The Role of Family Context in Moderating Treatment Effectiveness for an Adolescent Family-Based Health Intervention

Elizabeth Marie Schneider, University of South Carolina


Family context may serve as an important factor in whether families initiate, implement, and maintain key practices that are necessary for long-lasting changes in health status. Family-based treatments taking a behavioral approach are common interventions for pediatric obesity and improving adolescent health outcomes. However, the family context in which these behavioral practices occur has received less attention. The current study examined family context in a randomized trial of African American adolescents and their caregivers (N = 73 adolescent-caregiver dyads), exposed to either an 8 week novel family-based intervention focusing on key family context variables in regards to specific health behaviors (physical activity [PA], sedentary behavior [SB], fruit and vegetable [FV] intake, and total fat gram consumption) or a comparison group condition. Family context in the current study was conceptualized by the following constructs: parental responsiveness and demandingness (reflective of parenting style); caregiver stress; and emotional climate. Moderate to vigorous physical activity (MVPA) was assessed through 7-day accelerometry, SB was assessed through a validated child self-report measure, and diet was assessed via a series of 24-hour dietary recall measures. Specifically, it was hypothesized that greater improvement in family context variables would be demonstrated in the family based intervention condition compared to the general education comparison group from baseline to 8-weeks post-intervention. Furthermore, it was hypothesized that family context variables would significantly interact with treatment group on demonstrating favorable changes in PA, SB, and diet. These moderated effects were expected to be greater for the intervention than the comparison condition, as the intervention condition specifically targeted variables relevant to family context. Results revealed significant main effects of the treatment for SB and total fat consumption with large and moderate effect sizes, respectively, in the expected directions. Only one significant moderator effect was revealed, which suggested that parental responsiveness moderated the main effect of the intervention on MVPA such that as parental responsiveness increased, the treatment effect was reduced. Lastly, moderator as outcome analyses were conducted and demonstrated change in family context from baseline to post-intervention only for parental responsiveness, reflective of a significant decrease in the treatment condition. Implications, study limitations, and recommendations for future research are discussed.