Modeling and Predicting Heterogeneity of Physical Activity Trajectories in a Physical Activity Intervention for Endometrial Cancer Survivors

Previous research shows that physical activity (PA) can improve physical functioning and quality of life in cancer patients and survivors. The current recommendations for PA in healthy individuals and cancer survivors is 150-250 min/week in order to prevent weight gain and receive other health benefits. PA interventions are not always effective for all participants, and most research examining for whom these interventions are most effective have mainly looked at moderating factors such as age, gender, and psychological factors. Few studies have attempted to examine unmeasured heterogeneity in PA outcomes. This study attempts to model this heterogeneity using growth mixture modeling (GMM) and predict different PA trajectories using baseline measures of somatic sensations, modeling, social support, and self-efficacy.

Participants included 98 women with Stage I, II, or IIIa endometrial cancer who were at least 6 months post-treatment and had no evidence of disease. Longitudinal data were collected over a 6 month time period at six different time points using ecological momentary assessment data for at home self-report measures of antecedents of self-efficacy, self-efficacy, and PA as well as accelerometry data for PA. After baseline assessment, participants were given individually tailored PA recommendations; engaging in moderate-intensity exercise for 30 minutes a day, 5 days a week. Between measurement time points, participants received telephone counseling which reviewed exercise goals and barriers, and provided brief teaching of behavioral and cognitive skills to support their PA behaviors. Data were analyzed using GMM to determine if there are different PA trajectories and logistic regression to predict class membership of the PA trajectories using baseline measures of somatic sensations, modeling, social support, and self-efficacy.

Results from the GMM of PA suggest participants can be classified into one of two PA trajectories; a high PA trajectory, where participants engage in a high level of PA that is stable over time and a rising PA trajectory, where participants engage in low levels of PA at baseline and increase over time. Logistic regression predicting these trajectories shows those with high levels of social support at baseline are over six times more likely to be part of the high PA trajectory.

These results suggest endometrial cancer survivors with low levels of social support are less likely to engage in PA at baseline, and may be most likely to benefit from an intervention that provides support through telephone coaching. Results also indicate certain participants are already engaging in high levels of PA and were unaffected by the intervention. Future studies should see if changes over time in social support lead to increases in PA.