Determinants of risk for adolescent self-harm and suicide: a systematic review and diagram of shared and unique factors in non-clinical samples

Suicide ranks as the second leading cause of death among individuals aged 10–19. Estimates suggest that one-quarter of adolescents have self-harmed at least once, with 1 in 25 of them dying by suicide within 10 years of their initial self-harm presentation. Despite these alarming statistics, most risk assessment tools are designed for adults in clinical settings; a gap in the understanding of shared and unique risk factors for self-injurious behaviours (SIBs) in community adolescents. To address this, we systematically reviewed observational studies examining risk factors for self-harming and suicidal behaviours in non-clinical adolescents. Our protocol adhered to PRISMA guidelines and was registered in PROSPERO (CRD42023463134), with the research question structured using the PICOS framework. We systematically searched EMBASE, Scopus, PubMed, PsycINFO, and Web of Science, supplemented by snowball sampling, and evaluated quality with the Newcastle-Ottawa Scale. Two reviewers independently screened and extracted data, with a senior researcher resolving conflicts. Data were analysed through narrative synthesis. From 1651 articles identified, 8 were included, with sample sizes ranging from 70 to 73,648. Thirty-seven risk factors were identified and visually conveyed through a Sankey diagram. Key factors include female sex, history of SIBs, and vulnerabilities across psychological, neural, familial and social domains. Although many of the studies were cross-sectional and lacked robust ethnic representation, our findings reinforce previous literature linking adolescent SIB to specific individual and contextual factors, contributing a systematic overview that underscores non-clinical adolescent risk and the need for both early identification and tailored prevention and treatment strategies.