Did legal marijuana lead to a reduction in opioid deaths in Colorado? Evidence is hazy

Opioid overdose deaths have been increasing in Colorado, so why did a recent article in The American Journal of Public Health suggest that legalization of marijuana has slowed the trend?

I imagine my Facebook newsfeed looks a lot like yours. Pictures of babies you’ll never meet, event pages for LBJ lunch talks, and the occasional shared article from a questionable website. Sometimes, however, the article seems legitimate enough to warrant a click. Recently, the headline of one such article was “Recreational Marijuana is Reversing the Opioid Crisis in Colorado.”

Given that I spent the last two years writing about the increase in Colorado’s opioid overdose death rate, I clicked.

Before coming to the LBJ School, I was a policy analyst at the Colorado Health Institute – a nonprofit, nonpartisan health policy research shop based in Denver. I led the Institute’s behavioral health portfolio, which included research on opioid data and policy, and specifically analyzed data on opioid overdose deaths. After many hours spent analyzing Colorado’s opioid data, I can tell you with certainty: the opioid overdose rate in Colorado has been increasing steadily since 2000. In fact, opioid overdose deaths increased by over 250 percent in Colorado from 2001 to 2015.

So, why does this article claim that the “legalization of cannabis in Colorado was associated with short-term reductions in opioid-related deaths”?

The referenced analysis comes from an article in a reputable journal, the American Journal of Public Health, after researchers ran a regression model with the data. Essentially, they looked at the pre-legalization trend and compared it with the post-legalization trend starting in 2014. When they compared these trends, they found an estimated 6.5 percent reduction in opioid-related deaths after legalization.

Their methods are sound. However, there are some important confounders to consider. The authors acknowledged that there were changes to Colorado’s Prescription Drug Monitoring Program (PDMP) during this time, which is a way for providers to monitor the prescriptions their patients already received from other providers. Yet, they failed to acknowledge that Colorado has made strides in harm reduction legislation from 2014 through this latest legislative session, including a bill that increased access to the overdose antidote Naloxone. When administered to someone experiencing an overdose, Naloxone can reverse the effects of the opioid and save the person’s life. A series of legislation has made it easier for friends and family, along with first responders, to carry the life-saving drug.

The more questionable aspect of the research is that they did not specify a mechanism for their findings. If opioid deaths in fact were decreasing due to marijuana legalization, what exactly was the correlation between the two? One hypothesis might be that chronic pain patients are now substituting opioids like Oxycontin for marijuana. However, medical marijuana has been available in Colorado since 2000. Could an increase in the number of dispensaries after recreational legalization motivate more medical users to obtain prescriptions? Another hypothesis might be that those who were obtaining opioids illegally transitioned to legal marijuana. However, how do we relate this to the trend between heroin and marijuana? These questions point to the need for more research – and more time – to make the case that the legalization of marijuana has had an impact on opioid overdoses.

While I am open and optimistic about marijuana’s potential to change the way we treat pain in the United States, I worry about how the message is delivered. When we reduce articles to a single headline, it also reduces the complexity of our country’s opioid overdose epidemic. As policy professionals, it is our duty to make sure we not only read the research methodology and analysis, but also understand the underlying nuances.

Leave a Reply

Your email address will not be published. Required fields are marked *

Social Widgets powered by AB-WebLog.com.