Next is our new addition to the annual meeting, actually a very anticipated session, highlighting the 2023 highest impact publications. And we hope that it will be valued and well-received and we can continue to bring it back again and again. With that, I welcome our presenters, Carla Marrenfield, Brian Hurley, and Joshua Lee, and we have discussants, Izmini Patrakis, Fran Levin, and Ed Nunes. And the panel, this session will be moderated by John Renner. Let me give them a minute to settle and, all right. So let me introduce Dr. Marrenfield. She is board certified in psychiatry, addiction psychiatry, and addiction medicine. She is a professor in the Department of Psychiatry at UCSD. Her research looks at health outcomes for individuals with substance use disorders and involves monitoring as well as implementation of evidence-based practices. Her focus is primarily on buprenorphine treatment, effective interventions, and motivational interviewing. Prior to joining the UC San Diego, she was also an assistant professor in the Department of Psychiatry at Yale University School. And she is a co-chair of the MOC committee and a very valued member of AAAP. So I welcome Dr. Marrenfield. She will introduce the rest of the panel. Thank you. All right, well, thank you so much, Dr. Desai. I'm very excited that we're trying this out today. I think it's going to be a fun thing. I've been sort of informally calling it like Journal Club for Adults. I don't know if that's the right title, but the idea is to try to make it kind of a fun, fast, high yield way of looking at articles that have captured the public attention. So these may not be the most impactful to our day-to-day life, but these are the ones that are making an impact. And we'll describe a little bit more about the Ultimetric Score and what that is and how we did this when we go through our methodology. But before we jump in and start talking about some different articles, I want to introduce and say thank you to our panel of discussants who have kindly agreed to do their homework and read a bunch of articles and think about ways in which they might be relevant to discuss in terms of clinical implications, what it means for our society, what it means for our profession, et cetera. I also want to thank Dr. Joshua Lee, who really created this concept and idea and allowing us at AAAP to borrow with permission the concept and the idea and to help us make it a reality, as well as Dr. Brian Hurley, my co-collaborators on putting all of this together. So with that, I'm going to introduce each of our presenters. So Dr. Lee is a professor at the NYU School of Medicine. He's the program director of Addiction Medicine Fellowship and co-director of the section on tobacco, alcohol, and drug use in the Department of Population Health. His research focuses on medications for opioid and alcohol use disorders in criminal justice and primary care settings. Dr. Brian Hurley is an addiction psychiatrist based in Los Angeles. He's the medical director of the Los Angeles County Department of Public Health's Bureau of Substance Abuse Prevention and Control and is the current president of the American Society of Addiction Medicine. He's a volunteer assistant clinical professor of addiction medicine for the UCLA Department of Family Medicine. So I'm going to introduce them now, and then once we switch to our panels, I'll introduce our panelists just to keep things rolling. And with that, I'll turn it over to Josh. Thank you. Thank you, panel. Hi, Ned. I'm not a psychiatrist. I have given this talk a bunch at ASAM. So you know Brian is the president of ASAM. It is true. I think it all started with, like, last year's ASAM, and the idea was to bring a version of the talk here. I want to credit Sarah Wakeman, MD, at Mass General Hospital in Boston, who is my co-author and co-presenter at ASAM, and we'll be doing it again in the spring. That meeting is April 5, 6, 7 in Dallas, Texas. Let's start clicking. So we do have kind of a well-known format that we're going to repeat here, but we haven't had such a great panel, so we're super excited to add the panel. We'll have to be cutthroat in time management as this talk can otherwise kind of roll on and on, and we'll run out of time. Disclosures. We had a few. Check the app. We're going to do these learning objectives. I'm not going to read these out. You'll get some information, and we had to make up some learning objectives. Methods. What is the highest impact paper in your field this year? Well, what do you mean? What do you mean by impact? What has been most clicked, most downloaded, most viewed? What changed clinical practice overnight? What's going to change your clinical practice over the next 10 years? There's no really way of defining that that we all kind of accept. We don't have like Spotify downloads, album fails, kind of the billboard top 20, which has some kind of metrics that you could kind of draft off of. So we tried to do the same thing here. It boils down to the last point, which is altmetric scores, and then our own kind of opinions. So altmetric, I'm not trying to sell you stock in this company. I think it's privately held. It's a for-profit business. They're trying to rank publications with a score. I'm not showing you their methodology, but their score is you get points for tweets, which have been declining, as Elon Musk has completely ruined Twitter, Facebook posts, which nobody does that, but that could add to your score, and then media hits. So when media is picking up an article, and FYI, if you want people to read your article, one thing to consider is a press release that is going to get picked up and regurgitated verbatim by media wires, and that will drive your altmetric score. And then other stuff can come on later, like citations. That's a classic metric, which article got cited the most, but it takes years for the citations to build up. So if you want to know this month what people were reading, it's a decent proxy, but it's not the end of the world. It's what we're going to use in this one. The biggest impact journals get you the most eyeballs, period. That still seems to be the case when I was in med school, and it seems to be the case today. So the higher scores do translate to the bigger journal titles, most of the large, well-known general medicine journals. They're easy to search every year, every month, because they don't publish that much in our field every issue. And then you get down to specialty journals, and we search all of them. There's a widget you can use from altmetric where you just do PubMed, JSAT, the year, the publication date. You just get the entire table of contents for the year, and it displays a score. And you can see very quickly that most of my publications in JSAT had a score of zero. So a lot of what we do is kind of for each other, and maybe someone will read it in six months or cite it in three years, but it doesn't generate a ton of reads the month it comes out. So that's really up to you if you're in the publication world. But we make a point of searching these specialty journals, because that's our home. That's where most of our publications are going. And of course, we want to support and kind of celebrate that industry. So we make a point of trying to pick out some of the top papers from all these smaller journals that are our field, including AJA, the official journal of AAAP. And then we will scan newsletters to make sure we're kind of cross-referencing what other sources have been reviewing and citing throughout the year. Two good ones, subventions, are BU's NIDA-sponsored Alcohol, Drugs, Use, and Health. That's every two months. I highly recommend it. And ASAM Weekly is a great kind of newsletter, which is quite deep and on a weekly basis kind of doing the same type of review. So we ranked 121 papers. This version of the slides cut off a link to the actual results. If you email me, hit me on Twitter, Google me to get my email, we didn't display that either. But if you want a list of what we're presenting here and all the other articles that we're not going to present, we're happy to share that. The total number of articles reviewed is many times 121, because again, we go through the entire table of content of all the specialty journals. We reviewed pretty much every addiction article that showed up in Lancet, and there were about three of them. So that one is pretty simple. But then there's many more times that in terms of the addiction journals, and the scores range from zero, meaning nobody's reported on your article or tweeted about it, up to the top one was around 2,500, and we'll be reviewing that one. And what is driving the Altmetric score? It's ultimately media coverage right now, as opposed to like everybody's tweeting about it, because that has really diminished over the last, like, since the pandemic, basically. And what does the media cover? It covers what people will click on, right, and read and engage and subscribe to. So anything that intersects with big public events, like war, famine, COVID, is going to potentially play in there. Like during COVID, the top, like, articles in science were all about COVID. If you were publishing on, like, a new dialysis technique, nobody was reading about it that year. And then in our own field, the common themes and clickable topics are cannabis, anything about cannabis, really, like, it's illegal now, the boomers are using it, my kids are using it, it's good for arthritis, whatever, that is going to go in, the news is going to pick that up. Overdose crisis, probably there's been a little bit of fatigue, but that still runs a lot of headlines. And then what I think matters most to us are these kind of micro headlines about actual changes in our field. And we're going to try and highlight some of those important trends. And then this is a table of the top 10. And I'm not going to read it, but let me hand it back to go through that. So we have the top 10 in terms of what the ultimetric scores were. And we went through them, and we were like, okay, are these interesting articles? Do we want to discuss them or not? And so we decided to use a little bit of our editorial prowess to decide which ones we wanted to talk about and maybe not cover all of them. But just so in the interest of inclusion, these were the top 10. And I'm just going to run through them super quickly. So with like a one or two take home points of things that grabbed people's attention. So the number one article is one that we will discuss. So this one we'll talk about in more detail. But the take home point for this article by Zhao is that people who drink alcohol daily have an increased risk of death from all causes. And the reason this got picked up is that there were a lot of articles around alcohol not having any alcohol being for the first time a recommendation from the Canadian guidelines. Different things like that. And so there's more and more data about no alcohol is okay. And so this got picked up. These are a couple of other articles. Opioid analgesia for low acute back pain and neck pain. So this was an RCT that basically got