The number on a roadside THC breath or blood test was never built to answer the question that actually matters on the road: is this person’s driving compromised right now. A new randomized clinical trial published in JAMA Network Open makes that gap impossible to ignore, showing that cannabis edibles dose-dependently impair simulated driving performance even at blood THC concentrations well below the thresholds police departments use to make arrests.
Key Points
- A JAMA Network Open trial found THC edibles impaired simulated driving in a dose-dependent way, even when blood THC sat below common per se roadside thresholds.
- CAMH researchers describe a “clear disconnect” between measured blood THC and actual driving impairment after edible consumption.
- Edibles behave pharmacologically unlike smoked cannabis, producing delayed, longer-lasting impairment that outlasts detectable blood-level spikes.
- Earlier studies found impaired willingness and ability to drive lasting up to six hours after an edible dose, and simulator deficits persisting close to two hours.
- Research on edible-specific driving risk was nearly nonexistent before the last few years, leaving enforcement policy built almost entirely on smoked-cannabis data.
What the New Trial Actually Found
The JAMA Network Open study administered commercially available cannabis edibles to participants at varying doses and then tested them on a driving simulator, tracking performance against blood THC concentration drawn at intervals. The finding, restated in the Centre for Addiction and Mental Health’s release of the same research, was unambiguous: edibles “significantly impair driving performance on a simulator in a dose-dependent manner”. Higher doses produced worse driving outcomes. That part is unsurprising. What made the study notable is the second finding layered on top of it — impairment showed up even when blood THC readings sat under the concentrations that many jurisdictions treat as the legal cutoff for presumed impairment.
CAMH researchers called this a “clear disconnect between driving impairment and blood THC levels” — a polite way of saying the number on the test strip and the person’s actual capacity to drive a car were not tracking each other. That disconnect is the headline finding, and it lands squarely on the machinery that traffic enforcement has relied on for over a decade: a blood concentration threshold, drawn from research on smoked cannabis, applied uniformly to a product that behaves nothing like smoked cannabis in the body.
Why Edibles Don’t Play by the Same Pharmacological Rules
Smoked or vaporized cannabis delivers THC to the bloodstream almost instantly through the lungs, producing a sharp spike that peaks within minutes and clears within a few hours — a curve that correlates reasonably well, if imperfectly, with subjective and functional impairment. Edibles work through an entirely different route. THC absorbed through the gut passes through the liver before reaching systemic circulation, a process called first-pass metabolism that converts a meaningful share of it into 11-hydroxy-THC, a metabolite that is itself psychoactive and, by some measures, more potent than THC itself. The result is a slower onset, a lower and more gradual blood-concentration peak, and — critically — a duration of subjective and functional effect that can outlast the blood signal by hours.
That mismatch is precisely what a 2024 study on edibles and driving documented: participants reported altered subjective experience for seven hours and reduced willingness or ability to drive for up to six hours after consumption, describing the edible dose as genuinely intoxicating well past the window in which blood THC would register as elevated. Not every simulator metric moved in lockstep — that same study found reduced mean speed at two hours but no significant change in lane-position variability, top speed, or reaction time, a reminder that edible impairment doesn’t announce itself uniformly across every measurable driving behavior. An earlier simulator study found decreased speed, more lane departures, and greater lane-position variability persisting for nearly two hours post-consumption. Taken together, the pattern is consistent even where the specific metrics vary: functional impairment lingers well past the point where a roadside blood draw would look reassuring.
A Threshold System Built for a Different Product
Per se THC limits — the 1, 2, or 5 nanogram-per-milliliter blood thresholds adopted by various U.S. states and other jurisdictions — were derived overwhelmingly from research on inhaled cannabis, where blood concentration and impairment rise and fall together on a predictable timeline. As recently as 2021, reviews of the scientific literature noted there had been “virtually no testing of driver impairment following oral consumption of edibles”, with a companion review stating flatly that no edible-specific driving studies had even been published. Enforcement policy, in other words, was extrapolated from a product category that edibles don’t resemble pharmacokinetically. The new trial is among the first to test that extrapolation directly, and it found the assumption wanting: the blood number and the functional impairment simply come apart.
What It Means for Drivers, Regulators, and the Cannabis Market
The practical implication cuts in an uncomfortable direction for both enforcement and consumers. A driver who ate an edible three hours earlier may test clean on a roadside blood check while still showing measurable simulator-level impairment; conversely, a driver stopped soon after inhaling cannabis may register above threshold while past the peak of functional effect. Neither scenario serves the goal that per se laws were designed to achieve — a proxy that reliably separates impaired drivers from unimpaired ones. The honest caveat here is that this evidence comes from a driving simulator rather than instrumented on-road testing, so it measures laboratory performance rather than directly observed crash risk. But the dose-response pattern, replicated across multiple independent studies using different simulator protocols, is not the kind of result that dissolves under scrutiny. It points toward a genuine policy problem: edibles occupy a growing share of the legal cannabis market, and the enforcement tools built for smoked cannabis were never designed to catch what they now need to catch.
Sources:
zerohedge.com, pmc.ncbi.nlm.nih.gov, earth.com, camh.ca, frontiersin.org
