Last week, a surprising new study suggested that medical cannabis laws were not, as previously believed, associated with a decrease in opioid-related fatalities. This was a departure from a widely-cited 2014 JAMA Internal Medicine study that indicated that in the states that had legalized medical cannabis, there was a 25% lower rate of opioid-related deaths than in states that banned cannabis.
That 2014 JAMA opioid study has been cited hundreds of times, including by cannabis industry groups, often in the context of the national debate for changing cannabis laws. But the new research, published in Proceedings of the National Academy of Sciences, sought to replicate that 2014 study and found unexpected results: When the time frame, originally 1999-2010, was expanded to include changes in state laws through to 2017, researchers found a startling reversal: states that had passed medical cannabis laws had 23% more opioid-related fatalities than those without such laws.
One part of the study really caught our eye: “The nonrobustness of the earlier findings also highlights the challenges of controlling scientific messages in controversial policy areas,” the study read. “Corporate actors (e.g., the medical cannabis industry) with deep pockets have substantial ability to promote congenial results, and suffering people are desperate for effective solutions. Cannabinoids have demonstrated therapeutic benefits… but reducing population-level opioid overdose mortality does not appear to be among them.”
Cannabis Wire followed up with Chelsea Shover, one of the lead researchers on this study, about that very topic. Here is what Shover said: (This interview has been lightly edited for space and clarity.)
Alyson Martin, Cannabis Wire: Part of the study that caught my eye was about the “nonrobustness of the earlier findings,” which the researchers like you said highlighted “the challenges of controlling scientific messages in controversial policy areas.” What prompted that part of the discussion?
Chelsea Shover, Stanford: There were four of us on the paper. We did discuss all aspects, including that particular sentence. They [the researchers on the 2014 study] did everything right in terms of saying in the paper that ‘this is not causal, this is a correlation,’ and sort of explaining why ecological studies can’t really be used to understand individual level behaviors. But, since then, that study’s been cited over 300 times, it’s informed a lot of industry campaigns. Most recently, I’ve seen the Weedmaps billboard and their articles about it. And so, it’s definitely been a talking point for advocates and industry. And that’s understandable. But it’s also got a lot of traction in the science community. And so what we wanted to do was say, you know, this is something that has gotten a ton of attention and it’s been repeated a lot—let’s see if it still holds up.
CW: Do you have any thoughts about how research is used for industry purposes?
Shover: I think, given the history between industry and science in the United States, I think it is wise for the consumer to be skeptical of research-related claims made by industries, or lobbying groups, or just general advocates. I think it would be a mistake to get all your information from an actor that has a very vested interest in something going a certain way. And I think that’s the role where science could step in. We recognize that people who like results in one direction or the other are going to seize on research findings that seem congenial to their goals and report them somewhat out of context, or make them sound more causal than they are.
CW: What other studies would you like to see replicated or pushed forward now that we have a little bit of hindsight after legalization in the first two states?
Shover: If the question is whether or not cannabis can be helpful in helping people stop using opioids, or reduce their use, I would want to see a prospective study with a large sample of people who are using opioids, and some who are exposed to medical cannabis and some who aren’t. And tracking it over time. We’ve seen a little like that, but I would love to see that in the US context in a big, robust study. I think that’s an important thing to consider.