Order 924380: critically appraised topic and implementation of the evidence

A Randomized Trial to Promote Physical Activity Among Breast Cancer Patients

Bernardine M. Pinto The Miriam Hospital, Providence, Rhode Island, and W. Alpert

Medical School of Brown University

George D. Papandonatos Brown University

Michael G. Goldstein VHA National Center for Health Promotion and Disease Prevention, Durham, North Carolina

Objective: Physical activity (PA) has been shown to provide health benefits for breast cancer patients. The effects of augmenting oncology health care provider (HCP) advice for PA with 3 months of telephone counseling versus contact control were evaluated in a randomized trial. Methods: After receiving brief HCP advice to become physically active, 192 women (age in years: M � 60.0, SD � 9.9) who had completed treatment for Stage 0-IV breast cancer were randomized to telephone counseling to support PA (n � 106) or contact control (n � 86). Their PA, motivational readiness, fatigue, and physical functioning were assessed at baseline (before receiving HCP advice), 3, 6, and 12 months. Results: Telephone counseling produced significant effects on the primary outcome of moderate-intensity PA of about 30 min/week at both 3 months (95% CI � 0.44, 57.32) and 6 months (95% CI � 3.06, 61.26). Intervention participants were also more than twice as likely as control participants to report improvements in achieving PA guidelines of at least 150 min/week at 3 (OR � 2.43, 95% CI � 1.18, 4.98) and 6 months (OR � 2.11, 95% CI � 1.00 – 4.48). Telephone counseling was significantly more effective than contact control in increasing motivational readiness for PA at all follow-ups (ORs � 3.93– 6.28, all ps �.003). No between-groups differences were found for fatigue, while differential improvements in physical functioning did not remain significant past 3 months (p � .01). Conclusion: HCP advice plus telephone counseling improved PA among breast cancer patients at 3 and 6 months and also differentially improved patients’ motivational readiness at all follow-ups, suggesting the potential for exercise promotion in cancer follow-up care.

Keywords: breast cancer, physical activity, exercise, counseling

Supplemental materials: http://dx.doi.org/10.1037/a0029886.supp

A growing number of cancer survivors face impairments in physical functioning, increased fatigue and reduced quality of life (QOL), and increased risk for cardiovascular disease, obesity, osteoporosis and future cancers (Institute of Medicine and the National Research Council, 2006). Evidence suggests that partic-

ipating in moderate-intensity physical activity (PA) for at least three months improves physical functioning, QOL, and mood and reduces fatigue among cancer survivors (Agency for Healthcare Research and Quality, 2004; Galvão & Newton, 2005; Knols, Aaronson, Uebelhart, Fransen, & Aufdemkampe, 2005; Speck, Courneya, Masse, Duval, & Schmitz, 2010). Cancer treatments require frequent follow-up appointments that provide oncology health care providers (HCPs) with opportunities to encourage patients to change health risk behaviors. However, Sabatino and colleagues (2007) found that only 25% of a national sample of cancer survivors reported receiving a recommendation about ex- ercise from their physicians.

HCPs have played a minimal role, if any, in PA interventions for cancer patients. One study involved breast cancer patients seen at adjuvant treatment consultation. Participants received either: a) a recommendation to exercise, b) a recommendation plus a referral to an exercise specialist, or c) usual care (Jones, Courneya, Fairey, & Mackey, 2004). PA assessments at 1 and 5 weeks revealed greater PA participation in the group that received a recommen- dation to exercise versus usual care. In our trial, HCPs were asked to provide PA advice to patients who had completed surgery and adjuvant chemotherapy/radiation. Evidence suggests that it is not practical to rely on physicians to provide more intensive interven-

Bernardine M. Pinto, Centers for Behavioral and Preventive Medicine, The Miriam Hospital, Providence, Rhode Island, and W. Alpert Medical School of Brown University; George D. Papandonatos, Center for Statis- tical Sciences, Brown University; Michael G. Goldstein, Office of Patient Care Services, VHA National Center for Health Promotion and Disease Prevention, Durham, North Carolina.

This research was funded by a grant from the American Cancer Society and Rays of Hope (RSGPB-03-243-01 PBP). We gratefully acknowledge the contributions of the research staff (Susan Abdow, Stephanie Berube, Christopher Breault, Jennifer Correia, Kelly Greenwood, and Joyce Lee). We thank the physicians who participated in the study and assisted with patient recruitment. The trial is registered in the Clinical Trials Registry (NCT 002 30711).

Correspondence concerning this article should be addressed to Bernar- dine M. Pinto, Centers for Behavioral and Preventive Medicine, The Miriam Hospital, One Hoppin St., Coro Bldg., Suite 314, Providence, RI 02903. E-mail: [email protected]

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

Health Psychology © 2013 American Psychological Association 2013, Vol. 32, No. 6, 616 – 626 0278-6133/13/$12.00 DOI: 10.1037/a0029886

616

tions, and that instead we should involve nonphysician staff such as telephone counselors (Marcus et al., 1998) and incorporate interactive health technology (de Vries & Brug, 1999) in our interventions. Hence, we extended brief HCP advice with a 3-month telephone counseling program for PA.

The use of telephone-based interventions to promote PA in a general population has been well documented (see reviews by Castro & King, 2002; Eakin, Lawler, Vandelanotte, & Owen, 2007; Goode, Reeves, & Eakin, 2012). The studies reviewed showed convincingly that such interventions are not only effica- cious, but they also offer unique advantages of increased conve- nience and access. There are also increased opportunities for contact anywhere a telephone is accessible and increased time efficiency. These advantages, together with the counselor’s skills and resources, can help promote PA among individuals who may not be receptive to face-to-face contact or printed materials. Telephone-based PA interventions over 6 –12 weeks have been tested among small samples of breast cancer patients (Matthews et al., 2007; Mock et al., 1997) with positive effects on PA (Mat- thews et al., 2007) and reductions in patients’ anxiety, fatigue, and sleeping difficulties (Mock et al., 1997). Telephone calls have also been used in PA interventions offered over 6 months and longer to breast, prostate, and other cancer survivors (Bennett, Lyons, Winters-Stone, Nail, & Scherer, 2007; Demark-Wahnefried et al., 2006; Morey et al., 2009) with one study showing favorable effects at the end of a 6-month intervention (Bennett et al., 2007) and another study with a 12-month intervention showing significant group effects on PA and physical functioning (Morey et al., 2009). In sum, there is evidence to support the efficacy of telephone- based interventions at postintervention in promoting PA among cancer survivors. However, these PA interventions did not involve HCPs and a majority did not assess PA outcomes in the long-term. In this study, we used a telephone counseling program whose efficacy had been previously tested among breast cancer patients (Pinto, Frierson, Rabin, Trunzo, & Marcus, 2005) to extend the HCP advice. The comparison group also received HCP advice and telephone calls to control for contact as a more conservative test of the intervention. In addition, final assessment of outcomes oc- curred 6 months after all intervention contact ended.

