Oregonians voted to decriminalize drugs three years ago. Now, a new study from the NYU Grossman School of Medicine found that M110 did not lead to an increase in drug overdoses. Corey Davis is an adjunct assistant professor at NYU Langone Health’s Department of Population. He joins us to share more on the results and implications of the study.
This transcript was created by a computer and edited by a volunteer.
Dave Miller: This is Think Out Loud on OPB. I’m Dave Miller. It has been long enough since Oregon voters decriminalized drugs under Measure 110 that we’re starting to have some data about the measure’s impact. A study last month found that decriminalization led to 182 additional unintentional drug overdose deaths in Oregon in 2021. But last week, a team of researchers came to a very different conclusion. They looked at Oregon and Washington, which had its own version of decriminalization. They found that removing criminal penalties for drug possession did not cause an increase in fatal overdoses in the first year of implementation. Corey Davis is one of the authors of this new study. He’s the director of the Harm Reduction Legal Project at the Network for Public Health Law. He’s also an adjunct assistant professor at NYU Langone Health’s Department of Population Health. Corey Davis, welcome.
Corey Davis: Hey, glad to be here.
Miller: It’s good to have you on. So, our Oregon listeners have heard a lot about Measure 110 recently and for a number of years now. But can you remind us why you looked at 2021 data from Washington State as well?
Davis: Sure. As a policy researcher, I think that Washington is actually a really interesting case. As your listeners probably know there were a series of things that happened in Washington, beginning with the Washington Supreme Court finding the drug possession statute unconstitutional and then the legislature, relatively shortly thereafter, recreating the drug possession statute, but in a sort of much different way. Previously, possession of drugs had been a felony. It became a misdemeanor.
But I think most importantly, they instituted this policy whereby law enforcement officers were required to provide a person, that they would otherwise have arrested for simple possession of small amounts of drugs, with a referral to services two times before arresting them. So, we refer to this, in the paper, as partial decriminalization. It is not the kind of decriminalization that Oregon had. But it is a sort of mandatory nonjudicial deferral system, which is interesting and unique in this country.
Miller: And so these two states together, neighbors, gave you a kind of natural experiment to look at the first year of data for fatal overdoses. How exactly did you do the study? What did you look at and what did you compare?
Davis: I would note that I am a policy expert. I am not an epidemiologist. So I can just describe it in broad terms. But essentially what we did for both of the states, separately, is that we looked at the trends in fatal overdoses for every state in the country in the years before. It turns out that Measure 110 in Oregon and the legal change in Washington happened at almost the same time, within a month or two of each other. So we looked to see, in that pre period, which other states in the country have a trend that is similar to both Oregon and, separately, to Washington.
We statistically analyzed all of the other states in the union to see which of them has a trend line that most matches both Oregon and Washington. And we picked the ones that are closest. There’s a statistical cut-off that results in, I think, 12 or 13 states for each of them. And we used that as a kind of comparison. The statistical model assumes that because those states have had the most similar trajectory in the pre period, absent any change, they would continue to have a similar trajectory after the policy change.
So we basically looked to see if there was a statistically significant change in that trend between that bucket of states, that dozen or so states, after Measure 110 in Oregon, and after that court decision, called the Blake decision, in Washington. And we just compared them. We kind of extrapolated what the model predicts what Oregon and Washington would have looked like if there wasn’t the policy change. And we looked at what actually happened after the policy change. Can we just compare them statistically? I mean it’s, as you might imagine, a little more complicated than that. But that’s sort of an accurate description at a high level for what we did.
Miller: What do you think is the most important finding from this data that essentially looked at year one of implementation?
Davis: I think that the most important finding is the headline finding, which is that we did not find that the rate of fatal drug overdoses in either state was statistically significantly different from what we would have expected had the policy changes not been enacted. Now, note that that is a separate thing from, [whether] overdose deaths go up. They did go up. But the question is, did they go up more than we would have expected, absent Measure 110, and absent the Blake decision. And we found that they didn’t. The rate was increasing beforehand. It continued to increase afterwards.
Miller: The data that is at the heart of this for you is with one year of data of fatal overdoses, not including non-fatal ones. Why not and what does that mean for the totality of what we’re talking about?
Davis: I mean, unfortunately it’s a data problem. We would have loved to have been able to look at non-fatal overdoses. Unfortunately, it’s really difficult to even define, if you’re looking for data on that means. So it is, to some extent, possible to get measures of people who showed up in the emergency department with a non-fatal overdose. But of course, that’s just a small subset. It’s actually quite difficult to get a measure of people who EMS responded to, after experiencing a non-fatal overdose. There’s all kinds of definitional and encoding problems there. You’ve got thousands of different EMS agencies ranging from very well-funded fire department-based agencies to rural volunteer agencies. And the data is just difficult.
