Psychological, social and physical environmental mediators of the SCORES intervention on physical activity among children living in low-income communities

Objectives: This study examined mediation effects of socio-ecological variables on accelerometer determined physical activity among children in the Supporting Children's Outcomes using Rewards, Exercise and Skills (SCORES) intervention. Design: Randomized control trial mediation analysis. Methods: Primary school children were randomly assigned the SCORES multi-component intervention or control group for the 12-month study. The outcome was accelerometer derived total moderate-to vigorous physical activity (MVPA). Hypothesized mediators measured in children were physical activity enjoyment and perceived sport competence and social support. Hypothesized mediators measured in parents were family social support, access to physical activity facilities and equipment at home, and perceived access to physical activity opportunities in the local community. Mediation analyses were conducted using multi-level linear analysis in MPlus. Results: There were significant intervention effects for social support from teachers (A ¼ 1.73, SE ¼ 0.88, p ¼ 0.048) and parental perceived access to physical opportunities in the local community (A ¼ 2.69, SE ¼ 1.12, p ¼ 0.016). There were significant associations between changes in perceived sport competence (B ¼ 0.48, SE ¼ 0.36, p ¼ 0.027), parental perceived access to physical activity opportunities in the local community (B ¼ 0.60, SE ¼ 0.26, p ¼ 0.021), and changes in total MVPA. Parental perceived access to physical activity opportunities in the local community was found to have a significant mediating effect on total MVPA (AB ¼ 1.61, 95% CI ¼ 0.06 to 3.95, proportion 13%). Conclusions: Parental perceived access to physical activity opportunities in the local community is a potential mechanism of behavior change for children and mediated the effect of the SCORES intervention on objectively measured physical activity