference to TTM or rationale for why stageof change was assessed [51]. Most studies reported using goalsetting [48, 51–54, 56–58, 77]; however, few specificallyreported action plans [54] or review of goals [56–58]. Themost common strategy to increase self-efficacy was to providea pedometer [49, 50, 52, 53, 56, 57] and/or a log sheet for self-monitoring of PA behavior [49–53]. Social support or socialcomparison was the most common outcome expectancytargeted [48, 51, 54, 56, 58], and two trials reported strategiestargeting environment [51, 54] or relapse prevention [49, 53].Five trials incorporated identification and discussion of bar-riers and how to overcome them [50, 51, 53, 57, 58], but onlyone prompted a focus on past successful strategies [49].

In PA-only trials, improvements in self-efficacy were asso-ciated with increased PA in three studies [49, 50, 52]. Moder-ation analyses identified that intervention participants withhigh self-efficacy increased their PA levels faster over the6 month assessment period compared to intervention partici-pants with low self-efficacy [57]. Mediation analyses identi-fied that improvements in barrier interference and barrier self-

Table 2 Risk of bias (assessed using McMaster Quality Assessment Tool) [47]

Study (a)Selectionbias

(b)Studydesign

(c)Confounders

(d)Blinding

(e) Data collectionmethod

(f)Withdrawalsand dropouts

Globalrating

PA-only studies

Short et al. [54, 55] Weak Strong Strong Moderate Strong Strong Moderate

Valle et al. [56] Weak Strong Strong Moderate Strong Moderate Moderate

Rogers et al. [51, 68] Moderate Strong Moderate Moderate Strong Strong Strong

Pinto et al. [50, 71, 72] Moderate Strong Strong Moderate Strong Strong Strong

Bennett et al. [57] Weak Strong Strong Weak Strong Moderate Weak

Matthews et al. [58] Weak Strong Strong Weak Moderate Weak Weak

Ligibel et al. [49] Moderate Strong Strong Moderate Strong Moderate Strong

Wang et al. [52] Moderate Strong Strong Moderate Strong Strong Strong

Pinto et al. [53] Weak Strong Strong Moderate Strong Strong Moderate

Hatchett et al. [48] Moderate Strong Strong Moderate Strong Strong Strong

Diet only

Parsons et al. [59] Moderate Strong Weak Moderate Strong Strong Moderate

Multiple behavior studies

Demark-Wahnefried et al.—STRENGTH [60]

Strong Strong Strong Moderate Strong Strong Strong

Campbell et al. [61] Weak Strong Strong Moderate Strong Strong Moderate

Von Gruenigen et al. [65, 73] Weak Strong Strong Moderate Strong Moderate Moderate

Von Gruenigen et al. [67] Weak Strong Strong Weak Strong Moderate Weak

Demark-Wahnefried et al.—FRESH START[62, 63, 78]

Weak Strong Strong Moderate Strong Strong Moderate

Djuric et al. [64] Weak Strong Strong Moderate Strong Moderate Moderate

Djuric et al. [66] Weak Strong Weak Moderate Strong Strong Weak

J Cancer Surviv (2015) 9:305–338 319

efficacymediated 39 and 19% of the intervention effect on PAmaintenance 3 months after the intervention [70]. There wereno significant changes in decisional balance pros, cons, orexperiential processes of change [71, 72], or task self-efficacy,social support, outcome expectations, or fear of exercise [70].Two trials assessed but did not report results for self-regula-tion, outcome expectancy values, exercise self-efficacy, exer-cise role identity, behavioral capability, or social support [48,

54, 55]. Intervention effects on stage of change results weremixed, with one trial that reported a medium-to-large effect[51], one reported significant postintervention improvementsthat declined over subsequent follow-ups [53], and one re-ported no effect on stage of change [71].

The diet-only trial reported that the telephone counselingprotocol “used strategies adopted from SCT”; however, nofurther detail was provided [59]. In multiple behavior studies,

Fig. 2 Meta-analysis examiningthe effects of SCT-based inter-ventions on physical activity im-mediately postintervention

Table 3 SCT constructs mappedto behavior change techniquesusing the CALO-RE taxonomy[75]

SCT construct Behavior change techniquenumber

Behavior change technique description

Knowledge 1 Provide information on consequences of behavior ingeneral

2 Provide information on consequences of behavior to theindividual

Self-efficacy 16 Prompt self-monitoring of behavior

17 Prompt self-monitoring of behavioral outcome

21 Provide instruction on how to perform the behavior

22 Model/demonstrate the behavior

26 Prompt practice

27 Use of follow-up prompts

Goals 5 Goal setting (behavior)

6 Goal setting (outcome)

7 Action planning

10 Prompt review of behavioral goals

11 Prompt review of outcome goals

Outcomeexpectations

16 Prompt self-monitoring of behavior

17 Prompt self-monitoring of behavioral outcome

23 Teach to use prompts/cues

24 Environmental restructuring

28 Facilitate social comparison

29 Plan social support/social change

31 Prompt anticipated regret

35 Relapse prevention/coping planning

Facilitators/impediments

8 Barrier identification/problem solving

18 Prompting focus on past success

29 Plan social support/social change

320 J Cancer Surviv (2015) 9:305–338

Tab

le4

Socialcognitive

theory

constructsoperationalized

Study

Theoreticalbasis

SCTconstructsoperationalized

How

constructswereoperationalized

Constructsmeasured(no.of

items)

Results

PA-onlytrials

Shortetal.

[54,55]

One

interventiongroup(G

2)received

computer-tailo

red

newslettersbasedon

SCT

G3(targeted-print)interventio

nreceived

aTheoryof

Planned

Behavior-basedbooklet(previ-

ouslyevaluated)

Knowledgeof

PAguidelines,

beneficialoutcom

esof

PA,action

planning,feedbackon

PAperformance,socialsupport,role

modeling,physicalenvironm

ent

G2:tailored-printnew

sletters(n=3)

tai-

loredusinginform

ationfrom

individ-

ualassessm

entsatbaseline,and“up-

datecards”

assessingPA

andgoal

setting

behavior

overthelastmonth.

New

sletter1strategies

wereadvice

formeetingthePA

guidelines

for

cancersurvivors,inform

ationabout

thebeneficialoutcomesofPA

,advice

onexercising

safely,and

actionplan-

ning.N

ewsletter2strategies

were

expertadvice

from

abehaviorchange

expert,feedback

onPA

performance,

atestimonial,advice

onenhancing

socialsupport,andactionplanning.

New

sletter3containedexpertadvice

from

anexercise

physiologist,feed-

back

onPA

performance,tipson

changing

thePA

environm

ent,infor-

mationon

gainingfurthersupport,

andactionplanning

Outcomeexpectations

(11items);

outcom

eexpectancies

(1item);task

self-efficacy(7

items);b

arrier

self-

efficacy

(17items);b

ehavioralca-

pability(6

items);socialsupport

(15items);p

erceived

built

environ-

ment(7items);self-regulatio

n(12

items);actionplanning

(4items)

Not

reported

Valleetal.[56]

SCTwith

focuson

strategies

toenhanceself-efficacy,

behavioralcapability,self-

monito

ring,and

socialsupport

Socialsupport,problem

solving,self-

monito

ring,m

aintaining

PA,goal

setting,personalized

feedback

FITNETinterventio

ngoalwas

tomeet

PArecommendatio

nforcancer

survivors(150

min

moderate

intensity

PA/week).B

ehavioral

capabilitywas

operationalized

throughlin

ksto

publicly

available

websitesrelatedtoPA

and/orcancer

survivorship,12weeklyFacebook

messageswith

expanded

behavioral

lessonson

PAtopics

andbehavioral

strategies;self-efficacy

was

opera-

tionalized

bypedometer

which

provides

feedback

ondaily

walk-

ing,websitewith

weeklygoalset-

tingandchartsprovidingfeedback

onperformance

relativ

eto

weekly

exercise

goal,previousweeks

and

overallinterventiongoal;self-

monito

ring

was

operationalized

with

apedometer

tomonito

rsteps,

websitewith

diaryto

record

walk-

ingstepsandPA

type,duration,and

intensity

;and

socialsupportw

asoperationalized

throughthe

Facebookgroupwith

moderated

Nonereported

J Cancer Surviv (2015) 9:305–338 321

Tab

le4

(contin

ued)

Study

Theoreticalbasis

SCTconstructsoperationalized

How

constructswereoperationalized

Constructsmeasured(no.of

items)

Results

discussion

prom

ptsto

encourage

support,lin

ks,and

weeklyre-

minders

Rogersetal.

[51,68–70]

SCTself-efficacy,em

otional

coping,reciprocaldeterminism,

perceivedbarriers,outcome

expectations,behavioral

capability,goalsetting,

environm

ent,observational

learning,and

self-control

Socialsupport,exercise

barriers,self-

efficacy,goalsetting,environm

ent,

self-m

onito

ring,barrier

self-

efficacy,taskself-efficacy,barrier

interference,outcomeexpectations,

value(outcomeim

portance),

enjoym

ent,fear

ofexercise,role

model,exercisepartner

Participantsattended

6discussion

groupsessions

with

aclinical

psychologistwho

encouraged

socialsupport,provided

breast

cancer

survivor

exercise

role

models,andcoveredthefollo

wing

topics:journaling,tim

emanagem

ent,stress

managem

ent,

dealingwith

exercise

barriers,and

behavior

modification.The

specific

SCTconstructsaddressedby

the

groupsessions

included

self-

efficacy,emotionalcoping,

reciprocaldeterm

inism,perceived

barriers,outcomeexpectations,

behavioralcapability,goalsetting,

environm

ent,observational

observationallearning,andself-

control.Participantsalso

attended

12individualsupervised

exercise

and3individual“face-to-face”

updatecounselingsessions

with

anexercise

specialistthattapered

toa

home-basedprogram

bytheendof

theinterventio

n.The

specificSC

Tconstructsaddressedby

theindi-

vidualsessions

included

self-effi-

cacy,outcomeexpectations,behav-

ioralcapability,perceived

barriers,

andgoalsetting

with

self-m

onito

r-ing.To

furtherenhanceself-m

oni-

toring,participantswereencour-

aged

to“convert”theminutes

spent

inPA

recorded

ontheirweeklyex-

ercise

logs

into

“miles”

(i.e.,

1min

=2miles),w

hich

were

graphedon

amap

PAstageof

change

(5items);b

arrier

self-efficacy(9

items);taskself-

efficacy

(4items);b

arrier

interfer-

ence

(21items);socialsupport(4

items),positive

expectations

(14

items);n

egativeoutcom

eexpecta-

tions

(3items);fearof

exercise

(1item);PA

enjoym

ent(1item);ex-

ercise

rolemodels(3

items);exer-

cise

partner(1

item)

Medium-to-largeeffectsize

increase

was

notedforstageof

change

(meandifference=0.95;9

5%

CI=

0.75–1.83;

d=0.71;P

=0.034).

Com

paredwith

usualcare,the

interventio

ngroupreported

lower

barriersinterference

(mean

difference=−7

.8;P

=0.04)and

greaterPA

enjoym

ent(mean

difference=0.7;

P=0.06).

Statistically

nonsignificant

small-

to-m

edium

positiv

eeffectsize

in-

creaseswerenotedforbarrierself-

efficacy,fam

ilysocialsupport,and

totalsocialsupport,w

hilepositive

outcom

eexpectations,negative

outcom

eexpectations,and

negative

outcom

evalues

demonstrated

small-to-m

edium

negativ

eeffect

size

changesfortheintervention

comparedto

theusualcaregroup.

Littleto

nochange

was

notedfor

task

self-efficacy,friend

socialsup-

port,importance

ofpositiv

eout-

comes,fearof

exercise,exercise

partner,androlemodels

Mediatio

n:Barriersinterference

mediated39

%(P=0.004)

ofthe

interventio

neffecton

PA3months

postinterventio

n.PA

enjoym

ent

was

nota

significantm

ediator.

Reducingbarriersto

PApartially

explainedinterventio

neffect

Pintoetal.[50,

71,72]

Interventio

nbasedon

transtheoreticalmodel(TTM)

andSC

T.Pintoetal.2005[50]

statethatinterventio

nisbased

onTTM

only

TTM:counselingtailo

redto

participant’s

stageof

readinessto

change,S

CT:self-efficacy,goals,

PAbarriers

EachPA

participantreceivedin-person

instructions

onhowto

exercise

ata

moderateintensity

level,howto

monito

rheartrate,andhowto

warm

upbefore

exercise

andcool

downafterexercise.T

heywere

givenhomelogs

tomonitorPA

Decisionalb

alance

pros

andcons

(16

items),exerciseself-efficacy(5

items),stage

ofmotivationalreadi-

ness

forPA

(4items)

Nosignificantchanges

indecisional

balancepros,decisionalb

alance

cons,orstageof

change.B

aseline

self-efficacywas

asignificantp

osi-

tivepredictorof

meanminutes

ofweeklyexercise

(B=79.46min;

P=0.004),m

eanpedometer

steps

322 J Cancer Surviv (2015) 9:305–338

Tab

le4

(contin

ued)

Study

Theoreticalbasis

SCTconstructsoperationalized

How

constructswereoperationalized

Constructsmeasured(no.of

items)

Results

participationandapedometer.E

ach

participantreceivedaweeklytele-

phonecallover

12weeks

from

re-

search

staffto

monitorPA

partici-

patio

n,identifyrelevant

health

problems,problem

solveanybar-

riersto

PA,and

reinforcepartici-

pantsfortheirefforts.Finally,a

feedback

lettersummarizingthe

participant’s

progress

(e.g.,number

ofPA

sessions,average

durationof

each

session,andtheparticipant’s

barriersto

PAandsuggestions

toovercomethem

)was

sent

tothe

patient

atweeks

2,4,8,and12.A

ttheweeklycalls,subjectsreported

onthePA

recorded

onhomelogs,

andthey

received

feedback

perweek(B=2636.9steps;

P=0.0006)

Bennettetal.

[57]

TTM

andperceivedself-efficacy

from

SCT

Self-efficacy,goals

Duringtheinitialcounselingsession,

theparticipantw

asencouraged

toidentifybarriersto

engaging

inregularexercise,and

thePA

counselorandtheparticipant

workedtogethertodevelopideasto

overcomebarriers.A

goalof

30min

ofmoderateintensity

plannedPA

onmostd

aysof

the

week,butsom

eparticipantsstarted

with

moremodestg

oals.E

ach

interventionparticipantreceiveda

pedometer

andwas

show

nhowto

useitas

amotivator

forwalking

exercise,but

participantswerenot

required

towalkifthey

preferred

anotherform

ofmoderateintensity

exercise.T

elephone

calls

were

plannedto

lastabout2

0min,and

theconversatio

nincluded

motivationalstrategiesdirected

atsolvingproblems,offering

encouragem

ent,andreform

ulating

goals,ifneeded

Self-efficacyforregularPA

(6items);

stageof

change

forexercise

(6items:baselineonly)

Self-efficacywas

tested

asamoderator

ofinterventioneffects.Individuals

with

high

self-efficacyin

theinter-

ventionincreasedPA

levelsfaster

over

6monthsthan

lowself-

efficacy

individualsin

theinter-

ventiongroup.In

thecontrolgroup,

self-efficacyhadno

impacton

PAlevels(B=121.35;P

<0.05)

Matthew

setal.

[58]

Structured

behavioralcounseling

grounded

inSC

T(using

semistructuredscript)

Goals,PAenjoym

ent,positiv

ereinforcem

ent,self-rew

ard,

personalmotivation,barriers,

problem

solving,socialsupport,

The

initialcounselingsession

emphasized

goalsetting

andPA

safety.S

ubsequentcounselingcalls

weredesigned

tomonitor

None

J Cancer Surviv (2015) 9:305–338 323

Tab

le4

(contin

ued)

Study

Theoreticalbasis

SCTconstructsoperationalized

How

constructswereoperationalized

Constructsmeasured(no.of

items)

Results

goalreview

,self-efficacy,self-

monito

ring

participantsafetyandenhance

adherencethroughstructured

behavioralcounselin

gthatwas

grounded

inSC

T.Asemistructured

scriptwas

used

bythecounselorsin

each

ofthecalls

toinitiate

discussion

with

participantsabout

theirexperience

inmeetin

g(ornot)

theirw

alking

goalsthatwereagreed

upon

attheprevious

intervention

contact.Taking

theircues

from

the

inform

ationprovided

bythe

participantsin

theseconversatio

ns,

thestaffthen

delivered

appropriate

interventio

nmessages.When

participantsmettheirgoals,

individualized

positive

reinforcem

entw

asprovided

inthe

form

ofadiscussion

ofenjoym

ent

associated

with

beingactiv

eand

relevantself-rew

ards.D

iscussionof

personalmotivations

thathelped

theindividualmeettheirwalking

goalswas

also

emphasized.In

contrast,iftheparticipantd

idnot

meettheirwalking

goals,the

conversatio

nnaturally

ledto

the

barriersparticipantsexperiencedin

theperiod,and

thecounselor

initiated

aconversatio

nabout

problem

solvingstrategies

that

might

help

overcomeanticipated

barriersin

thecomingweek(s).

Whenappropriate,participants

wereencouraged

toelicitsocial

supportfrom

theirfamily

and

friendsthatmight

help

them

meet

theirgoals(e.g.,awalking

partner,

helpwith

othertim

ecommitm

ents).

Callswereendedwith

arecapofthe

conversatio

n(bythecounselor)that

included

areview

oftheagreed

upon

goalforthenext

week(s),a

review

ofthebehavioralissues

that

werediscussedduring

thecall(e.g.,

positivereinforcem

entsor

barriers/

problemsolving),and

anindicatio

n

324 J Cancer Surviv (2015) 9:305–338

Tab

le4

(contin

ued)

Study

Theoreticalbasis

SCTconstructsoperationalized

How

constructswereoperationalized

Constructsmeasured(no.of

items)

Results

ofwhenthenext

counselin

gcall

would

occur

Ligibeletal.

[49]

SCTandclient-centered

counseling

Goalsettin

g,self-efficacy,self-

monito

ring

Initialcalls

focusedon

goalsetting

and

performanceassessmentsoastobuild

self-efficacyforexercise

behaviors,

whilelatercallsconcentratedupon

the

adequacy

ofplansforrelapsepre-

vention.Eachcallreview

edperfor-

mance

onthebehaviorspreviously

discussedandencouraged

thepartici-

panttokeep

usingself-regulatory

skillstoachievechange.T

hetele-

phonecalls

weresupplementedby

aParticipantWorkbook,which

includ-

edadditionalinformationregarding

theimportanceof

exercise

incancer

populations,guidelines

forexercise

safety,and

journalpages

totrack

weeklyexercise.Participantswere

provided

with

apedometer.Instruc-

tions

forusingthepedometerwere

included

intheParticipantWorkbook

andwerereview

edduringthefirst

counselingsession.Participantswere

askedtorecord

thenumberof

mi-

nutesof

exercise

they

performed

and

stepsthey

completed

each

dayin

journals,w

hich

werereview

edduring

thetelephonecounselingcalls

Self-efficacy(5

items)

Interventio

nparticipantsreported

trends

towardim

provem

entin

exercise

self-efficacy(0.1±1.2vs

−0.3(±

0.8)

(P=0.06),as

compared

with

controls

Wangetal.

