When an entirely new method of intoxication becomes legal, one might assume that people crash their cars more. According to data from the Highway Loss Data Institute (HLDI), that's exactly what's happening.
In Colorado, Oregon and Washington -- three states with legal, recreational marijuana -- collision claims are approximately 3 percent higher than they should have been without legal weed, according to the HLDI's analysis.
The HDLI didn't just look at those three states. It also tracked the same information in control states: Idaho, Montana, Nevada, Utah and Wyoming, each of which has varying levels of legalization between full prohibition and medical use. Researchers also looked at prelegalization data in Colorado, Oregon and Washington.
The data covered collision claims between January 2012 and October 2016 for vehicles between 1981 and 2017 model years, and controlled for population, urban-rural mix, unemployment and even weather. Since it only investigated collision claims with insurers, crashes kept off the books are obviously not included.
The result of this research shows a higher incidence of collision claims following legalization. Colorado saw a 14-percent claim frequency jump compared to Nebraska, Utah and Wyoming. Washington jumped by 6.2 percent compared to Montana and Idaho, and Oregon's claims went up by 4.5 percent compared to Idaho, Montana and Nevada.
This is just the beginning. The Insurance Institute for Highway Safety has started a large case-control study in Oregon to see how legal cannabis affects the risk of injuries in collisions. The results of that one should be ready in 2020, so I hope you don't mind waiting.
It bears mentioning that this only looks at crashes, not injuries or fatalities. A 2016 study from Columbia University looked at fatalities in 19 different states before and after medical marijuana was legalized. There was an average 11 percent reduction in fatalities, but on a state-by-state basis, only seven saw a reduction, while 10 saw no change and two actually saw higher fatality rates.
It's also important to note that correlation does not imply causation. There could have been a number of other factors involved, and researchers can't possibly control for every single factor. But based on the data the HLDI crunched, perhaps it's best if drivers in Colorado, Washington and Oregon started walking and biking a bit more.