Educationright arrow
Myths & Risks

Doctors for Drug Policy Reform

Do motor vehicle accidents and fatalities increase following cannabis legalization?

As cannabis legalization has increased rates of adult cannabis use, it likely has increased rates of driving under the influence of cannabis. Presently, it is difficult to say by how much.

As more U.S. states legalize medicinal and recreational cannabis, reported rates of driving under the influence of the drug are increasing significantly. However, this observation is often based on the presence of THC in the blood, which can be detected for weeks following cannabis use, even when users are not currently intoxicated or impaired. In addition, there are no well-established set of roadside procedures or tests at this time for detecting drivers who are impaired by cannabis use.

The best epidemiologically derived estimates are that cannabis-impaired motor vehicle drivers have double the odds of being involved in a motor vehicle crash. This should be contrasted with those of alcohol which are approximately 20-times higher than baseline at a blood/breath alcohol concentration of 0.10.

Because of the public health implications, states and law enforcement agencies are seeking guidance on the development valid and reliable field tests for detecting cannabis-impaired drivers, analogous to those used currently to identify alcohol intoxication. However, the significantly different pharmacokinetics and behavioral impairment profiles produced by THC versus alcohol make achieving this goal extremely difficult. As a shortcut, some states have adopted so-called “per se” laws, whereby a specified blood THC level in and of itself equates to cannabis-related impairment. However, there is little scientific foundation for such legislation.

There are many, often unacknowledged, research gaps in these and related areas that need to be addressed to provide a more coherent basis for public policies. Issues of driver impairment following THC-containing edibles, and impairment levels following combinations of alcohol, cannabis, and other drugs are almost completely unresearched at present.

The situation for cannabis mirrors that of many prescription medications, such as benzodiazepines, opioids, and hypnotics (e.g., Ambien). There is presently no empirically valid measure of quantifying the level of acute intoxication and impairment from these medications. Society has accepted this limitation and has adapted by training police officers to evaluate individuals for impairment by drugs and other causes to the best of their abilities.

An assessment of traffic safety effects solely based on accident data does not provide reliable results. Scientific standards must be applied to such data, including case-control with matched pairs regarding vehicle type, traffic situation, driving direction, time of day, day of the week, age, and gender. There are well-known weaknesses of the accident criterion: Because we often do not know if there was a cannabis-induced impairment in case of an accident, it is necessary to orient the evaluation of road safety more directly to “normal behavior,” i.e., to systematically investigate the events preceding the accident. This is also accompanied by a prioritization of studies in real world settings in order to overcome discrepancies with potentially overestimated, theoretical laboratory results on the effects of cannabis. Thus, there is a need for gathering systematic epidemiological data.

To date, informal expert consensus reveals that no reliable statements can be generated on the influence of cannabis legalization on road safety as measured by the accident criterion. A review of the studies overall indicates that the risk of accidents is only moderately increased, and cannabis thus does not appear to contribute significantly to the occurrence of accidents. However, this does not mean that the use of cannabis in active participation in road traffic during the intoxicated phase can be regarded as harmless and unproblematic. This is especially true for "high-dose" cannabis users and people who mix cannabis with other substances. Available evidence suggests that behavior may pose a risk to road safety. Many unanswered questions remain. To develop a full understanding of potential long-term effects of the cannabis market, continuous monitoring is needed.

We must put these uncertainties in perspective. Driving under the influence of any substance is already illegal in every U.S. state, and cannabis legalization does not change that. Legalization will, however, foster honesty between physicians and patients about cannabis use and its potential effect on driving and other activities.


1. Pearlson, G. D., Stevens, M. C., D'Souza, D. C., 2021. Cannabis and Driving. Front Psychiatry 12, 689444.

2. Rogeberg, O., Elvik, R., 2016. The effects of cannabis intoxication on motor vehicle collision revisited and revised. Addiction 111, 1348-1359.

3. Voy, A., 2023. Collisions and cannabis: Measuring the effect of recreational marijuana legalization on traffic crashes in Washington State. Traffic Inj Prev, 1-9.

4. Walker, M., Carpino, M., Lightfoot, D., Rossi, E., Tang, M., Mann, R., Saarela, O., Cusimano, M. D., 2023. The effect of recreational cannabis legalization and commercialization on substance use, mental health, and injury: a systematic review. Public Health 221, 87-96.


Support Our Mission