[52]

Bandura’sself-efficacytheory

Self-efficacy

Discuss

program

with

wom

enand

maketheirow

nweeklywalking

goalforexercise;encourage

wom

ento

documentw

eekly

walking

logs

sothey

canseetheir

ownprogress

during

theprogram;

storytelling/rolemodelstoryin

booklet;theresearcher

will

make

weeklyphonecalls

tounderstand

wom

en’sfeelings,the

effects,and

thecountereffectsof

exercise,and

will

praise

wom

en’sperformance

andencouragewom

ento

keep

progressingintheprogramfortheir

personalgoals;self-m

onitoring

with

theheartratering

andpedom-

eter

during

exercise;introduce

the

walking

program

with

written

Exerciseself-efficacyscale(18items)

Subjectsin

theexercise

grouphad

significantly

betterexercise

self-

efficacy

than

thoseintheusualcare

groupover

theinterventionperiod.

Atb

aseline,theinterventio

ngroup

was+13.5pointshigher,and

attim

e4,thedifference

hadincreasedto

+31.3(P<0.001)

J Cancer Surviv (2015) 9:305–338 325

Tab

le4

(contin

ued)

Study

Theoreticalbasis

SCTconstructsoperationalized

How

constructswereoperationalized

Constructsmeasured(no.of

items)

Results

materialand

verbalexplanationby

theresearcher

includingwarm

up,

cool

down,andprogressivelyin-

creasing

intensity,frequency,and

durationover

time

Pintoetal.[53]

Transtheoreticalmodelandthe

SCT

Self-efficacy,outcom

eexpectations,

stim

ulus

control,reinforcem

ent

managem

ent,self-m

onito

ring,

goals,planning

Participantsreceived

in-personin-

structions

onhowto

exercise

ata

moderateintensity

level,howto

monito

rheartrate,andhowto

warm

upbefore

exercise

andcool

downafterexercise.T

heywere

givenhomelogs

tomonitorPA

participationandapedometer.E

ach

participantreceivedaweeklycall

over

12weeks

from

research

staff

tomonito

rPA

participation,identi-

fyrelevant

health

problems,prob-

lem

solveanybarriersto

PA,and

reinforceparticipantsfortheiref-

forts.Activity

counselin

gwas

basedon

thetranstheoreticalmodel

andthesocialcognitive

theory

and

tailo

redto

each

participant’s

moti-

vationalreadiness.T

hecounseling

focusedon

strengtheningself-

efficacy

forexercise,onsetting

re-

alistic

outcom

eexpectations,and

ontraining

participantsin

using

behavioralprocessesof

change

such

asstim

ulus

controland

rein-

forcem

entm

anagem

entand

inusingtechniques

such

asself-

monitoring

ofexercise

behavior,

setting

exercise

goals,andplanning

forexercise.A

fter

the12

week

program

was

completed,m

onthly

phonecalls

wereprovided

for

3monthsto

reinforceprogress,

identifylapses

from

PA,and

re-

coverfrom

anylapses

thatmay

have

occurred.F

inally,a

feedback

lettersummarizingparticipants’

progress

was

sent

atweeks

2,4,8,

and12

Stageof

motivationalreadinessforPA

(5items)

The

interventionproduced

strong

effectson

participants’m

otivational

readinessat3months(O

R=5.26,

95%

CI=

1.32–20.93;P

=0.018)

thatwereattenuated

at6months

(OR=3.81,95%

CI=

0.90–16.71;

P=0.070)

andweakenedfurtherat

12months(O

R=1.89,95%

CI=

0.52–6.86;

P=0.335)

Hatchettetal.

[48]

SCT

Self-efficacy,goalsetting,anticipated

resultof

exercise,tim

eThe

e-counseloroffered

adviceregarding

exercise

andPA

.The

researchersbe-

lievedthatifaparticipantwereasked

SCTvariables:self-regulation(20

items);o

utcomeexpectancy

values

(19items);exerciseself-efficacy

Not

reported

326 J Cancer Surviv (2015) 9:305–338

Tab

le4

(contin

ued)

Study

Theoreticalbasis

SCTconstructsoperationalized

How

constructswereoperationalized

Constructsmeasured(no.of

items)

Results

managem

ent,self-m

onito

ring,

barriers,relapse

preventio

ntoofferinformationregardingherb

e-havior

duringtheintervention,she

wouldbe

morelikelytoengage

inthe

desiredbehavior.T

hetopics

ofeach

emailareas

follows:week1:goal

setting,anticipated

resultof

exercise;

week2:goalsetting,timemanage-

ment,self-monitoring;week3:self-

monitoring,descriptionof

anexercis-

er,overcom

ingbarriers;w

eek4:self-

monitoring,barrierstoexercise;w

eek

5:self-monitoring,overcomingbar-

riers,describetheanticipated

out-

comes

ofexercise;w

eek7:goalset-

ting,self-monitoring,tim

emanage-

ment,relapseprevention;week9:

overcomingbarriers,goalsetting,

self-monitoring,tim

emanagem

ent,

relapseprevention;week11:proper-

tiesof

anexerciser,results

ofcancer

(14items);exerciseroleidentity(9

items)

Diet-only

trials

Parsonsetal.

[59]

Strategies

adoptedfrom

SCT

Not

described

The

principlestrategy

toprom

otedietary

change

intheinterventionarm

was

atelephonecounselingprotocolwith

individualized,directassistance

tailoredtoeach

participant.T

hetelephonecounselingprotocol

followed

astepwise,phased

approach

thatused

strategies

adoptedfrom

SCT.Motivationalinterview

ing

techniques

wereused

tohelp

participantsassumeandmaintain

responsibilityfortheirbehavioral

change.N

ootherdetails

reported

Not

reported

Multiplebehavior

trials

Dem

ark-

Wahnefried

etal.—

STRENGTH

[60]

SCT(key

conceptsof

prom

oting

self-efficacyandbehavioral

monito

ring)

Self-efficacy,behavioralmonito

ring

Writtenandverbalinstructionbased

onSC

T(key

conceptsof

prom

oting

self-efficacyandbehavioralmoni-

toring)(a

workbookandtelephone

counselin

g).N

ootherdetails

re-

ported

Confidence(self-efficacy)in

making

changesin

theirdietaryor

exercise

practices

(didnotspecify

numberof

items)

Not

reported

Cam

pbelletal.

[61,111,

112]

TTM andSC

TStages

ofchange,socialsupport,

barriersto

change,knowledge,role

models,self-efficacy

G2received

tailo

red-

printexpert

feedback

driven

bybaselin

edata.

G3received

motivational

interviewingtelephonecalls

that

encouraged

participantsto

over-

comeam

bivalenceandidentify

Self-efficacy—

eatin

gfruitand

vegetables,and

engaging

inPA

(2items).S

ocialsupportforhealthy

eatingandexercise

(4items).

Perceivedbarriersto

behavior

Mediatio

n:Nonemediateddietary

change.H

igherself-efficacywas

associated

with

greaterfruitand

vegetableconsum

ptionatboth

baselineandfollo

w-up,butincrease

inself-efficacydidnotp

redict

J Cancer Surviv (2015) 9:305–338 327

Tab

le4

(contin

ued)

Study

Theoreticalbasis

SCTconstructsoperationalized

How

constructswereoperationalized

Constructsmeasured(no.of

items)

Results

theirow

nstrategies

forchange.G

4received

both

thetailored-print

feedback

andmotivational

interviewingtelephonecalls

change

(6items),knowledgeof

recommendatio

ns(1

item)

greaterchange

infruitand

vegeta-

bleconsum

ption

There

wereno

interventioneffectsfor

colorectalcancer

survivors

Von Gruenigen

etal.[65,73]

SCT

Establishshort-term

goals,build

self-

efficacy,reinforcement,individual

progresstowardgoals,em

phasison

long-term

change,patient

feedback

The

protocol

follo

wed

astepwise,

phased

approach

usingstrategies

outlinedby

SCT,indicatingthatthe

optim

alinterventio

nforamajor

behavior

change

should

focuson

establishing

short-term

goals,and

enablingtheperson

tobuild

self-

efficacy.P

articipantswere

contactedby

theresearch

dietician

byphoneor

newslettereveryweek

thatthegroupdidnotm

eet.Ph

one

calls

werestructured

incontentand

included

reinforcem

entand

discus-

sion

regardingtheprevious

week’s

topic.Participantswerealso

given

feedback

onindividualprogress

to-

wardPA

andnutrition

goals.

New

slettertopics

included

thefol-

lowing:

holid

ayrecipes,reinforce-

mento

fnutrition

goals,waysto

increase

PAandstep

count,restau-

rant

menumakeovers,and

eatin

gon

therun

Self-efficacyusingtheWeight

EfficacyLife-Style(W

EL)ques-

tionnaire

(20items).S

elf-efficacy

specificto

eatin

gbehaviorsin

five

situationalfactors:n

egativeem

o-tio

ns,foodavailability,socialpres-

sure,physicald

iscomfort,and

pos-

itive

activ

ities

Significantd

ifferencein

“social

pressure”subscale(P=0.03).

Increase

inself-efficacyrelatedto

negativeem

otions

(P<0.01),food

availability(P=0.03),andphysical

discom

fort(P=0.01)in

those

wom

enwho

lostweightd

uringthe

year.A

t12months,self-efficacy

scores

remainedhigh

(6months

afterinterventio

nhadconcluded).

Morbidlyobesepatientshadsig-

nificantly

decreasedself-efficacy

whenfeelingphysicaldiscom

fort

anddecreasedtotalself-efficacy

score.Therewas

asignificanteffect

forself-efficacyrelatedto

social

pressure

andrestraintimproved.

Forself-efficacyrelatedto

negativ

eem

otions,there

was

ameanin-

crease

of8.9in

wom

enwho

lost

weightv

ersus0.6in

thosewhose

weightw

asstableor

who

had

gained

weight

Von Gruenigen

etal.[67]

SCTwith

afocuson

establishing

short-term

goals,enablingthe

person

tobuild

self-efficacy

The

interventionfollowed

astepwise,

phased

approach

with

afocuson

establishing

short-term

goals,en-

ablin

gtheperson

tobuild

self-effi-

cacy

Individualexpertphysiciancounseling,

individualgoalsetting,goal

reinforcem

entinnewsletters,social

supportand

eatinginsocialsituations,

planning

mealsandgrocery

shopping,how

toread

food

labels,

pedometersprovided

feedback

and

reinforcem

entofPA

goals.