picked up because it's telling us that opioids are not effective for the treatment of acute low back pain and neck pain. So something we've had a lot of signals for for a while that we can continue to talk to people about when they ask for opioids for these reasons and whether or not it's indicated. The next article evaluating the impact of alcohol minimum unit pricing on deaths and hospitalizations in Scotland. So not only if you increase the price per unit of alcohol do people buy less, but it actually has secondary outcomes on reduction in death and hospitalizations. So we're not going to talk in detail about this, but more on that theme about reduction of alcohol use and ways to go about reducing alcohol use that impacts big outcomes like death. Next article, number four on the top ten list, cannabis use disorder and subsequent risk of psychotic and non-psychotic unipolar depression and bipolar depression. So we're going to talk a little bit more in depth on this one. Number five on the list, treatment used among adolescent residential addiction treatment facilities in the U.S. So for the child and adolescent psychiatrists out there, this study actually went through and looked at what are kids getting treatment with. They aren't necessarily getting as much medication-assisted treatment or as I like to now call it medications for addiction treatment, not medication-assisted treatment as I may like, but they are getting some. So this study really looked at what's actually out there that people are receiving. Number six on the top ten, population level health effects of involuntary displacement of people experiencing unsheltered homelessness who inject drugs in U.S. cities. We're going to talk a little bit more in depth about this one. Number seven, one-year association of drug possession law change with fatal drug overdose in Oregon and Washington. So this was a little bit more specific. We're not going to go into details, but basically this study was looking clearly in Oregon and Washington and they found no evidence of an association between legal changes that were occurring in these two particular states that removed or substantially reduced criminal penalties for drug possession. So there were a lot of law changes that reduced or removed these criminal penalties and there was no evidence of an increase in fatal drug overdose rates and so you can imagine why this one caught a lot of headlines. Number eight, charting the fourth wave, geographic, temporal, race and ethnicity, and demographic trends in polysubstance fentanyl overdose deaths in the United States between 2010 and 2021. So that's a title that's a mouthful, a couple of these are. This paper we're not going to go into details on. This was published by Friedman in the journal Addiction, but basically we are seeing a lot more headlines around the co-occurring sort of fourth wave as it's called of the opioid epidemic between co-occurring opioid and stimulant use and this paper looked at the widespread concurrent use of fentanyl and stimulants as well as other polysubstance formulations, right? So xylosine and stimulants and fentanyls and et cetera all being together and presents some of the novel health risks and public health challenges. And so I think this is an important thing for us to be aware of as it's happening and impacting us, but we didn't go into details in this paper. The ninth article of the top 10, Association of Receipt of Opioid Use Disorder-Related Telehealth Services and Medications for Opioid Use Disorder with Fatal Drug Overdoses Among Medicare Beneficiaries Before and During the COVID-19 Pandemic. That's a mouthful. We're going to go into more details on that paper. And then the 10th paper, Trends in Alcohol-Related Deaths by Sex in the United States. So this is an interesting one. Again, alcohol was a big theme and we'll talk a little bit more about alcohol in terms of what people are experiencing, but this paper found that when we look at the fact that alcohol-related deaths have been increasing in the U.S., which has been going on for a few years, the increase interestingly has been more pronounced among men than women and I think this has interesting implications when we think about the telescoping effect that we sometimes think about with women and alcohol use disorder, and yet we're still seeing higher rates of deaths amongst men. All right. So that's the rapid-fire top 10 in review. So now we're going to go back and take a little bit more of a deeper dive into a few theme areas for discussion. Yeah, we're actually, now we're like blowing up the whole format. We wanted to present a paper from last year because it was so fun to talk about it, Sam, and it was about double the kind of score and kind of, you know, double the impact, if you will, than papers we saw this year. So we thought we'd keep it in the slide deck that we had it evolve from last year. And then those two papers, well, the first one is the one we're really focused on. The second one I'll get to, but the first one, I have a conflict, it's from some colleagues at NYU. This is a double-site study at Bellevue Hospital, where I work, and University of New Mexico, Albuquerque, psilocybin for AUD. This was a big paper, one of the biggest in science last year, just again, based on the score. And it showed that psilocybin with guided therapy and quite intensive therapy, actually, was pretty good compared to Benadryl, which was the placebo in this one, for helping people drink less during the trial and then post-trial. Probably most of you saw this, it came out around, like, August, September of last year. The results, this is just part of figure two, but basically, people are drinking a lot at baseline in both groups, and they look about the same. They both have good reductions, both arms of the study, they get a lot of psychotherapy. And then even if you got Benadryl, you're drinking a lot less during the trial than you were at baseline. But if you got the two active dosing sessions with psilocybin, you drank even less. And that was true for rates of abstinence, rates of reduced heavy drinking, and rates of kind of drinks per day overall, some of the standard self-reported metrics. They also threw in some biomarkers, and I think those did correlate also with the self-report. Some limitations you can already see and probably know about, and we'd love to hear the panel's thoughts on. It's intensive. This isn't, like, ready for prime time. Obviously, it's not FDA approved yet for this indication. There's a lot of politicking about how you bring psychedelics to market, psilocybin in particular. And then people are in the dosing day for a long time. There's double therapists in these protocols. It's, like, heavily protocolized. It's not how any of you used acid on your hiking trip in college. And so how do we kind of, like, make it more feasible, accessible to, you know, poor people and less kind of time intensive if we can get away with it in terms of getting more people in therapy? That's a lot of fun stuff to think about with psychedelics. This table is from the same paper, just showing how people did in these. I was highlighting the longer follow-up period here. So week 33, 36 is 10 plus weeks since any active, or even 20 plus weeks since you got any active dose. You did get some more therapy on your way out of the trial, and the gains, or kind of the benefits, seem to be pretty sustainable and longer, long lasting. So good stuff. Some limitations in this type of research. It's mostly attracting more educated white volunteers to the research cohort, which does not represent, in total, kind of the burden of alcohol disease in the United States. And then they're not funded by NIH yet. NIH, I think, is getting revved up, but of course these are Schedule I compounds, so there's a lot of problems with traditional federal science funding. And then you're left to kind of the donors and the foundations that have been driving this research for decades. But that, in a sense, is kind of a conflict, because some of those same foundations are the ones that really want to bring these to market. So another topic for discussion, perhaps. And then just to highlight, there's a lot of commonality between the classic psychedelics and the non-psychedelic psychedelics, like ketamine, for this same condition. And this was another trial from last year. This one was in the UK. They also had a kind of psychoeducation versus psychotherapy component to this study. But it was ketamine infusions, I think it was once a week for three weeks, versus saline infusions. And people drank less, and they drank less with ketamine plus psychotherapy, versus ketamine just with psychoeducation, kind of like, here's a pamphlet, here's some basic advice. And that was good stuff. So is ketamine and psilocybin the same compound? No. Here's a cartoon I found on the internet. That's kind of looking at the mechanism of action and the downstream effects. But they all seem to, this is LSD, psilocybin, DMT, MDMA, and ketamine. There's a strong signal so far that they help with depression. Not here is, well, alcohol use disorder is down there for MDMA. So this study didn't contribute to the cartoon. But why did these all kind of like work differently and yet maybe have kind of a common effect is something fascinating to think about. How long does it take for yours to treat pain? Get to that in the panel. Yeah, pointing out the anti-pain or pain treatment effects. Yeah. Yeah. Many other conditions, of course, are under study for these compounds. And then blending this in with this current year's top article, this is the kind of number one billboard hit for 23. And it was already mentioned up top as looking at essentially all the studies we already have to look at mortality risk with moderate drinking. And they did kind of a new meta-analysis. This came from a group in Canada at one of the Institutes for Addiction in British Columbia. And what it showed was, let me see if I can get the laser to work, yeah. So this was, da-da-da-da-da-da-da, this is kind of the old news. And this figure actually, the line of crossing one got weirdly adjusted in this graphic. But up top, when you fully adjust for what goes into mortality in addition to your drinking status, you kind of erase many associations with moderate drinking and lower mortality. And look at the confidence interval, includes one. On my little figure here, it looks like it's to the left of one. That is just because of how we cut and pasted it in. If you look at the article, that's crossing one. But then the older story that I grew up with, having gone to medical school in like 2000, was that there was a J-state curve, that moderate drinking was potentially healthy and reduced mortality in moderation. That's red wine with dinner, Mediterranean diet, that kind of stuff. And then increasingly over the last 10 years, we've seen reanalysis, new data, new cohorts, and people don't, they're really challenging this. This article got a lot of headlines for, as the Wall Street Journal put it, it doesn't really help one way or the other, the moderate