CannEpil Driving Trial: Speed Variability Is Different From Average Speed
A controlled trial of CBD-dominant cannabis oil found increased variability in driving speed, selected lane-position changes and some eye-movement differences. Its speed-variability endpoint should not be summarized as a demonstrated increase in average driving speed.
A low-THC combination still required functional testing
Thirty-one healthy licensed drivers received a single sublingual dose containing 100 mg CBD and 5 mg THC, or placebo, in a randomized crossover study. A 40-minute highway simulator drive began 90 minutes after dosing.
The main driving measures were standard deviation of lateral position, standard deviation of speed and steering variability. Eye tracking was usable for only 14 participants because of calibration and equipment limitations. Blood cannabinoid results were reported in plasma, separately from oral-fluid THC.
What the speed result means
Standard deviation of speed measures how much speed fluctuates around a driver’s average. The active condition had significantly greater speed variability overall, with a particularly clear difference during the third ten-minute driving segment.
That is a concern about consistency of control, rather than proof that participants drove faster on average. A driver can fluctuate more while maintaining the same mean speed. Preserving the actual endpoint avoids changing the behavioral meaning of the finding.
The lane and eye findings were mixed
Lane-position variability did not show a significant overall treatment effect or treatment-by-time interaction, although a post hoc comparison found greater variability during the third segment. Steering variability did not show a treatment effect.
The smaller eye-tracking sample showed longer fixation duration and shorter blink duration in selected comparisons. Other gaze measures and microsleeps did not significantly differ. Self-rated driving quality also did not significantly distinguish active treatment from placebo.
These results support reporting the full pattern. Neither the isolated lane comparison nor the overall lane null justifies declaring every aspect of performance impaired or unaffected.
Concentrations did not identify the driving deficit
The study did not find significant correlations between measured cannabinoid concentrations and driving-performance outcomes. An association between oral-fluid THC and blink rate was a separate finding in the small ocular sample, not validation of a driving-impairment classifier.
Participants were healthy and largely nonregular users, received one dose and performed a simple simulator task. The design did not estimate real-world collision risk or establish a safe concentration or waiting period. It also cannot determine whether CBD reduced THC effects because it lacked a THC-only comparison at the same dose. A high CBD:THC ratio describes composition; it does not replace direct evidence about performance.
Related analysis: Concentration Is Not Impairment: THC Blood Levels in California Driving Cases.
Primary research
Brooke Manning and colleagues. Effect of CannEpil® on simulated driving performance and co-monitoring of ocular activity: A randomised controlled trial. Journal of Psychopharmacology (Oxford, England) (2023). DOI: 10.1177/02698811231170360.
