Original Articles
Three independent factors predicted adherence in a randomized controlled trial of resistance exercise training among prostate cancer survivors

https://doi.org/10.1016/j.jclinepi.2003.11.010Get rights and content

Abstract

Objective

To examine predictors of adherence in a randomized controlled trial of resistance exercise training (RET) in prostate cancer survivors receiving androgen deprivation therapy.

Study design and setting

A randomized controlled trial conducted at fitness centers in Ottawa and Edmonton, Canada. Prostate cancer survivors (n = 155) completed measures of social cognitive variables, quality of life (QOL), behavior, and fitness before being randomized to either an exercise (n = 82) or control (n = 73) group. The exercise group was asked to perform supervised RET three times per week for 12 weeks.

Results

The exercise group attended 28.2 of the 36 (78.3%) RET sessions. Univariate analyses revealed eight different significant (Ps < .05) predictors of exercise adherence including exercise stage of change, intention, age, QOL, fatigue, subjective norm, leg-press test, and perceived behavioral control. A multivariate analysis indicated that there were three independent predictors of adherence that explained 20.4% of the variance: exercise stage of change (β = 0.26; P = .013), age (β = −0.22; P = .037), and intention (β = 0.19; P = .073).

Conclusion

Exercise adherence in the trial was very good but not optimal. Adherence was predicted by variables from many different categories including social cognitive, QOL, behavioral, fitness, and demographic. These findings may have important implications for maximizing adherence during clinical trials of exercise in prostate cancer survivors.

Introduction

Researchers have used randomized controlled trial (RCT) methodology to test the utility of exercise interventions in cancer survivors (e.g., [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11]). A critical component to the success of these trials is achieving high levels of exercise adherence in the exercise group and low levels of exercise contamination in the control group [12], [13]. Exercise adherence may be defined as the extent to which the exercise group has performed the exercise prescription. Conversely, exercise contamination refers to the extent to which the control group has performed the exercise prescription. Theoretically, to provide an optimal test of the efficacy of an exercise intervention it is necessary to achieve 100% adherence in the exercise group and 0% contamination in the control group. Any deviation from these exercise rates may have profound implications for estimating the efficacy of the intervention. Unfortunately, research to date has reported significant problems with exercise adherence and contamination in RCTs of exercise in cancer survivors.

More specifically, exercise adherence rates in cancer survivors have typically ranged from 60 to 85% for both home-based (e.g., [2], [3], [7], [14], [15]) and supervised (e.g., [3], [4], [9], [11]) RCTs. These rates are comparable to those reported in exercise RCTs of older adults without cancer [16]. Exercise contamination rates, on the other hand, are rarely reported in RCTs, but have ranged from 20–50% in several home-based exercise programs in cancer survivors [2], [7], [15]. These data suggest that RCTs in cancer survivors have been hampered by less than optimal exercise adherence and contamination rates. Before any corrective action can be taken, however, it is necessary to have an understanding of the determinants of exercise adherence and contamination in RCTs of cancer survivors.

To date, only two studies have examined this issue in cancer survivors [12], [17]. Unfortunately, the studies were limited by small sample sizes, unsupervised home-based exercise programs, and reliance on self-reports of exercise. Moreover, given the heterogeneous nature of cancer survivor groups, cancer treatments, and exercise programs, the generalizability of the findings is unclear.

We have recently completed an RCT that examined the effects of a 12-week, supervised resistance exercise training (RET) program on fatigue, quality of life (QOL), and muscular strength in prostate cancer survivors receiving androgen deprivation therapy (ADT) [4]. The trial demonstrated clinically meaningful changes in these outcomes that favored the exercise group. We also reported a 78.3% adherence rate to the supervised RET program. The primary purpose of the present study was to examine predictors of adherence to the supervised RET program in the trial. A secondary purpose was to report the contamination rate. Such information should aid in the design and conduct of future trials in this population.

We adopted Ajzen's [18] Theory of Planned Behavior (TPB) to assess the social cognitive determinants of exercise. We also included traditional demographics (e.g., age, marital status), medical variables (e.g., cancer stage, treatments), behavioral/fitness indicators (e.g., past exercise, smoking, muscular strength), and QOL as additional predictors. Many of these variables have been associated with exercise behavior in the general population [19] and with exercise adherence in previous RCTs involving noncancer populations [16].