The primary purpose of this study was to examine the effects of HCP advice to become physically active plus Telephone Counsel- ing (Intervention) versus HCP advice plus Contact Control (Con- trol) on self-reported minutes of PA (leisure and occupational activity) of at least moderate-intensity at 3 months among women who had completed breast cancer treatment. We hypothesized that extending brief HCP advice by providing telephone counseling specific to PA would produce stronger increases in PA at 3 months than telephone contact of the same frequency that provided health monitoring. Secondary aims included examining maintenance of intervention effects on PA at 6 and 12 months. We also hypothe- sized that the increased PA among intervention participants would maintain over time. Other goals included examining intervention effects on the proportion of participants who met PA guidelines and on participants’ motivational readiness for PA at 3 months, 6 months, and 12 months. We hypothesized that a larger proportion of intervention participants would meet PA guidelines, and that the intervention group would progress further in motivational readi- ness for PA. Finally, we sought to examine intervention effects on self-reported physical functioning and fatigue at follow-up. We

hypothesized that the intervention group would report improved physical functioning and reduced fatigue at follow-ups compared with the control group.

Methods

Design

We conducted a randomized trial offering all participants HCP advice for PA and then compared: (a) 12 weeks of additional Telephone Counseling, and (b) Contact Control. Assessments were conducted at baseline, posttreatment (3 months), at 6 months and 12 months. Institutional Review Boards at the Miriam Hospital and Women and Infants Hospital approved the study. The study was conducted in accordance with the Helsinki Declaration from 2004 –2009.

Recruitment

Participants were recruited by informational letters sent by oncologists and surgeons to their patients, and by in-person re- cruitment at a hospital-based oncology clinic. HCPs were asked to review their nonurgent follow-up care schedules and to identify women who had completed breast cancer treatment, had no current evidence of disease, and were expected to live � 12 months. Letters were mailed to these patients approximately three months before their next visit. If patients were interested in the study, they were asked to contact the study staff who conducted an eligibility screen by telephone. Eligibility criteria: 1) female aged � 18 years, 2) completed primary and adjuvant treatment for breast cancer (patients on hormone treatment such as Tamoxifen were eligible), 3) � 5 years since treatment completion, 4) able to read and speak English, 5) provided consent for medical chart review, 6) able to walk unassisted, 7) were relatively inactive (�30 min/ week of vigorous-intensity exercise or �90 min/week of moderate-intensity exercise), and 8) had access to a telephone. Participants were excluded if they had a prior history of cancer or if they had a medical or current psychiatric illness (e.g., cardio- vascular disease, diabetes) that could hinder compliance with the study protocol.

We completed 351 initial telephone screens to determine study eligibility (see Figure 1). Of those screened, 192 (54.7% of phone screens, 71% of eligible respondents) were eligible, interested, and eventually randomized. The study was designed to have 80% power to detect a between-groups difference in change scores of 0.35 SD units at the 5% level of significance, based on cross- sectional comparisons at 3 months. Due to recruitment difficulties, the study goal of 300 based on N � 125/group at 3 months (starting from N � 150/group at baseline) could not be met within the time available. Based on 83 control and 88 intervention par- ticipants with valid 7-day physical activity recall (PAR) measures at 3 months (see Figure 1), the minimum detectable between- groups difference in change scores rose to 0.42 SD units. Given the observed 3-month change-score SD of 106 min/week, this trans- lates to a 45-min difference in 3-month change scores, before taking into account the additional power offered by the repeated measures design.

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

617PHYSICAL ACTIVITY INTERVENTION

Procedure

After providing informed consent, participants obtained medical clearance from their oncologist. All participants received PA ad- vice from an oncologist/surgeon during a clinic visit (n � 100) or advice documented in a letter (n � 92) after they were referred for study participation during a clinic visit. After receiving HCP advice, they were randomly assigned to the two study arms using a centrally administered randomization procedure that stratified on prior chemotherapy status (yes/no) and PA level (participants classified as active vs. not based on a PA threshold of 30 min/ week). HCPs and staff conducting the assessments were blinded to participants’ group assignments. Participants and intervention co- ordinators were not blinded to group assignments.

HCP Advice

Oncologists and surgeons (n � 14, 29% women, mean years in practice � 15.6, SD � 8.9, mean age � 50.8, SD � 9.6) at three local hospitals and two private practices who were invited to participate in the study received training (15–30 min) in providing

brief PA advice (�5 min). The brief motivational counseling protocol was derived from the 5As counseling strategy (address the agenda, assess, advise, assist and arrange follow-up) used previously for training physicians (Goldstein et al., 1999; Pinto, Goldstein, Ashba, Sciamanna, & Jette, 2005). The HCP’s role was to provide patients a brief message about PA benefits, recommend 30 min of moderate-intensity PA on most days of the week, and arrange for follow-up with study staff.

Participants who were recruited via informational letters re- ceived HCP advice at the next regularly scheduled clinic visit. At this visit, providers were cued by prompts placed on patients’ charts to deliver PA advice. Documentation of message delivery was recorded on the chart prompt. Providers were allowed to drop patients from the study if the goal of moderate-intensity PA would be unsafe for the patient. After completing the clinic visit, each participant was met by research staff, the chart prompt was col- lected and her randomization status was determined. For partici- pants recruited on-site (n � 92), HCPs recommended the study to patients seen in clinic. If interested, eligible and enrolled in the study, the participant was given a letter from her HCP document-

Initial phone screen for eligibility, n=351

Ineligible: 23.1% (n=81) Too active=36 Medical issues=16 >10 years postdiagnosis=2 Ongoing psychological issues=3 No English fluency=2 Not able to exercise= 3 Enrolled in another study=6 HCP not participating=1 Other=12

Eligible at phone screen: 76.9% (n=270)

Eligible and randomized: 71.1% (n=192)

Not randomized: 28.9% (n=78) No interest=18, Too busy=18 Lost contact=8, Family issues=5 Medical issues=4, Other reason=11 Reason unknown=14

TC Group (n=106) CC Group (n=86)

12-week PA Counseling 12-week Contact

Post-treatment assessment: 83.9% (n=89) Attrition=17 (Lost contact=8, family issues=4, cancer=2, no interest=2, too busy=1) Primary outcome analyzed: 83.0% (n=88)

Post-treatment assessment: 97.6% (n=84) Attrition=2 (Lost contact=2) Primary outcome analyzed: 96.5% (n=83)

Monthly PA calls for 3 months Monthly calls for 3 months

Oncology HCP advice (in-person or by letter)

Assessment at 6 months: 81.1% (n=86) Attrition=3 (Lost contact=1, family issues=1, no interest=1) Primary outcome analyzed: 80.2% (n=85)

Assessment at 6 months: 93.0% (n=80) Attrition=4 (No interest=2, too busy=1, surgery=1) Primary outcome analyzed: 89.5% (n=77)

Assessment at 12 months: 79.2% (n=84) Attrition=2 (Lost contact=1, cancer=1) Primary outcome analyzed: 77.4% (n=82)

Assessment at 12 months: 90.6% (n=78) Attrition=2 (too busy=1, death=1) Primary outcome analyzed: 88.4% (n=76)

Figure 1. Flow diagram of participant recruitment, randomization, and retention.