drinking. And that's kind of overturning the J-state curve. And then some context, there was a lot of other studies in the last five years or so that were arguing there was no safe drinking. And then Canada recently, this was I think a year ago, went with, you shouldn't drink at all, Canada. You know, basically, forget about it. It can't be helpful, it can only hurt you. And then just to highlight, these are older papers, but really good stuff from the Global Burden of Disease Consortium, which has been kind of doing this. And one of their take-homes from another article from last year was that it's really, really harmful in young people. That's where the clearest kind of messaging should probably go. If you're young and healthy, nothing can kill you except for drinking a lot, crashing your car, overdosing, firearms, suicide. When you're older, the plane's already crashing, so you might as well have a gin and tonic. That was kind of the messaging in this article. So there may still be a J-shaped curve when you're older, but when you're a young person, alcohol is really only harmful, was one of the take-homes there. All right. Okay, panel, what do you think of psychedelics and about drinking and mortality? What do you want to start with? Let's start with psychedelics for alcohol. Can we go back and show the slide with the graph of the data? Yeah, I was going to look at the graph. That's the first figure. That one. And that looks a lot like the depression data that I've seen. Same thing, big placebo response, but then there's a certain amount of separation. So it's probably a new tool. In fact, it's not new, right? It was widely used in the 1950s. For treatment of alcoholism, actually, right? And particularly in Canada, British Columbia, there was a whole network of, you read Michael Pollan's book, he goes into this. And then it was all, there were a lot of papers published, and it was all buried. I went to medical school, you didn't learn anything about hallucinogens, except that they produced funny effects and they had no medical value, right? That was the message. So one of my most important mentors always said, make clinical observations. Look at what the patients are doing and what seems to be helping the patients. And this is one of them. That's how it started in the 50s, right? People just started taking it and finding it seemed to help with different things. So I say, when the patient says something's helping them, I pay attention. It's also the case, this is not a miracle cure. The number needed to treat is probably about the same as our other treatments. Yeah, here, they put a number needed to treat here. What happened? Uh-oh. There we go, sorry about that. In the low single digits, which is a great number needed to treat, but it meant what roughly treating four or five people with psilocybin got you one great success story, which that doesn't sound as good to the lay public. Well, that's not like Prozac. Yeah. Yeah, yeah, it's better than Prozac. That's actually very, very good. I mean, naltrexone is up at 10 or 12 or 16. And one of the arguments that's made why even naltrexone is very useful is that people put people on antihypertensives to prevent stroke, but the N and T of that is like 20 or 30 or 40. It's very high. So actually, a four or five is quite good. I think one of the questions that you pull out from this paper is that you can't blind this. That was one of the things. And so the question, if you're a strict trialist, you might say, oh, that's really problematic because you can't blind it. But I think of Izmimi's work with disulfiram, you can't really blind disulfiram either. And yet, for the patients it works for, it's very effective. So I think we can go around the edges and criticize this paper because it's not blinded. It was a group that really wanted treatment and all this stuff. But it's still another tool. And there really is not that many tools in terms of medications for alcohol use disorder. The one question I have is how much do we have to attend to this eight hours of therapy sitting with patients? And my assumption is that it's being used more for safety and that that's the only way they're gonna move towards approval of psilocybin or other psychedelics. But it's gonna really limit feasibility. And so the question is can people do future studies doing less of that? So those are my comments. I would follow up on what Ned said in terms of, pay attention to what the patients are actually doing. Because we don't really know the best way to use this therapeutically at this point. But I'm concerned that patients are doing all sorts of things on their own. And much as our patients often do, if one pill is good for you, 10 is better. And I've seen some people who have really gotten into serious trouble because they have overdosed on psilocybin for long terms. Mm-hmm. And I think ketamine then is the example of what people are using right now. Less so I think for alcohol, but in a lot of different ways. And stripping out most of the therapy in terms of like telehealth and home use. So maybe that's another example to think about. I think the only thing I'd like to add is that there's just so many questions. And there's need for a lot of research. I mean people talk about the difficulty with implementation, but I think we just aren't sure which psychotherapy is the best. How many doses do you need? What's the dose even? And so it would be good if they funded research to actually answer some of those questions before we get to, you know. Thank you. the NIAA panel again, maybe we can make that comment. Getting back to the placebo thing, I don't really care if it's a placebo, if it's like a really good placebo. Yeah. Yeah. Who cares? Right. A safe placebo. Safe. Great, I mean, fascinating in alcohol studies, placebo arm does good in almost every large RCT because people are ready for change and that is probably the biggest driver, who's gonna do good in treatment no matter what we give them. Great, thank you panel for that first session. All right, thank you guys. So we're gonna switch. So again, we tried to group things a little bit into themes kinda sorta when we were selecting articles. So next we're gonna talk a little bit about cannabis and psychiatric disease. So as Josh mentioned, there's 8 bazillion things about cannabis in the news. So these were just a couple of articles to talk on. So this first article, the lead author was Jephson, it was published in JAMA Psychiatry. And basically it's looking at cannabis use and cannabis use disorder, both in the subsequent risk of either psychotic or non-psychotic mood disorders. And they look specifically at unipolar depression and bipolar depression. And so all kinds of different papers looking at the association of cannabis use with affective disorders. And particularly there's been so much focus on psychotic components, but less focus on mood symptoms, co-occurring anxiety, co-occurring other psychiatric disorders. And so this paper was looking specifically at affective disorders. It was a prospective population-based cohort study using a Danish nationwide registry of approximately 6.5 million individuals. So pretty good sample size. The question around generalizability to US populations or other populations outside of Denmark always is one. So overall they looked at cannabis use disorder and they first looked specifically at unipolar depression. And when you look at cannabis use disorder and unipolar depression, we see an overall increased risk of unipolar depression. And we see an overall increased risk of both psychotic and non-psychotic presentations of unipolar depression. And they're relatively similar rates actually. Then they secondarily looked at cannabis use and bipolar disorder. And they had a little bit of differences between men and women. Although again, still within the same ballpark. But the risk, the association with cannabis use and bipolar disorder was higher overall than the association with unipolar disorders. And then much higher for psychotic bipolar disorder associated with cannabis use. And then again, we see a little bit of this breakdown between men and women in non-psychotic bipolar disorder. So the other thing that's really interesting is that cannabis use was associated with higher risk of psychotic than non-psychotic subtypes of bipolar disorder overall, but not unipolar depression. All right, so in unipolar depression the rates were roughly similar. They were higher overall, but they were roughly similar. Bipolar really had a little bit more of a nuance around that difference between psychotic and non-psychotic presentations. So overall, the conclusion, cannabis use was associated with an increased risk of psychotic and non-psychotic bipolar disorder, as well as an increased risk of unipolar depression. So these are some of the Kaplan-Meier curves from this study. Basically it breaks it down into the various types, unipolar, bipolar, psychotic, non-psychotic, et cetera. The scales do change quite a bit on these slides, and I know it's hard to see in this scene here, but overall you do see a significant difference. The second table there is showing unipolar depression at the top and bipolar at the bottom, and then the bipolar again is broken down into male and females. And they did do a fair amount of adjustments for this. So they adjusted the data for sex, any co-occurring alcohol use disorder or co-occurring other substance use disorder, people who were born in Denmark versus those who were not born there, the calendar year associated with it, parental education level, co-occurring diagnoses of substance use disorders, sorry, parental co-occurring substance use disorders and parental affective disorders in this. And so it's, you know, further evidence around cannabis and psychiatric disease. The second paper we're gonna talk about about cannabis has to do with state cannabis laws. So not terribly dissimilar from the Oregon-Washington thing. This was looking at all 50 states and they grouped the states into different categories, but it's looking at state cannabis legalization and psychosis-related healthcare utilization. So they looked at two outcomes. One outcome was around increased rates of psychosis-related diagnoses. And the second outcome was increased rates of prescribed antipsychotics as sort of a proxy for worsening psychosis potentially in the state. So the concern, of course, is that legalization of cannabis could be associated with an increased rate of psychosis and that there would be more healthcare utilization for this psychosis. So this was a retrospective cohort. It used commercial and Medicare Advantage claims data from a long period of time, 2003 through 2017 for people aged 16 and above, all 50 states and Washington, D.C. Oh, they did the analysis in 2021, 2022. It's almost 64 million people who are included in this study. And so they grouped the states into four groups. So the states were either patients, states that had medical cannabis and no retail outlets, medical cannabis and retail outlets, recreational cannabis and no retail outlets, and recreational cannabis and retail outlets. And what you see is that there was no statistically significant increase in rates of psychosis-related diagnoses in any of those groups. And there was no increase in prescribed antipsychotics in any of those groups. So they did some other sort of secondary analyses and kind of looked through and looked at rates of psychosis-related diagnoses in certain populations and certain things. They did find some increase in men overall versus women. People aged 55 to 64, so I don't know, maybe your plane is crashing and you're getting a little psychotic from your marijuana that you're increasingly using now at this age range. And then Asian beneficiaries. So overall though, the conclusion really was that there wasn't a statistically significant increase in psychosis-related outcomes. So these are just some of the diagrams that show the things they adjusted for a lot of different things, including state-level confounders, percentage of Hispanic, non-Hispanic black populations. And I'm going to see what our discussants would like to say about these things. It's like me to start. Let's start on the other end, yeah. On the other end. Maybe you want to start. Well, I was just, because you brought up like how this might be perceived by the media, it is a little bit interesting that this might be misinterpreted a little bit because even though there is an increase in diagnosis and prescription, there were still some signal there. That just might be too big of a, too gross of an outcome. Do you know what I mean? That it's not enough. So that's one thought I had. And in terms of the other paper, it does just seem to continue to show that there are difficulties with cannabis. And the mood disorders may be less than the psychosis, but that may also be, you know, our diagnosis. symptoms may be more affected. I did not understand why being born in Denmark wasn't important. I didn't know what that was. I think they were looking at immigration, immigrant populations versus us. It would have been nicer if they had actually said what that meant, but anyway, that's my two cents. That was what I took from it, was whether or not people who were immigrants might be affected differently. I think, you know, from my past life in global mental health stuff, there's interesting things around like psychosis with immigration and increased rates of schizophrenia in populations that move, so I wonder if that was the thought behind it. Yeah, okay. I think the story is still out on cannabis. Can you speak to the microphone? I think the story is still out on cannabis and we're getting signals, we're learning a little bit, but I don't think we really know whole things and I'm particularly concerned about the high potency cannabis that's available now and what that's doing to adolescent brains and young people and we don't have that kind of information at the moment, which I think is really critical. Yeah, and I think in particular, we worry about what is it doing and then we also worry about how much it mediates or moderates any of the psychotic experience and or comorbid psychiatric stuff. I think I have a slide that I stole from you from like a presentation probably 15 years ago, Dr. Levin, that shows like the co-occurring rates of psychiatric disorders in people who have a consistent cannabis use and the association's been clear forever. For a time, yeah. So I'm curious if you think this adds anything to that. Well, it definitely do. I mean, you could look at the two papers and say, oh, they seem somewhat inconsistent because one is saying cannabis use doesn't cause more psychosis and doesn't have, and that you're not seeing more prescribing for antipsychotics and the other study suggests that in fact, there is this risk for developing a psychiatric disorder, but I think the devil's in the details and that the first paper was cannabis use and there was so much cannabis use. In the United States, you've got like 60 so million people over the age of 12 who've tried it in the past year. So you're getting a wide berth of people using, it's not, you know, it wasn't, it didn't seem from the article unless I missed it, like heavy users of high potency cannabis. It was, you know, all cannabis use. So somebody who's used once or twice and a low potency is being lumped in with others. So I don't think they're really that inconsistent. In the article, it was interesting because sometimes, and I don't know if this was an editing thing or a true thing, but they had differences where they would sometimes say cannabis use disorder and sometimes just say cannabis use. Right. And it wasn't really clear, like clear are these people who are diagnosed with a problem. That would be a different article or a different thing. Exactly, or is it just exposure? And so I had some questions about that too. Yeah, and then the second article, I mean, it sort of goes along. People have done reviews already in higher associations. And I think, you know, daily users are more likely to have cannabis use disorder in that group. You're gonna see higher rates of prevalence of psychiatric conditions in that group. What's interesting is I think as drug, as cannabis gets normalized or is normalized in society, then users are not gonna have as much psychopathology. It's sort of the opposite of what it is with tobacco. People who are heavy tobacco users, the group that is still using and smoking cigarettes have higher rates of psychopathology than the people who were smoking in the 1960s. So I think that in fact, we might not see these associations because it's becoming so normative. Yeah, and the population that's using it and why and how are they using it and how that is impacting it. Any other closing thoughts? I'm gonna play devil's advocate a little bit. So in the discussion, I'm hearing a lot of causal inference. Do we have causality here from these data? I'm very specific in saying associations all the time. I appreciate that. And I looked at the paper and one of the things I liked is that if I understand it right, can you go back to the graphs? So in the methods, they say that they use cannabis use disorder and cannabis use as a time dependent covariate, which I think that means they do actually pay attention to the order of onset, which but the graphs don't show that. They show age in years along the horizontal axis. So I'm confused by it a little bit. I would think that the graph would show time since initiation of cannabis and show the hazards on that basis. And maybe it's just how they show the data. But I've always been skeptical that the people that are smoking, what we're seeing as an association and a lot of it is that people with brewing, particularly young people with brewing prodromal cell pathology are drawn to cannabis for a variety of reasons. And that's why you see the association. Now, we're never gonna do the experiment where you randomly assign people to cannabis or no cannabis. The other thing is that, I listen to patients, we all listen to patients. You hear patients say that cannabis helps. appreciate your saying association they assumed a time analysis though the article was designed they did but even even even so you're not sure whether that reflects a diathesis that's emerging right it could be that there that it's all it's just that this that cannabis use, any cannabis use, was a marker, again, association, for school failure, problems, you know, more depression, more suicidal thoughts, and it was not the diagnosis, it was sort of stepwise. No cannabis use was the lowest. Cannabis use, and then it was cannabis use disorder. Still, it's association, but it's interesting that sort of this progression, and that it suggests maybe an underlying diathesis towards developing. What I'm hearing, and what I'm taking away, is that, you know, my overly simplified understanding of it is not changed by these papers, which is, we have associations, we still have a lot of these important questions about. But I think people misinterpret it as quasi, or over-interpret it as quasi. I think they do over-interpret it, because we have so many papers looking at the associations, but it doesn't, none of the papers really contribute that much new to the discussion, I don't think. Has anybody done a study with a big sample of teenagers treating half of them for cannabis? No, I think the closest is like NCANDA and ABCD. So, you know, ABCD is looking at adolescents, and so they're seeing adolescent sort of risk and prodromal-ish symptoms, maybe, prior to onset of selection of use of cannabis. Right, so we're still not randomizing kids to use cannabis or not, and so, you know, there are predictors of who's gonna end up using it. Who's gonna randomize kids to treatment? That's an observational study, but potentially. Well, no, it's randomized, right? You could randomize them to, let's say, If it's causal, you would expect to see. That's a long-term study. It's a big study. All right, so we're going to switch topics. I appreciate the. We'll be retired. We'll get one of the other institutes and let them know to do that study for us. All right, so next we're going to switch and talk a little bit about opioids. So big topic, opioids always in the media. Another big topic, COVID, switch to telehealth services. A number of articles had come out around this, but this is the one that has a title that's a big mouthful. I already read it, so I'm not going to read it again. This is a study by Jones. It was in JAMA Psychiatry. Brief background, so during the COVID pandemic, there was a lot of expansion of telehealth. There was some lessening of various rules around treatment for opioid use disorder, et cetera. So this was an exploratory longitudinal study. It used Medicare fee-for-service beneficiaries who are 18 and above, and then they looked at outcomes through the National Death Index data from the CDC, and so they were comparing two different cohorts, one was the pre-pandemic cohort for people from 2018 through 2020 that had about 105,000 people, and then the pandemic cohort, which they actually started at the end of 2019, going through 2021, that had about 70,000 people. So in this study, basically the rates of all-cause mortality was higher in the pandemic cohort. The rates of fatal drug overdoses was higher in the pandemic cohort. Interestingly, the percentage of deaths due to a fatal drug overdose was similar between the two groups. So even though the gross numbers, the rates were higher in the pandemic cohorts, the percentage of deaths due to fatal overdose was similar. But the receipt of extended-release naltrexone in office-based settings was not associated with a lower odds for fatal drug overdose. And when they did a multivariable analysis of the pandemic cohort specifically, there were some things that were associated with a significantly lower adjusted odds ratio for a fatal overdose. So what was associated with a lower odds ratio for a fatal drug overdose? If you had received opioid use disorder-related telehealth, so receiving telehealth treatment, you had a lower risk. Receipt of medications for opioid use disorder from an opioid treatment program. And receipt of buprenorphine in an office-based setting. So those were the three things in this study of roughly 70,000 people in this pandemic cohort that were associated with a lower adjusted odds ratio for fatal overdose. So the conclusion is that the receipt of these opioid use disorder-related telehealth services was associated with reduced risks for fatal drug overdose, as was receipt of MOUDs from opioid treatment programs and buprenorphine in office-based settings. All right, so here's a bunch of tables looking at all the different things that they looked at. So they broke it down in terms of age and race and other co-occurring medical disorders and other things like that. And then table two there looks at the all-cause mortality and the drug overdose data. So I'm not gonna belabor all those little details, but I think this will be an interesting one to discuss. In parallel to this idea around who's getting medications and are medications helping people, this was, I think, a really interesting study looking at racial inequity in receipt of MOUDs by Barnett that was published in the New England Journal of Medicine. Background briefly, since 2010, we've really seen this increased rate of opioid overdose-related mortality, specifically amongst black persons in the United States, but also in other racial and ethnic groups. So this study looked at Medicare claims data from 2016 to 2019, so pre-pandemic. And they had some sort of an index event related to opioids use disorder. So usually an opioid-related overdose, but some sort of an index event that was documented and coded. And then they followed outcomes for 180 days after this index event. And they looked at about 23,000 people. So in this 180 days, six months after whatever this index event was, the patients who received buprenorphine, the percent of patients who received buprenorphine was 13% in black patients, 19% in Hispanic patients, and 23% in white patients. So quite clearly, differences there. And then those that received naloxone, similar 14%, 21%, 23%. Patients that received a benzodiazepine, interestingly, 23%, 30%, 37% after. And they noted that racial differences in the receipt of medications to treat opioid use disorder did not change over time, over that 180 days. And this conclusion found that black and Hispanic people with opioid use disorder were less likely to receive medications for opioid use disorder than white patients after an incident event. So here's some of the slides and tables for that. It kind of breaks it down for the numbers based on, across the top, black, Hispanic, and white. Going down with the medications, buprenorphine, naloxone, naltrexone, and opioid analgesic and a benzodiazepine. The graphs on the right are just showing the differences between white and black patients with filled buprenorphine prescriptions within 180 days after. And then the B table there is filled naloxone prescriptions within 180 days after. And our third article on opioids, enhancing patient choice using self-administered intranasal naloxone for novel rapid buprenorphine initiation. So this one we chose, it's kind of a different format, it's a case report, but we chose it because there's been all kinds of talk on is there precipitated withdrawal with buprenorphine, is there not, so the panel yesterday, Dr. Hawk was kind of talking about the very low rates of precipitated withdrawal. Then she was talking about people who have a rocky course. I know clinically a lot of clinicians talk about some of the difficulties here. And so we chose this case report because it did get picked up in the altmetrics, so people were reporting on it by Adam Randall in the Journal of Addiction Medicine. But overall they basically say that with illicitly manufactured fentanyl, starting buprenorphine is more likely to precipitate withdrawal and that this can impact things. And the patient who was able to transition, so this patient basically elected to give themselves intranasal naloxone as a way of sort of getting through the precipitated withdrawal and then transitioning on to buprenorphine. And so after the naloxone, the transition took 31 minutes, including 14 minutes of some moderately severe withdrawal. And when you give Narcan, that's what you get. And the patient then was able to transition and remain on treatment with buprenorphine. So I think, you know, lots of discussion, low dose, high dose, all these different approaches, what works, what doesn't. We're all using our clinical experience, trying to capture it as best as we can. The table on the top just kind of goes through some of the methods for transitioning on to buprenorphine, very low dose, low dose, high dose, and then this sort of case study. And then table two kind of gives the chronology, last use of fentanyl, cow scores, various events of what happens, some, you know, vomiting, intranasal naloxone was given, GI upset, cows vomiting, et cetera, goes through and you can kind of walk through it as an example. So with that, what's new in opioids? Why don't we start in the middle? Maybe John, Dr. Renner, would you like to start and then we'll circle around? Just some general reservations or comments. I mean, first of all, any treatment's probably better than no treatment. So if we can get people into some kind of treatment for opiate addiction, they clearly will be doing better and will drop the death rates. But we look at this other article and I think it makes the point that the social determinants of health are really important and it's not only important as to whether people get treatment at all, but I think it's whether people are consistent with their treatment. You know, if you're not housed well, if you're having difficulty with money, you may be on methadone or you may be on buprenorphine, but are you taking it every day or are you taking it the way you should be taking it? You know, this isn't a research trial where we're guaranteeing that people aren't taking the meds that we think we're guaranteeing taking it as prescribed. Doing an R21 right now on unhoused people and they are not taking the buprenorphine like they're supposed to, even in a trial setting, it's really, really difficult. So I think it just shows that we have a long way, we know where we need to go, but we've got a long way to go to make sure that the treatments are available to everyone and that they're able to take advantage of them correctly. Yeah, yeah. Yep. Okay, yeah. One paper that I just wanna make sure I mention this because it's highlighted a couple of times, so the Jones paper in which in the discussion section they say something like, we have found the receipt of MOD from OTPs and receipt of buprenorphine and it lowers the risk of fatal drug overdose. But then in contrast, consistent with prior research, receipt of ER naltrexone was not associated with a lower risk for overdose death or with a lower risk for non-fatal overdose in our prior study. So there's definitely camps out there that are becoming more and more anti-naltrexone. And I think that when you look at this paper, there were actually not that many patients who actually received naltrexone. So was it an effect that it didn't work or was it an effect that there wasn't enough patients on it? And I know in the New York Times, Maya Salsitz wrote that it's almost like malpractice to give naltrexone. I shouldn't use those words, but very strongly against naltrexone. And I think, I hate to see that the field gets polarized into in different camps because we don't have that many options and we should be considering all of them and really adjusting it by patient. And we have these large data sets, making patient-informed decisions, patient-to-patient. It's over-interpreted. It's over-interpreted, but no one should ever be using naltrexone. So that was one of my concerns. And then in terms of this intranasal idea, it's okay if, and I don't know how many patients are gonna be willing to throw themselves into significant withdrawal. Maybe some will, and it's an interesting thing. But again, like everything else needs to be studied. I just wanna make a plug for PCSS has a fentanyl guidance, which goes into the different options and why, in general, overriding- Like the different buprenorphine initiation. Yeah, I mean, the conclusion we reach is while all these approaches are reasonable, just overriding and giving high enough doses will probably solve your problem with buprenorphine if you're using buprenorphine. So anyway, so I recommend the guidances, which are online. Awesome, I was just gonna say, I had a patient who came in and was like, I don't want buprenorphine, I wanna be on Vivitrol. I want the long-acting injectable naltrexone. There's no way I can quit beforehand and I'm not going to detox. Just give me the shot. And I was like, you're gonna go into withdrawal, you're gonna be miserable. I can give you some symptomatic meds, but that's all we can do. And we documented it. And he was like, it was the worst 24 hours of my life. I'm so happy. And then I had him for three years after that and he was great. I think I, too, was very naive around some of these political comments on naltrexone and I kind of didn't pay attention to the impact of some of these things because I do think that I, too, kind of got lost in that a little bit and Dr. Nunez and I talked about that. And I do worry that sometimes these ideas prevent people from accessing things that can be really, really helpful for them. And I know that patient, as an example, was one where that medication was life-saving for him. Yeah, I agree with both of you about the naltrexone. And I think it's a general point about interpreting these kinds of papers. The statistics that you're seeing are the averages and are subject to the limitations, but they're just the averages. They're not showing the individual differences. So it doesn't mean that there isn't a subgroup of patients like your patient that are gonna do really well. Or this one who wanted to give himself Narcan. Right, right, right, right. The data actually show that naltrexone does a better job than buprenorphine of suppressing use and craving while it's on board. So if you can keep it on board. But I think the general methodologic point is you read papers and you see effects like no alcohol is good for you. Well, probably true, but there are probably some people for whom alcohol is good. You read the papers about people living in places where they live to be 100 years old. They all drink red wine. So you have to be careful reading papers and you see an odds ratio of two. Bear in mind that's just the average and hidden within that is a lot of individual variability that's probably not accounted for. And then my other thought was the paper about the racial ethnic differences. Yeah, the first, yeah. So my reaction to that is, I mean, of course the racial differences are what's emphasized, but how many people think all those three numbers are way too low? Yeah, yeah, yeah. Well, I mean, 23% in white people is still really bad. So the message I would take away from this is, yes, the social determinants make an important difference, but it's overall still pretty dismal. Everybody needs more. Other comments? The only thing I wanna add is these are people that had an index event too. So you can only imagine what the racial differences are in people who are just seeing a doctor for other reasons. These are people who overdosed, ended up in the ER and they still have incredibly low rates overall. And there's this huge difference and that's just really terrible. Yeah, absolutely, absolutely. All right. I guess so. All right, good afternoon. I will be bringing us home. Adverse childhood experiences and SUD treatment response. This was a paper published in the American Journal of Addiction. It tracked the association between the number of adverse childhood experiences, measures of depression, anxiety, stress, PTSD, and abstinence, self-efficacy at intake, 30 days and discharge. And it found people with more adverse childhood experience scores enter treatment with more depression, post-traumatic stress disorder, symptoms, generalized anxiety symptoms, and alcohol or drug abstinence self-efficacy. All patients improved at 30 days and discharge, but there was this separation at the 30-day mark between people with fewer adverse childhood experiences and people with more. So this is what the graphs look like. And I will actually go on this side. If you look at the separation, this is generalized anxiety scale, there was a clear difference at 30 days. A trend towards that here, but there was like a clear demarcation. And if you go to PCL-5, which is like a PTSD screening score, again, you saw a difference. Interestingly though, from my perspective, this study does show a convergence at the end of treatment. So that isn't to say that treatment is less effective for people with more adverse childhood experience, but it does imply it has to last long enough for people to experience a sufficient improvement in scores. And so this just highlights the importance of duration in treatment. Next study, displacement of people experiencing homelessness, PEH for short, who inject drugs. This is a modeling study. So it looks at population data, and it looked at 23 separate cohorts of people experiencing homelessness in geographies across the United States, pulling census data, 2018 National HIV Behavioral Health Surveillance data. And then there are two studies, one from LA and one from San Francisco, that actually stratified health outcomes for people who were involuntarily displaced from people who weren't. And that was then modeled onto all of the other variables that were tracked from all of the other cohorts to model what do we know about the impact of involuntary displacement on a number of health outcomes for people experiencing homelessness across the jurisdictions. Okay, so this is essentially the chart that they found. And then again, it's a modeling study. You'll notice there was not a huge separation over overdose deaths in and of themselves between the continuous displacement and non-displacement, but it's certainly a trend. Hospitalization was different, but there was a particularly striking difference between the combination of fatal and non-fatal overdoses and a clear separation in non-fatal overdoses. So clearly overdose risk factors go way up related to involuntary displacement. I live in Los Angeles. There are 75,000 unsheltered individuals living in my county. And there is this big question, what do you do? And when you do a, for lack of a better term, sweep, when you take people who are residing in a space and you relocate them, that does not come without health consequences is essentially what my view of this study showed. And they also looked at serious injection-related tissue infections or serious injection-related infections. And with that, I will now turn it over to our discussants on social determinants, adverse childhood experiences and substance use. I'm happy to go back to any slide you like. Thoughts? I just say I agree with you about the study of adverse childhood experiences. You look at those graphs, everybody's getting better. Everyone's getting better. So, and a lot better. So adverse childhood experiences are adverse prognostic features, but people still get better. And actually, if you look at it, you could maybe see that the people with adverse childhood experiences get more improved, get more better, more change. Yeah. There's actually more change. That's right. When there's greater severity, there's more potential for change. So it's actually a optimistic take on the relationship between adverse childhood experiences and treatment. But you often see that in the treatment literature, that the impact of treatment is greatest in the subgroup of the population with the most severe problems. So I think it's pretty optimistic. I think that's also clinically an important thing to use with patients when they come in and they're talking about all the different things. I mean, we have our motivational interviewing lens sometimes, but reframing it around people in your situation, higher rates of SUD, higher problems to start with, have the greatest rate of improvement. Yeah, and what is treatment providing that helps mitigate some of this? I mean, it's nice to see this data because you could get fairly nihilistic and say, oh, somebody's so overwhelmed with all this trauma from their past, and yet they have an opportunity to really still get better. And I think that's very, very positive. In the setting of this particular case was an academic medical center in Southeastern United States. I'll just suggest that patients there were probably receiving a combination of medications, evidence-based psychotherapies, and support. Like that being the core domains of what we do. I would just raise the question of, how long do we need to run a study to evaluate whether we're really having long-term effective benefit from treatment? That's a provocative question, Dr. Brower. How long do we run a study to show the long-term effectiveness? As long as you're funded, no. Well, is that going to give us the information that we really need? I forget who was saying in one of the earlier sessions that we have the idea that substance use disorders are different from most of the rest of the disorders in medicine, that you can stop treating them and they should stay better. Right. Right. I mean, this is the kind of Tom McClellan literature, right? Like, you know, addiction treatment. You're in treatment. You get better. You leave treatment, and, you know, the behaviors resume. You are hypertensive. You take your antihypertensive, your blood pressure goes down. You stop your hypertensive, it goes up. You stop your hypertensive, it goes up. You know, like that kind of analogy. Or why do we see the studies that show 12-step participation has a very strong effect? Well, because it keeps going on forever, and it doesn't end at three months and the patient is cured. I'll call it Big Book-style 12-step does not have no predefined endpoint. Right. Yeah. So, really, people should stay in treatment even when they're better. Yeah. And you would wonder how resilient people are with that type of trauma history, and they'd be fine as long as nothing goes wrong, per se. And they've had the support. Right. They get the support. Well, it oftentimes depends, you know, to what extent does the treatment provide a platform for ongoing support. You know, not everyone needs to be in nine-hour-a-week treatment indefinitely, right? But is somebody maintaining a therapeutic alliance with, you know, a clinician? Is somebody addressing their housing, vocational, social milieu? I mean, you know, and those things all actually are impactful. Yeah, I mean, it's interesting. You know, many of our faculty at Columbia, we supervise residents, and they sometimes get frustrated because they're not that motivated. They go back and forth in terms of relapsing. And you really have, in your clinical practice, you have the long game. You know, you're with patients and you have developed a rapport, and it can take years. But over years, you see your relationships so that, you know, when you're my age, you have a lot of patients that have been in recovery 10, 20, 30 years. But that wasn't the case the first few years I had them in practice. And yet, we expect research studies to find a difference and say that it's going to maintain itself post-study, and that's unrealistic. Right. Yeah. I don't have much to add, except for that I wouldn't overthink this adverse childhood experiences. It's a small end. These studies that look at associations of something that happened in childhood and treatment, you need more time. You need more people. And I think it's a little bit, you know, the role of trauma-informed programming would be important, but this doesn't really address that at all. The other study I thought was very interesting, this modeling, I don't really completely understand that. But it also, both of these are sort of in the category of things we think we know. They kind of confirm it. Like, you think people that have childhood traumatic events should affect their treatment. Or if you think it does make sense that if you disrupt someone's care or their housing with sweeps, it's going to make their substance abuse worse. It does make sense. I think it's important to do it, to show it. But anyway, I thought that was interesting. All right. So on to tobacco use and novel approaches. So full disclosure, I am a co-author on this one, so that may have impacted how it ended up in our list. Journal of Addiction Medicine published. I will promise you, ASAM has journalistic integrity. I was uninvolved with the review of this manuscript. So this was a study looking at a cluster-randomized implementation trial where the grant paid for technical assistance, but the actual delivery of care was delivered by the employee county health system. And then it tracked patient-level outcomes, although the study is on prescribing behavior and not patient-level outcomes, but it tracked patient-level outcomes in response to health system-delivered care impacted by technical assistance and training. So tobacco use rates, tobacco product rates remain higher among under-resourced individuals. In L.A. County, we found 18 L.A. County operating safety net clinics, 11 of them were community health centers, physical health centers like primary care clinics, and seven of them were community mental health centers. And they were cluster randomized in a four to one ratio to tobacco use disorder training with a monthly echo and as needed technical assistance versus treatment as usual. And then this particular paper looked at prescribing behavior. What do we see around the clinician's prescribing behavior of medications for tobacco use disorder? And then just one other caveat, you might notice that bupropion is not listed on here, mostly because of the difficulty in tracking the indication for which bupropion was used. So it's looking at just nicotine replacement therapy and varenicline. Not that you couldn't do that, that just how, that was a limitation of the data. So you'll notice, and I