Did Oregon’s Drug Decriminalization Policy Increase Overdose Deaths?

In this video interview, Brandon del Pozo, PhD, MPA, MA, of Rhode Island Hospital and Brown University in Providence, discusses his new article in JAMA Network Open, which assessed whether drug decriminalization in Oregon was the cause of the increase in the state’s overdose deaths.

The following is a transcript of his remarks:

My name is Brandon del Pozo. I’m an assistant professor of medicine and public health at Brown University, and the article we published is “Drug Decriminalization, Fentanyl, and Fatal Overdoses in Oregon.”

The reason why we conducted this study is we knew that Oregon was planning on rolling back drug decriminalization. It was the first state in the country to try to decriminalize personal drug possession, and when it did, overdoses skyrocketed. One of the big critiques of its decriminalization efforts was this apparent association between doing it and a big spike in overdose deaths.

But what we realized was that fentanyl is probably — it is, not probably — it’s the number one driver of overdose death in America. And fentanyl did not spread throughout the country all at once; it made its way across the country.

So the objective of the study was to figure out when fentanyl hit Oregon and if fentanyl accounted for the increase in overdose deaths more so than decriminalization did.

What we found is really interesting. We found number one, that overdoses did skyrocket in Oregon around the time that decriminalization occurred. But we also found that when you look at the way fentanyl made its way across the country — and we used the state-level drug lab records submitted to the federal government as a proxy for fentanyl traveling across the U.S. — we found that around 2014 it started really hitting New England, but it really didn’t reach the Pacific Northwest until about 2021. And 2021, the beginning, the first part of 2021 was when Oregon decriminalized, but it was also when fentanyl tipped the illicit drug market in Washington, in Oregon, in California. Right at the same time — like terrible, unfortunate timing.

Then when we built that into our models, we found that number one, we replicated the existing studies. Some of them found this big association between criminalization and overdose; some found none or just were inconclusive. But we found that really looking at when fentanyl hit Oregon and looking at its spread nationally and its effect — so, this is the subtlety, but it’s important — looking at its effect state-by-state on overdose deaths, when you account for how it affects each state and then account for how it hit Oregon, it hard nulls the association between decriminalization and overdose. Really, it was fentanyl that explains so much of what happened, really all of what happened, in terms of an increase in overdose.

Now, this doesn’t mean that there weren’t other issues with how decriminalization was implemented or other concerns about the law, but it does say that if you’re concerned about it’s effect on overdose, you cannot say that decriminalization is what led to Oregon’s increase in overdose.

This study shows us something that I think is really interesting about drug policy in the U.S. in general, which is that fentanyl is what I’ve called the apex predator of illicit opioids, but I also called it the apex predator of drug policy. Meaning, if you are not accounting for the effects of fentanyl in your laws and your policies, but also in your analytical models, you’re missing the key ingredient to what’s driving death in our drug crisis in America. You are missing the main variable, right?

So when you look at studies that argue, for example: Does more naloxone [Narcan] create a moral hazard? And then they do a study that shows, well, overdoses went up when naloxone distribution went up. Or a study that looks at the effectiveness of syringe service programs at preventing overdose — or whatever law or policy you’re looking at. If you’re not modeling for when fentanyl hit that jurisdiction and for the effects of fentanyl on death, you’re not doing the model properly.

And then I guess I’ll restate that if you’re constructing laws and policies and interventions that don’t target the substance that is doing the killing, then it’s not good public policy.

This study is important because people are going to look back for quite some time and try to figure out why Oregon ended up undoing its decriminalization law and re-criminalizing drugs. And they’re going to find a lot of possible reasons. Some are about public order, some are about the inability of the state to really ramp up access to treatment in the way that it promised, and there will be concerns about whether or not Oregon’s [decriminalization] increased fatal overdose because it did go up sharply when Oregon decriminalized.

So I think the importance of the study is it shows once again that there just is no evidence that Oregon’s [decriminalization] increased overdose. When we look at why that happened, we have the very plausible explanation that is also why it happened in the 49 other states where overdoses skyrocketed over time. That’s because fentanyl hit that state.

I think it’s going to be quite some time before a state seriously considers decriminalization again, because the political fallout of Oregon is going to be lasting. It’s perceived as something that didn’t work, right? And some people perceived it very hostilely as a failure.

But based on our study, I think the takeaways are one, when you are doing modeling, you have to build fentanyl into your model to truly understand what might be happening in the community with the drug law or policy.

Two, when you’re making drug laws or policies, you really need to understand that fentanyl is what’s doing the killing, causing the fatal overdoses. And your policies have to attack that directly, address that, right? If you can’t explain how it’s addressing that, then maybe you need to rethink the policy.

The third is going forward, we need to be a lot more cautious about what we ascribe cause to, right? Causation and correlation are two different things. In the political dust settling, everybody pointing fingers at Oregon for increasing overdose deaths just isn’t borne out by the evidence. If you’re driven by evidence as a policymaker or voter or citizen, we need to pay attention to things like this.