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

618 PINTO, PAPANDONATOS, AND GOLDSTEIN

ing “brief advice” elements (advise, assist and arrange follow-up/ referral to study staff) and randomized. Advice documented in a letter was used to reduce delays in study enrollment since the next clinic visit may have been more than 3 months later.

HCP Advice Plus Telephone Counseling (Intervention)

These participants received in-person instructions on how to exercise at a moderate-intensity level, monitor heart rate, and how to warm up before and cool down after PA. They were given home logs to monitor PA participation and a pedometer (Digiwalker, Yamax Corporation, Tokyo, Japan). The intervention was individ- ualized to the participant’s baseline PA (and motivational readi- ness) such that, inactive participants were encouraged to be phys- ically active for at least 10 min on at least 2 days/week (these goals were higher for those who were physically active at baseline), and the goals were gradually increased over the 12 weeks to 30 min/day on at least 5 days/week (U.S. Department of Health and Human Services, 1996). For participants who reported some level of PA at baseline, the exercise goals negotiated by the interventionist were higher. Hence, starting points and rates of PA progression varied across participants because these were individualized to increase the motivation and confidence of the participants. The counseling promoted moderate- intensity aerobic PA at 55– 65% maximum heart rate such as brisk walking, biking, or swimming.

Each participant received eight telephone calls over 12 weeks (weekly for 4 weeks, biweekly for 8 weeks) from Intervention Coordinators to support PA adoption. Counseling was based on the Transtheoretical Model and Social Cognitive Theory (Bandura, 1986; Prochaska & DiClemente, 1983), and it was tailored to each participant’s motivational readiness (Marcus & Simkin, 1993). The counseling focused on strengthening self-efficacy for PA, and it trained participants in techniques such as self-monitoring of PA, setting PA goals, and planning for exercise. Cognitive processes of change were emphasized for participants in Contemplation, and behavioral processes were emphasized for those in Preparation (Marcus & Simkin, 1993). Specific components from motivational interviewing (conviction of the importance of PA to cancer recov- ery and confidence in becoming/staying active) were also assessed during the calls.

The PA counseling followed a structured format covering the following topics: assessment of the past week’s PA (and motiva- tional readiness), assessment of health problems, exploration of barriers to PA, assessment of the participant’s conviction of the importance of PA, negotiation of PA goals for the following week(s), assessment of the participant’s confidence in achieving the goals, and review of the tip-sheets that were sent to the participant. If participants reported physical symptoms such as chest pain, they were referred to their physician for clearance to resume study participation.

Participants were mailed a PA and a cancer survivorship tip-sheet on topics such as body image, each week over the 12-week intervention. Finally, a letter summarizing the partic- ipant’s progress was sent to her at weeks 2, 4, 8, and 12. After the 3-month assessments were completed, monthly phone calls over the next 3 months were provided to reinforce regular PA and prevent lapses.

HCP Advice Plus Contact Control Group (Control)

These participants received eight calls over 12 weeks (weekly for 4 weeks, biweekly for 8 weeks) during which the Symptom Questionnaire (Winningham, 1993) was administered to monitor problems such as headaches. Interventionists were trained not to discuss PA with this group. If the participants reported PA, the interventionist listened but did not provide any counseling related to PA. The goal was to match contact frequency with the inter- vention group, with no attempt made to match call duration across groups. In addition, participants received cancer survivorship tip- sheets. After the 3-month assessment, they also received monthly phone calls for 3 months, during which the Symptom Question- naire was administered.

Intervention Delivery

All telephone calls to study participants were audio-taped, and 25% of these tapes were randomly selected for review by the principal investigator and a co-investigator to ensure fidelity to protocol. In addition, participant issues were discussed during weekly staff meetings.

Measures

Disease and treatment variables (from medical records) and demographic information were obtained at baseline. At baseline and subsequent assessments, body weight and height were mea- sured. Participants received small incentives (e.g., $10 gift cards) for completing the assessments which included:

Seven-Day Physical Activity Recall (7-day PAR;Blair et al, 1985). This interviewer-administered measure (Sallis et al., 1985; Sarkin, Campbell, & Gross, 1997) assesses hours spent in sleep as well as moderate, hard, and very hard activity (leisure and occupational) over the past week. We were interested in the weekly minutes of at least moderate-intensity PA, which we ana- lyzed as a continuous outcome (primary outcome) and as a dichot- omous indicator of whether participants met recommendations (U.S. Department of Health and Human Services, 1996) of at least 150 min/week of moderate-intensity PA.

Stage of Motivational Readiness for PA (Marcus, Rossi, Selby, Niaura, & Abrams, 1992). This reliable and valid mea- sure assesses an individual’s motivational readiness for PA (Mar- cus & Simkin, 1993). It classifies individuals into one of five stages: precontemplation (individuals who do no PA and do not intend to start), contemplation (those who do not participate in PA but intend to start), preparation (those who participate in some PA but not regularly), action (those who currently participate in reg- ular activity, but have done so for less than 6 months), and maintenance (those who have participated in regular PA for 6 months or longer). For the purposes of this study, regular PA was defined as at least 30 min of moderate-intensity exercise on � 5 times per week. Since movement into Action/Maintenance has been significantly associated with fitness improvements (Marcus & Simkin, 1993), we modeled motivational readiness as dichoto- mous, contrasting those who successfully transitioned into Action/ Maintenance with those that did not.

MOS 36-Item Short Form Health Survey (SF-36; McHor- ney, Ware, & Raczek, 1993; Ware & Sherbourne, 1992). This assesses eight health concepts (e.g., physical functioning, bodily

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

619PHYSICAL ACTIVITY INTERVENTION

pain). We used the Physical Functioning subscale (PF), as cancer survivors who adopted exercise have shown improvements on this subscale (Pinto, Trunzo, Reiss, & Shiu, 2002). This measure yields a continuous variable that ranges from a low score of 0 (limitations in physical activities) to a high score of 100 (no limitations).

Functional Assessment of Cancer Therapy Scale-Fatigue (FACT-F). This 13-item scale is a brief, reliable, and valid measure of the physical and functional effects of fatigue. It has strong internal consistency, and it shows a significant positive relationship with other measures of fatigue (Yellen, Cella, Web- ster, Blendowski, & Kaplan, 1997). Scores on this measure range from 6 (high fatigue) to 52 (low fatigue).

Analyses

T tests for continuous variables and �2 tests for categorical variables were used to examine the success of the randomization procedure in balancing participants’ characteristics, including baseline values of the outcomes of interest (see Table 1). Similar analyses were used to compare retained participants versus drop- outs.

