Potential moderators and mediators of intervention effects in an obesity prevention program for adolescent boys from disadvantaged schools

The purpose of this study was to explore potential moderators and mediators of intervention effects in the Physical Activity Leaders (PALs) obesity prevention program for adolescent boys from disadvantaged secondary schools. Design: group randomized controlled trial. One hundred adolescent boys (mean age = 14.3 (0.6) years) from four schools were randomized to the PALs intervention or a control group for the 6-month study period. The primary outcome was change in BMI and secondary outcomes were physical activity assessed using pedometers and constructs from Social Cognitive Theory (SCT) assessed using a questionnaire. Moderation analyses revealed a significant weight status interaction for BMI (p = 0.04), indicating stronger intervention effects for youth classified as overweight/obese at baseline. The intervention had a significant effect on resistance training self-efficacy (p < 0.001), but none of the SCT constructs satisfied the criteria for mediation. The number of intervention sessions attended was associated with changes in BMI (r = −0.38, p = 0.001), resistance training self-efficacy (r = 0.42, p < 0.001) and physical activity behavioral strategies (r = 0.26, p = 0.018). Changes in BMI were also associated with changes in resistance training self-efficacy (r = −0.21, p = 0.06) and physical activity behavioral strategies (r = −0.29, p = 0.009). A school-based intervention incorporating student leadership increased adolescent boys’ resistance training self-efficacy, but changes in physical activity were not detected and none of the SCT constructs satisfied the criteria for mediation. Baseline weight status was a moderator of intervention effect with the strongest intervention effects observed among overweight and obese adolescent boys.