Why “Study” Linking Medical Marijuana with Driving Fatality Reductions is Flawed

By Gen. Arthur Dean, CADCA Chairman & CEO

A study that has not been peer-reviewed but rather published as part of a “discussion series” by a German labor journal is getting a lot of publicity. Time Magazine, Join Together, the Denver Post, and other outlets have published stories about the study. The study concludes that traffic fatalities are reduced in medical marijuana states because, as the author’s assert, “alcohol and marijuana are substitutes.” So, as substitutes, the authors contend that medical marijuana laws cause people to drink less and therefore not drive drunk. This is most certainly a spurious and coincidental relationship, however, as a large body of data points to other reasons why we have witnessed historic reductions in road fatalities over the last 20 years. Just because A happened when B happened does not mean that A caused B. The study seems to completely ignore the fact that crash fatalities were already coming down far before the introduction of medical marijuana.

CADCA takes issue with this non-reviewed study for three primary reasons:
(1)    The study methodology is greatly flawed; and,
(2)    The study’s authors disregard a large body of evidence showing that marijuana and alcohol are    compliments; and,
(3)    The study’s authors disregard mounting evidence that marijuana use is linked with impaired driving

In greater detail:

(1)    The study methodology is greatly flawed
First, the authors examine only three states as being representative of “medical” marijuana, Rhode Island, Vermont, and Montana. To test effects of the medical marijuana law, they divide each state into a period they call “before” medical marijuana (1999-2003) and after (2005-2009). These three states, however, are not representative of widespread medical marijuana programs – indeed in 2009 Rhode Island and Vermont had only about 300 and 550 members of their medical marijuana programs, respectively. Montana, the authors contend “had more than 27,000” members in 2011. What the authors neglect to mention, however, is that in 2009, when their study period ends, Montana had only 6,000 members, up from a mere 800 in 2008. The vast majority of the growth in the medical marijuana program occurred after the time period studied by the authors.

Furthermore, the methodology ignores some very important recent findings. First, the study ignores previously established data showing increases in marijuana use in medical marijuana states when they state that “Although opponents of legalization argue that it encourages recreational use among teenagers we find no evidence that the use of marijuana by minors increased.” They ignore two major studies by Cerda and colleagues as well as Wall and that don’t conform to their conclusions.   Finally, in almost every section where arguments advanced by “proponents of legalization” appears, references to “clinical studies” are brought up. Authors say that legalization advocates “cite clinical research showing that marijuana relieves chronic pain, nausea, muscle spasms and appetite loss and note that neither the link between medical marijuana and youth consumption, nor the link between medical marijuana and criminal activity, has been substantiated.” Besides those being false statements not backed up by science, it is interesting that when arguments from legalization opponents are discussed, the authors say that they are claims they cannot substantiate.

Finally, the study completely ignores other possible reasons why there may be a reduction of alcohol use fatalities in these states. The study does not measure the effect of anti-drunk driving programs, MADD campaigns, coalition work, or other such possible reasons. The study seems to completely ignore the fact that crash fatalities were already coming down far before the introduction of medical marijuana.

(2)    The study’s authors disregard a large body of evidence showing that marijuana and alcohol are compliments, not substitutes.
It is absurd to contend that even when controlling for other factors (the authors did control for various alcohol laws) medical marijuana could cause a reduction in fatalities. How do the authors explain this? Their guess is that alcohol and marijuana are substitutes, not complements. But although they cite four studies showing that alcohol and marijuana are complements, they simply disregard those in favor of their own conclusion that the drugs are substitutes. 

(3)    The study’s authors disregard mounting evidence that marijuana use is linked with impaired driving.
Finally, the study’s authors explicitly state that “neither simulator nor driving-course studies provide consistent evidence that impairments to driving-related functions (resulting from marijuana) lead to an increased risk of collision.” This statement flies in the face of the bulk of the evidence on marijuana and driving. Perhaps the most robust evidence linking cannabis use and driving comes from a new meta-analysis of nine studies conducted by researchers at Columbia University’s College of Physicians and Surgeons. After reviewing these epidemiologic studies from the past twenty years, they found that marijuana use was linked to heightened risk of crash involvement, even when controlling for multiple different variables. Furthermore, they found that the risk of crash involvement increased along with an increase in marijuana potency (tested through urinalysis) and self-reported frequency of use. The researchers commented that “the results of this meta-analysis suggest that cannabis use by drivers is associated with a significantly increased risk of being involved in motor vehicle crashes.”

Furthermore, in the past decade, researchers from all corners of the world have documented the problem of marijuana use and driving. Marijuana is linked to neurological deficits, including the impairment of motor coordination and reaction time, and therefore its use can increase the risk of road accidents in drivers who are under the influence.   Marijuana remains the second most cited drug after alcohol in car crashes. A U.S. Department of Transportation nationally-representative sample found that more than 8 percent of weekend, nighttime drivers tested positive for marijuana, nearly four times of the percentage of drivers with an alcohol BAC of 0.08 or greater.  In a study of seriously injured drivers admitted to a Level-1 shock trauma center, more than a quarter of all drivers tested positive for marijuana.  Research conducted at the University of Auckland, New Zealand also showed that marijuana use and auto crashes are strongly linked. The research found that habitual marijuana users were 9.5 times more likely to be involved in crashes, with 5.6 percent of people who had crashed having taken the drug, compared to 0.5 percent of the control group.  Research has indeed revealed a marijuana and impaired driving link, and therefore states who care about car crashes should be wary of medical marijuana laws, which increase use and in turn increase the chances of driving under the influence.