Longitudinal trajectory modeling of continuous outcomes was conducted using Linear Mixed Effects (LME) models, as imple- mented in Splus 8.2 (Insightful Corporation, 2007). Mean change scores from baseline were adjusted for baseline values of each outcome, and they were calculated separately by treatment group at each follow-up. Any variables showing significant between- groups differences at baseline were also included as potential confounders. Subject-specific random intercepts were used to ac- commodate within-subject correlation across time.

Of note, LME models employ likelihood-based estimation pro- cedures that use all available data to produce consistent estimates of the regression coefficients (Daniels & Hogan, 2008; Little & Rubin, 2002). Although they remain sensitive to drop out patterns that depend on the missing outcome itself, they are superior to completers-only analyses or intention-to-treat approaches that as- sign a prespecified score to the missing data.

Longitudinal binary outcomes were analyzed using Generalized Estimating Equation (GEE) methodology, as implemented in the Correlated Data Library of Splus 8.2 (Insightful Corporation, 2007). Logistic regression models with a working independence correlation matrix were used to estimate the effect of baseline PA levels and study arm on the odds of meeting or exceeding PA guidelines at each follow-up (U.S. Department of Health and Human Services, 1996, 2008). A similar GEE procedure was used to analyze movement into Action/Maintenance by study arm, controlling for stage of change at baseline (Contemplation vs. Preparation).

Results

Sample Characteristics

As seen in Table 1, 192 women (mean age � 60.0 years, SD � 9.9, mean time since diagnosis � 2.9 years, SD � 2.1) were assigned to either intervention (n � 86) or control (n � 106), using a stratified randomization scheme. Overall, 22 intervention and eight control participants withdrew or were dropped from the trial (see Figure 1). Attrition in the control group was consistently low

across time, whereas the intervention group experienced higher dropout at 3 months (n � 17), and limited losses thereafter. Within-group comparisons in the intervention arm, in terms of baseline characteristics, showed that 26% of dropouts had a mas- tectomy at 3 months versus 12% of retained participants (p � .1). Two participants sustained minor injuries related to falling off a treadmill, and one died during the trial for reasons unrelated to study participation.

Analyses revealed no statistically significant between-groups differences on demographic variables or outcomes at baseline. However, intervention versus control differences in chemotherapy rates (55% vs. 66%) and full-time employment (FTE) status (55% vs. 47%) were deemed meaningful enough to warrant further examination of these variables as potential confounders of the treatment-outcome relationship. Results suggested that chemother- apy did not affect any outcome of interest. However, FTE status affected all outcomes other than fatigue, at least during the 12- week intervention period. Therefore, longitudinal trajectories of study participants were adjusted not only for baseline values of each outcome, but also for FTE status, where warranted.

PA Outcomes

Seven-day PAR. Intervention participants outperformed con- trol participants by about 30 min/week of at least moderate inten- sity PA at both 3 months (p � .048) and 6 months (p � .032), but this beneficial telephone counseling effect dissipated at 12 months (p � .574). For illustrative purposes, we also included in Table 2 covariate-adjusted intervention and control change score trajecto- ries for a reference group of participants not in FTE with baseline PA levels set at the overall sample mean (45 min/week). These can be combined with the reported baseline PA and FTE effects to construct anticipated PA trajectories for any study participant of interest. For every additional hour by which a participant’s base- line PA level exceeded the sample mean, anticipated PA increases at follow-up were reduced in both study arms by 16 min at 3 months (p � .03), 35 min at 6 months (p � .001), and 28 min at 12 months (p � .001). In addition, FTE status increased weekly PA levels by 46 min at 3 months (p � .002), but its effect was attenuated at both 6 months (p � .604) and 12 months (p � .643).

Meeting PA guidelines. Given the sensitivity of average PA levels to the presence of outliers, we also estimated a logistic regression model in which the binary response was an indicator of whether a participant was able to meet or exceed guidelines of 150 min/week of PA at follow-up (U.S. Department of Health and Human Services, 2008). Results in Table 3 suggest beneficial intervention effects at 3 months (OR � 2.43, p � .016) and 6 months (OR � 2.11, p � .05), but not at 12 months (OR � 1.16, p � .704) for a reference group of participants not in FTE report- ing mean PA levels at baseline. As expected, higher PA at study entry made it even more likely that a participant would succeed in meeting guidelines at follow-up: For every hour by which a participant’s baseline PA exceeded the sample mean of 45 min/ week, the odds of meeting guidelines at follow-up rose by 11% to 23% across study arms, depending on time point. Finally, FTE status more than doubled the odds of meeting guidelines at 3 months (OR � 2.33, p � .02), but its effect was attenuated at both 6 months (p � .366) and 12 months (p � .477).

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

620 PINTO, PAPANDONATOS, AND GOLDSTEIN

Table 1 Sample Characteristics at Baseline (N � 192)

Characteristic/Category

Groupsa

CC (n � 86) TC (n � 106)

p-valueNo. % No. %

Race/Ethnicity Non-Hispanic White 80 93 100 95 .79 Non-Hispanic Black 4 5 3 3 Hispanic 2 2 2 2

Marital status Single 7 8 6 6 .81 Married/Living with partner 59 69 79 75 Divorced/Separated 12 14 12 11 Widowed 8 9 9 8

Employment status Employed full-time 47 55 50 47 .27 Employed part-time 10 12 20 19 Unemployed 4 5 8 8 Retired 20 24 18 17 Homemaker/Medical leave 4 5 10 9

Educational level High School Diploma 16 19 19 18 .99 Vocational/Trade School 5 6 6 6 Some college 24 28 28 26 Associate Degree 10 12 11 10 Bachelor Degree 14 16 20 19 Graduate School 17 20 22 21

Household income Less than $29,999 8 10 13 13 .39 $30,000–$39,999 7 9 11 11 $40,000–$49,999 15 19 10 10 Over $50,000 49 62 63 65

Age in years Mean (SD) 55.9 (9.9) 56.1 (9.9) .89

Body mass index Mean (SD) 28.7 (5.1) 29.6 (6.2) .28

Cancer stage 0 12 14 12 11 .89 I 33 38 41 39 II 34 40 44 42 III/IV 7 8 9 8

Cancer treatmentb

Lumpectomy 66 77 76 73 .68 Lumpectomy with dissection 44 51 53 50 .96 Mastectomy 28 33 34 33 .91 Mastectomy with reconstruction 6 7 6 6 .95 Radiation 63 73 76 72 .94 Chemotherapy 47 55 69 66 .16 Hormone treatment 70 81 78 74 .32

Years since diagnosis Mean (SD) 2.9 (2.1) 3.0 (2.2) .72

Motivational readiness Contemplation 67 78 81 76 .13 Preparation 13 15 23 22 Action/Maintenance 6 7 2 2

PA guidelines �150 PAR min/week 79 92 100 94 .70 �150 PAR min/week 7 8 6 6

7-day PAR (min/week) Mean (SD) 46.8 (62.5) 42.9 (59.4) .67

FACT-F Mean (SD) 38.1 (11.6) 39.3 (9.9) .47

SF-36 PF Mean (SD) 72.8 (22.8) 77.2 (19.5) .15

Note. TC � Telephone Counseling; CC � Contact Control; PA � Physical Activity; PAR � 7-day PAR; FACT-F � Functional Assessment of Cancer Therapy Scale-Fatigue; SF-36 PF � MOS 36-Item Short Form Health Survey: Physical Functioning subscale. a Percentages have been calculated on cases with available data. b Each patient may have received more than one treatment; percentages do not add to 100.

