Testing motivational and self-regulatory mechanisms of action on device-measured physical activity in the context of a weight loss maintenance digital intervention: A secondary analysis of the NoHoW trial

Abstract
Background: Weight-loss maintenance is challenging. To date, few weight loss maintenance digital behavior change interventions that focused on physical activity promotion used a sound intervention design grounded on a logic model underpinned by behavior change theories. The current study is a secondary analysis of the NoHoW trial and investigated whether changes in the hypothesized motivational and self-regulatory mechanisms of action operated by the intervention content at 6 months accounted for changes in objectively measured physical activity (PA) and sedentary behavior (SB) outcomes at 12 months. Methods: A subsample of 855 participants (M Age = 45.69 ± 11.37; 67.1% female) completed all questionnaires on motivational and self-regulatory variables and had all objectively measured PA and SB data available for 0, 6 and 12 months. We examined the direct and the indirect effects of Virtual Care Climate on post intervention changes in PA/SB (6 to 12 months) through changes in the motivational and self-regulatory mechanisms of action during the intervention period (0 to 6 months), as conceptualized in the NoHoW logic model. Results: Model 1 tested the mediation processes on Steps and presented a marginally poor fit to the data: Model 1 𝝌2/df = 12.72, p = 0.001; CFI = .770; RMSEA = .117. Model 2 tested mediation processes on SB and PA intensity levels and presented reasonable fit to the data: Model 2: 𝝌2/df = 8.12, p = 0.001; CFI = .959; RMSEA = .091. Specific indirect effects shown that Virtual Care Climate was positively and indirectly associated with self-regulatory capacities via two distinct paths: i) Intrinsic Goals and Autonomous Motivation, and ii) Basic Psychological Needs Satisfaction and Autonomous Motivation. In turn, Autonomous Motivation was positively and indirectly associated with increases in PA time (except for light intensity PA) and decreases in SB via Action Control. Only Action Control operated to directly predict long-term PA/SB outcomes. Conclusion: This is one of the first attempts to test a serial mediation model considering psychological mechanisms of change and objectively measured PA/SB in a 12-month longitudinal trial. Post intervention changes in PA/SB time (12 months) were predicted by a part of the overall hypothesized logic model. We found different pathways of influence on long term increases in PA and decreases in SB via motivational and self-regulatory mechanisms. These psychological mechanisms may represent possible targets for future interventions focusing on the promotion of long-term physical activity in adults trying to maintain their weight loss.
Year
2023
Category
Refereed journal