Longitudinal associations between violence exposure and adolescent conduct problems in a high‐adversity, South African setting

Background Violence exposure is a well‐established risk factor for adolescent conduct problems, yet longitudinal research in high‐adversity, low‐ and middle‐income countries (LMICs) remains limited. This study investigated whether early adolescent violence exposure predicts concurrent and longer‐term conduct problems, and explored potential bidirectional associations and sex differences in a peri‐urban South African community with high rates of poverty and violence. Methods Data were drawn from the Thula Sana birth cohort ( n = 357; 51.5% female), a longitudinal intervention study in Khayelitsha, South Africa. Adolescents were assessed at early (ages 12–14) and late adolescence (ages 16–19). Violence exposure was measured using adolescent self‐report. Conduct problems were measured using adolescent and caregiver report in early adolescence and adolescent self‐report in late adolescence. Multiple linear regressions tested cross‐sectional associations, and cross‐lagged panel models examined longitudinal and bidirectional associations, adjusting for contextual adversity and intervention status. Missing data were addressed using multiple imputation, and findings were confirmed through sensitivity analyses. Results Violence exposure was associated with higher concurrent conduct problems in early adolescence (β = .15–.19, p .05). Interaction analyses did not provide evidence that associations differed by sex. Conclusions Violence exposure in early adolescence represents a prospective risk factor for conduct problems in a high‐adversity South African setting. Findings highlight the importance of early, contextually grounded violence prevention and the need for further research to test sex‐specific pathways and inform the development of gender‐responsive intervention strategies.