Cannabis use is associated with increased risk of violence: A systematic review and meta-analysis.

Trotta G.; Rodriguez V.; Marino P.; Gurmesa M.; Spinazzola E.; Li Z.; Alameda L.; Di Forti M.; Murray R.; Vassos E. · Psychological medicine · 2026

**Background & Methods** This PRISMA-compliant meta-analysis synthesized evidence from 63 peer-reviewed studies (N=265,079) identified through MEDLINE, EMBASE, and PsycINFO searches through November 2024 to examine the association between cannabis use and violence as either a perpetrator or victim. Random-effects models were employed to estimate odds ratios across both psychiatric and general populations, stratified by study design (longitudinal vs. cross-sectional) and violence outcome severity. **Key Findings** • Cannabis users demonstrated significantly elevated risk of perpetrating violence in both psychiatric patients (OR 2.49, 95% CI 1.72-3.61) and general populations (OR 2.05, 95% CI 1.74-2.41), with effects consistent across longitudinal (OR 1.17, 95% CI 1.07-1.28) and cross-sectional (OR 2.37, 95% CI 1.66-3.40) study designs. • Violence resulting in criminal conviction showed substantially larger effect sizes, with odds ratios of 3.75 (95% CI 2.54-5.53) in general populations and 4.21 (95% CI 2.86-6.22) in psychiatric populations. • Cannabis use was significantly associated with increased victimization risk (OR 1.49, 95% CI 1.38-1.60), with disproportionate effects observed in female and mixed-gender cohorts compared to male-only samples. **Dosage & Administration** Not reported. **Safety & Adverse Effects** Not reported. (Note: The study focuses on violence outcomes rather than direct adverse health effects. Cannabis-related psychosis and aggression are mentioned as contextual mechanisms.) **Evidence Quality** This meta-analysis represents high-level evidence synthesizing 63 studies with substantial sample size (N=265,079). However, limitations include: heterogeneity inherent in aggregating cross-sectional and longitudinal designs with varying violence definitions; potential for unmeasured confounding in observational studies; publication bias; inability to establish definitive causality; and variable adjustment for confounders across included studies. The notably larger effects in criminal conviction outcomes versus general violence perpetration warrant cautious interpretation regarding measurement reliability. The inverse relationship between cannabis use and male victimization specifically requires further mechanistic investigation. These findings support routine cannabis assessment in psychiatric settings but necessitate replication and investigation of dose-response relationships and causal pathways.

More recent cannabis research