Increm

entalgoals(for

months1–2,

months5–6),m

odelingof

resistance

exercise.T

heinterventionfocusedon

theadoptionof

lifelongchanges

ratherthan

caloric

restriction.

Educationandskilldevelopm

entto

increasePA

andPA

self-efficacywere

included

usingaguidepreviously

developedforbreastcancersurvivors.

Patientswereencouraged

toaddac-

tivitiesthatthey

enjoyedortobegina

walking

programor

otherexercise

Not

reported

328 J Cancer Surviv (2015) 9:305–338

Tab

le4

(contin

ued)

Study

Theoreticalbasis

SCTconstructsoperationalized

How

constructswereoperationalized

Constructsmeasured(no.of

items)

Results

activity.L

ong-term

changesinevery-

dayactivities

(for

exam

ple,climbing

stairsinsteadof

taking

elevators)and

moderateaerobicactivity

wereem

-phasized.Participantsweregivenpe-

dometerstoprovideimmediatefeed-

back

andreinforcem

enttopatients

andtoprovideobjectiveassessment

ofPA

.Patientsweregiven3lbhand

andadjustableankleweightsand

instructed

intheproperform

and

procedureforperformingresistance

exercises.Heartratemonitorswere

provided

tofacilitatemonitoringof

targetheartrategoals.Physician

counselingvisits(conducted

bythe

PI)at3,6,and

12monthsfocusedon

nutritionandPA

goalsforSU

CCEED

participantsinordertoaugm

entthe

groupsessions

andprovideindividu-

alized

attention

Dem

ark-

Wahnefried

etal.—

FRESH

START[62,

63,78]

SCT:cuesto

action,self-efficacy,

skill

developm

ent,goals,goal

reinforcem

ent.Messageswere

custom

ized

tostages

ofchange

(TTM)

Benchmarkbehavior,goal,behavior

logs,behavioralcues,tailo

redto

stages

ofchange,goal,testim

onial,

overcomingbarriers,benefits,

progress

togoal

The

FRESH

STARTinterventionwas

basedon

theSC

Tthatem

phasizes

confidence

buildingandskills

developm

ent;thetranstheoretical

modelalso

was

used

tofram

emessageson

participants’stageof

readinesstomotivatebehavior

change.Participantsareencouraged

tosetsmallincrementalgoals,w

hich,

whenachieved,arereinforced

tobuild

self-efficacy.To

build

upon

self-

efficacyincrem

entally,participants

areassisted

inmakingchangesinone

behavioraldomainatatim

e.Partici-

pantsarefirstassigned

thebehavior

with

thehighestself-efficacyscore,

andbehaviorswith

lowerscores

are

presentedsubsequently(with

thepre-

misethatafterthe

participantachieves

successfulbehaviorchange

inthefirst

area,heor

shecangeneralizethis

successtothenexthealthdomain).In

situations

whereself-efficacyscores

areequalfor

thetwobehaviors,the

mostadvancedstageofreadinesswill

dictatethefirstdomaintargeted.For

Self-efficacy(PAanddiet)(3

items),

stageof

readiness(range

3–12

items,dependingon

responses),

socialsupport(11

items),barriers

(37items)

The

interventionwas

notsignificantly

associated

with

self-efficacyfor

exercise;how

ever,therewas

apos-

itive

correlationobtained

between

self-efficacyforexercise

andtotal

minutes

perweekof

exercise

atfollow-up

Mediatio

n:Resultssupportthe

hypothesisthatchangesin

self-

efficacy

forfatrestrictionandeatin

gmorefruitand

vegetables

partially

mediatetheeffectsof

theinterven-

tionon

dietquality

(37.7%

vari-

ance,P

<0.001).F

urthermore,

change

inself-efficacyforfatre-

strictionpartially

mediatedthein-

terventio

n’seffectson

thepercent-

ageof

kilocalories

from

fat(30.1

%variance,P

<0.001),and

change

inself-efficacyforfruitand

vegetable

consum

ptionpartially

mediatedthe

interventio

n’seffectson

daily

serv-

ings

offruitand

vegetables

J Cancer Surviv (2015) 9:305–338 329

Tab

le4

(contin

ued)

Study

Theoreticalbasis

SCTconstructsoperationalized

How

constructswereoperationalized

Constructsmeasured(no.of

items)

Results

participantsreporting

3deficientbe-

havioralareas,theinitialintervention

materialstargetthebehavior

associat-

edwith

thehighestself-efficacyscore,

andthesecond

behavioralarea

isse-

lected

atrandom

.Intheinitialmail-

ing,participantsreceiveapersonal-

ized

workbookthatincludes

thefirst

unitmaterials,and

asecond

install-

mentofworkbookmaterialsarrives

midway

throughtheintervention.For

each

unit,thefirstpage

isafeedback

form

inwhich

theparticipant’sbe-

havior

iscomparedwith

goalbehav-

ior,andencouragem

entisprovided

toachievethegoal.E

achinstallmentof

theworkbookincludes

personalized

behavior

record

logs

thatcorrespond

tothecontentareas

tohelppartici-

pantstrack

behavior

(toprom

ote

change

andimproveself-reporting

accuracy).In

addition,each

install-

mentoftheworkbookincludes

items

thatserveas

behavioralcues

[i.e.,a

pany

theexercise

unit].N

ewsletters

are4pagesof

colorfulgraphics

and

textthatincludethefollowingcom-

ponents:(1)apersonalized

greeting

tailoredtostageof

readiness;(2)a

goalstatem

entthatreflectsengage-

mentingoalsetting

behavior;(3)

atestimonialtailoredon

age,race,and

cancercoping

style;(4)an

advice

columnthatprovides

guidance

for

overcomingbarriers—tailoredtoa

subject’s

reportedbarriers;(5)

a“Fun

Facts”

section—

untailored;(6)a

benefitssectionthatisuntailoredand

emphasizes

thebenefitsof

practicing

goalbehavior;(7)astatus

sectionthat

features

agraphdepictingthepartici-

pant’sprogressinrelationtogoaland

accompanyingtailoredmessages[i.e.,

achievem

entofgoal(praise),progress

towardgoal(praiseandencourage-

ment),no

progress(encouragement),

330 J Cancer Surviv (2015) 9:305–338

pedometerandTherabands®

accom-

Tab

le4

(contin

ued)

Study

Theoreticalbasis

SCTconstructsoperationalized

How

constructswereoperationalized

Constructsmeasured(no.of

items)

Results

ortheabsenceof

data(encourage-

menttosubm

itupdateddata)]

Djuricetal.

[64]

SCT—thetelephonecounseling

approach

blendedmotivational

interviewing(M

I)with

SCT

Goals,self-monito

ring,self-efficacy

The

telephonecounselingapproach

blendedMIwith

SCT.Theyalso

received

pedometers,adaily

food

and

exercise

log,andexam

plemenus

atindividuallyappropriatecalorie

levels.

The

counselingplan

was

forthe

dieticiantocontactsubjectsw

eeklyfor

thefirsttwocalls,biweeklyforthe

next5months,andmonthlyforthe

last6months,foratotalof19

calls.

The

self-monitoringlogs

were

review

edduringthecalls.T

hecounselingapproach

combinedprin-

ciples

ofSC

TandMI.Subjectswere

involved

inderivingtheirow

nshort-

term

goalsandevaluatingtheirprog-

resstowardgoals.To

build

self-effi-

cacy,any

positivechangeson

theself-

monitoringsheetswereidentifiedand

praised

Self-efficacy(6items),self-confidence

formaintaining

ahealthylifestyle

(6items)

Not

reported

Djuricetal.

[66]

SCT

Self-m

onitoring,goalsettin

g,self-

efficacy,considerationof

body

image,socialsupport,removing

roadblocks,positive

thinking,

dealingwith

high-risksituations

andslips,andcueelim

ination

G3(individualized

arm):Monthly

writteninform

ationwas

prepared

onvarious

weightlosstopics

(environm

entalcontrol,servingsize

control,exercise,m

otivation,goal

setting,holiday

eating,seasonal

foods)andeitherpresentedtothe

wom

enatthemonthlymeetingor

mailedtotheirhomes.Pedom

eters

wereprovided

forself-monitoringand

goalsetting.Itw

asrequestedthatex-

ercise

anddietarylogs

bekeptdaily,

andthesewerereview

edtogetherwith

each

subject.Contactswereby

phone

orinperson,and

food

andexercise

recordsweremailedtothedietician

beforethescheduledcontact.The

counselingsessionvariedinlength

dependingon

individualneeds.The

dieticianfirstverifiedwhetheror

not

theparticipantwas

meetingbehavior

change

goalssetintheprevious

week.

Ifnot,theproblem

was

delineated,

andthedieticianhelped

thesubject

devise

aplan

thatwouldbe

used

to

Nonereported

Not

reported

J Cancer Surviv (2015) 9:305–338 331

one did not report how SCT constructs were operationalized[60]. All other trials reported goal setting, self-monitoring,building self-efficacy (for PA and diet) [60–63, 67], or diet[65, 73], or for maintaining a healthy lifestyle [64], overcom-ing barriers, and social support [61–66, 73]. Goal setting [48,63–67] and review of goals [63–66] were commonly opera-tionalized. Self-monitoring was commonly operationalizedthrough providing a pedometer [63, 64, 66, 67] or log sheet[63, 64, 66]. Few trials reported how they operationalizedoutcome expectations, with only three that reported socialsupport [63, 66, 67], one that included environment [66],and one that reported relapse prevention [48]. Four studiesincluded identification of barriers and how to overcome them[48, 61, 63, 66]. Self-efficacy did not appear to be related toPA behavior change [60, 63]. Improvements to diet qualitywere partially mediated by changes in self-efficacy for fatrestriction and eating more fruit and vegetables [62, 63, 78].Only the social pressure subscale of self-efficacy was signif-icantly related to eating behaviors (P=0.03) [65, 73]. Twotrials reported that self-efficacy was not associated with dietchanges [60] or fruit and vegetable consumption [61]. Socialsupport for healthy eating, perceived barriers to behaviorchange, and knowledge of recommendations were assessed,but none mediated fruit and vegetable consumption [61].

Discussion

Overview of findings

The aims of this review were to synthesize the existing liter-ature relating to PA and diet interventions based on SCT thattarget cancer survivors and to identify successful strategies toassist cancer survivors in making positive PA and diet behav-ior change. This review supports the efficacy of SCT-basedinterventions in changing PA and diet behavior in cancersurvivors. Our effect size of 0.33 for PA interventions can bedefined as a small-to-medium effect [79] and is similar toother meta-analyses [5, 80, 81] that reported effect sizes of0.32–0.38 for PA interventions (including cancer survivorsboth during and after treatment). Our positive results for PAbehavior mirror the results reported in two recent reviewsexamining PA behavior change in breast cancer survivorsand also found that trials were mostly PA only, few includedobjective measures, and few reported postintervention main-tenance [82, 83]. While evidence has been building to supportthe effects of PA and diet behavior on health outcomes, thereremains a need to focus on behavior change trials to under-stand how to promote sustainable healthy behaviors.

Our findings that the majority of included trials reportedstatistically significant improvements to at least one aspect ofdiet quality and weight loss are supported by other reviewsT

able4

(contin

ued)

Study

Theoreticalbasis

SCTconstructsoperationalized

How

constructswereoperationalized

Constructsmeasured(no.of

items)

Results

circum

venttheproblem.T

hetech-

niques

taughtincluded

goalsetting,

menuplanning,self-efficacy,self-

monitoring,considerationof

body

image,socialsupport,socialeating,

removingroadblocks,positive

think-

ing,dealingwith

high-risksituations

andslips,andcueelimination

G4(com

prehensive

arm):Subjects

received

theindividualized

counselingdescribed

aboveandwere

askedtoattend

weeklyweight

watchersmeetings

usingfree

coupons

332 J Cancer Surviv (2015) 9:305–338

with the general adult population [80, 81]. Due to considerableheterogeneity in the dietary outcomes assessed, it was notpossible to conduct a meta-analysis, although self-reportedimprovements to diet quality were evident in six of eightstudies. The two trials that did not find improvements todietary quality had a primary aim of decreasing weight, ratherthan behavior change [65, 66]. More evidence is requiredfrom behavior change trials that have an emphasis on dietarychange rather than weight.

The limited number of trials and the heterogeneity ofincluded studies in this review prevented any formal subgroupanalyses in our review. The effect appeared strongest for PA-only interventions compared to multiple behavior interven-tions; however, this should be interpreted with caution as PA-only interventions included smaller sample sizes and shorterfollow-up periods. A review of single compared to multiplebehavior interventions in older adults also reported that PAeffects appeared strongest in single behavior change interven-tions. However, there were inadequate multiple health behav-ior change interventions to compare to [84]. Similar reviews[80, 81] examined intervention setting, duration, person de-livering the intervention, delivery mode, age of target group,and intervention effectiveness and found that only increasedcontact frequency was associated with increased PA and dietbehavior change [80].

Social cognitive theory

Few trials conducted mediation analyses or reported changes intheoretical constructs. In those that did report the impact ofinterventions on theoretical constructs, results were inconsis-tent. Self-efficacy was the only construct that appeared to beassociated with positive behavior change for both PA and diet[52, 57, 61–63, 65, 73]; however, mediation analyses in twotrials identified that theoretical constructs only partially medi-ated the intervention effects [62, 63, 70]. Other reviews exam-ining individual SCT constructs have concluded positive out-come expectations, and intentions are associated with behaviorchange [85, 86]. Self-efficacy and goal setting were commonlyoperationalized, but there was limited reporting of how otherconstructs were operationalized as part of the intervention.Recent reviews identified that self-efficacy, self-monitoring ofbehavior, prompting intention formation, planning, specificgoal setting and review, and feedback on performance wereassociated with increased effectiveness in PA and diet behaviorchange [80, 81, 87]. However, given the crossover betweentheoretical constructs and behavior change techniques, the pos-itive results for SCT-based trials in this review may be a resultof the individual behavior change techniques employed, such asself-regulatory behaviors, rather than the theoretical constructs.

Recent reviews have questioned the value of theory indeveloping and evaluating interventions, with two recent be-havior change reviews concluding that interventions based on

theory were no more effective than atheoretical interventions[80, 88] and another two reviews that supported the efficacy oftheory-based interventions [89, 90]. The conflicting resultsmay be due, at least in part, to the inadequate description ofhow theory is implemented and evaluated in interventions andalso due to the overlap with specific behavior change tech-niques, which have been associated with interventioneffectiveness.

There were differences in the risk of bias assessment. Themajority of PA-only studies received a strong global rating(n=5), and multiple behavior trials received a moderate globalrating (n=5). Trials received a weak global rating becausefewer than 60 % of potentially eligible participants agreed totake part before randomization, which is used as an estimate ofthe external validity of the study [47, 91], and because both theoutcome assessor and study participants were aware of theresearch question [47]. Unlike in clinical trials where partici-pants are unaware of their exposure status, behavior changetrials present significant problems with blinding and recruit-ment as participants are expected to actively engage with theintervention. Self-selection bias is a likely issue in behaviorchange trials.

Strengths and weaknesses of review methods

Although this is a comprehensive review of the publishedliterature, there are some limitations that should be noted.Search results were screened for eligibility by only one re-viewer, despite recommendations that this step is conductedby two independent reviewers [92]. This review comprehen-sively searched a number of databases; however, it made noattempt to search for non-English publications or unpublishedliterature. Potentially eligible study protocols were obtained;however, no attempt was made to contact trial authors toobtain unpublished results of these studies [93–95]. The re-view included a broad definition of cancer survivors, includ-ing those both during and after completion of active treatment.While this increases the breadth of evidence, it likely contrib-uted to the heterogeneity of the included studies.

There are a number of SCT-based health outcomes trials,including the WINS [96–98] and WHEL [99, 100] trials,Active for Life trial [101], and RENEW trial [102, 103], thatinitially met criteria for inclusion in the review [104]. How-ever, the study team agreed not to include these trials asbehavior changewas either not reported [101, 104] or reportedas a secondary outcome only [96, 99, 102]. Due to the hetero-geneity of these trials and the inconsistency in reportingbehavior change outcomes, these trials were not included inthe review. The definition of SCT-based intervention wasrelatively broad and not limited by how well SCT was de-scribed or operationalized in the intervention. Studies neededto explicitly state that the intervention was based on SCT,which may or may not have included other theories, and was

J Cancer Surviv (2015) 9:305–338 333

dependent on the author description of the trial. Recent pub-lications have detailed a checklist for evaluating the extent towhich an intervention is theory-based [105], whichwill enablegreater clarity in the role of theory in the development of anintervention. There was one trial where it was unclear if thestudy was based on a theoretical framework. One paper [50]had been screened and judged not eligible due to notreporting that the trial was based on SCT. However,subsequent eligible and included trial papers [71, 72]reported that the intervention was based on SCT, andtherefore, this trial was included in the review. It maybe that due to publication size restrictions, authors havelimited space to fully describe intervention development.Alternatively, theoretical frameworks may be appliedpost hoc to an intervention. With journals requiringadherence to CONSORT [91], this is likely to improvethe consistent and transparent reporting of RCTs.

Limitations of the included trials

Studies demonstrated moderate heterogeneity, although mostinterventions included breast cancer patients and were con-ducted with patients after completion of active treatment. Thislimits the generalizability of findings to males, to survivors ofother cancer types, and to patients undergoing active treat-ment. Most studies involved small sample sizes and only fourtrials reported a sample size greater than 100 [49, 54, 61, 63].Only three of the trials included in this review focused onpromoting resistance training [54, 60, 67], despite a recentreview and meta-analysis that concluded resistance traininghas benefits on body composition and muscle strength incancer patients during and after cancer treatment [106].Meta-analysis used objectively assessed data, where available[51, 60]; however, the majority of data was self-reported. PAand diet outcomes were predominantly based on self-reportdata, while weight was frequently measured or objectivelyassessed [59–62, 64–66]. Participants were not blinded tointervention aims in any trials, so there may be inherentdifferences between those participants recruited for PA-onlyinterventions, compared to participants interested in diet,weight control, or multiple behavior interventions.

Future research

This review supports the efficacy of SCT-based interventionsin changing PA and diet behavior in cancer survivors. Whileinterventions reported a theoretical basis, these constructswere often inadequately operationalized or reported and rarelymeasured or tested [89, 107, 108]. Despite a large body ofcross-sectional data [33–38] linking SCT constructs with dietand PA behavior, there remains a need to test whether changesin these constructs predict behavior change in interventions.

Comparison between health theories would also be a usefulgap to address.

Development of the taxonomy assessing the extent towhich interventions are theory-based and use of behaviorchange techniques [81, 105] will contribute evidence to helpresearchers understand the intervention components that areessential to behavior change. Whether these are related tospecific theoretical constructs, or to behavior change tech-niques, such as self-regulatory techniques, requires furtherresearch. Research assessing whether single or multiple healthbehavior interventions have the greatest benefit to improve PAand diet behaviors is required.

There is a large evidence base supporting the efficacy of PAinterventions, and these are predominantly based on breastcancer patients, using short-term, self-reported outcomes. Fu-ture studies need to consider how to translate this research intoongoing support and programs to assist cancer survivors toincrease and maintain PA levels. Further work should alsoinclude trials which focus on resistance training, as there arespecific guidelines for cancer survivors to undertake resis-tance training. The field of dietary interventions is much lessdeveloped, with interventions demonstrating that cancer sur-vivors are willing and able to make improvements to diet [13,63, 80, 99, 102]. As diet quality is comprised of a complex setof behaviors, there is a need to examine the co-occurrence ofchanges in different aspects of diet [109]. Future studieswould benefit from considering the impact of behavior changefrom a healthy lifestyle perspective, such as considering com-pliance with World Cancer Research Fund guidelines [14].

