A new systematic review and meta-analysis in Psychological Medicine puts hard numbers on the cannabis violence risk. The authors pooled 63 peer reviewed studies covering 265,079 people. Cannabis users had significantly higher odds of committing violent acts. Giulia Trotta and colleagues from King’s College London and the South London and Maudsley NHS Foundation Trust searched the literature up to November 2024. Anyone debating cannabis policy should read their conclusion.
The timing matters. Professor Bobby Smyth, writing about the paper on LinkedIn, notes a recent flurry of reports and opinion pieces. They urge societies to dismantle drug laws in the name of women’s wellbeing. He calls this the latest gambit of a billionaire funded legalisation lobby. Whatever one makes of that framing, the new evidence on violence linked to cannabis use lands in the middle of the argument.
Inside the Cannabis Violence Risk Evidence
The headline numbers agree with each other. In the general population, cannabis users had odds of perpetrating violence of 2.05 (95 per cent confidence interval 1.74 to 2.41). Among psychiatric patients the figure was 2.49 (1.72 to 3.61). In plain terms, cannabis users had roughly twice the odds of committing violence as non-users in the same populations.
Cannabis use was also tied to a higher chance of becoming a victim of violence, with an odds ratio of 1.49 (1.38 to 1.60). That risk was higher for women (1.60) than for men (1.30). Across the populations studied, the picture is consistent.
Earlier reporting, including a July 2024 WRD News article, cited a meta-analysis of 30 studies and nearly 300,000 participants. It also found roughly twice the odds of violent behaviour among users. The new review draws on more than twice as many studies. It examines perpetrators and victims, yet it points the same way.
Convictions and Clinical Settings Tell the Starkest Story
Violence that led to a criminal conviction showed the largest effects. The odds ratio was 3.75 (2.54 to 5.53) in the general population and 4.21 (2.86 to 6.22) among psychiatric patients. Each figure rests on three studies. The authors call this finding particularly important. They point to the severity of the violence and the likely greater reliability of conviction data than self-report. By comparison, self-reported violence produced odds ratios of 2.02 and 2.18.
The clinical picture is just as serious. Four longitudinal studies of psychiatric populations produced a pooled odds ratio of 3.71 (2.73 to 5.05). The authors recommend that mental health professionals routinely assess cannabis use in people presenting with violent tendencies. They single out those with severe mental illness. Clinicians should also treat continued use as a modifiable risk factor. The paper adds that in England and Wales, substance misuse was common among people with schizophrenia who committed homicide. Cannabis was among the most frequently reported substances.
For clinicians, then, the cannabis violence risk is something to ask about and act on.
What the Authors Say About Causation
Honest reporting means noting what the paper does not claim. The authors state that the link probably reflects several interacting pathways, with no single direct cause. Effect sizes shrank in longitudinal studies of the general population, to 1.17 (1.07 to 1.28). Cross sectional work showed 2.37. Statistical heterogeneity was high, and tests suggested publication bias. The authors cannot exclude residual confounding by alcohol or other drugs. Most studies also lacked detail on how often people used cannabis and at what strength, so dose response could not be tested.
Yet the cannabis violence risk held up under the main checks the authors ran. Results stayed significant after they adjusted for possible publication bias. The longitudinal analyses, they say, suggest the association is not simply down to recall bias or reverse causation. They also note that most included studies predate the rise of high potency products. A risk factor does not need to be the only cause of harm to deserve action.
Heavier Use, Withdrawal and Rising Potency
Professor Smyth argues that drug use escalates when a drug such as cannabis is legalised. He points to North America, where intensive patterns of use rose in particular. Heavier patterns, he says, are the ones more strongly linked to violence. If he is right, the cannabis violence risk grows as use intensifies.
He also highlights cannabis withdrawal, now described in DSM 5 and ICD 11. Irritability is its most common symptom. In his view it likely drives much of the aggression tied to use. That includes violence against women, from intimate partner violence to aggression towards mothers in the home.
The review itself offers supporting mechanisms. Withdrawal can bring irritability, anxiety and restlessness. Regular use also raises the likelihood of paranoia. People may then believe they must defend themselves.
Potency adds to the concern. Writing for the Wall Street Journal, Allysia Finley observed that cannabis today is on average about four times as potent as in 1995. She added that concentrated dabs can contain 20 times as much THC as joints did in the 1960s. Finley also cited the finding that one in six people who start using before the age of 18 will develop cannabis addiction.
Where Harm Reduction Falls Short
Harm reduction asks how a person can use a drug more safely. The Trotta review raises a different question: who else pays? The authors point out that the health effects of cannabis on users have been extensively investigated. Considerably less attention has gone to harms to others, including interpersonal violence.
That gap undercuts the harmless plant message. Roughly twice the odds of violence among users cannot be squared with a claim that cannabis harms no one. Violence linked to cannabis use lands on partners, parents and strangers who never chose the risk.
Raised odds of violence appeared in intimate settings and beyond them. Perpetration within intimate or dating relationships carried odds of 1.22. Violence in other contexts reached 2.79. Smyth adds a sobering comparison. Alcohol, a legal drug, plays an immense role in violence across society, including gender based violence. In his view, that shows legalisation does not protect against it.
Why the Cannabis Violence Risk Demands Accountability
Smyth says the legalisation lobby pushes a narrative that most drug harm flows from the illegal status of drugs. His answer is that the reverse is true. Societies make behaviours such as drug use illegal because of demonstrated risk of harm to the user and to others.
Finley makes a related point. She observes that the stigma once attached to marijuana has vanished. Politicians of both parties, she writes, increasingly treat it as harmless. The public, however, needs an honest discussion of its social and health risks.
That discussion is overdue. Over the same years the paper describes a push towards commercialisation, rising use and rising potency. Meanwhile the evidence on harm to others received comparatively little attention. Those who promoted cannabis as benign now face a plain question. What is their answer to this evidence?
If industry and advocates believe the association is overstated, they can show it. They can fund and publish research with proper exposure detail. The authors name exactly what is missing: frequency, cumulative use, and the balance of THC and CBD. Until that work appears, the burden of proof on the cannabis violence risk has shifted.
Demand Reduction and Prevention Must Lead
The authors describe cannabis use as a potential target for interventions to cut the cannabis violence risk. They say violence should sit alongside other recognised cannabis harms in prevention strategies, public education campaigns and regulatory policies. This applies particularly where cannabis has been legalised or decriminalised. They also call for research testing whether reducing use cuts violent behaviour.
That agenda points upstream. Finley notes that young people now use cannabis more often and in higher doses than earlier generations did. That raises their addiction risk. Prevention that delays or stops uptake can help. So can education that tells the full story, and treatment pathways that support cessation. Each reduces demand before violence linked to cannabis use reaches a household or a street.
The researchers frame violence as a harm to other individuals and society, as well as to the user. Policymakers can keep treating cannabis as a lifestyle product. Or they can treat the cannabis violence risk as the public health matter the evidence says it is.
Dalgarno Institute
Sources: