Abstract
<title>Abstract</title> <p> Background Most research on moderators of preventive interventions focuses on between-group effects, providing limited insights into within-arm heterogeneity. This study investigates predictors of within-arm change in youth-reported internalizing symptoms in both arms of the PRODO trial, a randomized controlled depression-prevention trial for offspring of depressed parents. Methods The present study reanalyzed data from the PRODO trial, in which families were randomly assigned to GuG-Auf ( <italic>N</italic> = 50) or a no-intervention control condition ( <italic>N</italic> = 50). Assessment occurred at baseline and at 6, 9, and 15 months. Child and family factors predicting change in children’s internalizing symptoms were examined with Bayesian linear mixed models. Hierarchical clustering was employed to explore potential symptom trajectory patterns among children in the intervention arm. Results For children in the intervention arm, higher baseline symptoms were associated with more favorable symptom courses. In the control arm, lower baseline symptoms and male gender were associated with steeper symptom increases, albeit with substantial uncertainty. Age and parental depressive symptoms were unrelated to rates of symptom change in either arm. Gender showed no association in the intervention arm. Exploratory clustering of children’s internalizing symptoms in the intervention group revealed two clusters (increasing symptoms vs. stable or decreasing symptoms), although cluster solutions varied across imputations. Conclusion Baseline youth symptoms may be more informative for understanding treatment response to GuG-Auf than demographic or parental factors. Trial registration NCT02115880 (registered July 4, 2014). Clinical Trial Registration The trial was registered on July 4, 2014 at ClinicalTrials.gov (https//clinicaltrials.gov/study/NCT02115880). </p>