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

621PHYSICAL ACTIVITY INTERVENTION

Motivational readiness. All but eight participants were in either the Contemplation or Preparation stage at study entry, and a secondary study goal was to move them (N � 184) to Action or Maintenance stage at follow-up (Prochaska & DiClemente, 1983). Telephone counseling appears to have produced long-lasting ef- fects on motivational readiness among a reference group of par- ticipants not in FTE that joined the study while in Contemplation: As seen in Table 3, such participants were much more likely to have reached Action/Maintenance at 3 months (OR � 4.45, p � .001) and 6 months (OR � 3.93, p � .003) if assigned to the

intervention than the control arm, and these intervention effects were strengthened further at 12 months (OR � 6.28, p � .001). Participants entering the study in Preparation were significantly more likely to move to Action/Maintenance than those in Contem- plation, whether at 3 months (OR � 3.76, p � .002), 6 months (OR � 2.57, p � .033), or 12 months (OR � 2.64, p � .041). In contrast, FTE status more than doubled the odds of reaching Action/Maintenance at 3 months (OR � 2.58, p � .02), but its effect was attenuated at both 6 months (p � .373) and 12 months (p � .725).

Table 2 Point Estimates and 95% Confidence Intervals for Change Scores From Baseline to Follow-Upa

Outcome/Group

Follow-up

3 Months 6 Months 12 Months

Mean 95% CI Mean 95% CI Mean 95% CI

7-day PAR (min/week) TC 59.70 (35.59, 83.80) 56.64 (32.22, 81.07) 44.06 (19.22, 68.89) CC 30.82 (5.13, 56.51) 24.48 (�1.43, 50.40) 35.61 (9.04, 62.17) TC vs. CC 28.88 (0.44, 57.32) 32.16 (3.06, 61.26) 8.45 (�20.95, 37.86) Baseline PARb �15.83 (�30.36, �1.30) �35.25 (�49.85, �20.64) �27.76 (�42.40, �13.13) FTE vs. not 46.10 (17.67, 74.52) 7.70 (�21.34, 36.73) �6.96 (�36.32, 22.41)

SF-36 PF TC 3.73 (�0.39, 7.86) 4.79 (0.64, 8.95) 3.87 (�0.32, 8.06) CC �2.74 (�7.14, 1.65) 1.09 (�3.42, 5.59) 1.11 (�3.40, 5.62) TC vs. CC 6.48 (1.60, 11.35) 3.71 (�1.27, 8.69) 2.76 (�2.26, 7.77) Baseline SF-36 �0.40 (�0.52, �0.29) �0.35 (�0.46, �0.23) �0.35 (�0.47, �0.23) FTE vs. not 6.49 (1.60, 11.38) 4.08 (�0.93, 9.09) 1.75 (�3.29, 6.80)

FACT-F TC 4.53 (2.88, 6.18) 3.84 (2.17, 5.51) 3.69 (1.98, 5.39) CC 3.41 (1.70, 5.13) 1.95 (0.18, 3.72) 1.44 (�0.31, 3.20) TC vs. CC 1.12 (�1.26, 3.50) 1.89 (�0.54, 4.33) 2.44 (�0.20, 4.69) Baseline FACT-F �0.40 (�0.51, �0.29) �0.37 (�0.48, �0.26) �0.40 (�0.51, �0.29)

Note. TC � Telephone Counseling; CC � Contact Control; FTE � Full-time employment; PAR � 7-day PAR; SF-36 PF � MOS 36-Item Short Form Health Survey Physical Functioning subscale; FACT-F � Functional Assessment of Cancer Therapy Scale-Fatigue. a Boldface estimates denote p-values significant at � � .05. b Baseline PAR expressed in hours/week.

Table 3 Longitudinal Logistic Regression Models Predicting Achievement of PA Guidelines and Movement to Action/Maintenance at Follow-Upa

Outcome/Coefficientb

Follow-up

3 Months 6 Months 12 Months

OR 95% CI OR 95% CI OR 95% CI

PA guidelines TC 0.43 (0.23, 0.82) 0.39 (0.21, 0.73) 0.33 (0.17, 0.65) CC 0.18 (0.09, 0.35) 0.18 (0.09, 0.36) 0.29 (0.15, 0.54) TC vs. CC 2.43 (1.18, 4.98) 2.11 (1.00, 4.48) 1.16 (0.54, 2.52) Baseline PAR 1.23 (1.08, 1.39) 1.12 (1.02, 1.23) 1.11 (1.03, 1.19) FTE vs. not 2.33 (1.14, 4.76) 1.41 (0.67, 2.98) 0.76 (0.35, 1.63)

Action/Maintenancec

TC 0.27 (0.13, 0.57) 0.28 (0.14, 0.59) 0.35 (0.16, 0.74) CC 0.06 (0.02, 0.15) 0.07 (0.03, 0.19) 0.06 (0.02, 0.15) TC vs. CC 4.45 (2.02, 9.80) 3.93 (1.57, 9.80) 6.28 (2.29, 17.24) Prep. vs. Con 3.76 (1.59, 8.86) 2.57 (1.08, 6.14) 2.64 (1.04, 6.70) FTE vs. not 2.58 (1.16, 5.74) 1.45 (0.64, 3.26) 1.16 (0.50, 2.70)

Note. TC � Telephone Counseling; CC � Contact Control; FTE � Full-time employment; Con � Contemplation; Prep. � Preparation; PA � Physical Activity; PAR � 7-day PAR. a Boldface estimates denote p-values significant at � � .05. b Baseline PAR expressed in hours/week. c Model estimated among N � 184 participants in Contemplation or Preparation at study entry.

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

622 PINTO, PAPANDONATOS, AND GOLDSTEIN

Psychosocial Outcomes

Physical functioning. Intervention participants outperformed control participants by 6.48 units on the SF-36 PF scale at 3 months (p � .01), but group differences narrowed at 6 months (p � .147) and 12 months (p � .497). Table 2 displays covariate- adjusted intervention and control change score trajectories for a reference group of participants not in FTE reporting with average SF-36 levels at baseline (75.21 units). Trajectories for other study participants can be constructed by noting that for every additional unit by which a participant’s baseline SF-36 score exceeded the sample mean, anticipated SF-36 PF increases in both study groups were reduced by 0.35– 0.40 units at follow-up across study arms (all ps � .001). In addition, FTE status increased physical func- tioning by 6.49 units at 3 months (p � .01), but its effect was attenuated at both 6 months (p � .112) and 12 months (p � .497).

