Abstract
BACKGROUND: Understanding the mechanisms of change has recently been emphasized as crucial for advancing research on preventive interventions. This study, embedded within a pilot trial, aimed to explore the mediating effects of resilience for three preventive interventions focusing on reducing the risk of adolescent depression and anxiety in Nepal. We hypothesized that the self-regulation, economic, and combined interventions would differentially increase internal and external resilience, which in turn would reduce adolescent depressive and anxiety symptoms among adolescents in Nepal, supporting to development of a testable, replicable mediation model of prevention intervention pathways. METHODS: Data were collected as part of a feasibility cluster-randomized controlled trial. A total of 229 adolescents aged 13-15, identified as living in poverty and at risk of depression or anxiety, were cluster-randomized by school into three intervention arms (self-regulation, economic, combined) and a control group. Assessments for depression and anxiety symptoms, and external and internal resilience were performed at three timepoints, with a six-month interval between each. Twenty school-based group intervention sessions were conducted weekly post-baseline. Linear mixed modeling explored changes in resilience within groups. Exploratory mediation analyses were performed to examine the association between interventions (as predictors), resilience at 6 months (as mediator), and symptoms of depression and anxiety at 12 months as outcomes, through two separate parallel mediation models. RESULTS: While no significant sensitivity to change effects for resilience was found, hypothesized directional time-related improvements were observed in external resilience for males in self-regulation and economic arms, and for females in the self-regulation arm; internal resilience showed positive trends for males in economic and combined arms, and for both males and females in the self-regulation arm. No significant effects of the interventions through resilience on mental health outcomes were detected, though internal resilience at 6 months predicted lower anxiety at 12 months, and for males in all intervention arms, higher internal resilience was significantly associated with lower depression. CONCLUSIONS: The findings warrant the testable conceptual mediation model with resilience as a mechanism in larger, fully powered prevention trials for adolescents. TRIAL REGISTRATION: ISRCTN14601588 https//doi.org/10.1186/ISRCTN14601588.