Mediators of Physical Activity Behavior Change in the "Girls on the Move" Intervention

Background The minimal effect of interventions to date on increasing young adolescent girls' physical activity (PA) may be due to inadequate understanding of the mechanisms underlying behavior change, yet sparse research testing a PA intervention has examined the capacity of theories to explain PA, particularly when using objective measures. Objectives The aim of the study was to examine whether constructs from the health promotion model and self-determination theory mediated changes in moderate-to-vigorous physical activity (MVPA) following a 17-week intervention. Methods The study was a secondary analysis of data from a group randomized trial, including 12 intervention and 12 control schools in the Midwestern United States. Data were collected in 2012-2016. Girls (fifth- to eighth-grade, N = 1,519) completed surveys on perceived benefits and enjoyment of PA, PA self-efficacy, social support and motivation for PA, and barriers to PA and wore accelerometers. Results The final path model had a good fit: chi(2)(4) = 2.48, p = .648; goodness-of-fit index = 1; comparative fit index = 1; root-mean-square error of approximation = 0; standardized root-mean-square residual = 0.01. For MVPA change from baseline to postintervention, enjoyment (B = 24.48, p < .001) and social support (B = 30.48, p < .001) had a positive direct effect, whereas the intervention had a positive indirect effect through enjoyment and social support (B = 9.13, p < .001). Enjoyment (B = -13.83, p < .001) and social support (B = -17.22, p < .001) had a negative indirect effect on MVPA change from postintervention to follow-up. Discussion Enjoyment of PA and social support for PA may be important mediators of PA in young adolescent girls and warrant consideration when designing interventions.