Cannabis and traffic collision risk: findings from a case-crossover study of injured drivers presenting to emergency departments

This study examined whether acute cannabis use leads to an increased collision risk. Participants were 860 drivers presenting to emergency departments in Toronto and Halifax, Canada, with an injury from a traffic collision, between April 2009 and July 2011. Cannabis and other drug use were identified either through blood sample or self-report. A case-crossover design was employed with two control conditions: a fixed condition measuring substance use during last time driving, and whether the driver typically uses cannabis prior to driving. Collision risk was assessed through conditional fixed-effects logistic regression models. Results revealed that 98 (11 %; 95 % CI: 9.0-13.1) drivers reported using cannabis prior to the collision. Regression results measuring exposure with blood and self-report data indicated that cannabis use alone was associated with a fourfold increased (OR 4.11; 95 % CI: 1.98-8.52) odds of a collision; a regression relying on self-report measures only found no significant association. Main findings confirmed that cannabis use increases collision risk and reinforces existing policy and educational efforts, in many high-income countries, aimed at reducing driving under the influence of cannabis.