Driving while dangerously impaired on marijuana is reckless and irresponsible, but that doesn’t make a recent study published online in the American Journal of Epidemiology any less full of shit when it claims 'Fatal Car Crashes Involving Pot Use Have Tripled in the US,' as Web MD and countless other outlets ran the headline.

“Currently, one of nine drivers involved in fatal crashes would test positive for marijuana," co-author Dr. Guohua Li, director of the Center for Injury Epidemiology and Prevention at Columbia, told the press. "If this trend continues, in five or six years non-alcohol drugs will overtake alcohol to become the most common substance involved in deaths related to impaired driving."

Looking at six states where officials routinely perform toxicology tests following traffic fatalities, researchers from Columbia University's Mailman School of Public Health found that alcohol has contributed to about the same percentage of traffic fatalities over the last decade (40%), while marijuana has moved from 4% to 12% -- the “tripling” touted throughout the media.

What most of those stories don’t tell you, however, is that the study included an endnote admitting that since marijuana can be detected in the blood for a week after use, "the prevalence of non-alcohol drugs reported in this study should be interpreted as an indicator of drug use, not necessarily a measurement of drug impairment."

Or, in laymen’s terms: Our entire study was basically a bunch of malarkey. Because we don’t make any distinction between someone who took twenty dabs five minutes before jumping in the car and someone who smoked half a joint six days ago.

And so, for those interested in actually learning the relative risks of marijuana and driving, NORML’s authoritative review of the scientific evidence remains the best available source.