Purpose: The objective of this investigation was to examine the mechanisms of physical activity and dietary behavior change in the Program X intervention. Methods: Program X involved a clustered randomized controlled design with six schools (N = 124 participants, mean age = 14.1 ± 0.8 years) randomized to intervention or control conditions for the 6-month study period. Physical activity and fruit and vegetable consumption were assessed using pedometers and questionnaires, respectively. The theoretical framework of the intervention was assessed using structural equation modeling, mediation, and moderation analyses. Results: The model explained 56% of the variance in physical activity at 6-months, but did not represent a good fit to the data, χ2 = 87.43, df = 19, p < .001. The model explaining fruit and vegetable consumption explained 31% of the variance and provided a good fit to the data, χ2 = 12.40, df = 10, p = .259. None of the variables satisfied the criteria for mediation or moderation in the physical activity model, but gender moderated the effects of intervention on fruit and vegetable consumption. Conclusions: None of the hypothesized mediators were responsible for behavior change in the Program X intervention. Future studies should address the limitations of existing psychosocial scales and continue to explore the mechanisms of behavior change using model testing, mediation, and moderation analyses.
Journal of Adolescent Health Vol. 47, Issue 1, p. 83-91