A study published by the American Journal of Epidemiology last month found that 12.2 percent of drivers killed by car crashes in six states tested positive for cannabinol, a marijuana metabolite, in 2010, up from 4.2 percent in 1999. Here is how NBC News translated that finding in the headline over a story posted on Saturday: "Pot Fuels Surge in Drugged Driving Deaths." The article, which begins by describing the deaths of a Colorado woman and her infant son in a crash caused by "a driver who admitted he smoked pot that day," links the purported surge in marijuana-related traffic fatalities to laws allowing medical use of cannabis. "As medical marijuana sales expanded into 20 states," writes health reporter Bill Briggs, "legal weed was detected in the bodies of dead drivers three times more often during 2010 when compared to those who died behind the wheel in 1999." There are several problems with reading the trend described by this study as evidence that legalizing medical marijuana causes an increase in fatal car crashes:
1. The fact that cannabinol was detected in a driver's blood does not mean he was under the influence at the time of the crash, let alone that marijuana caused the crash. "It is possible for a driver to test positive for cannabinol in the blood up to 1 week after use," the researchers note. "Thus, the prevalence of nonalcohol drugs reported in this study should be interpreted as an indicator of drug use, not necessarily a measurement of drug impairment."
2. Only three of the six states included in the study (which were chosen because they routinely do drug testing on drivers killed in crashes) have medical marijuana laws: California, Hawaii, and Rhode Island.
3. Traffic fatalities fell by more than 20 percent nationwide during the study period, even as "medical marijuana sales expanded." Between enactment of its medical marijuana law in 1996 and 2010, California saw a 31 percent drop in traffic fatalities. The number of traffic fatalities also fell in Hawaii and Rhode Island after they legalized medical marijuana—by 14 percent and 21 percent, respectively.
4. A study published last year by the Journal of Law & Economics found that adoption of medical marijuana laws is associated with a decline in traffic fatalities, possibly because people in those states are substituting marijuana for alcohol, which has a more dramatic impact on driving ability. Briggs mentions that study in the 17th paragraph of his article.
It is important to keep these points in mind as more states liberalize their marijuana laws, especially since "preventing drugged driving" is one of the "enforcement priorities" that the Justice Department says might justify federal interference with legalization in Colorado and Washington. If "drugged driving" means operating a motor vehicle with any detectable amount of cannabinol in your blood, "drugged driving" inevitably will rise after legalization as consumption rises. But having cannabinol in your blood is not the same as being intoxicated. And even if the share or absolute number of traffic fatalities caused by marijuana-related impairment rises, the total number of traffic fatalities could still drop thanks to substitution effects. Regardless of what happens with traffic fatalities, the possibility of marijuana-related accidents is a reason for discouraging people from driving while impaired, not a reason for prohibiting the drug altogether.