Research on violent behavior among patients with schizophrenia has mainly focused on risk factors, while protective factors have been less systematically synthesized. Guided by the social-ecological model, this study integrated and examined evidence on protective and risk factors associated with violent behavior. Ten Chinese- and English-language databases were systematically searched from inception to January 26, 2026. Factors were classified according to the social-ecological model. Three-level random-effects meta-analyses were performed, and methodological quality assessment, publication bias assessment, moderator analyses, and sensitivity analyses were conducted. A total of 40 studies were included. Across these studies, 45 factors were identified, and 39 were included in the meta-analysis, including 20 protective factors and 19 risk factors. Overall, 22 factors were significantly associated with violent behavior: 11 protective factors with lower violent behavior and 11 risk factors with higher violent behavior. The strongest risk associations were observed for history of violence or criminal offending, depressive symptoms, history of suicide attempts, and substance/alcohol use problems. Family-level risk factor included family history of psychiatric disorders. Protective factors associated with lower violent behavior included longer illness duration, longer education, female sex, higher social functioning, not being single, employment, not living alone, timely treatment, treatment adherence, longer interval between clinical visits, and active help-seeking from professionals; treatment adherence showed the strongest protective association. Significant risk factors were mainly concentrated at the individual and family levels, whereas significant protective factors spanned individual, family, community/social, and health service levels. Three significant moderator effects were identified. This systematic review and three-level meta-analysis showed an asymmetric social-ecological distribution of factors associated with violent behavior among patients with schizophrenia. Classic risk exposures had the strongest associations, while protective associations were distributed across multiple levels and require further longitudinal evaluation regarding their stability, contextual dependence, and incremental value. CRD420261287987.
No takes yet. Share an insight, caveat, or question.
Yan et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: