Impact of Canada’s Cannabis Act on drug- and alcohol-related collisions in Québec: an interrupted time-series analysis of five major cities

This study examines the impact of non-medical cannabis laws (NMCLs) on road safety outcomes, specifically focusing on drug- and alcohol-related traffic crashes. Using cannabis sales data as a proxy for consumption trends, the study aims to assess how changes in cannabis availability may influence road safety outcomes, particularly exploring the potential for drugs and alcohol to have distinct yet related influences on impaired driving. An interrupted time-series design was used to assess the impact of NMCLs on daily drug- and alcohol-related traffic crashes, including fatalities and severe injuries (KSI). The analysis covered five cities in the province of Québec—Montréal, Québec, Laval, Longueuil, and Sherbrooke—using data from January 1, 2015; to December 31, 2022. The dependent variables included KSI, drug-related crashes, and alcohol-related crashes, while the independent variables were daily cannabis legal sales (kg) and total legal and estimated illegal cannabis sales. Control variables accounted for temperature, time trends, and the COVID-19 non-pharmaceutical interventions’ index for the province of Québec (QCnPI-Index). To estimate effects, we applied Generalized Linear Models using Negative binomial regression, followed by a random-effects meta-analysis to assess overall effects across cities. Higher total cannabis sales were significantly associated with a 12% increase in drug-related crashes (IRR: 1.12; 95% CI: 1.01–1.27) and a 12% rise in alcohol-related crashes (IRR: 1.12; 95% CI: 1.06–1.18) across all cities combined. In Montréal, cannabis sales were linked to an 87% increase in drug-related crashes (IRR: 1.87; 95% CI: 1.54–2.28) and a 93% increase in alcohol-related crashes (IRR: 1.93; 95% CI: 1.58–2.36). In Longueuil, drug-related crashes rose by 76% (IRR: 1.76; 95% CI: 1.02–3.02) and alcohol-related crashes by 43% (IRR: 1.43; 95% CI: 1.08–1.92). Québec City only showed a 44% increase in alcohol-related crashes (IRR: 1.44; 95% CI: 1.28–1.64). No significant associations were found in Laval or ...