A study published in the American Journal of Public Health determined traffic fatality data did not significantly increase in states where cannabis is socially legal.

Researchers analyzed data from the states of Washington and Colorado and used eight control states without social cannabis legislation. The control states were nonadjacent, which allowed for a stronger analysis than if adjacent or randomly selected states were chosen.

Year-after-year changes in traffic fatality rates compared the per billion miles vehicles traveled before and after social cannabis was legalized to data gathered from the eight control states. 

The study showed Washington and Colorado did not statistically differ from control states, where cannabis is not yet legal.

Between 2009 and 2015, 60,737 motor vehicle crash fatalities occurred in Washington, Colorado, and the 8 control states. During this period, annual crash fatality rates decreased from 12.8 fatalities per billion miles in 2009 to 11.4 fatalities in 2015.

Data was collected using the U.S. Fatality Analysis Reporting System (FARS), a nationwide census which reports traffic fatality and crash data. Colorado and Washington are the only states which have post-legalization FARS data, with 2015 being the most recent year the data was reported.

This study contradicts an earlier study conducted by an insurance-industry non-profit which reported higher collision claims in states where cannabis is socially legal.