OBJECTIVE: To assess the effectiveness of mental health and psychosocial support interventions for addressing post-traumatic stress symptoms in children and adolescents in humanitarian settings, with separate analyses of targeted and universal interventions. METHODS: We systematically searched MEDLINE, Web of Science, PTSDpubs and PsycInfo databases for relevant randomized controlled trials that involved individuals aged 18 years or younger in humanitarian settings and compared psychological interventions with passive controls. Random-effects meta-analyses were performed separately for interventions targeted to specific symptoms and for more general universal interventions. The review was registered on PROSPERO (CRD42023434878). FINDINGS: The meta-analysis included 16 trials of targeted interventions ( = 2356) and 11 of universal interventions ( = 3378) that met inclusion criteria. Children and adolescents who received targeted interventions reported significantly fewer post-traumatic stress symptoms after the intervention than controls. These positive effects were sustained at follow-up. Benefits were also observed for depressive symptoms. In addition, universal interventions were associated with significantly fewer post-traumatic stress symptoms. Moderator analyses showed that outcomes were not significantly influenced by country income level, the use of lay or professional therapists, or whether trauma was caused by human-made or natural disasters. However, considerable heterogeneity and a high risk of bias were noted across studies. CONCLUSION: Both targeted and universal mental health and psychosocial support interventions were associated with fewer post-traumatic stress symptoms among children and adolescents in humanitarian settings compared with passive controls. As interventions delivered by non-specialists were also successful, task-sharing could be considered where resources are scarce.