The study, by a former University of Colorado graduate student, looked at data on twins to answer a longstanding question.
In the decade since Colorado voters passed Amendment 64 legalizing adult use of marijuana in state law, a slew of studies have tried to estimate the impact.
They’ve looked broadly and more narrowly, and have concluded — depending on how you might read them — that legalization either did or did not lead to an increase in cannabis use.
But a new study, conducted by a former Ph.D. student in Colorado, takes perhaps the most novel approach yet and comes to the conclusion that legalization may, in fact, lead to people using marijuana more frequently.
The student, Stephanie Zellers, was studying neuroscience at the University of Colorado before she followed her adviser to finish up her doctorate at the University of Minnesota. She was interested in studying the effects of substance use on the brain, but a lot of studies on the brain work the same way: you have to crack open the craniums of lab animals. That didn’t sit well with Zellers.
So she went looking for a different method, and she found it in a massive dataset on the lives of twins born in Colorado or Minnesota.
The data are longitudinal, meaning they follow the same people over long periods of time. And twins make for intriguing research subjects because they are not only genetically similar but they also generally grow up in the same household, meaning they typically share the same socioeconomic background, education experience and other social factors.
That helped solve two big problems other studies on legalization have run into. How can you be sure that aggregate changes in marijuana use are the result of individual people changing how they use marijuana and not population changes that occurred post-legalization? And how can you fairly compare the behaviors of two people when they might differ in so many ways?
The dataset also neatly straddled a dividing line — Colorado voters have legalized marijuana for recreational use, while Minnesota, which is otherwise similar to Colorado in a number of ways, has not.
Using the data, Zellers looked at more than 3,400 adult twins born in Colorado and Minnesota and charted where they were living just prior to legalization. Some were still living in the state of their birth, some had moved; some pairs were living in the same state while other pairs had split and were living in separate states. The variability allowed her to effectively create a randomized controlled trial, something normally not possible when you’re talking about government policies.
“I think we got really lucky with legalization happening the way it did and happening to be legal in one state and not in the other,” Zellers said in a video interview from Finland, where she is conducting post-doctoral research on tobacco use. “It’s pretty rare to have this amount of data on this many people going this far back.”
So, the results?
A twin living in a legal state, on average, reports using marijuana more frequently than their co-twin living in a state where recreational marijuana use is illegal, according to the study.
Zellers said the difference is about 20% — which amounts to consuming cannabis on five more days per every six months. Twins living in states where recreational cannabis use is illegal, on average, consumed on 13 days in the previous six months, while twins living in legal states consumed on 18 days. (Zellers said she and her coauthors weighted the data for age and other factors to standardize it, which is why the difference between those numbers isn’t exactly 20%.)
“This is about as conclusive as we can say that the policy does cause more use,” Zellers said.
The study’s finding is in line with those from a large-scale state report released last year, which found that cannabis use has increased among adults in Colorado since legalization. (The trends around youth use are more muddled, but there is ultimately no conclusive evidence that legalization caused more kids to use cannabis.)
The novel approach of the research, though, provides it with extra heft, another of the study’s authors said.
“This is the first study to confirm that the association between legal cannabis and increased use holds within families in genetically identical individuals,” John Hewitt, a professor in CU’s Department of Psychology and Neuroscience, said in a statement. “This makes it much more likely that legalization does, in itself, result in increased use.”
Beyond the headline result, there are still a lot of unanswered questions in the new study.
For one, these are self-reported statistics. So it’s possible that twins living in states where marijuana use is illegal were less likely to report their true levels of use.
There’s also the issue of motivation — if this increase is legit, why did it happen?
“We don’t know why someone might have changed their use,” Zellers said.
Lastly, the study doesn’t explore potential consequences to the apparent increase in cannabis use. Did people suffer worse health problems? Did they cut back on their use of alcohol or other substances?
Zellers said that is the subject of a follow-up study that she hopes will be published by the end of the year.