Cannabis policy bundles and traffic fatalities in the American States over time

Since 1996, 38 US states have legalized access to cannabis (medical and/or adult-use recreational). We aimed to estimate the effect of three dimensions of state cannabis policy design - pharmaceutical, permissive and fiscal - on levels of overall, alcohol-impaired, occupant, light truck and pedestrian fatality rates.

Observational study of US states' overall, alcohol-impaired, occupant, light truck and pedestrian fatalities between 1994 and 2020.

The unit of analysis was at the state level, consisting of 50 states and 27 years of time series data, resulting in a total of 1350 state-year observations.

Fatality rates associated with alcohol-impaired, pedestrian, total occupant, passenger car and light truck fatality rates were obtained from the Fatality Analysis Reporting System of the National Highway Traffic Safety Administration and normalized per 10 billion vehicle miles traveled. State cannabis policies are measured in three bundles (scales): pharmaceutical, permissive and fiscal.

The pharmaceutical bundle was associated with increases in all fatality rates [β = 0.145; 95% confidence interval (CI) = 0.116-0.173; P < 0.000]. The permissive bundle was associated with lower overall fatality rates (β = -0.319; 95% CI = -0.361 to -0.277; P < 0.000). The fiscal bundle was generally associated with higher fatality rates (β = 0.062; 95% CI = 0.043-0.081; P < 0.000), occupant (β = 0.070; 95% CI = 0.042-0.098; P < 0.000), light trucks (β = 0.049; 95% CI = 0.026-0.072; P < 0.000).

US state cannabis regulations influence traffic safety. Greater permissiveness in US state cannabis regulations does not appear to correlate with traffic fatality rate increases, but greater medicalization and fiscal operation does.