Fatigue. No significant group differences in fatigue levels were found at follow-up. Illustrative intervention and control change score trajectories are depicted in Table 2 for a reference group of participants not in FTE reporting mean FACT-F scores of 38.76 units at baseline. Trajectories for other participants can be calculated by noting that for every additional unit by which a participant’s baseline FACT-F score exceeded the sample mean, anticipated FACT-F increases in both study groups were reduced by 0.37– 0.40 units at follow-up (all ps � .001).

Intervention Delivery

The proportion of participants receiving in-person HCP advice was balanced across study arms, with negligible intervention ver- sus control differences (51.12% vs. 52.83%, p � .93). In-person HCP advice, as evidenced by completed chart prompts, was de- livered to 98% of the participants who received in-person advice (mean duration of advice � 4.7 min, SD � 1.4). Eighty-six percent of the participants reported that their HCPs explained the health benefits of PA, and 96% were satisfied with the advice. During the 3-month intervention phase, a mean of 6.7 calls (SD � 1.81) were delivered to intervention participants and 7.1 calls (SD � 1.3) to control participants (p � .07; max. possible � eight calls). As expected, calls in the intervention arm were of longer duration (M � 15.0 min, SD � 5.8) than calls in the control arm (M � 9.0 min, SD � 3.9, p � .001).

Discussion

Our primary goal was to examine the effects of HCP advice plus Telephone Counseling (Intervention) versus HCP advice plus Con- tact Control (Control) on participants’ PA at 3 months. HCPs were able to provide brief exercise advice, which the participants found satisfactory. We found that intervention participants outperformed control participants by about 30 min/week of at least moderate intensity PA at 3 months and 6 months, but that this effect dissipated at 12 months. In practical terms, this translates to PA increases of one additional day/week in terms of USDHHS guide- lines (U.S. Department of Health and Human Services, 2008) that recommend moderate-intensity PA of at least 30 min/day on five or more days/week, or a minimum of 150 min/week overall. Results were consistent across continuous and binary measures of PA (average 7-day PAR levels vs. proportion meeting PA guide-

lines of 150 min/week), which is reassuring, since the former can be susceptible to the influence of outliers.

On motivational readiness for PA, the outcome most closely related to the theoretical basis underlying the intervention, we found strong intervention effects that were maintained throughout the 12-month study period. In particular, intervention participants outperformed control participants in terms of moving from Con- templation/Preparation at study entry to Action/Maintenance at follow-up, a change in motivational readiness previously associ- ated with fitness improvements (Marcus & Simkin, 1993). The apparent discrepancy in the strength of intervention effects on self-reported PA levels and on motivational readiness for PA at 12 months may be due to differences over the reference assessment period (the previous week in the PAR vs. the previous 6 months for moving to the Action/Maintenance stage of motivational readiness for PA). As PA levels were elevated in the intervention group at 6 months relative to 12 months, motivational readiness at 12 months may be capturing PA increases at the previous assessment point not included in the 7-day PAR administered at 12 months.

The only other known study in which HCPs provided PA advice to breast cancer patients, had effects assessed at 1 and 5 weeks (Jones et al., 2004), so it is difficult to compare the results across studies, but it is clear that our study—which followed patients for much longer—found positive effects of HCP advice plus telephone counseling on PA at 3 months and 6 months. When considering telephone-based interventions and short-term effects (3 months), stronger effects on PA were found in our previous 12-week tele- phone counseling intervention among breast cancer patients (Pinto, Frierson, et al., 2005). Significant effects on PA were also found in previous telephone counseling studies among breast cancer pa- tients at 6 weeks (Mock et al., 1997) and at 12 weeks (Matthews et al., 2007). In studies using other intervention approaches such as the effects of exercise recommendations alone, print materials alone, pedometers alone, and a combination of print materials and pedometers among breast cancer survivors (Vallance, Courneya, Plotnikoff, Yasui, & Mackey, 2007), larger group differences (39 to 57 min/week across groups) were found at 3 months. When considering PA outcomes at 12 months, a group difference of 13 min was achieved in a sample of 641 overweight, older long-term cancer survivors who received a 12-month PA and dietary inter- vention via telephone and print materials or a delayed intervention (Morey et al., 2009). These interventions did not involve the HCP, and overlooking the HCP may present a missed opportunity for supporting a healthy behavior such as exercise.

It is clear that the significant intervention effects in helping breast cancer survivors meet PA guidelines at 3 months and 6 months dissipated at 12 months. One call/week over 12 weeks had produced significant increases in PA in a prior study among breast cancer survivors (Pinto, Frierson, et al., 2005). We had reduced the number of calls to eight in this trial which may account for weaker effects. Another explanation is that the inability to detect between- groups differences was driven by the increased PA reported by control participants over time. Though intriguing, this increase should not be interpreted to suggest that brief advice from HCPs is sufficient to increase long-term PA, because control participants received not only HCP advice, but also similar frequency of contact with research staff as intervention participants. This was done to provide a more conservative test of the intervention. However, it is possible that the contacts kept PA salient for control

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

623PHYSICAL ACTIVITY INTERVENTION

participants and reduced the ability to detect differential interven- tion effects. The true test of this explanation would involve a 3-arm study: HCP advice plus Telephone Counseling, HCP advice plus Contact Control, and HCP Advice alone.

Study goals included examining intervention effects on psycho- social outcomes. Group differences in fatigue were nonsignificant, and the intervention effects on self-reported physical functioning were not maintained past 3 months. Our study participants were not screened for high levels of fatigue and/or poor physical func- tioning. Mean fatigue scores at baseline were similar to those in another PA trial for breast cancer patients initiating adjuvant chemotherapy in which significant improvement in fatigue in the PA group was not found (Courneya et al., 2007). Both study groups showed improvements in fatigue, and in the absence of a control group that received no intervention, these results are in- conclusive. The strength of the effect size of exercise interventions on cancer patients’ fatigue has been found to be inconsistent and highly heterogeneous across studies (0.06 –2.26), and it may be linked to a “take all comers” approach, that is, patients in the studies may have had low fatigue levels (Speck et al., 2010). Similarly, our study sample’s physical functioning was high at baseline (compared with normative data; Ware, Kosinski, & Dewey, 2000), suggesting possible “ceiling” effects.

The higher attrition at 3 months among participants receiving telephone counseling rather than contact control (16.1% vs. 2.3%) was surprising (see Figure 1 for reasons), and suggests that study demands may have been too burdensome for some breast cancer participants. Although higher attrition among intervention partic- ipants is not uncommon (Dubbert, Morey, Kirchner, Meydrech, & Grothe, 2008), retention was at 94% in a previous trial using telephone counseling (12 weekly calls in a 3-month intervention) to promote PA among breast cancer patients (Pinto, Frierson, et al., 2005).

The association of working full-time and increased PA at 3 months (but not thereafter) was surprising. Finding time to exer- cise is often a barrier for individuals who work, and this barrier may be stronger among women who also have household respon- sibilities (Dishman, 1990). But it is also possible that the women who worked full-time may have had better health and fewer comorbidities than those who were not working full-time.

