Psychosocial mediators of physical activity behavior among adults and children

Background: Researchers examining theory-based, physical activity (PA) interventions postulate that interventions are effective by changing theoretical constructs hypothesized to mediate the relationship between the intervention and PA behavior. Research indicates that PA interventions are effective for increasing PA behavior. However, whether effective interventions are due to predicted changes in theoretical constructs remains poorly understood. Methods: Studies that examined theoretical constructs (i.e., mediators) in PA interventions of adults or children, which used experimental designs and met other criteria for evaluating mediation, were collected via literature searches, personal searches of files, and personal communications. Only studies examining the direct effect of the intervention on the hypothesized mediator were considered relevant for this study. Results: Based on our criteria, the adult literature search yielded ten studies and the child literature search yielded two studies. The most common mediators examined included behavioral processes of change, cognitive processes of change, self-efficacy, decisional balance, social support, and enjoyment. Research indicates that behavioral processes are likely mediators. There was some support for the importance of self-efficacy as a mediator. Conclusions: Few studies have used statistically recommended methods to examine mediators in PA intervention studies. Therefore, definitive conclusions about the importance of the mediators reviewed are not possible at this time. Additional PA mediator-intervention studies using recommended statistical methods are necessary to truly test if theory-based PA interventions are effective due to predicted changes in theoretical constructs.