In late December, Research and Practice, a gut-suckingly dry peer-reviewed science journal, published an article that briefly turned the marijuana world on its head. Cosigned by a half-dozen Columbia University PhDs, “U.S. Traffic Fatalities, 1985-2014, and Their Relationship To Medical Marijuana Laws” came to the following shocking conclusion:
“Both MMLs (medical marijuana laws) and dispensaries were associated with reductions in traffic fatalities, especially those among aged 25 to 44 years.”
In other words, these scientists said, when it comes to driving, marijuana laws save lives. As I read the coverage in the marijuana publications that I frequent for solely professional purposes, I thought, how can this be possible? It’s oft repeated in weed circles that driving stoned is better than driving drunk, but usually with a tone of, “I’m not drunk, I’m just buzzed.” I never considered that could be backed up by actual science.
To find out what was going on, I emailed one of the authors of the study, a Columbia professor named Dr. Silvia Martens. She referred me to a grad student named Julian Santaella, who’d come up with the idea. Santaella told me that in August 2014, for his epidemiology methods exam, one of the questions was, “Are medical marijuana laws associated with traffic fatalities?”
“This was a health outcome that people care about,” Santaella said. “People are concerned about recreational use. Everyone assumed there would be more deaths.”
“To me,” Santaella said, “That was a fascinating question.” He dug into the available literature, looking at medical marijuana survey data and traffic fatality lists from the National Traffic Safety Board, and framing his answer. “I got a good grade,” he said. “At that point, I said, okay, this could be a nice study.”
Santaella went to Professor Martins, who he knew was studying various outcomes of medical marijuana laws, and offered up his data. They showed it around to numerous faculty members, who agreed that it was important information that could be very much in the public interest. “This was a health outcome that people care about,” Santaella said. “People are concerned about recreational use. Everyone assumed there would be more deaths.”
Santaella found one paper, published in 2012, that concluded medical marijuana laws led to a reduction of traffic fatalities. But in true academic style, he found that paper’s methodology troubling, or at least limited. It only computed averages from year to year. Santaella wanted to introduce multiple variables. He went for what he calls, appropriately enough in car-related research, a “hybrid model.”
The researchers felt that the previous study hadn’t included enough variables. Even though overall fatalities were going down, it could be because states were putting more law enforcement on the roads once marijuana laws got on the books. The scientists hypothesized that as more and more people got access to marijuana, traffic fatalities would begin to rise again. But their data didn’t match their hypothesis.
“My original thinking was that when people use more marijuana you have more people driving under the influence. That tells me you should expect more people getting into accidents and dying from it,” Santaella says. “As I ran more and more analysis and data, models were still showing me there was a reduction. I was definitely surprised by that.”
The study was rigorous and thorough. They ran through 14 different variables, adjusting for unemployment rate, median household income, traffic laws, seat belt regulations, driver’s license rules, and alcohol and cell phone laws. The conclusion didn’t budge. In every state they surveyed, with the difficult-to-understand exceptions of Connecticut and Rhode Island, traffic fatalities had gone down significantly among young people when medical marijuana laws went on the books. It was a study worth publishing.
The scientists hypothesized that as more and more people got access to marijuana, traffic fatalities would begin to rise again. But their data didn’t match their hypothesis.
The reaction surprised and pleased the researchers, who generally don’t receive a lot of public attention. Public response was mostly positive. Santaella said, “We got emails that said, I’ve been smoking marijuana for 50 years and never got a ticket or never got into an accident. I didn’t want to encourage that behavior.”
Santaella and Martins were rewarded with one of Internet media’s most coveted gigs: A Reddit Ask Me Anything. “One explanation for our findings,” their introductory post went, “might be that states that legalized medical marijuana have seen reductions in drunk driving, in part because some people are substituting marijuana for alcohol.” People in drunk-driving crashes, they said, sometimes had cannabinoids (organic compounds contained in marijuana) in their system, but those often stay in the body, inactive, for days after consuming cannabis. In other words, correlation didn’t equal causation. From there, the AMA continued deep into the weeds, so to speak, of marijuana science, but also featured practical notes from commenters like “in states where you can legally buy pot there would be a lot less people smoking at their dealer's and then driving home.”
In the conversation, the scientists made it clear that they weren’t advocating for stoned driving. Impairment is bad at any level. Drivers should be as clear-headed as possible when they get behind the wheel. They were just saying that fatalities decreased when medical-marijuana laws went into effect. As to why, Santaella says, only science can provide those answers.
“To my understanding,” he says, “this is only the second paper trying to understand this question. I hope there are many more to come.”