Our primary objective was to examine predictors of adherence to the supervised RET program in the exercise group. Based on previous research [12], [17], we expected exercise stage of change, employment status, treatment protocol, and perceived behavioral control (PBC) to be important predictors of adherence to the supervised RET program. Our secondary objective was to document the level of exercise contamination (i.e., exercise performed outside the formal supervised program) in both the exercise and control groups. Based on our supervised exercise protocol, we expected the contamination rate to be minimal [1].

Section snippets

Participants

The methods of the trial have been described elsewhere [4]. Briefly, the trial was conducted at the Ottawa Regional Cancer Center in Ottawa, Canada (coordinating center) and the Cross Cancer Institute in Edmonton, Canada. Approval was obtained from the ethics committees at both sites and all participants provided written informed consent. Participants were 155 prostate cancer survivors recruited between September 1999 and August 2001. Eligibility criteria for the study were: (1) histologically

Flow of participants through the study

Flow of participants through the trial is presented elsewhere [4]. Briefly, we randomized 155 of 507 (30.6%) eligible participants—82 to the exercise group and 73 to the control group. We had 8 drop outs in the exercise group (9.8%) and 12 (16.4%) in the control group (P = .790). Baseline data were available on all 155 participants and we had at least one measure of self-reported nonprogram exercise (i.e., the 6 or 12 week assessment) for 151 of 155 (97.4%) participants. Last observation carried

Discussion

Our adherence rate to the supervised RET program during the trial was 78.3%. In univariate analyses, eight different variables predicted exercise adherence: preprogram overall exercise stage, intention, age, QOL, fatigue, subjective norm, leg-press test, and PBC. In a multivariate analysis, the independent predictors of exercise adherence were preprogram overall exercise stage, age, and intention. We also found minimal contamination in the control group but a significant decline in nonprogram

Acknowledgements

This study was funded by the National Cancer Institute of Canada (NCIC) with funds from the Canadian Cancer Society (CCS). K.S. Courneya was supported by an Investigator Award from the Canadian Institutes of Health Research and a Research Team Grant from the NCIC with funds from the CCS and the CCS/NCIC Sociobehavioral Cancer Research Network. R.D. Reid was supported by a New Investigator Award from the Heart and Stroke Foundation of Canada. We thank Todd Bobick, MSc, and Neil Eves, MSc, for

References (40)

  • V. Mock et al.

    Fatigue and quality of life outcomes of exercise during cancer treatment

    Cancer Pract

    (2001)
  • F. Dimeo et al.

    An aerobic exercise program for patients with haematological malignancies after bone marrow transplantation

    Bone Marrow Transplant

    (1996)
  • F.C. Dimeo et al.

    Aerobic exercise in the rehabilitation of cancer patients after high dose chemotherapy and autologous peripheral stem cell transplantation

    Cancer

    (1997)
  • F.C. Dimeo et al.

    Effects of physical activity on the fatigue and psychologic status of cancer patients during chemotherapy

    Cancer

    (1999)
  • K.S. Courneya et al.

    Correlates of adherence and contamination in a randomized controlled trial of exercise in cancer survivors: an application of the theory of planned behavior and the five factor model of personality

    Ann Behav Med

    (2002)
  • M. Pickett et al.

    Adherence to moderate-intensity exercise during breast cancer therapy

    Cancer Pract

    (2002)
  • A.L. Schwartz

    Exercise and weight gain in breast cancer patients receiving chemotherapy

    Cancer Pract

    (2000)
  • K.S. Courneya et al.

    A randomized trial of exercise and quality of life in colorectal cancer survivors

    Eur J Cancer Care

    (2003)
  • K.A. Martin et al.

    Who will stay and who will go? A review of older adults' adherence to randomized controlled trials of exercise

    J Aging Phys Activity

    (2001)
  • K.S. Courneya et al.

    Pedictors of adherence and contamination in a randomized trial of exercise in colorectal cancer survivors

    Med Sci Sports Exerc

    (2003)
  • Cited by (129)

    • Baseline Health-related Quality of Life Predicts Bladder Cancer–specific Survival Following Radical Cystectomy

      2022, European Urology Focus
      Citation Excerpt :

      These results further underline the independent prognostic value of general baseline HRQOL measured as GHS for CSS, OS, and RFS for patients with bladder cancer treated with RC. As previously hypothesised, the prognostic value might originate from the fact that higher HRQOL is associated with better adherence to treatment strategies and healthy behaviour by patients, with a conceivable positive impact on survival outcomes [22]. Another possible explanation is that HRQOL assessed via patient-reported outcome measures (PROMs) may better reflect survival-related patient functioning and wellbeing than common clinical measures.

    View all citing articles on Scopus
    View full text