This study, which is one of the first to promote PA in collabo- ration with oncology follow-up, clearly showed that motivated HCPs were able to provide brief advice to their patients (98% completed chart prompts). The duration of advice was brief, as intended, and participants were generally satisfied with the advice. The advice was associated with short-term and long increases in PA in both groups who received calls focusing either on PA or contact control. The improvement in PA in the control arm was surprising, but it may also represent the growing awareness of the relevance of PA for cancer recovery. Future studies may want to test the efficacy of HCP advice in the absence of a contact control arm on short-term and long-term outcomes. If such studies also focus on psycho-social effects such as fatigue and physical func- tioning, it is important to also recruit patients who report high levels of fatigue and low physical functioning in order to avoid “floor” and “ceiling” effects.

Study limitations include an actively recruited sample of pa- tients who were able to obtain physician consent and were willing to be randomized. The sample was relatively homogeneous with

regard to race/ethnicity and socioeconomic status limiting the generalizability of the findings. HCPs were asked to provide brief exercise advice to patients during a follow-up visit, but we were not able to assess whether advice was provided at subsequent follow-up visits, which may be a confounder. Another drawback is that the measures of PA were based on self-report. While we included a conservative contact control group (that may have inadvertently kept PA salient for the CC arm), there was no true control group in the study. Finally, it is possible that additional effects might have been detected on self-reported physical func- tioning had the sample included women with poorer functioning at baseline.

Strengths of the study include a large sample size of women within 5 years of a breast cancer diagnosis, documented delivery of HCP advice, use of several standardized measures of PA, motiva- tional readiness and psycho-social outcomes, a theoretically based intervention, and follow-up assessments at 6 months and 12 months. Our results show that among motivated volunteer HCPs, providing brief advice was feasible in the context of a follow-up visit, and when this advice was supplemented by telephone coun- seling, patients’ PA participation increased for at least 6 months. HCP advice is perceived as credible to patients and if the advice is kept brief and does not take valuable time from the HCP-patient encounter, it is not likely to be burdensome in the health care setting. While we cannot be sure that HCP advice alone would suffice (our study design does not allow us to draw that conclu- sion), our results suggest that the HCP advice will require supple- mentation to support the adoption and maintenance of PA in this patient population. There is scope for examining whether this type of intervention can be implemented in large health care systems where cancer patients are monitored for follow-up care.

References

Agency for Healthcare Research and Quality. (2004). Effectiveness of behavioral interventions to modify physical activity behaviors in general populations and cancer patients and survivors (Evidence Report/ Technology Assessment, No. 102). Retrieved from www.Ahrq.gov/ clinic/epcsums/pacansum.htm

Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice Hall.

Bennett, J. A., Lyons, K. S., Winters-Stone, K., Nail, L. M., & Scherer, J. (2007). Motivational interviewing to increase physical activity in long- term cancer survivors: A randomized controlled trial. Nursing Research, 56, 18 –27. doi:00006199-200701000-00003[pii]

Blair, S. N., Haskell, W. L., Ho, P., Paffenbarger, R. S., Jr., Vranizan, K. M., Farquhar, J. W., & Wood, P. D. (1985). Assessment of habitual physical activity by a seven-day recall in a community survey and controlled experiments. American Journal of Epidemiology, 122, 794 – 804.

Castro, C. M., & King, A. C. (2002). Telephone-assisted counseling for physical activity. Exercise and Sport Sciences Reviews, 30, 64 – 68. doi:10.1097/00003677-200204000-00004

Courneya, K. S., Segal, R. J., Mackey, J. R., Gelmon, K., Reid, R. D., Friedenreich, C. M., . . . McKenzie, D. C. (2007). Effects of aerobic and resistance exercise in breast cancer patients receiving adjuvant chemo- therapy: A multicenter randomized controlled trial. Journal of Clinical Oncology, 25, 4396 – 4404. doi:JCO.2006.08.2024[pii]10.1200/JCO .2006.08.2024

Daniels, M. J., & Hogan, J. W. (2008). Missing data in longitudinal studies. Boca Raton, FL: Chapman & Hall/CRC. doi:10.1007/S11749- 009-0141-2

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

624 PINTO, PAPANDONATOS, AND GOLDSTEIN

Demark-Wahnefried, W., Clipp, E. C., Morey, M. C., Pieper, C. F., Sloane, R., Snyder, D. C., & Cohen, H. J. (2006). Lifestyle intervention devel- opment study to improve physical function in older adults with cancer: Outcomes from project LEAD. Journal of Clinical Oncology, 24, 3465– 3473. doi:24/21/3465[pii]10.1200/JCO.2006.05.7224,1532928

de Vries, H., & Brug, J. (1999). Computer-tailored interventions motivat- ing people to adopt health promoting behaviours: Introduction to a new approach. Patient Education and Counseling, 36, 99 –105. doi:10.1016/ S0738-3991(98)00127-X

Dishman, R. K. (1990). Determinants of participation in physical activity. In C. Bouchard, R. J. Shephard, T. Stephens, J. R. Sutton, & B. D. McPherson (Eds.), Exercise, fitness and health (pp. 75–102). Cham- paign, IL: Human Kinetics.

Dubbert, P. M., Morey, M. C., Kirchner, K. A., Meydrech, E. F., & Grothe, K. (2008). Counseling for home-based walking and strength exercise in older primary care patients. Archives of Internal Medicine, 168, 979 – 986. doi:10.1001/archinte.168.9.979

Eakin, E. G., Lawler, S. P., Vandelanotte, C., & Owen, N. (2007). Tele- phone interventions for physical activity and dietary behavior change: A systematic review. American Journal of Preventive Medicine, 32, 419 – 434. doi:S0749-3797(07)00010-4[pii]10.1016/j.amepre.2007.01.004

Galṽao, D. A., & Newton, R. U. (2005). Review of exercise intervention studies in cancer patients. Journal of Clinical Oncology, 23, 899 –909. doi:10.1200/JCO.2005.06.085

Goldstein, M. G., Pinto, B. M., Marcus, B. H., Lynn, H., Jette, A. M., Rakowski, W., . . . Tennstedt, S. (1999). Physician-based physical activity counseling for middle-aged and older adults: A randomized trial. Annals of Behavioral Medicine, 21, 40 – 47. doi:10.1007/BF02895032

Goode, A. D., Reeves, M. M., & Eakin, E. G. (2012). Telephone-delivered interventions for physical activity and dietary behavior change: An updated systematic review. American Journal of Preventive Medicine, 42, 81– 88. doi:S0749-3797(11)00742-2[pii]10.1016/j.amepre.2011.08 .025

Insightful Corporation. (2007). S-Plus 8 for Unix user’s guide. Seattle, WA: Author.

Institute of Medicine and the National Research Council. (2006). From cancer patient to cancer survivor: Lost in transition. Washington, DC: National Academies Press.

