Cannabis edibles dose-dependently impair simulated driving performance hours after ingestion at blood concentrations of delta-9-tetrahydrocannabinol (THC) below limits commonly used for roadside enforcement, new data suggest.

“We were surprised to see that the level of THC in the blood was so low compared with smoked cannabis,” Bernard Le Foll, MD, PhD, a senior scientist at the Centre for Addiction and Mental Health (CAMH) in Toronto, told Medscape News Canada. “Many people will show impairment in their driving ability while being below the limit that’s used by law enforcement for impairment assessment.”

Indeed, the team found that lane-position variability, a validated measure of driving impairment, rose by about 2 cm at a 10-mg THC dose and 3.3 cm at 20 mg compared with placebo, for at least 5 (but not 24) hours after ingestion. Peak blood THC levels, however, reached only 1.6 ng/mL and 3.4 ng/mL at those doses, which is well under the 2-5 ng/mL thresholds many jurisdictions use to define impaired driving.

“Ultimately, the holy grail in the field will be to develop a validated way to measure impairment,” Le Foll said. “That’s the long-term objective of this line of research.”

The study was published on August 31 in JAMA Network Open.

Challenges for Roadside Enforcement

The researchers conducted a randomized, double-blind, placebo-controlled, crossover trial at CAMH that included 40 adults (mean age, 30.4 years; 50% women) who used cannabis at least weekly and edibles at least monthly.

Participants received THC soft-chew gummies at doses of 0 (placebo), 2 mg, 10 mg, or 20 mg, then completed driving-simulator tasks and blood draws at 2 and 5 hours after dosing and again at 24 hours.

The primary outcome was SD of lateral position (a measure of how well a driver stays centered in a lane) at 2 and 5 hours after THC exposure. That measure rose at the 10-mg dose (by 2.0 cm) and 20-mg dose (by 3.3 cm) compared with placebo, but not at 2 mg, indicating dose-dependent impairment.

Reaction time was longer at the 20-mg dose, and speed variability increased at 20 mg. Participants’ ratings of their willingness to drive and their perceived driving ability dropped as the dose increased.

A total of 22 adverse events (AEs) occurred (most commonly headache, nausea, and dizziness), eight of which were deemed unrelated to the study intervention. No serious AEs occurred, and no AEs were judged to be severe or moderate in intensity.

Overall, dose-related impairment of driving simulator performance was identified at 2 and 5 (but not 24) hours after ingestion. “We found decreased lane control, increased speed variability, and increased [reaction time], especially at the highest dose (20 mg), with minimal evidence of changes in driving performance at the low dose (2 mg),” the authors wrote.

“Objective measures of driving simulator performance aligned with self-reported data suggesting that participants were less willing to drive and perceived lower driving ability in a dose-dependent manner, even at the low dose,” they noted. “Whole-blood THC levels were below the commonly used 5 ng/mL per se threshold in all dose conditions, suggesting that THC edibles may present unique challenges for roadside enforcement.”

The study had limitations, however. The primary driving simulator scenario was a straightforward rural drive with minimal distractors, which may not be comparable with more complex driving environments, such as urban roads with heavy traffic. The focus of the study was limited to measures of lane control, speed, and reaction time. “Other driving measures, such as headway and braking dynamics, should be considered in future studies,” the authors wrote.

Exposure vs Impairment

Timing is an important factor in impairment, Le Foll said. “With edibles, the effect takes a bit of time to kick in compared with smoking cannabis, but then the effect is prolonged. Users may feel fine immediately after taking the edible, but after a certain delay, their driving abilities may be impaired. You don’t want to be caught in that situation where you feel you are fine taking the wheel but then realize 1 hour later that you are not in a good state to drive.”

Although the study measured impairment out to 5 hours, Le Foll’s safe margin would run longer. “My estimation would be probably at least 6-8 hours, probably more on the 8-hour side, knowing there also may be individual differences,” he said. “Some people may be impaired longer than others.”

In an invited commentary accompanying the study, José Ignacio Nazif-Munoz, PhD, associate professor of addiction services at the Université de Sherbrooke in Longueuil, Quebec, agreed that “a central challenge in cannabis-impaired driving research is…distinguishing cannabis exposure from functional impairment.”

Per se thresholds assume that a given blood concentration reflects a consistent level of impairment across people and routes of administration, an assumption that this study calls into question, he noted. Future policy should aim at identifying impairment directly rather than relying on detection of cannabis exposure alone.

Research priorities could include establishing “how laboratory measures correspond with actual driving outcomes, determining how tolerance and patterns of cannabis use modify impairment, and identifying combinations of biological and behavioral indicators that can better characterize driving risk,” he suggested.

“Research should also examine a broader range of driving domains, including cognitive processing, braking behavior, hazard perception, and decision-making under complex conditions, while considering enforcement feasibility and potential equity implications,” he concluded.

Using Cannabis Safely

Paxton Bach, MD, clinical assistant professor of social medicine at the University of British Columbia in Vancouver, who was not involved in the study, called it well designed, although artificial because it was done in a laboratory setting.

“This study is important because it provides more information in an area that we have not yet fully reckoned with,” Bach told Medscape News Canada. “Unlike for alcohol, there’s no point-of-care test that can measure cannabis intoxication on the spot.” Roadside functional assessments are prone to bias, and the blood tests used to confirm cannabis exposure are more difficult to obtain in the field than a breathalyzer.

“Even more challenging, the level of THC measured in the blood does not necessarily correlate with the level of intoxication or impairment,” said Bach, who is also a general internist and addiction medicine physician at St. Paul’s Hospital in Vancouver and co-medical director for the British Columbia Centre on Substance Use. “It depends on not just the level in the blood but also the route of [cannabis] administration.”

Smoking impairs drivers faster than edibles do, he said, because smoked cannabis reaches peak effect almost immediately, while oral THC takes effect after an hour or more. “The study underscores the fact that not only are we not particularly capable of assessing blood levels in the moment on the road, but even if we can, that’s still an imperfect measure of somebody’s degree of impairment.

In the clinical setting, “Cannabis use should come up like any other substance-use conversation, making sure that patients are using it safely,” he said. Clinicians should also watch for signs of cannabis use disorder and raise the driving risk directly with patients.

The broad message that cannabis can impair driving, whether smoked or eaten, shouldn’t surprise most clinicians, he said, but the disconnect between blood THC levels and impairment after edibles is a less appreciated finding.

“Cannabis-related driving charges and crashes still make up a small share of cases compared with alcohol, though the numbers are rising,” he noted. The study results should prompt reconsideration of how cannabis-impaired driving is defined, he suggested, although there’s no simple answer. “More research, more discussion is certainly needed. Defining and measuring intoxication with cannabis for the purposes of driving is deceptively challenging.”

The study was funded by a Public Safety Canada grant to Le Foll and a coauthor. Indiva provided the placebo edibles in kind. Le Foll reported receiving grants from the government of Canada during the conduct of the study; receiving grants from Indivior and Canopy Growth Corporation outside the submitted work; receiving personal fees from Shionogi, Third Bridge, Changemark, Department of Psychiatry, University of Toronto, Department of Family and Community Medicine Clinician Scientist Award, University of Toronto, the Centre for Addiction and Mental Health, and Waypoint Centre for Mental Health Care outside the submitted work; and serving on the national advisory board and the clinical trial steering board for Indivior Canada, the scientific advisory board for NFL Biosciences, and the Canadian Institutes of Health Research Governing Council. Bach reported no relevant financial relationships.

Marilynn Larkin, MA, is an award-winning medical writer and editor based in New York City whose work has appeared in numerous publications, including Medscape Medical News and its sister publication MDedge, The Lancet (where she was a contributing editor), and Reuters Health.