After inhaling THC, recreational cannabis users accurately predicted their driving impairment — self-assessed confidence dropped in proportion to dose and correlated with actual simulator performance.
Read this if you use cannabis and drive — this study suggests your own sense of 'am I okay to drive?' may be more reliable than you'd think, but there are important caveats about who was studied and how.
What the researchers found
One of the most important questions in cannabis and driving is whether users can tell when they're too impaired to drive. Unlike alcohol, where people notoriously overestimate their driving ability while drunk, this study found cannabis users were surprisingly accurate at assessing their own impairment.
Thirty healthy male volunteers — 15 chronic users (1–2 joints/day) and 15 occasional users (1–2 joints/week) — inhaled placebo, 10 mg, or 30 mg of THC mixed with tobacco in a controlled setting. They then rated their driving confidence, completed reaction time tests, and drove on a simulator.
Driving confidence dropped markedly after cannabis use, proportional to the THC dose. The EC50 (half-maximal effect concentration) was very low at 0.11 ng/mL, meaning self-perceived impairment kicked in at very low THC blood levels with a rapid onset (half-life of 37 minutes). Importantly, driving confidence correlated with both objective measures: reaction time and lane-keeping ability on the simulator.
Chronic users showed faster recovery of driving confidence than occasional users, consistent with pharmacological tolerance. At 30 mg THC, driving impairment (measured by standard deviation of lane position, SDLP) was significant in both groups, but chronic users recovered driving ability faster.
The practical implication is notable: cannabis users appear capable of recognizing their impairment, which could support self-regulation strategies — though the authors caution this shouldn't replace objective testing standards.
Why it matters
If cannabis users can accurately assess when they're too impaired to drive, this has major implications for public health messaging and policy. Rather than relying solely on arbitrary blood THC thresholds (which this study and RTHC-00092 show are problematic), campaigns could emphasize self-assessment as a first line of defense. However, the study only tested men, the sample was small, and real-world driving involves many more variables than a simulator.
The numbers in context
30 male participants (15 chronic, 15 occasional users). THC doses: 0, 10, and 30 mg inhaled. EC50 for self-perceived impairment: 0.11 ng/mL blood THC. Onset half-life for perceived impairment: 37 minutes. Driving confidence remained below baseline at 8 hours post-inhalation. Both 10 mg and 30 mg THC impaired driving ability and reaction time compared to placebo.
How the study worked
Randomized controlled trial with 30 healthy male volunteers (15 chronic, 15 occasional cannabis users, ages 18–34). Participants inhaled placebo, 10 mg THC, or 30 mg THC mixed with tobacco. Measurements before and at multiple timepoints after inhalation: self-assessed driving confidence (visual analog scale), vigilance (Karolinska Sleepiness Scale), reaction time (psychomotor vigilance test), driving ability (standard deviation of lane position on York driving simulator), and blood THC concentrations.
Who was studied
N=30 healthy male volunteers aged 18-34, 50% chronic consumers, 50% occasional consumers.
What this study cannot tell us
Male-only sample — women may assess impairment differently due to pharmacokinetic differences in THC processing. Small sample (30 total, split between use groups). Driving simulator, not real-world driving. THC was mixed with tobacco, which may modify effects. Laboratory setting may heighten self-awareness compared to social use settings. The study measured acute effects; fatigue, distraction, and other real-world factors weren't captured.
How to read the evidence
This is a well-designed randomized controlled trial with placebo control and objective outcome measures. The crossover design within chronic/occasional groups strengthens the findings, but the small, male-only sample limits generalizability.
When this study was published
Published in 2023. Cannabis and driving research is an active area; newer studies may include female participants and real-world driving measures.
The bigger picture
This pairs with RTHC-00092 (Italian DUI blood testing delays) to highlight the cannabis driving measurement problem from two angles: the Italian study shows blood tests are often too late to be useful, while this study suggests self-assessment may be surprisingly reliable. Together with RTHC-00082 (workplace impairment detection) and RTHC-00089 (standard THC units), these studies are building toward better, more practical approaches to measuring cannabis impairment.
Replication
Not stated in abstract.
Funding
Not reported in abstract.
Conflicts of interest
Not reported in abstract.
Questions still open
- Would women show similar self-assessment accuracy? Does self-assessment remain accurate in real-world social settings where peer pressure or urgency to drive exists? Could validated self-assessment tools (apps, standardized questions) supplement or replace per se THC blood limits? At what point after use do chronic users' driving abilities return to baseline?
Read the original research
Can inhaled cannabis users accurately evaluate impaired driving ability? A randomized controlled trial.
Frontiers in public health, 11, 1234765
Frontiers in Public Health is a peer-reviewed journal known for publishing high-quality research in public health.
Citation
Hartley, Sarah; Simon, Nicolas; Cardozo, Bibiana; Larabi, Islam Amine; Alvarez, Jean Claude. (2023). Can inhaled cannabis users accurately evaluate impaired driving ability? A randomized controlled trial.. Frontiers in public health, 11, 1234765. https://doi.org/10.3389/fpubh.2023.1234765