Mixing Edible Cannabis and Alcohol May Impair Driving More Than Scientists Expected

Cannabis edibles and alcohol together produced stronger driving impairment than either substance alone, while standard field sobriety tests missed cannabis-related impairment.

In a study supported by the National Institute on Drug Abuse of the National Institutes of Health, Johns Hopkins Medicine researchers strengthened evidence that combining cannabis edibles with alcohol can make driving impairment worse than using either substance by itself.

The study also found that cannabis (alone or with alcohol) did not impair performance on standard field sobriety tests.

Roadside tests miss cannabis

The findings, published in JAMA Network, point to a need for broader public awareness about the stronger effects that may occur when cannabis and alcohol are used together. They also suggest that roadside methods for detecting driver impairment need improvement. The study further found that the legal alcohol intoxication limit used across most of the U.S. (0.08% breath alcohol level, or BrAC) may be too high for drivers who have consumed both cannabis and alcohol.

“Our findings indicate that co-use of cannabis and alcohol produces significantly greater driving impairment and subjective intoxication than either substance alone,” says the study’s lead author, Austin Zamarripa, Ph.D., assistant professor of psychiatry and behavioral sciences at the Johns Hopkins University School of Medicine. “Importantly, these findings suggest that the interaction between cannabis edibles and alcohol is not merely additive, but may be synergistic in producing impairment, which has important implications for real-world risk.”

A controlled test of co-use

For the study, the research team used a carefully controlled design in which participants came in for multiple outpatient experimental sessions. In each session, they received either a cannabis-infused brownie (10 or 25mg THC) or a placebo brownie, followed by either an alcoholic beverage or a non-alcoholic placebo drink. Alcohol amounts were adjusted for each participant to produce BrACs of 0.05% or 0.08%.

Participants were healthy adults ages 21 to 55. Their eligibility was confirmed at the first visit through medical and psychiatric screening, a physical examination, and routine bloodwork. Thirty participants were enrolled and randomized, and 25 completed every study visit.

All participants reported binge drinking within the past 90 days, previous use of cannabis and alcohol together within the past year, and limited cannabis use (fewer than three times per week, with at least one use in the past year) to reduce the effect of tolerance. Urine drug testing confirmed that participants had no recent use of other illicit drugs.

After the initial screening confirmed medical eligibility, participants completed a second visit before the study sessions began. During that visit, they practiced with the driving simulator and other performance tests so that learning effects would be less likely to influence the results. After this training visit, participants completed seven experimental sessions in which they received cannabis alone, alcohol alone, cannabis and alcohol together, or placebo cannabis and placebo alcohol. The order of these sessions was carefully balanced across participants to help prevent bias.

Before dosing at each experimental session, participants completed baseline assessments. These included a simulated drive, standard field sobriety tests, cognitive and psychomotor performance tests, subjective drug effect questionnaires, and a blood sample to measure THC and THC metabolite levels. One hour after breakfast, participants ate either a cannabis-infused brownie or a placebo brownie.

Forty-five minutes later, they received either alcoholic beverages or placebo drinks, which were designed to provide similar sensory cues as the alcohol drinks and maintain dose blinding. The drinks were consumed over a 15-minute period. Participants then repeated the same driving and other assessments throughout the day for up to 7.5 hours after eating the brownie. Study sessions were separated by at least one week to allow enough time for the drug to clear from the body between visits.

Impairment outlasted detection

The findings showed that mixing cannabis edibles with alcohol caused greater and longer-lasting driving impairment and stronger feelings of intoxication than either substance alone. Yet standard field sobriety tests classified participants as intoxicated only in the high alcohol dose condition (0.08% BrAC) compared with placebo.

“We designed this study because people are increasingly co-using alcohol with edible cannabis products, yet controlled research has largely focused on smoked cannabis. This is the first controlled study to examine how cannabis edibles and alcohol interact, despite their growing combined use,” says Tory Spindle, Ph.D., the study’s principal investigator and associate professor of psychiatry and behavioral sciences at the Johns Hopkins University School of Medicine. “Consuming typical retail doses of cannabis edibles alongside even low doses of alcohol can produce driving impairment comparable to — or greater than — alcohol alone at the legal limit.”

Legal limits may fall short

As cannabis legalization expands and access to cannabis products grows, the researchers emphasize that public health messaging, policy discussions, and regulatory decisions should account for the added risks of using cannabis and alcohol together.

They also note that more research is needed to better understand impairment from combined use and to develop stronger biological or behavioral tools for detecting cannabis-related driving impairment.

Reference: “Impact of Cannabis Edibles Combined With Alcohol on Driving, Field Sobriety Performance, and Subjective Effects: A Within-Participant Crossover Trial” by C. Austin Zamarripa, Spencer Lin, McKenna Klausner, Kriti Rastogi, Daniel J. O. Roche, Matthew Novak, Denis Antoine, David Wolinsky, Thomas D. Marcotte, Elise M. Weerts, Ryan Vandrey and Tory R. Spindle, 1 May 2026, JAMA Network Open.
DOI: 10.1001/jamanetworkopen.2026.9842

Never miss a breakthrough: Join the SciTechDaily newsletter.
Follow us on Google and Google News.