A Blood-THC Driving Review Found That Task Complexity Mattered to the Correlation Question
A 2025 systematic review found that most included studies did not report significant linear relationships between blood THC and driving measures. The studies that did find relationships used complex driving situations, making task design important to interpretation.
The review asked a specific statistical question
Twelve studies met its criteria, and ten reported no significant linear correlations for the driving measures examined. A null linear relationship does not show that THC has no effect on driving. It means that a particular concentration-based linear model did not demonstrate the reported relationship under those conditions.
Driving outcomes were not one interchangeable measure
The synthesis considered lateral control, speed, following behavior, reaction time and overall performance. Different tasks may place different demands on a driver. The available studies did not establish that complexity alone caused the contrast in results; that requires systematic comparison.
A correlation cannot classify every driver
Even a group-level association does not supply a perfectly discriminating individual threshold. Conversely, a non-significant result is not proof that every concentration is harmless. Dose, timing, task and the underlying study design remain attached to the inference.
Interpreting the evidence
This review supports carefully stating which relationship was tested and which task was used. It does not justify substituting a blood THC number for a complete performance assessment or dismissing demonstrated driving effects because a simple linear correlation was absent.
Related reading: interpreting cannabis tolerance claims.
Research reference
Danial Behzad; Sampson Zhao; Reena Besa; Bruna Brands; Christine M Wickens; Marilyn A Huestis; Bernard Le Foll; Patricia Di Ciano (2025). Association of driving with blood delta-9-tetrahydrocannabinol: a systematic review. International Journal of Neuropsychopharmacology. doi:10.1093/ijnp/pyaf021.
This discussion draws on the abstract and introductory material in the supplied research synthesis.
