Cannabis consumption was associated with a 55% higher risk of fatal motor vehicle collisions and a doubled risk of injury-causing crashes, though the evidence certainty was low.
Traffic safety researchers, policymakers setting cannabis-impaired driving laws, and public health communicators.
55% higher risk of fatal car crashes and 2x higher injury risk associated with cannabis use
What the researchers found
Across 31 studies with 328,388 individuals, cannabis consumption was associated with increased risk of MVC fatality (OR=1.55, 95% CI: 1.20-1.98) with 14 more deaths per 100,000 MVCs, and increased risk of injury (OR=2.00, 95% CI: 1.31-3.07) with an absolute risk increase of 6.8%. Evidence for culpability/unsafe driving was very low certainty.
Why it matters
As cannabis legalization expands globally, quantifying the crash risk associated with cannabis use is essential for setting driving policy, THC legal limits, and public education campaigns.
The numbers in context
31 studies, 328,388 individuals. Fatal MVC: OR=1.55 (95% CI: 1.20-1.98), ARI 14 per 100,000 MVCs. Injury MVC: OR=2.00 (95% CI: 1.31-3.07), ARI 6.8%. Both rated as low certainty evidence by GRADE.
How the study worked
Systematic review and meta-analysis searching 8 databases through November 2024. Included 31 observational studies. DerSimonian and Laird random-effects model. GRADE approach for certainty assessment. Registered protocol CRD42022357478.
What this study cannot tell us
Low certainty evidence per GRADE. Heterogeneous methods for detecting cannabis use across studies (some used THC blood levels, others self-report). Cannabis remains detectable long after impairment resolves. Cannot distinguish acute impairment from chronic use. Confounding by polydrug use is common.
How to read the evidence
Large meta-analysis with preregistered protocol and GRADE assessment. Low certainty reflects observational study limitations and heterogeneous cannabis exposure measurement.
When this study was published
2025 publication with searches through November 2024.
The bigger picture
The absolute risk increase of 14 additional deaths per 100,000 MVCs is modest compared to alcohol-impaired driving. However, the doubled injury risk from 9 case-control studies suggests cannabis-impaired driving is a meaningful safety concern.
Questions still open
- What is the THC blood concentration threshold above which crash risk significantly increases?
- How does the crash risk from cannabis compare to the risk from legal blood alcohol levels?
Read the original research
Cannabis consumption and motor vehicle collision: A systematic review and meta-analysis of observational studies.
The International journal on drug policy, 142, 104832
Citation
Jin, Andrew; Darzi, Andrea J; Dargham, Amne; Liddar, Navroop; Bozorgi, Sepehr; Sohrevardi, Shamim; Zhang, Maurice; Torabiardakani, Kian; Couban, Rachel J; Khalili, Malahat; Busse, Jason W; Sadeghirad, Behnam. (2025). Cannabis consumption and motor vehicle collision: A systematic review and meta-analysis of observational studies.. The International journal on drug policy, 142, 104832. https://doi.org/10.1016/j.drugpo.2025.104832