The scientific debate around a legal THC limit for driving

Conflicting data makes makes addressing concerns with impaired driving and cannabis in a post-legalized world challenging

The federal government of Canada is currently discussing the implementation of a legal limit of THC for driving. Several other countries have already enacted policies of this nature. In Australia, it is illegal to drive with any detectable level of THC, while in Switzerland the limit is set at 2.2 ng/mL. In those U.S. states that have legalized cannabis, the limits imposed range from zero to 5 ng/mL.

The variety of policies reflects a broader lack of scientific consensus on what constitutes a safety limit of THC concentration for drugged driving. Of special interest in understanding this question is a 2013 article by Rebecca Hartman and Marilyn Huestis, from the U.S. National Institute on Drug Abuse and University of Maryland, respectively, which reviews the most recent research conducted on cannabis and driving impairment.

Cannabis is the most frequent drug found in the blood of drivers. A survey conducted in 2007 in the U.S. showed that more than 8.6% of weekend nighttime drivers test positive for THC, a value that rose to 12.6% in 2014. Even after acknowledging the risk of driving under the influence of cannabis, many drivers affirm they would still drive provided there is a good reason for doing so, or if they believe themselves to be tolerant to cannabis. Similarly, a new Canadian survey found that half of those who frequently drive under the influence of cannabis are convinced that they are able to drive safely, while some even think that smoking cannabis makes them drive better.

Some of the controversy regarding the effects of cannabis on driving comes from early epidemiologic studies that arrived at polar opposite conclusions. According to Hartman and Huestis, this can be explained by some methodological flaws or the usage of different measures and controls, such as studies measuring cannabis intoxication from blood samples of killed drivers and comparing it to oral fluid samples of living drivers.

Cannabix cannabis testing dui
The Cannabix marijuana breathalyzer image via

This confusion has largely been overcome in recent studies, which show a strong association between cannabis consumption and an increase in crash risk, even after controlling for other important factors such as age, sex, risky behaviours and polydrug profiles. Together, laboratory and epidemiological studies show consistent deleterious effects on driving ability that peak one hour after cannabis consumption.

Simulator studies unequivocally demonstrate that cannabis affects important skills such as reaction time and lane swaying. Regular smokers seem to be immune to some of these effects, but not to all of them. Several studies also show that THC blood concentrations are associated with slower driving speed and increased headway. This suggests that drivers might be employing compensatory strategies, but, according to the authors, it does not preclude a more serious driving impairment.

Regarding the question of THC blood levels, several laboratory studies show a dose-dependent relationship with impairment measures.

On the one hand, epidemiological studies regularly failed to find a linear relationship between THC concentrations and crash risk, despite showing a strong association between the two. One Norwegian study suggests that concentrations as low as 2.2 ng/mL render drivers unsuited for driving. Other authors have pointed out a strong interaction between cannabis and alcohol, in which small legal doses of each might lead to much stronger impairments than when taken alone. Finally, the fact that oral fluid measures of THC and blood levels vary drastically questions the legal validity of the first measure.

On the other hand, several studies argue against the practicability of a zero-tolerance policy. Researchers have found detectable traces of THC up to a week after monitored abstinence, and also that chronic daily users under three weeks of abstinence perform worse on validated driving indicators than occasional smokers. Finally, residual and withdrawal effects such as sleep disruption might still affect driving performance, among those who have not smoked for long periods of time.

Featured image via

Related: Canopy Growth Corporation, which owns Smiths Falls, Ont., cannabis producer Tweed and Toronto-based Bedrocan Canada Inc., will provide funding over three years to the Canadian Drug Policy Coalition and Mothers Against Drunk Driving Canada to develop and administer the campaign. More at

In this article

Join the Conversation

1 comment

  1. Robert Reply

    They need to do more research because at the low limits they want to use everyone who has used cannabis 3 days before testing is going to test positive even though they are completely sober.
    Sounds like there is some discrimination against cannabis users or this is just the governments way of discouraging people from using it.
    I mean you have to get to work and only the big cities have any sort of reliable public transit so if you don't want to lose your license you pretty much can't use cannabis at all.
    When are they going to start testing for people driving high on legal pharmaceutical drugs It's only fair the alcohol and cannabis users are not the only one's driving intoxicated. What about smartphone users these people are just as dangerous and so far the the cops and government have not been able to do anything about them.