Implications

Despite the limitations of this review, it appears that SCT-based interventions demonstrate promise for improving thePA and diet behaviors of cancer survivors. Interventions usinga range of delivery modes all demonstrated significant PAimprovements, with a small-to-medium effect size, after arelatively short intervention period (12 weeks). Diet and mul-tiple behavior component interventions tended to have ahigher number of intervention contacts and greater interven-tion duration (6 months). However, the increased contact timedid not appear to be related to the magnitude of change. Hightrial retention across both single and multiple behavior changetrials may be related to the low burden of predominantlyunsupervised interventions or that cancer survivors are moti-vated to improve their PA and diet behaviors. Unfortunately,there was little evidence to guide researchers in helping cancersurvivors to maintain health behaviors after completion ofinterventions, and this has been noted previously [83, 110].Improved description and reporting of intervention contentand the way in which theory-based interventions use theoryto guide the trial and intervention components remain neces-sary to understand what factors are driving the results of

334 J Cancer Surviv (2015) 9:305–338

theory-based interventions. Given the recent mixed findingson the efficacy of theory-based interventions, a greater under-standing of how theory is operationalized is necessary tounderstand what factors contribute to the success of interven-tions. Further evidence on theory-based trials is required tounderstand the crossover between theory-based constructs andbehavior change techniques and understand the impact of eachon improving health behaviors. Research expanding the rig-orous implementation and reporting of behavior change tech-niques is likely to improve understanding of the workingmechanisms that underpin how and why an interventionworks or does not work.

Conclusions

SCT-based interventions appear effective in improving PAand diet behaviors. No specific intervention characteristicsor theoretical constructs were associated with effectiveness.Future SCT-based interventions should describe the extent towhich theoretical or behavior change constructs are imple-mented and evaluated, in order to identify the successfulcomponents of SCT-based interventions.

Conflict of interest The authors have no potential financial or personalconflicts of interest to disclose in relation to this work.

Appendix 1: Medline search

1. cancer survivor*.mp.2. Survivors/3. cancer*.mp.4. exp Cancer/5. 3 or 46. 2 and 57. cancer patient*.mp.8. patient*.mp.9. exp Patients/

10. 8 or 911. 5 and 1012. 1 or 6 or 7 or 1113. exp Nutrition Surveys/ or exp Nutrition Policy/ or exp

Nutrition Assessment/ or exp Nutrition Therapy/14. Diet, Cariogenic/ or Diet Surveys/ or Diet, Carbohy-

drate-Restricted/ or Diet Therapy/ or Diet, Atherogenic/or Diet/ or Diet, Sodium-Restricted/ or Diet, Gluten-Free/ or Diet, Reducing/ or Diet, Fat-Restricted/ or Ke-togenic Diet/ or Diet Fads/ or Diet, Macrobiotic/ or Diet,Protein-Restricted/ or Diet, Vegetarian/ or Diet,Mediterranean/ or Diabetic Diet/ or Diet Records/

15. Food Habits/ or exp Food/ or Health Food/ or FoodPreferences/

16. 13 or 14 or 1517. physical activity.mp. or Motor Activity/18. exercise therapy/ or muscle stretching exercises/ or re-

sistance training/ or strength training19. Exercise/20. weight.mp.21. aerobic.mp. or Physical Exertion/22. running/ or swimming/ or walking/23. 17 or 18 or 19 or 20 or 21 or 2224. 16 or 2325. Health Behavior/ or Self Efficacy/ or social cognitive

theory.mp. or Psychological Theory/ or Social Support/26. social cognitive.mp.27. Motivation/ or Health Education/ or Health Promotion/

or social learning theory.mp.28. 25 or 26 or 2729. 12 and 24 and 28

Open Access This article is distributed under the terms of the CreativeCommons Attribution License which permits any use, distribution, andreproduction in any medium, provided the original author(s) and thesource are credited.

References

1. Campo RA, Rowland JH, Irwin ML, Nathan PC, Gritz ER, KinneyAY. Cancer prevention after cancer: changing the paradigm—areport from the American Society of Preventive Oncology. CancerEpidemiol Biomarkers Prev. 2011;20:2317–24.

2. Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Globalcancer statistics. CA Cancer J Clin. 2011;61:69–90.

3. Carmack CL, Basen-Engquist K, Gritz ER. Survivors at higher riskfor adverse late outcomes due to psychosocial and behavioral riskfactors. Cancer Epidemiol Biomarkers Prev. 2011;20:2068–77.

4. Cramp F, Daniel J. Exercise for the management of cancer-relatedfatigue in adults (review). Cochrane Database of SystematicReviews. 2008;CD006145.

5. Speck RR, Courneya KS, Masse LC, Duval S, Schmitz KH. Anupdate of controlled physical activity trials in cancer survivors: asystematic review and meta-analysis. J Cancer Surviv. 2010;4:87–100.

6. Ferrer RA, Huedo-Medina TB, Johnson BT, Ryan S, Pescatello LS.Exercise interventions for cancer survivors: a meta-analysis ofquality of life outcomes. Ann Behav Med. 2011;41:32–47.

7. McNeely ML, Campbell KL, Rowe BH, Klassen TP, Mackey JR,Courneya KS. Effects of exercise on breast cancer patients andsurvivors: a systematic review and meta-analysis. Can Med AssocJ. 2006;175:34–41.

8. Schmitz KH, Courneya KS, Matthews C, Demark-Wahnefried W,Galvao DA, Pinto BM, et al. American College of Sports Medicineroundtable on exercise guidelines for cancer survivors. Med SciSport Exerc. 2010;42:1409–26.

9. Rock CL, Doyle C, Demark-Wahnefried W, Meyerhardt J,Courneya KS, Schwartz AL, et al. Nutrition and physical activity

J Cancer Surviv (2015) 9:305–338 335

guidelines for cancer survivors. CA Cancer J Clin. 2012;62:242–74.

10. Ballard-Barbash R, Friedenreich CM, Courneya KS, Siddiqi SM,McTiernan A, Alfano CM. Physical activity, biomarkers, and dis-ease outcomes in cancer survivors: a systematic review. J NatlCancer Inst. 2012;104:1–26.

11. Demark-Wahnefried W, Clipp EC, Morey MC, Pieper C, Sloane R,Snyder DC, et al. Lifestyle intervention development study toimprove physical function in older adults with cancer: outcomesfrom Project LEAD. J Clin Oncol. 2006;24:3465–73.

12. Scheier MF, Helgeson VS, Schulz R, Colvin S, Berga S, BridgesMW, et al. Interventions to enhance physical and psychologicalfunctioning among younger women who are ending nonhormonaladjuvant treatment for early-stage breast cancer. J Clin Oncol.2005;23:4298–311.

13. Chlebowski R. Lifestyle change including dietary fat reduction andbreast cancer outcome. J Nutr. 2007;137:233S–5S.

14. World Cancer Research Fund/American Institute for CancerResearch. Food, nutrition, physical activity, and the prevention ofcancer: a global perspective. Washington: AICR; 2007.

15. Psaltopoulou T, Ilias I, Alevizaki M. The role of diet and lifestyle inprimary, secondary, and tertiary diabetes prevention: a review ofmeta-analyses. Rev Diabet Stud. 2010;7:26–35.

16. Hussain A, Claussen B, Ramachandran A, Williams R. Preventionof type 2 diabetes: a review. Diabetes Res Clin Pract. 2007;76:317–26.

17. Artinian NT, Fletcher GF, Mozaffarian D, Kris-Etherton P, VanHorn L, Lichtenstein AH, et al. Interventions to promote physicalactivity and dietary lifestyle changes for cardiovascular risk factorreduction in adults: a scientific statement from the American HeartAssociation. Circulation. 2010;122:406–41.

18. Robien K, Demark-Wahnefried W, Rock CL. Evidence-based nu-trition guidelines for cancer survivors: current guidelines, knowl-edge gaps, and future research directions. J Am Diet Assoc.2011;111:368–75.

19. Hayes SC, Spence RR, Galvão DA, Newton RU. Australian asso-ciation for exercise and sport science position stand: optimisingcancer outcomes through exercise. J Sci Med Sport. 2009;12:428–34.

20. Campbell A, Stevinson C, Crank H. The BASES expert statementon exercise and cancer survivorship. J Sports Sci. 2012;30:949–52.

21. Cancer Council Australia. Position statement on benefits of healthydiet and physical activity for cancer survivors. 2013 Available from:www.cancer.org.au/content /pdf/CancerControlPolicy/PositionStatements/PSbenefitshealthydietcancersurvivorsJUN06.pdf. Accessed 11 Aug 2013.

22. Courneya KS, Friedenreich CM. Physical activity and cancer con-trol. Semin Oncol Nurs. 2007;23:242–52.

23. Bellizzi KM, Rowland JH, Jeffery DD, McNeel T. Healthbehaviors of cancer survivors: examining opportunities forcancer control intervention. J Clin Oncol. 2005;23:8884–93.

24. Eakin EG, Youlden DR, Baade PD, Lawler SP, Reeves MM,Heyworth JS, et al. Health behaviors of cancer survivors: data froman Australian population-based survey. Cancer Causes Control.2007;18:881–94.

25. Blanchard CM,CourneyaKS, Stein K. Cancer survivors’ adherenceto lifestyle behavior recommendations and associations with health-related quality of life: results from the American Cancer Society’sSCS-II. J Clin Oncol. 2008;26:2198–204.

26. Glanz K, Bishop DB. The role of behavioral science theory indevelopment and implementation of public health interventions.Annu Rev Public Health. 2010;31:399–418.

27. Noar SM, Benac CN, Harris MS. Does tailoring matter? Meta-analytic review of tailored print health behavior change interven-tions. Psychol Bull. 2007;133:673–93.

