Cannabis involvement among mass shooters increased after medical legalization began | Research

The takeaway

The proportion of mass shooters with documented cannabis involvement more than doubled after 1996 (when medical cannabis legalization began), while no such increase occurred among those who committed mass murder by other methods.

Public health researchers studying violence, policymakers evaluating cannabis regulations, and mental health professionals assessing risk.

What the researchers found

Among mass shooters, cannabis involvement (use, possession, or distribution) was significantly higher from 1996 onward compared to before (11.2% vs 4.9%, p=0.002). No similar increase occurred among non-firearm mass murderers (4.8% vs 5.7%, p=0.76). Mass shooters with cannabis involvement were significantly younger (28.7 vs 33.5 years).

Why it matters

While this study cannot establish causation, it raises questions about whether increased cannabis availability may be relevant for a small subgroup of individuals vulnerable to violence, particularly given the known association between heavy cannabis use and psychosis in predisposed individuals.

The numbers in context

Mass shooters with cannabis involvement: 11.2% post-1996 vs 4.9% pre-1996 (p=0.002). Non-firearm mass murderers: 4.8% vs 5.7% (p=0.76). Cannabis-involved shooters were younger: 28.7 vs 33.5 years (p<0.001).

How the study worked

Analysis of publicly available media reports and court/police records for mass shooting events in the United States from 1900 to 2019. Cannabis involvement was compared before and after 1996 (California medical cannabis legalization). Non-firearm mass murderers served as a comparison group.

What this study cannot tell us

Reliance on publicly available reports likely underestimates cannabis involvement across all time periods. Cannot establish causation. Cannabis involvement includes use, possession, and distribution, conflating different relationships with the substance. Population-level cannabis use also increased post-1996.

How to read the evidence

Ecological-level retrospective analysis using publicly available data. Interesting association but multiple confounders and cannot establish causation.

When this study was published

2025 publication analyzing 1900-2019 data.

The bigger picture

This is a deeply controversial area. Cannabis use has increased dramatically since legalization, so higher rates among mass shooters could simply reflect population-level trends. The comparison with non-firearm mass murderers somewhat addresses this but cannot fully account for confounding factors.

Questions still open

Does the increase simply reflect rising cannabis use in the general population?
Is there a specific neurobiological pathway linking heavy cannabis use to violence in predisposed individuals?

Common questions

Does cannabis cause violence?

This study found an association, not causation. Most cannabis users never engage in violence. However, heavy cannabis use is associated with psychosis in vulnerable individuals, and psychosis can rarely be associated with violent behavior.

Could this just reflect more cannabis use in general?

This is a major limitation. Cannabis use has increased substantially since legalization. The comparison with non-firearm mass murderers (which showed no increase) somewhat addresses this, but cannot fully rule out population-level trends as the explanation.

Read the original research

Cannabis involvement and mass shooting events in the United States from 1900 to 2019.

East Asian archives of psychiatry : official journal of the Hong Kong College of Psychiatrists = Dong Ya jing shen ke xue zhi : Xianggang jing shen ke yi xue yuan qi kan, 35(1), 28-31

View on PubMed ↗Publisher’s paper ↗

Citation

Girgis, R R; Hesson, H; Appelbaum, P S; Brucato, G. (2025). Cannabis involvement and mass shooting events in the United States from 1900 to 2019.. East Asian archives of psychiatry : official journal of the Hong Kong College of Psychiatrists = Dong Ya jing shen ke xue zhi : Xianggang jing shen ke yi xue yuan qi kan, 35(1), 28-31. https://doi.org/10.12809/eaap2464