will now use this side, an upward trend in the primary care system. That's the figure on the left. In primary care, monthly prescriptions by medication type, and it looked again at the five FDA approved formulations of nicotine replacement therapy and then varenicline. You'll notice an upward trend. You'll also notice a dip in March of 2020, and I will leave it to you to figure out why that might have been. And community mental health centers also showed an upward trend, and this is just overall across the health system, right? This wasn't just looking like a net total, and this is actually unadjusted by total clinic volume, but I'll say the clinic volume was roughly stable. I mean, there was not a huge shift in clinic volume over this data period, and among the community mental health centers, again, an upward trend. Again, you'll notice a leveling out in March of 2020. So this was the scatter plots and the differences in rate of change over time, between the intervention group and the treatment as usual. There was a statistically significant association between a clinic having received enriched training for prescribing medications for tobacco use disorder and actual prescribing practices. It went up, actually, in both arms, but the slope was higher in the intervention arm. And last but not least, digital therapeutics for substance use disorders. And again, this was a American Journal on Addiction paper. Most patients with ICDs don't receive evidence-based treatment, and so the authors said, hey, is a digital therapeutic an option to help close that treatment gap? So the intervention was a digital therapeutic with 61 CBT modules that was based on the community enforcement approach with digitally delivered incentives after you complete modules, and many of them were set up with toxicology-verified abstinence. Not everybody, but what this study did is look at all patients, right? All patients that accessed this app from between January 2019 and March of 2021, from 28 states. There were 658 patients who received this digital therapeutic. 602 completed one or more module. There were 55% of patients active with technology the week 12, and a positive association between lesson completion and abstinence based on toxicology and self-report, when toxicology was available, but they also tracked self-report, and the conclusion was digital therapeutics can serve as an access point. There is no comparison arm. This was just a kind of observational study. This was the completion rate, so in weeks of treatment, who completed four or more lessons that week, who completed one to three lessons, and then as people crossed the eight-week mark, there were some people that actually just had completed all 61 lessons, so they were counted as a completer, and this was the association between average lessons per week in the first four weeks, treatment retention, that is people who were engaging the app in an ongoing way, and this was, again, percentage of patients and abstinence, so the more lessons per week were completed in the first four weeks, there was an association between retention in the tool and self-reported and toxicology-verified abstinence. Discussion on these approaches to treating substance use. I was just gonna make a point that that's one of our American Journal on Addiction highest ultimetric papers that we were covering. American Journal on Addiction was included in what we talked about here at this point. So, the digital therapeutic, one thing to know is that this was a study of the Reset app, which is gone now, because the company went bankrupt, and I'm acutely aware of that because we were in the middle of doing a big study. Oh, you were doing a study on it? In the clinical trials network that involved Reset as one of the randomized conditions, and we had to scramble around and figure out, but it's really a cautionary tale about these digital therapeutics. I think they made a mistake and sort of went like a big pharma company and hired a big staff and went into all kinds of debt, and they didn't have the income to back it up. They couldn't get enough reimbursement, and then it just went bust. In addiction treatment, business models are important. Yes, exactly. Wait, what did you say? In addiction treatment, business models are important. Oh, it's critical. Yeah. Matter. The second thing is I'm surprised at how good the adherence is here, that they get about, what, 50% of patients? That stick around. Stick around. In a couple of studies I've been involved with so far, it's not quite as good. My guess is that these are patients who put their hands up and volunteered to do this thing, so they were more psyched from the beginning as opposed to people who entered a study where they might or might not. It's not a randomized trial, right? So this is just sort of all comers, so you can't control for the person's intrinsic and just motivation. I think this is terrific because most of us are not gonna remember how to do community reinforcement approach, but this thing has 61 lessons that put into little 10-minute bits all the points in community reinforcement approach which are really useful. A lot of them would be useful to all of us, like how to get along with your friends and your spouse and your colleagues. There's a lot of useful stuff in there for anybody, but I actually think that the way to deliver it or one way to do it, patients using it and loves it, that's fine, but if the patients are not so, patient isn't using it or doesn't seem to be getting it, what I'm saying is it's like a prescription. You prescribe a medication, you've got to talk to the patient about the medication and how's it going, are you having side effects, is it helping you, and patient might say, I'm taking it every day and it's great or no, I'm not really using it, that if you're getting the message the patient isn't really using it, then another way to go would be to have the therapist actually monitor the session and talk to the patient. Maybe even bring the session into the visit because it's only 10 minutes. Listen to it and then talk about it because then you essentially get the patient to listen to the programming about how to have an argument without having a fight and things like this that are in there or how to say no to drugs and drug refusal skills, all these good CBT and CRA things that are in there, but it's not so easy to get the patients to do it. I think it raises the question, is it easier to change the therapist's behavior or is it easier to change the patient's behavior? It's really hard to change the therapist's behavior. Right. It's really hard. But this makes it easy for the therapist because the therapist doesn't have to remember anything, it's just the therapist can say, oh, you're having trouble getting along with your family members? Well, let's go listen to module so and so about how to have an argument without a fight and then you can listen to it and talk about it. But you hope there's some sort of outgrowth from this. I mean, the fact that this business model failed, I think part of it that I've learned being with a spouse who develops devices and whatever is you've got to have the business model up front and I hope it doesn't put a pall on this development. I mean, I know NIDA and other NIH are very interested in this still, but it's critical to take care of reimbursement and getting all that set up up front rather than just jumping into something that you know is great or think is great and not having it be put in, whether pieces of this could be used or repackaged so that it gets out to the community. I think it's a very open question about if you do have a novel digitally delivered treatment, do you attach it to the health system? Do you try to get your state pair to pay you to do it on top of everything else? Do you have insurance pay for it? I actually think we're kind of still in an emerging category where's the groove for these technologies and what's the business model gonna be? I forget the device, the auricular device which got FDA acceptance. They say cleared, I can't make a distinction between cleared and approval and the uptake I think has been nil, right? I mean, it's been and so what you're concerned about is sort of like what happens with medications where a company will say it's not worth going into this area because it's 10% or five. Can't get in. Or people won't prescribe it. So I think how to not have this become a neg, Paul on the whole development of these kinds of approaches is concerning. I think they were trying to charge $1,700 for 12 weeks. I'm thinking, come on, for an app? Yeah, yeah. Most apps are free. Yeah. There might've been some overhead built into that. Or heavily. But I mean, yeah. They were heavily trying to get into the VA for a while too. I don't know, they got on the formulary at the VA and they were really aggressively marketing it, let's just say. Well, Dr. Renner, I was gonna say, I appreciate you're talking about the difficulty changing therapists or clinicians' behavior. I am glad that there is some evidence that it's possible with a fair amount of effort. So with that, I think we're gonna transition to an audience Q&A. Dr. Maribeth. All right, so lots of different articles, lots of different things. Again, I think what's novel here is what gets picked up in tweets and media and press releases and how does that impact what we think, what we do, those types of things. So let's start with a question here. Hi, thank you for the discussion. I'm Julio, I'm a psychiatry resident at Yale. There was one article in particular that I read earlier this year, Barnett's study on New England Journal of Medicine. It's a refresh. It discusses how some racial ethnic minority groups had less access to treatment for OUD, particularly not Loxone. But what's interesting in that study is that there were no differences in follow-up rates between the groups. So the differences in prescription were not explained for reduced access, for example. Black, Latin American, and white people, they had the same follow-up rates. Well, and also I think sometimes people assume that certain disadvantaged populations don't follow up. And we actually published a paper on bipolar patients that they actually do follow up disproportionate to the perception of that. So that there's this idea that maybe they're just not following up, and then also the idea around access. But, sorry. Exactly, I'm curious on what your thoughts are that actually drive those differences. So what drives the differences between, the racial differences, since it's probably not an access issue or it's probably not a follow-up issue? Well, it's a prescription issue. Like, thinking of it and. So I would say it's a structural racism issue. But yeah, you're saying they're not getting the prescriptions based on race, yeah. That's what I would guess, yeah. Yeah, so you're saying the bias in terms of prescribing and people just not prescribing. And interestingly, bias against not giving the population, African American and minority populations benzos, that sort of in some ways is protective if you're worried about giving