Unique features of peer relationships in specific developmental periods predict borderline personality disorder and major depressive disorder in late adolescence.

It is well established that peer relationships and psychopathology reciprocally influence each other across development, but it is less clear how trajectories of specific peer functioning domains are associated with borderline personality disorder (BPD) and major depressive disorder (MDD). We hypothesized that trajectories reflecting greater relational aggression would show stronger associations with later BPD symptoms, consistent with developmental models linking interpersonal conflict strategies to borderline features, and that lower peer acceptance, particularly from middle childhood through adolescence, would show stronger associations with later MDD symptoms. Using data from a 17-year prospective longitudinal study that oversampled preschoolers (ages 3-5) with elevated depressive and disruptive symptoms from the St. Louis region (N = 189; 50.8% female; 54.0% White, 34.4% Black), we examined how parent-reported bullying victimization, peer acceptance, and relational aggression- assessed up to 8 times via the MacArthur Health and Behavior Questionnaire (HBQ-P) across early childhood (ages 3-8) and middle childhood through adolescence (ages 9+) - predicted self-reported BPD symptoms (Borderline Personality Features Scale for Children) and interviewer-assessed MDD symptoms (Kiddie Schedule for Affective Disorders and Schizophrenia) in late adolescence (ages 13-21). Multilevel models estimated individual intercepts and slopes for each peer domain within each developmental period, and these trajectory parameters were entered as predictors of adolescent outcomes in general linear models, covarying for age, sex, and income-to-needs ratio. False discovery rate correction was applied to account for multiple comparisons. Elevated relational aggression in early childhood (intercept at age 6) uniquely predicted greater adolescent BPD symptoms after FDR correction. Increases in peer acceptance from middle childhood through adolescence were associated with fewer adolescent MDD symptoms, though this finding did not survive FDR correction. Bullying victimization trajectories were not significantly associated with either outcome. These findings suggest that specific peer functioning domains during distinct developmental windows may differentially relate to BPD and MDD risk, highlighting the potential value of developmentally informed screening and monitoring of peer relationship difficulties.