When someone commits an act of violence during a psychotic break, the toxicology screen rarely tests for the one drug the science keeps pointing to.
That is the argument at the center of a new Frontiers in Psychiatry perspective from Kenneth Finn, Robin Murray, Elizabeth Stuyt, and Karen Randall, who reviewed the epidemiological literature alongside 29 real-world violence cases, 17 with confirmed THC-positive toxicology and 12 where THC was never tested or never reported.
Their conclusion is blunt. “The convergence of evidence from epidemiological, neurobiological, and clinical studies supports a causal relationship between cannabis use and psychotic disorders and violence,” the authors write.
The numbers they assemble are hard to wave off. Roughly 34% of cannabis-induced psychosis cases go on to become schizophrenia, compared with 10% for alcohol. Among adolescents aged 12 to 19, regular use was tied to an 11.2-fold increased risk of psychotic disorders — and a 26.7-fold risk for cases severe enough to require hospitalization. Daily users of cannabis stronger than 10% THC faced roughly five times the risk of a psychotic disorder.
The potency piece matters, they argue, because the product has changed. Cannabis flower that ran 2–4% THC in the 1970s now averages around 20%, and concentrates reach 80–90%. In Canada, 71% of dried cannabis sold in 2022 exceeded 20% THC.
On violence specifically, the review cites a dose-response pattern: 3% of occasional users, 6% of regular users, and 11% of frequent users displayed violent behavior. Persistent users showed a seven-fold increase in odds of violent convictions. Among veterans, 27% of those with cannabis use disorder reported difficulty controlling violence, versus 8% without.
Post-legalization data tracks the same direction. Colorado youth psychosis hospitalizations rose from 21.9 to 32.3 per 100,000, and Canadian emergency visits for cannabis-induced psychosis climbed roughly 40%.
The authors’ proposed fix is procedural: make THC a mandatory part of toxicology screening in violent incidents, and target prevention at adolescents and people with psychiatric histories.