For two decades, the opioid crisis has been accelerating. Between 1999 and 2010, overdose deaths related to opioids grew by 9.1% per year, and between 2010 and 2018, by about 12% per year. Between June of 2019 and May of 2020, the rate of overdose deaths grew by 18%.
In raw numbers, that translates into roughly 47,000 overdose deaths in 2018—or five people every hour.
With this expanding epidemic in the background, Balázs Kovács, associate professor of organizational behavior at Yale SOM and Greta Hsu of the University of California Davis were examining questions of strategic positioning through the study of cannabis dispensaries, and wondered if the data he had collected could shed light on the way in which legalized cannabis and opioid deaths interact.
“I’m not an expert on opioids, but what seems very clear is that the action of choosing one drug is not independent of other drugs,” Kovács says. “Substitutes like alcohol or cannabis might affect how people use other drugs.”
Kovács and Hsu were not the first to make this connection. Since 1996, when California became the first state in the country to legalize medical marijuana, researchers have wondered how such a law might affect the consumption of other drugs. Today, the question holds particular pertinence given nearly half of the 50 states have legalized either recreational or medical cannabis. How, researchers have wondered, does state-level cannabis legalization influence the ballooning problem of opioid overdoses?
But, as Kovács points out, the passage of state laws provides a rather coarse lens: once cannabis is legalized, it still takes a long time for dispensaries to get permits, open storefronts, and develop a customer base. Years after legalization, many areas across a given state still have no dispensaries present; many will never have one.
To sharpen the picture, Kovács and Hsu used data describing the number of dispensaries that have had a sustained business presence in a given county. “Going to the local level like this allowed us to produce a much more fine-grained analysis,” says Kovács.
An increase from one available dispensary in a county to two is associated with a 17 percent reduction in opioid-related overdose deaths; an increase from two to three is associated with a further 8.5 percent reduction.
Looking at medical and recreational dispensaries in 812 counties across 23 states and Washington D.C., they found that an increase from one available dispensary to two is associated with a 17% reduction in opioid-related overdose deaths; an increase from two to three is associated with a further 8.5% reduction. These figures vary slightly depending on whether medical or recreational cannabis is available and on the type of opioid-related death. (Deaths from fentanyl, for example, decline most sharply in counties with one or more dispensaries.) But the overarching result raises an intriguing prospect.
Kovács is quick to acknowledge that these results, published in the British Medical Journal, don’t provide causal evidence: the legalization of cannabis and the establishment of dispensaries doesn’t necessarily cause fewer opioid deaths. But he sees at least three possible explanations for the correlation that are worth further investigation. First, when cannabis is legal and available, some people may opt to purchase and use it rather than opioids. Second, people who have started using opioids may wean themselves from their addiction with legal cannabis. And third, legal cannabis, through regulation, is safer than illicit product.
“In the Northeast at least, we see a lot of cannabis on the black market that gets laced with fentanyl,” Kovács says. “It could be that people die because they buy a bad batch of cannabis and they OD on fentanyl. Even if it’s unintentional, that would show up in the data.” Purchasing through legal channels would eliminate this risk.
Kovács hopes policymakers will start to consider jointly what may, at first, appear to be disparate concerns: the opioid epidemic and legalized cannabis. “There are obvious benefits to having state control of certain things,” he says. “Even when we’re talking about drugs, the state has to think about whether to ban them, allow them, or control them, and there are important secondary effects with all of these decisions.”