too many benzos. I think that that was an interesting sub-finding. There's all of this controversy around benzos, should we give them, should we not? Is there a harm reduction role for benzos instead of using illicit benzos? Is there a role for them to play in precipitated withdrawal, et cetera? And I do think that there's interesting racial differences in who gets something that is very closely guarded. That was true with the prescription opioids, it was that there was less of an inroad in African American populations compared to white populations, at least initially. Yeah, I think that's true. And I don't know the data for like stimulants for ADHD, but I would guess we see similar things there too. We'll switch sides of the room, there you go. Just a further word on the post-mortem for reset. Important to remember that part of its impact was the CM element, not just the CRA. And I think that part of the difficulty and uptake for reimbursement was just part of this broad skepticism that reimbursers and the field and government has about paying for CM, which was built into the cost and part of its effectiveness. So that's still up in the air, broadly. The second thing is that- That's a great point, I just wanna say thank you. And then the second thing is that the user interface was lame. And that's what my patient told me. So I had patients use it who loved it. Yeah, yeah, but it really diverged. Half really loved it, half said, blah, anyway. But I think your point around contingency management speaks to the larger point around the business model, right, so we're still trying to figure out a business model for contingency management. Unrelated to reset, our state, the state of California, the one we're in right now, our Medicaid program pays for contingency management. So like there are, in other words, like it's evolving. I think there now will be examples of payers that are ready to actually implement technology-supported CM. But I think- It's too soon. But there was also like New Jersey and Oklahoma Medicaid that paid for reset, and that wasn't enough. So it's true, figuring out these business models. Other thoughts on that? Okay, yes? Hi, thanks for this review. I'm Paul Grechen, I'm Medical Director of Evergreen Treatment Services, which is a large non-profit OTP in Western Washington. And I wanted to just point out what I think was Maya Solisovitz's main point about naltrexone. You know, we use drug use as an outcome measure in our studies, but I think when you're talking about a highly fatal condition like opioid dependence, there's no more important outcome measure than mortality. And given that there is no data to support that naltrexone is as effective as buprenorphine and methadone, there are still courts and various programs that require patients be on naltrexone if they're gonna be on any medication at all. And I think that her point is that there is no excuse to deny patients to a more effective medication. Or to deny the patients the medicine that's best for them. Yeah, shared decision making. I think I would assert that there's some, it was like, Carla, like your patient, where that's what they want, that's what they'll take, that may be what they'll do well on. So I wouldn't deny naltrexone to a patient that really wants it and won't consider something else after a shared decision making discussion. But yeah, I wouldn't push naltrexone on anybody either. I'd probably lean in favor of buprenorphine in an office-based setting, or methadone in an OTP. Methadone's more effective than buprenorphine, frankly. Yeah. If you look at the data. It's just it's not available to everybody. So I agree with you. I think when in doubt, or the patient's willing to take buprenorphine, give them buprenorphine. It's safer. But then your point also around what is legally required and what people are allowed and not allowed to have access to, and how that doesn't track with the data either. That's really a shame, you know, that there's legal stuff like that, yeah. Yeah. Thank you. Some systems like New York, now in the carceral system, require that all three medications be available. That it's not just naltrexone, yeah. That's great. Yeah. Yeah, hi, this was really fabulous. Oh, sorry. Oh, sorry, go ahead. Go ahead, doctor. I was just saying, you may get a chance to make an adjustment to a different med if the patient has tried their first choice, and you maintain a relationship. And then if they have trouble, you may be in a better position to recommend something new rather than try to force it on them in the first place. Yeah, I think establishing that rapport is critical. Absolutely, for success. Just so wonderful, and I had so many thoughts, but I'll try to keep it to two. One is, I just wanted to make a point more than a question about the benzodiazepine issue that y'all were just discussing. This is an evidence-based risk factor for overdose death among chronic pain patients who receive prescription opioids. And so it's striking to me in that paper, so prescribing benzodiazepines to someone who's just had an overdose event would really be malpractice. And it was striking that there were more patients receiving the contraindicated medication than those receiving the indicated medication. So that was one of the most striking things about that paper to me. I don't know for sure, but I wonder if the benzos was after they were given naloxone and then they were having opioid withdrawal, but I don't know for sure the role, but your point is well taken. Yeah, but at least looking in the VA system as we were trying to do our opioid safety initiative and one of our important points was getting providers not to prescribe benzos when someone was taking opioids. And it's not easy to get folks off benzodiazepines as we all know. But it was especially striking that we would re-prescribe the benzodiazepines even after it contributed to an overdose. So I wondered about that. My other question for you though was about a different paper looking at MOUD and finding that telehealth and in-person office-based buprenorphine had similar outcomes in terms of overdose risks during the pandemic, which was wonderful news. But it seems to me that the OTPs had a strikingly larger effect size on that outcome. And I would love to hear from the panel what y'all think about that. I'm a big pro-OTP person. I don't know if the panel has thoughts on. Was that in, was the OTP thing in the data that we looked at? So that was the third point in the data was that the receipt of medications through an OP, or the receipt of treatment through an OTP was in there. So. The relative risk was 0.4 as opposed to 0.6. So receipt of. Was more effective, right? Is that what you're saying? Yeah. Yeah, it's more effective. So receipt of OUD-related telehealth was the first one with an adjusted odds ratio of 0.67. Receipt of MAUD from an OTP with an adjusted odds ratio of 0.41. And then receipt of buprenorphine in an office-based setting was 0.62. So it actually had a receipt of MAUD from an OTP. And so. Which is probably the more likely methadone, right? Potentially more likely methadone and potentially more in-person contact at some point. More in-person contact monitoring, all of that. So it's also selected for patients that are willing to, more motivated and willing to undergo that, right? That's the other thing. Well, or who are there, yeah. I mean, people who are in a clinic in an established sort of framework. And even if they do the telehealth through the OTP or other things like that, I think that setting has something to do with it. Yeah. Oh, yeah. All right. So I think we don't have anyone immediately following us. So I'm gonna go ahead since you've been waiting patiently. All right. Go over time. Thank you. I'm from Wisconsin. Thanks for a wonderful panel and discussion. My comment of sort is about the digital therapeutics paper. I'm glad that was one of the papers selected. I personally was one of the very early adopters of digital therapeutics. Like I had been using both Reset and Reset-O for about three and a half years till they filed for bankruptcy on April 7th. I had a whole panel of about 100 and close to 200 prescriptions that I had written. I was very impressed with the different nuances of what we could get from this. I'm very glad that Dr. Fishman brought up the issue about not only did it had CBT and CRA, but also the contingency management, which honestly, despite all the best evidence that we have, delivering it in the real world remains a challenge. And these apps were helpful in bridging that gap. So we integrated it as part of the treatment. In Wisconsin, we were very close to having it integrated into the Medicaid as a formulary-like product, but- When you prescribed it, how did you work with the patient? How did you talk to the patient about it? I introduced it as a product, which was available. I gave a brief, but I think one of the key things was you couldn't just prescribe and leave it alone. You actually had to be very involved. And I think that was the key. And I think that's what led to its downfall that people thought I would just prescribe and that'll be the end of it. It didn't work that way. So when I got data from PAIR, they showed that the engagement level and the percentage of my patients who were doing the modules were actually close to 80%, which wasn't anywhere close to the national norm. So my point is we shouldn't throw out the baby with the bathwater. I mean, there's lots of things we learned from this. We can learn from it. And I'm happy to say there's actually a few other digital therapeutic companies who have adopted a different business model. The, I don't know if you're aware of Chess Health. They have gone the route of more like discussing directly with organizations as opposed to having this prescription. The FDA clearance didn't do a darn thing. Of course, we know that. But there's Dynamic Air Health. So there's a few other players. There's CBT for CBT. So I think we should start talking more and more about how we incorporate all these different technologies because they've been incredibly helpful. And I'm just very interested in the fact that you were very involved with delivering this to your patients. Absolutely. You didn't just prescribe it and say, good luck. I was seeing the results of inter-appointment, what my patients were doing when they were engaging in the product. And then would you talk with them about what they were learning from the app? Absolutely, and I think that was key. Thank you. Awesome, yeah, thank you. I mean, I think that was the idea, though, is that when we talked about this theme of novel approaches, novel therapeutics, so the business model, what does it mean for our practice? How do we kind of think about integrating these things is really critical. So yeah, that's helpful. All right, thank you all. All right.

cannabis

opioids

telehealth services

psychiatric disease

depression

bipolar disorder

psychosis-related healthcare utilization

opioid use disorder

fatal drug overdoses

telehealth

pandemic cohort

digital therapeutics

addiction treatment