Mediators of behavior change in two tailored physical activity interventions for adolescent girls

Objectives: It has been suggested that the lack of knowledge regarding the mechanisms responsible for behavior change may be responsible for the low levels of effectiveness in physical activity interventions among youth. While it is common for physical activity interventions to cite a theoretical framework, few test the validity of their constructs using an established mediation analysis technique. The purpose of this study was to identify mediators of physical activity behavior change in two tailored interventions for adolescent girls. Design: This study involved an experimental design. Methods: Participants (N = 161) were randomly allocated to a control (CON) group, an intervention based on the Health Promotion (HP) Model or an intervention developed from the HP Model that included two processes from the Transtheoretical Model (THP). Both interventions included school-based education sessions, individual counseling sessions, and two physical activity sessions completed with the participants' mothers. Measures were assessed prior to the intervention, at post-intervention and at a 6-month follow-up. The following constructs were included in the mediation analyses: perceived benefits, perceived barriers, self-efficacy, exposure to models, social support, interpersonal norms, planning, stimulus control, and counterconditioning. Results: Perceived benefits, perceived barriers, self-efficacy, and commitment to planning satisfied the criteria for mediation in the THP intervention. Self-efficacy and commitment to planning were identified as mediators in the III? intervention. Conclusion: The results of this study provide evidence that both interventions were successful in increasing physical activity through changes in the theoretical constructs. (c) 2007 Elsevier Ltd. All rights reserved.