28. Abraham C, Michie S. A taxonomy of behavior change techniquesused in interventions. Health Psychol. 2008;27:379–87.

29. Nigg CR, Allegrante JP, Ory M. Theory-comparison and multiple-behavior research: common themes advancing health behavior re-search. Health Educ Res. 2002;17:670–9.

30. Lubans DR, Foster C, Biddle SJH. A review of mediators ofbehavior in interventions to promote physical activity among chil-dren and adolescents. Prev Med. 2008;47:463–70.

31. Bandura A. Health promotion by social cognitive means. HealthEduc Behav. 2004;31:143–64.

32. Bandura A. Health promotion from the perspective of social cogni-tive theory. Psychol Health. 1998;13:623–49.

33. White S, Wojcicki T, McAuley E. Social cognitive influences onphysical activity behavior in middle-aged and older adults. GerontolPsychol Sci Soc Sci. 2011;67B:18–26.

34. Phillips S, McAuley E. Social cognitive influences on physicalactivity participation in long-term breast cancer survivors.Psychooncology. 2012;22:783–91.

35. Rovniak LS, Anderson ES, Winett RA, Stephens RS. Socialcognitive determinants of physical activity in young adults: aprospective structural equation analysis. Ann Behav Med.2002;24:149–56.

36. Ayotte B, Margrett J, Hicks-Patrick J. Physical activity in middle-aged and young-old adults: the roles of self-efficacy, barriers, out-come expectancies, self-regulatory behaviors and social support. JHealth Psychol. 2010;15:173–85.

37. Anderson-Bill ES, Winett RA, Wojcik JR. Social cognitive deter-minants of nutrition and physical activity among web-health usersenrolling in an online intervention: the influence of social support,self-efficacy, outcome expectations, and self-regulation. J MedInternet Res. 2011;13:e28.

38. Anderson E, Winett RA, Wojcik JR. Self-regulation, self-efficacy,outcome expectations, and social support: social cognitive theoryand nutrition behavior. Ann Behav Med. 2007;34:304–12.

39. Graves KD. Social cognitive theory and cancer patients’ quality oflife: a meta-analysis of psychosocial intervention components.Health Psychol. 2003;22:210–9.

40. Courneya KS, Segal RJ, Mackey JR, Gelmon K, Reid RD,Friedenreich CM, et al. Effects of aerobic and resistance exercisein breast cancer patients receiving adjuvant chemotherapy: a multi-center randomized controlled trial. J Clin Oncol. 2007;25:4396–404.

41. Courneya KS, Sellar CM, Stevinson C, McNeely ML, Peddle CJ,Friedenreich CM, et al. Randomized controlled trial of the effects ofaerobic exercise on physical functioning and quality of life inlymphoma patients. J Clin Oncol. 2009;27:4605–12.

42. Rowland JH, Hewitt M, Ganz PA. Cancer survivorship: a newchallenge in delivering quality cancer care. J Clin Oncol. 2006;24:5101–4.

43. Courneya KS. Efficacy, effectiveness, and behavior change trials inexercise research. Int J Behav Nutr Phys. 2010;7:81.

44. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC,Ioannidis JPA, et al. The PRISMA statement for reportingsystematic reviews and meta-analyses of studies that evaluatehealthcare interventions: explanation and elaboration. BMJ.2009;339.

45. Review Manager (RevMan). The Cochrane Collaboration,Copenhagen, Denmark. 2011.

46. Deeks JJ, Higgins JPT, AltmanDG. Analysing data and undertakingmeta-analyses. In: Higgins JPT, Green S, editors. Cochrane hand-book for systematic reviews of interventions. Chichester: Wiley;2008.

47. Thomas BH, Ciliska D, Dobbins M, Micucci S. A process forsystematically reviewing the literature: providing the research evi-dence for public health nursing interventions. Worldviews EvidBased Nurs. 2004;1:176–84.

336 J Cancer Surviv (2015) 9:305–338

48. Hatchett A, Hallam J, Ford M. Evaluation of a social cognitivetheory-based email intervention designed to influence the physicalactivity of survivors of breast cancer. Psychooncology. 2012;22:829–36.

49. Ligibel JA, Meyerhardt J, Pierce JP, Najita J, Shockro L, CampbellN, et al. Impact of a telephone-based physical activity interventionupon exercise behaviors and fitness in cancer survivors enrolled in acooperative group setting. Breast Cancer Res Treat. 2011;132:205–13.

50. Pinto BM. Home-based physical activity intervention for breastcancer patients. J Clin Oncol. 2005;23:3577–87.

51. Rogers L, Hopkins-Price P, Vicari S, Pamenter R, Courneya K,Markwell S, et al. A randomized trial to increase physical activityin breast cancer survivors. Med Sci Sports Exerc. 2009;41:935–46.

52. WangY-J, BoehmkeM,WuY-WB, Dickerson SS, Fisher N. Effectsof a 6-week walking program on Taiwanese women newly diag-nosed with early-stage breast cancer. Cancer Nurs. 2011;34:E1–E13.

53. Pinto BM, Papandonatos GD, Goldstein MG, Marcus BH, FarrellN. Home-based physical activity intervention for colorectal cancersurvivors. Psychooncology. 2013;22:54–64.

54. Short CE, James EL, Girgis A, D’Souza MI, Plotnikoff RC. Mainoutcomes of the Move More for Life trial: a randomised controlledtrial examining the effects of tailored-print and targeted-print mate-rials for promoting physical activity among post-treatment breastcancer survivors. Psychooncology. 2014.

55. Short CE, James EL, Girgis A, Mcelduff P, Plotnikoff RC. Movemore for life: the protocol for a randomised efficacy trial of atailored-print physical activity intervention for post-treatment breastcancer survivors. BMC Cancer. 2012;12:172.

56. Valle CG, Tate DF, Mayer DK, Allicock M, Cai J. A randomizedtrial of a Facebook-based physical activity intervention for youngadult cancer survivors. J Cancer Surviv. 2013;7:355–68.

57. Bennett JA, Lyons KS, Winters-Stone K, Nail LM, Scherer J.Motivational interviewing to increase physical activity in long-term cancer survivors: a randomized controlled trial. Nurs Res.2007;56:18–27.

58. Matthews CE, Wilcox S, Hanby CL, Der Ananian C, Heiney SP,Gebretsadik T, et al. Evaluation of a 12-week home-based walkingintervention for breast cancer survivors. Support Care Cancer.2007;15:203–11.

59. Parsons JK, Newman VA, Mohler JL, Pierce JP, Flatt S, Marshall J.Dietary modification in patients with prostate cancer on activesurveillance: a randomized, multicentre feasibility study. BJU Int.2008;101:1227–31.

60. Demark-WahnefriedW, Case LD, Blackwell K,Marcom PK, KrausW, Aziz N, et al. Results of a diet/exercise feasibility trial to preventadverse body composition change in breast cancer patients onadjuvant chemotherapy. Clin Breast Cancer. 2008;8:70–9.

61. Campbell MK, Carr C, Devellis B, Switzer B, Biddle A,Amamoo MA, et al. A randomized trial of tailoring andmotivational interviewing to promote fruit and vegetableconsumption for cancer prevention and control. Ann BehavMed. 2009;38:71–85.

62. Demark-Wahnefried W, Clipp E, McBride C, Lobach D, Lipkus I,Peterson B, et al. Design of FRESH START: a randomized trial ofexercise and diet among cancer survivors. Med Sci Sports Exerc.2003;35:415–24.

63. Demark-Wahnefried W, Clipp EC, Lipkus IM, Lobach D, SnyderDC, Sloane R, et al. Main outcomes of the FRESH START trial: asequentially tailored, diet and exercise mailed print interventionamong breast and prostate cancer survivors. J Clin Oncol.2007;25:2709–18.

64. Djuric Z, Ellsworth JS, Weldon AL, Ren J, Richardson CR,Resnicow K, et al. A diet and exercise intervention during chemo-therapy for breast cancer. Open Obes J. 2011;3:87–97.

65. von Gruenigen VE, Courneya KS, Gibbons HE, Kavanagh MB,Waggoner SE, Lerner E. Feasibility and effectiveness of a lifestyleintervention program in obese endometrial cancer patients: a ran-domized trial. Gynecol Oncol. 2008;109:19–26.

66. Djuric Z, DiLaura NM, Jenkins I, Darga L, Jen CK-L, MoodD, et al. Combining weight-loss counseling with the weightwatchers plan for obese breast cancer survivors. Obes Res.2002;10:657–65.

67. von Gruenigen V, Frasure H, Kavanagh MB, Janata J, Waggoner S,Rose P, et al. Survivors of uterine cancer empowered by exerciseand healthy diet (SUCCEED): a randomized controlled trial.Gynecol Oncol. 2012;125:699–704.

68. Rogers L, Hopkins Price P, Vicari S, Markwell S, Pamenter R,Courneya K, et al. Physical activity and health outcomes threemonths after completing a physical activity behavior change inter-vention: persistent and delayed effects. Cancer EpidemiolBiomarkers Prev. 2009;18:1410–8.

69. Rogers L, Vicari S, Courneya K. Lessons learned in the trenches:facilitating exercise adherence among breast cancer survivors in agroup setting. Cancer Nurs. 2010;33:E10–7.

70. Rogers LQ, Markwell S, Hopkins-Price P, Vicari S, Courneya K,Hoelzer K, et al. Reduced barriers mediated physical activity main-tenance among breast cancer survivors. J Sport Exerc Psychol.2011;33:235–54.

71. Pinto BM, Rabin C, Dunsiger S. Home-based exercise amongcancer survivors: adherence and its predictors. Psychooncology.2009;18:369–76.

72. Rabin CS, Pinto BM, Trunzo JJ, Frierson GM, Bucknam LM.Physical activity among breast cancer survivors: regular exercisersvs participants in a physical activity intervention. Psychooncology.2006;15:344–54.

