Cannabis Alone Was Not Linked to More Crashes in a 2024 Emergency-Department Study—Alcohol Was

A 2024 study of drivers treated at emergency departments found no evidence that cannabis use by itself was associated with higher odds of a motor-vehicle collision. Alcohol, however, was consistently associated with increased crash odds—both when used alone and when combined with cannabis.

The study, published in Accident Analysis & Prevention, analyzed 1,397 patients recruited from emergency departments in Denver, Portland and Sacramento. The researchers compared drivers involved in collisions with patients who had sought emergency care for other reasons. Participants were interviewed about substance use, while researchers tested blood samples for THC and its metabolites. Alcohol use was assessed through breath tests or measurements collected during clinical care.

After adjusting for demographic characteristics and other factors, the researchers found that cannabis use alone was not associated with increased odds of a collision. Acute alcohol use alone was associated with higher odds, as was using alcohol and cannabis together.

The analysis also found no clear dose-response relationship between cannabis measures and crash risk. Participants reporting higher levels of acute cannabis use did not show higher odds of a collision; in one analysis, heavy self-reported use was associated with lower odds, with an odds ratio of 0.18. That result should not be interpreted as evidence that cannabis protects drivers. It may reflect differences between participants, reporting or recruitment patterns, and the study’s observational design.

The findings speak to a longstanding difficulty in cannabis-impaired-driving research: a positive THC test does not necessarily establish that a person was impaired when driving. THC can remain detectable after the period of acute effects, and blood concentrations do not map neatly onto impairment in the way blood-alcohol concentration is used for alcohol. The National Highway Traffic Safety Administration’s report on marijuana-impaired driving similarly notes that toxicology can confirm the presence of THC without demonstrating impairment.

For that reason, the study’s authors argued that evaluations should place greater emphasis on driving behavior and clinical signs of intoxication rather than relying on a single THC threshold. The issue is particularly important in jurisdictions that use “per se” cannabis laws, which treat a specified THC concentration as evidence of impaired driving. Such limits can be difficult to interpret because results vary with the timing of testing, route and amount of consumption, and a person’s pattern of cannabis use.

The study has important limitations. It included only people who presented for emergency care and agreed to participate, so the participants may not represent all drivers. Self-reported substance use may also be inaccurate. In addition, the study was designed to identify associations, not to prove that cannabis has no effect on driving or that alcohol caused any individual collision. Its results also do not rule out impairment immediately after cannabis consumption.

That distinction matters because laboratory and public-health evidence indicates that cannabis can affect reaction time, coordination, judgment and other skills required for safe driving. The Centers for Disease Control and Prevention’s overview of cannabis and driving notes that research has linked acute cannabis use with car crashes, while also emphasizing that the relationship between THC concentration and individual impairment remains difficult to measure. Combining cannabis with alcohol can increase impairment further.

Earlier research has reached similarly nuanced conclusions. A 2019 prospective study of injured drivers in British Columbia found no increased crash-responsibility risk below 5 nanograms of THC per milliliter of blood. At or above that level, the estimated increase was not statistically significant, while alcohol at or above 0.08% was strongly associated with crash responsibility. The study did not establish a safe THC level.

Overall, the 2024 findings add to evidence that cannabis-positive test results and actual driving impairment are not interchangeable. They also reinforce a more consistent conclusion across the research: alcohol is a clear contributor to collision risk, and using alcohol and cannabis together is especially concerning. The safest approach for drivers remains avoiding driving after using either substance, particularly in combination.