Safety, adherence, and compliance in a RCT of supervised exercise training among persons with multiple sclerosis who have slowed cognitive processing speed

Background: Safety, adherence, and compliance have been poorly-characterized in randomized controlled trials (RCTs) of exercise training within multiple sclerosis (MS) research. The MoXFo initiative recently proposed standard criteria for defining exercise safety, adherence, and compliance, yet these criteria have infrequently been applied within an exercise training RCT involving people with MS. Objective: This study applied published criteria for characterizing safety, adherence, and compliance within a RCT that compared 12-weeks of supervised treadmill walking exercise (TMWX) training (intervention condition) with 12-weeks of stretching-and-toning (active control condition) among fully-ambulatory persons with MS who demonstrated slowed cognitive processing speed (CPS). Methods: 37 fully-ambulatory persons with MS with slowed CPS were randomly assigned into the TMWX condition (n = 18) or active control condition (n = 19). We defined safety as number of adverse events (AEs); adherence as number and rate of sessions attended; and compliance as number and rate of sessions completed as prescribed. We further examined potential baseline predictors of adherence and compliance rates using correlations and regressions. Results: There were 10 non-serious AEs that were generally unrelated to the conditions (e.g., non-injurious falls that took place outside of the study, temporary illnesses). On average, participants attended 25.8 of 36 possible sessions (i.e., adherence rate=71.5 %). On average, participants completed 25.2 of the 36 possible sessions as prescribed (i.e., compliance rate = 70.0 %) and 97.7 % of attended sessions as prescribed. Of note, adherence did not differ by condition, though relative compliance rate was significantly higher among those who underwent the control condition (p = .03). In the overall sample, regressions identified no significant predictors of adherence, but higher agreeableness predicted higher relative compliance rate. Within the TMWX condition, longer disease duration and lower anxiety predicted higher adherence, and lower impact of MS on physical and mental domains and higher agreeableness predicted higher relative compliance rate. Within the control condition, lower brainstem functional systems scores and lower neuroticism predicted higher adherence, and no baseline variables predicted relative compliance rate. Conclusions: The safety, adherence, and compliance data, as defined based on recent recommendations, suggest that the TMWX and stretching conditions are safe, feasible, and attainable among fully-ambulatory, persons with MS with slowed CPS. The heterogeneous pattern of baseline variables that predicted adherence and compliance rates overall and in the respective experimental conditions warrant replication in larger MS samples.