States with medical marijuana laws have fewer traffic fatalities than those without, especially among younger drivers, a new study found.

You would think crash rates might be higher, supposing that more drivers are, too – especially around midnight, when a run to a 7-Eleven becomes necessary.

But, no. Researchers at Columbia University’s Mailman School of Public Health found an 11 percent reduction in traffic fatalities on average when examining places that have enacted medical marijuana laws – 23 states and the District of Columbia. The presence of medical marijuana dispensaries also correlated with fewer traffic fatalities, the study found.

Silvia Martins, a physician and associate professor who was the study’s senior officer, theorized that lower traffic fatality rates in states with marijuana laws might be related to lower levels of alcohol-impaired driving: People, especially younger people, began substituting weed for booze.

There was little evidence of reductions in the traffic fatality rate for people 45 or older, who are disproportionately represented in larger numbers among people enrolled in state medical marijuana programs. The largest reduction in traffic fatality rates in states with medical marijuana laws occurred among drivers ages 15 to 44.

“We found evidence that states with the marijuana laws in place compared with those which did not, reported, on average, lower rates of drivers endorsing driving after having too many drinks,” Martins said in a written statement. She said other factors that might partially explain the correlation could be the “strength of public health laws related to driving, infrastructure characteristics, or the quality of health care systems.”

The researchers cautioned, however, that not all states with medical marijuana laws experienced reductions in the traffic fatality rates. California and New Mexico experienced initial reductions of 16 percent and 17.5 percent, respectively, after the passage of medical marijuana laws but then saw gradual increases in their traffic fatality rates.

The study – which was published online in the American Journal of Public Health – analyzed National Highway Traffic Safety Administration (NHTSA) data on traffic fatalities from 1985 to 2014. The researchers took into account factors such as whether the states had graduated driver licensing laws, as well as median household income, unemployment rates, laws increasing the speed limit to 70 mph or more, laws on enforcing the use of seat belts, and bans on using cellphones and texting while driving.

The researchers undertook the study as more states passed medical marijuana laws – nine between 2010 and 2014 – and concerns were raised about the possible effect on public safety. Previous research, for example, found that marijuana use caused drivers to weave in traffic more often and have slower reaction times; studies also found that drivers who were high tended to reduce speed and increase headways – which suggest that the users were aware of marijuana-related impairment and were trying to compensate.