73. von Gruenigen V, Gibbons H, Kavanagh M, Janata J, Lerner E,Courneya K. A randomized trial of a lifestyle intervention in obeseendometrial cancer survivors: quality of life outcomes and media-tors of behavior change. Health Qual Life Outcomes. 2009;7:17.

74. Christy SM, Mosher CE, Sloane R, Snyder DC, Lobach DF,Demark-Wahnefried W. Long-term dietary outcomes of theFRESH START intervention for breast and prostate cancer survi-vors. J Am Diet Assoc. 2011;111:1844–51.

75. Michie S, Ashford S, Sniehotta FF, Dombrowski SU, Bishop A,French DP. A refined taxonomy of behaviour change techniques tohelp people change their physical activity and healthy eating behav-iours: the CALO-RE taxonomy. Psychol Health. 2011;26:1479–98.

76. Michie S, Richardson M, Johnston M, Abraham C, Francis J,Hardeman W, et al. The behavior change technique taxono-my (v1) of 93 hierarchically clustered techniques: buildingan international consensus for the reporting of behaviorchange interventions. Ann Behav Med. 2013;46:81–95.

77. Norris JM, Culos-Reed SN, Carlson LE, Aldous SH. Utilizing theTPB for understanding yoga participation in cancer survivors. JSport Exerc Psychol. 2007;29:S194-S.

78. Mosher CE, Fuemmeler BF, Sloane R, Kraus WE, LobachDF, Snyder DC, et al. Change in self-efficacy partially me-diates the effects of the FRESH START intervention oncancer survivors’ dietary outcomes. Psychooncology.2008;17:1014–23.

79. Cohen J. Statistical power analysis for the behavioral sciences. 2nded. New Jersey: Lawrence Erlbaum Associates; 1988.

80. Greaves CJ, Sheppard KE, Abraham C, Hardeman W, Roden M,Evans PH, et al. Systematic review of reviews of interventioncomponents associated with increased effectiveness in dietary andphysical activity interventions. BMC Public Health. 2011;11:119.

81. Michie S, Abraham C, Whittingham C, McAteer J, Gupta S.Effective techniques in healthy eating and physical activityinterventions: a meta-regression. Health Psychol. 2009;28:690–701.

J Cancer Surviv (2015) 9:305–338 337

82. Short CE, James EL, Stacey F, Plotnikoff RC. A qualitative synthe-sis of trials promoting physical activity behaviour change amongpost-treatment breast cancer survivors. J Cancer Surviv. 2013.

83. Spark LC, Reeves MM, Fjeldsoe BS, Eakin EG. Physical activityand/or dietary interventions in breast cancer survivors: a systematicreview of the maintenance of outcomes. J Cancer Surviv. 2013.

84. Nigg CR, Long CR. A systematic review of single health behaviorchange interventions vs. multiple health behavior change interven-tions among older adults. TBM. 2012;2:163–79.

85. Webb TL, Sheeran P. Does changing behavioral intentions engenderbehavior change? A meta-analysis of the experimental evidence.Psychol Bull. 2006;132:249–68.

86. Williams DM, Anderson ES, Winett RA. A review of the outcomeexpectancy construct in physical activity research. Ann BehavMed.2005;29:70–9.

87. Anderson ES, Winett RA, Wojcik JR, Williams DM. Social cogni-tive mediators of change in a group randomized nutrition andphysical activity intervention: social support, self-efficacy, outcomeexpectations and self-regulation in the guide-to-health trial. J HealthPsychol. 2010;15:21–32.

88. Prestwich A, Sniehotta FF, Whittington C, Dombrowski SU,Rogers L, Michie S. Does theory influence the effectiveness ofhealth behavior interventions? Meta-analysis. Health Psychol.2013;33:465–74.

89. Avery KNL, Donovan JL, Horwood J, Lane JA. Behavior theory fordietary interventions for cancer prevention: a systematic review ofutilization and effectiveness in creating behavior change. CancerCauses Control. 2013;24:409–20.

90. Broekhuizen K, Kroeze W, Poppel MNM, Oenema A, Brug J. Asystematic review of randomized controlled trials on the effective-ness of computer-tailored physical activity and dietary behaviorpromotion programs: an update. Ann BehavMed. 2012;44:259–86.

91. Moher D, Hopewell S, Schulz KF, Montori V, Gotzsche PC,Devereaux PJ, et al. CONSORT 2010 explanation and elaboration:updated guidelines for reporting parallel group randomised trials.BMJ. 2010;340:c869.

92. Higgins JPT, Green S. Cochrane handbook for systematic reviewsof interventions version 5.1.0 [updated March 2011]: The CochraneCollaboration; 2011. Available from: www.cochrane-handbook.org.

93. Livingston PM, Salmon J, Courneya KS, Gaskin CJ, Craike M,Botti M, et al. Efficacy of a referral and physical activity programfor survivors of prostate cancer [ENGAGE]: rationale and designfor a cluster randomised controlled trial. BMCCancer. 2011;11:237.

94. Rogers LQ, McAuley E, Anton PM, Courneya KS, Vicari S,Hopkins-Price P, et al. Better exercise adherence after treatmentfor cancer (BEAT Cancer) study: rationale, design, and methods.Contemp Clin Trials. 2012;33:124–37.

95. James EL, Stacey F, Chapman K, Lubans DR, Asprey G, SundquistK, et al. Exercise and nutrition routine improving cancer health(ENRICH): the protocol for a randomized efficacy trial of a nutritionand physical activity program for adult cancer survivors and carers.BMC Public Health. 2011;11:236.

96. Blackburn GL, Wang KA. Dietary fat reduction and breast canceroutcome: results from the Women’s Intervention Nutrition Study(WINS). Am J Clin Nutr. 2007;86(suppl):878S–81S.

97. Chlebowski RT, Blackburn GL, Thomson CA, Nixon DW, ShapiroA, Hoy MK, et al. Dietary fat reduction and breast cancer outcome:interim efficacy results from the Women’s Intervention NutritionStudy. J Natl Cancer Inst. 2006;98:1767–76.

98. Hoy MK, Winters BL, Chlebowski RT, Papoutsakis C, Shapiro A,LubinMP, et al. Implementing a low-fat eating plan in theWomen’sIntervention Nutrition Study. J Am Diet Assoc. 2009;109:688–96.

99. Pierce JP, Natarajan L, Caan BJ, Parker BA, Greenberg RE, FlattSW, et al. Influence of a diet very high in vegetables, fruit, and fiberand low in fat on prognosis following treatment for breast cancer:TheWomen’s Healthy Eating and Living (WHEL) randomized trial.J Am Med Assoc. 2007;298:289–98.

100. Pierce JP, Natarajan L, Sun S, Al-Delaimy W, Flatt SW, Kealey S,et al. Increases in plasma carotenoid concentrations in response to amajor dietary change in the Women’s Healthy Eating and Livingstudy. Cancer Epidemiol Biomarkers Prev. 2006;15:1886–92.

101. Carmack Taylor CL, Demoor C, Smith MA, Dunn AL, Basen-Engquist K, Nielsen I, et al. Active for Life After Cancer: a ran-domized trial examining a lifestyle physical activity program forprostate cancer patients. Psychooncology. 2006;15:847–62.

102. Morey MC, Snyder DC, Sloane R, Cohen HJ, Peterson B, HartmanTJ, et al. Effects of home-based diet and exercise on functionaloutcomes among older, overweight long-term cancer survivors:RENEW: a randomized controlled trial. J Am Med Assoc.2009;301:1883–9.

103. Snyder DC, Morey MC, Sloane R, Stull V, Cohen HJ, Peterson B,et al. Reach out to Enhance Wellness in Older Cancer Survivors(RENEW): design, methods and recruitment challenges of a home-based exercise and diet intervention to improve physical functionamong long-term survivors of breast, prostate, and colorectal cancer.Psychooncology. 2009;18:429–39.

104. Twiss JJ,WaltmanNL, Berg K, Ott CD, Gross GJ, Lindsey AM. Anexercise intervention for breast cancer survivors with bone loss. JNurs Scholarsh. 2009;41:20–7.

105. Michie S, Prestwich A. Are interventions theory-based?Development of a theory coding scheme. Health Psychol.2010;29:1–8.

106. Strasser B, Steindorf K, Wiskemann J, Ulrich CM. Impact ofresistance training in cancer survivors: a meta-analysis. Med SciSports Exerc. 2013;45:2080–90.

107. Hutchison AJ, Breckon JD, Johnston LH. Physical activity behaviorchange interventions based on the transtheoretical model: a system-atic review. Health Educ Behav. 2008;36:829–45.

108. Painter JE, Borba CPC, Hynes M, Mays D, Glanz K. The use oftheory in health behavior research from 2000 to 2005: a systematicreview. Ann Behav Med. 2008;35:358–62.

109. Lawler SP, Winkler E, Reeves MM, Owen N, Graves N, Eakin EG.Multiple health behavior changes and co-variation in a telephonecounseling trial. Ann Behav Med. 2010;39:250–7.

110. White SM, McAuley E, Estabrooks PA, Courneya KS. Translatingphysical activity interventions for breast cancer survivors into prac-tice: an evaluation of randomized controlled trials. Ann BehavMed.2009;37:10–9.

111. Reedy J, Haines PS, Campbell MK. The influence of health behav-ior clusters on dietary change. Prev Med. 2005;41:268–75.

112. Ko LK, Campbell MK, LewisMA, Earp J, DeVellis B.Mediators offruit and vegetable consumption among colorectal cancer survivors.J Cancer Surviv. 2010;4:149–58.

113. Wilkinson AV, Barrera SL, McBride CM, Snyder DC, Sloane R,Meneses KM, et al. Extant health behaviors and uptake of standard-ized vs tailored health messages among cancer survivors enrolled inthe FRESH START trial: a comparison of fighting-spirits vs fatal-ists. Psychooncology. 2012;21:108–13.

338 J Cancer Surviv (2015) 9:305–338