Study debunks link between Oregon drug decriminalization and overdose deaths.

It's the fentanyl...

PATRICK T. FALLON/AFP via Getty Images

Back in March, I wrote an Inside Medicine about Oregon’s drug decriminalization law. At the time, some observers were linking a precipitous rise in overdose deaths to the rollout of a law that decriminalized possession of small amounts of all drugs. There was a drumbeat to repeal the law. Many experts like Maia Szalavitz (an Inside Medicine reader) argued that this was not the right move.

My take was that the association between the rise in overdose deaths in Oregon and the new law was a flimsy one at best. It looked to me as if the increase in overdose deaths preceded the law going live. Also, a similar pattern was seen in Oklahoma, a state with some of the strictest drug laws in the US. Here’s that piece:

They're blaming deaths on Oregon's drug decriminalization. I'm not sure that's fair.

New study debunks law link, strongly suggests the real cause…

A new study in JAMA Network Open published last week rigorously studied this question. More than just agreeing with their conclusion, I found the approach the researchers took to be rather brilliant.

The researchers started with a simple question: Was Oregon’s drug decriminalization law “M110” the driver of increased overdose deaths in Oregon? Beyond that, they wanted to test another important hypothesis: Was the real culprit behind overdose mortality increases the penetration of fentanyl—a strong synthetic opioid that has supplanted heroin everywhere it has shown up—into the unregulated (i.e., illegal) drug market?

To determine that, the researchers tracked drugs identified in a national law enforcement repository over many years. In a stunning and simple set of graphs, they showed that as fentanyl became more common in drugs seized by law enforcement, the overdose mortality rate rose dramatically in lockstep.

But that wasn’t enough to answer their questions. To finish the inquiry, the researchers also determined when fentanyl took over in each US state. They did this “agnostically,” meaning that they asked computational software to pinpoint exactly when fentanyl gained supremacy in any particular state. Once they had determined the “changepoint” (i.e., the Rubicon moment when fentanyl had become the dominant opioid in the region), they could also see a sudden and horrific change in the mortality slope. Those changes were reliable; fentanyl would take over in a state, and mortality would double or triple in just a few years.

The “non-effect” of Oregon M110.

Enter Oregon’s drug possession decriminalization law, M110. When M110 went live in 2021, fentanyl had just begun its takeover in Oregon. In fact, Oregon was one of the last states in the US to succumb to fentanyl’s invasion onto its unregulated/illegal drug scene. The researchers understood what the arrival of fentanyl in Oregon meant for mortality; it portended a huge increase, just as had happened in dozens of other US states that had been through the same transition. Their model projected that overdose deaths in Oregon would rise from around 11 to around 18 per 100,000 people from 2021 to the end of 2022. That is exactly what happened. But if the researchers had not taken the fentanyl market penetration into account, the predicted rate by the end of 2022 would have only been around 14 per 100,000 people. In other words, unless fentanyl market penetration was taken into account, researchers might have incorrectly concluded that M110 was responsible for the increase. This is why study methods matter so much. Had the researchers been lazy, uncreative, or naive, they easily could have come to the opposite (and incorrect conclusion) that Oregon M110 was behind a spike in drug overdose mortality. In reality, it probably had no effect during the study period.

Washington state data back the theory up elegantly.

Oregon’s neighbor to the north, Washington State, provided a fascinating alternative angle which the researchers explored. There, drugs were decriminalized from 2018-2021. The effect on mortality? Nada. Why? Because fentanyl had not yet flooded the Washington market. Then, in 2021, Washington State recriminalized drugs. That policy changed just so happened to coincide with fentanyl’s takeover there. Guess what happened? Drug overdose deaths in Washington State spiked massively.

By taking fentanyl into account, the researchers had avoided another error, this time in the opposite direction. Recriminalization of drugs in Washington State did not cause the spike in overdose deaths that seemed to come with it. The driving force behind the increased overdose deaths there after 2021 was—say it with me—fentanyl.

This superb study, conducted and written by Brown University School of Medicine researchers, is a tour de force. We should be grateful it exists—and it should remind us to think carefully about data and policies, and to assess dispassionately before we react. Indeed, Oregon has backtracked on key provisions in M110, in large part because people incorrectly jumped to conclusions, rather than being data driven. We should demand better than that.

What should we do?

Now we know that neither drug decriminalization nor drug recriminalization was responsible for increases in drug overdose deaths in Oregon and Washington. Fentanyl was (and is) the problem.

How can we save lives there and elsewhere? It’s an important question. For one, border security would help, as fentanyl is indeed coming from Mexico and China. Second, we need to destigmatize treatment for substance use disorders (like buprenorphine and methadone, both of which replace dangerous IV drugs like fentanyl, helping people stay alive and productive) and make naloxone (“Narcan”) antidotes easier to get and ubiquitous. Recent data from my own state of Massachusetts showed the latter has worked.

Questions? Comments? Feedback? Please participate in the Comment section.