But even if you could get all of those, even if you had a good measure of those, you’d run into this sort of analytical question of: one of the goals of Measure 110 was to encourage people to call 911 in the event of an overdose. So even if we were able to get a good measure of those data and we found that they did go up, to say that we didn’t look at them. But if they did there’s still this question of what that is telling us? Is that actually a measure of non-fatal overdoses or is that a measure of people who had a non-fatal overdose and somebody called 911? So all that said, I would love to have looked at those data. It’s just that the data wasn’t available.
Miller: I want to turn to the next year of data from the CDC that we now have access to. Because it really does seem to show a very different story. It says that between April 2022, right after your paper stops, and April 2023 Washington and Oregon were estimated to have had the 1st and 2nd highest rates of fatal overdoses in the country, I should say, the highest increases in those rates. In other words, the latest data seems to potentially undercut your conclusion in a dramatic way. What is your hypothesis for what happened in year two?
Davis: Well, I would say, I don’t know that that is an accurate representation because, again, the question that we are looking at is not, were deaths going up? [The question is] was there an inflection point, right around Measure 110 or the legal change in Washington, such that it seems like they started going up significantly, at a significantly greater rate after that happened?
Miller: But what’s special about the first year? I mean we’re talking about a policy that has been ramping up potentially. There are various ways to hypothesize what happened. But isn’t an entire second year of data, a really important thing to now pay attention to and reckon with? Just to give folks a sense for the numbers here, the CDC says that in Oregon, we saw a 23% increase in fatal overdose deaths in the last year. Washington saw a 34% increase, while the country as a whole, went up 1.7%. The rates of increase are wildly different and the only thing I can think of that unifies Washington and Oregon is decriminalization.
Davis: Well, there are a couple of other things. Again, it seems like you’re saying the only thing you can think of. Well, I can think of a number of other things. Again, we’re looking at rates. So Oregon and Washington both historically actually had relatively low rates and numbers of overdose fatalities. So even in 2020, Oregon ranked something like 40th, 38th, 39th, in the country as far as overdose deaths. And in 2021, the latest year for which we have final data, those data that you’re referring to are provisional data, Oregon was still well into the lower half of the country as far as actual overdose deaths. So, you’re right that we are, to some extent, interested in those changes. But I think that this is where some of this recording is kind of conflating those two things. It is true that the rates have increased, but they are increasing from a very low level…
Miller: But OK, just to be clear, what we’re both talking about and focusing on is the rate of increase, not where they started. And that was the entire point that you were talking about earlier with comparing Oregon and Washington to those other states to look at the trajectory. The entire point of the trajectory is the rate of increase. Let me put it to you this way because I’m truly curious. Would you commit to doing a follow up study with this latest data and publishing it regardless of what you find? I mean, if what you find is that Measure 110 is the reason for an increase in fatal overdoses, would you tell the world?
Davis: Yes, of course. It sounds like you’re suggesting that this study that was funded by the Centers for Disease Control and Prevention, that had two CDC scientists on it, had a number of scientists from New York University, that there’s something fishy about it, that maybe we wouldn’t have published the results if we had found something other than what we found. I would say the finding that we found and which we published is not what the proponents of Measure 110 were hoping for. I mean Measure 110 was kind of pitched, at least in part, to the voters as saying we think this is going to reduce overdose deaths.
If we were somehow in the bag for big Measure 110, as I would say, as a number of people have suggested to me in my Twitter replies, would we want to publish the study? Because it goes against that hope. I think that Measure 110 proponents had… I think that this is a study that doesn’t make anybody happy.
I think that those who oppose Measure 110, I don’t think that they are bad people who actually hope that more people would have overdosed and died. I’m not saying that. But I think they would have found it politically helpful to their cause if we had found Measure 110 actually seems to have increased overdoses, just as the proponents of Measure 110 probably would have liked to have had a study that says it seems to be decreasing them. We didn’t find either of those. And with my policy hat, I should say-
Miller: We have 20 seconds left for the policy hat.
Davis: Yeah, I mean, it is actually uncommon for a policy, even a relatively big change like Measure 110, to significantly move a big outcome like overdose deaths. So I’m actually not terribly surprised that we didn’t find that, in the first year that there was no significant change.
Miller: Corey Davis, thanks very much for your time. I appreciate it.
Davis: Sure.
Contact “Think Out Loud®”
If you’d like to comment on any of the topics in this show or suggest a topic of your own, please get in touch with us on Facebook, send an email to thinkoutloud@opb.org, or you can leave a voicemail for us at 503-293-1983. The call-in phone number during the noon hour is 888-665-5865.