Jones, L. W., Courneya, K. S., Fairey, A. S., & Mackey, J. R. (2004). Effects of an oncologist’s recommendation to exercise on self-reported exercise behavior in newly diagnosed breast cancer survivors: A single- blind, randomized controlled trial. Annals of Behavioral Medicine, 28, 105–113. doi:10.1207/s15324796abm2802_5

Knols, R., Aaronson, N. K., Uebelhart, D., Fransen, J., & Aufdemkampe, G. (2005). Physical exercise in cancer patients during and after medical treatment: A systematic review of randomized and controlled clinical trials. Journal of Clinical Oncology, 23, 3830 –3842. doi:10.1200/JCO .2005.02.148

Little, R. J. A., & Rubin, D. B. (2002). Statistical analysis with missing data (2nd ed.). New York: J. Wiley & Sons.

Marcus, B. H., Bock, B. C., Pinto, B. M., Forsyth, L. H., Roberts, M. B., & Traficante, R. M. (1998). Efficacy of an individualized, motivationally-tailored physical activity intervention. Annals of Behav- ioral Medicine, 20, 174 –180. doi:10.1007/BF02884958

Marcus, B. H., Rossi, J. S., Selby, V. C., Niaura, R. S., & Abrams, D. B. (1992). The stages and processes of exercise adoption and maintenance in a worksite sample. Health Psychology, 11, 386 –395. doi:10.1037/ 0278-6133.11.6.386

Marcus, B. H., & Simkin, L. R. (1993). The stages of exercise behavior. Journal of Sports Medicine and Physical Fitness, 33, 83– 88.

Matthews, C. E., Wilcox, S., Hanby, C. L., Der Ananian, C., Heiney, S. P., Gebretsadik, T., & Shintani, A. (2007). Evaluation of a 12-week home- based walking intervention for breast cancer survivors. Supportive Care in Cancer, 15, 203–211. doi:10.1007/s00520-006-0122-x

McHorney, C. A., Ware, J. E., Jr., & Raczek, A. E. (1993). The MOS 36-Item Short-Form Health Survey (SF-36): II. Psychometric and clin- ical tests of validity in measuring physical and mental health constructs. Medical Care, 31, 247–263. doi:10.1097/00005650-199303000-00006

Mock, V., Dow, K. H., Meares, C. J., Grimm, P. M., Dienemann, J. A., Haisfield-Wolfe, M. E., . . . Gage, I. (1997). Effects of exercise on fatigue, physical functioning, and emotional distress during radiation therapy for breast cancer. Oncology Nursing Forum, 24, 991–1000.

Morey, M. C., Snyder, D. C., Sloane, R., Cohen, H. J., Peterson, B., Hartman, T. J., . . . Demark-Wahnefried, W. (2009). Effects of home- based diet and exercise on functional outcomes among older, overweight long-term cancer survivors: Renew: A randomized controlled trial. Jour- nal of the American Medical Association, 301, 1883–1891. doi:301/18/ 1883[pii]10.1001/jama.2009.643,2752421

Pinto, B. M., Frierson, G. M., Rabin, C., Trunzo, J. J., & Marcus, B. H. (2005). Home-based physical activity intervention for breast cancer patients. Journal of Clinical Oncology, 23, 3577–3587. doi:23/15/ 3577[pii]10.1200/JCO.2005.03.080

Pinto, B. M., Goldstein, M. G., Ashba, J., Sciamanna, C. N., & Jette, A. (2005). Randomized controlled trial of physical activity counseling for older primary care patients. American Journal of Preventive Medicine, 29, 247–255. doi:S0749-3797(05)00278-3[pii]10.1016/j.amepre.2005 .06.016

Pinto, B. M., Trunzo, J. T., Reiss, P., & Shiu, S. (2002). Exercise partic- ipation after diagnosis of breast cancer: Trends and effects on mood and quality of life. Psycho-Oncology, 11, 389 – 400.

Prochaska, J. O., & DiClemente, C. C. (1983). Stages and processes of self-change of smoking: Toward an integrative model of change. Journal of Consulting and Clinical Psychology, 51, 390 –395. doi:10.1037/0022- 006X.51.3.390

Sabatino, S. A., Coates, R. J., Uhler, R. J., Pollack, L. A., Alley, L. G., & Zauderer, L. J. (2007). Provider counseling about health behaviors among cancer survivors in the United States. Journal of Clinical On- cology, 25, 2100 –2106. doi:10.1200/JCO.2006.06.6340

Sallis, J. F., Haskell, W. L., Wood, P. D., Fortmann, S. P., Rogers, T., Blair, S. N., & Paffenbarger, R. S., Jr. (1985). Physical activity assess- ment methodology in the Five-City Project. American Journal of Epi- demiology, 121, 91–106.

Sarkin, J. A., Campbell, J., Gross, L., Roby, J., Bazzo, S., Sallis, J., & Calfas, K. (1997). Project grad seven day physical activity recall inter- viewer’s manual. Medicine and Science in Sports and Exercise, 29, S91–102.

Speck, R. M., Courneya, K. S., Masse, L. C., Duval, S., & Schmitz, K. H. (2010). An update of controlled physical activity trials in cancer survi- vors: A systematic review and meta-analysis. Journal of Cancer Survi- vorship: Research and Practice, 4, 87–100. doi:10.1007/s11764-009- 0110-5

U.S. Department of Health and Human Services. (1996). Physical activity and health: A report of the Surgeon General. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, U.S. Government Printing Office.

U.S. Department of Health and Human Services. (2008). 2008 physical activity guidelines for Americans. Retrieved from www.health.gov/ paguidelines

Vallance, J. K., Courneya, K. S., Plotnikoff, R. C., Yasui, Y., & Mackey, J. R. (2007). Randomized controlled trial of the effects of print materials and step pedometers on physical activity and quality of life in breast cancer survivors. Journal of Clinical Oncology, 25, 2352–2359. doi: 25/17/2352[pii]10.1200/JCO.2006.07.9988

Ware, J. E., Jr., & Sherbourne, C. D. (1992). The MOS 36-Item Short- Form Health Survey (SF-36). I. Conceptual framework and item selec- tion. Medical Care, 30, 473– 483. doi:10.1097/00005650-199206000- 00002

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

625PHYSICAL ACTIVITY INTERVENTION

Ware, K. E., Kosinski, M., & Dewey, J. E. (2000). How to score version 2 of the SF36® Health Survey. Lincoln, RI: QualityMetric Incorporated.

Winningham, M. (1993). Developing the symptom activity 27: An instru- ment to evaluate perception of symptom effects on activity. Oncology Nursing Forum, 20, 330.

Yellen, S. B., Cella, D. F., Webster, K., Blendowski, C., & Kaplan, E. (1997). Measuring fatigue and other anemia-related symptoms with the functional assessment of cancer therapy (FACT) measurement system.

Journal of Pain and Symptom Management, 13, 63–74. doi:10.1016/ S0885-3924(96)00274-6

Received July 27, 2011 Revision received February 28, 2012

Accepted March 9, 2012 �

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

626 PINTO, PAPANDONATOS, AND GOLDSTEIN