Association Between Self-reported Physical Activity And Social Cognitive Theory Constructs Among People With Multiple Sclerosis

This study was an exploratory, secondary analysis of data from a pragmatic, multisite, randomized controlled trial on a 3-month tele-exercise intervention for people with multiple sclerosis (MS). PURPOSE: To examine the association between (1) baseline social cognitive theory (SCT) variables and self-reported physical activity (PA) level at 3-month (post-intervention) and 6-month (follow-up); and (2) 3-month SCT variables and PA level at 6-month. METHODS: The analysis included 377 persons with MS who completed a 3-month intervention and received SCT-based educational materials and automated calls over a 1-year period. Outcomes included self-reported moderate-to-vigorous PA using the Godin Leisure-Time Exercise Questionnaire (GLTEQ), and SCT variables of exercise self-efficacy, outcome expectations, and goal setting. The associations were examined using Spearman (rs) and partial Spearman correlations (prs), adjusted for age, sex, MS type, disease severity, and clinic allocation. The prediction analysis was performed using logistic regression and receiver operating characteristic (ROC) analyses to determine the optimal cutoff points for each predictor. Binary anchor groups were defined using the GLTEQ <24 vs. ≥ 24 (i.e., active vs. moderately/insufficiently active) at 3- and 6-month. RESULTS: Baseline and 3-month goal setting were the strongest (moderate effect) and statistically significant variable with PA level at 3- (prs = 0.28) and 6-month follow-ups (prs = 0.32), respectively (p < 0.001 for all). Next, SCT variables showing significant associations with PA level were entered into multivariable logistic regression models. A covariates-adjusted ROC analysis demonstrated that baseline goal setting had an area under the curve (AUC) of 0.784 with an estimated cut-off point of 26, whereas 3-month goal setting had an AUC of 0.710 with an estimated cut-off point of 33. CONCLUSION: This study identified that goal setting was associated with PA outcome and the level of success (i.e., active vs. moderately/insufficiently active) after an intervention. The findings may inform the design and delivery of future interventions by optimizing program levels and types of support based on SCT to maximize treatment efficacy for people with MS.