A new pharmacokinetic model for oral cannabis showed that a 1 ng/mL blood THC driving limit produces too many false positives, while 2 and 5 ng/mL limits remain inconclusive without better data linking blood levels to impairment.
Policymakers, cannabis consumers who use edibles, and anyone interested in cannabis driving law reform.
The 1 ng/mL THC driving limit had the highest false positive risk across all simulated scenarios
What the researchers found
Using data from 10 published studies, researchers built a model simulating THC blood levels in frequent and occasional users after oral doses from 2.5 to 100 mg. The 1 ng/mL per se limit was the least effective due to high false positive risk. The 2 and 5 ng/mL limits could not be validated because there is insufficient data connecting specific blood THC levels to actual driving impairment after oral consumption.
Why it matters
With edibles and oral cannabis products growing in popularity, driving laws based on blood THC levels need to account for how oral consumption produces very different blood level patterns than smoking. Current laws were largely designed around inhalation data.
The numbers in context
Model built from 10 published studies. Simulated doses: 2.5 mg to 100 mg oral THC. Three per se limits evaluated: 1, 2, and 5 ng/mL. The 1 ng/mL limit showed the highest false positive risk across both user types.
How the study worked
Semi-mechanistic population pharmacokinetic model developed from 10 published studies of intravenous or oral cannabis administration. Simulated THC and metabolite concentrations across a range of doses in frequent and occasional users to evaluate existing per se driving limits.
What this study cannot tell us
Model was built from published study data, not new clinical trials. Pharmacodynamic (impairment) data for oral cannabis remains scarce. Simulations may not capture all real-world variability in absorption and metabolism.
How to read the evidence
Moderate: well-constructed pharmacokinetic model from multiple studies, but limited by lack of pharmacodynamic (impairment) validation data.
When this study was published
2025 study.
The bigger picture
Cannabis-impaired driving laws remain one of the most contentious policy areas. This study highlights a fundamental problem: existing blood THC thresholds were developed primarily from inhalation data and may be inappropriate for the growing population of oral cannabis users.
Questions still open
- What blood THC level actually correlates with impaired driving after oral consumption? Should driving laws use different thresholds for different consumption methods? Are there better biomarkers than blood THC for assessing impairment?
Common questions
Why do current THC driving limits not work well for edible users?
Is there a good blood THC level that reliably indicates impairment?
Read the original research
Evaluation of Cannabis Per Se Laws: A Semi-Mechanistic Pharmacometrics Model for Quantitative Characterization of THC and Metabolites in Oral Users.
Journal of clinical pharmacology, 65(5), 535-549
Citation
Li, Peizhi; An, Guohua. (2025). Evaluation of Cannabis Per Se Laws: A Semi-Mechanistic Pharmacometrics Model for Quantitative Characterization of THC and Metabolites in Oral Users.. Journal of clinical pharmacology, 65(5), 535-549. https://doi.org/10.1002/jcph.6181