Good afternoon everybody. I'm Michael Dawes. We're getting ready to start. This is paper session number three, and just to give you a brief overview, we're going to have 10-minute presentations, and please hold your questions until the end, and I'm going to go ahead and get started with Wasim. Wasim Stefani is going to do the first presentation. Welcome everybody. Thank you. Good afternoon, everyone. It's Dr. Stefani from BronxCare PGY-3. Today we're going to be talking about substance-induced agitation management, and we're going to talk about the study we did. This is my first slide. I don't have any disclosures, and as we said, we're going to be talking about the study we did. I added some substance-induced agitation and the management. So as we know, agitation is a state of restlessness and irritability the patient has, and it's very common in the psychiatric field. Also, it's most common in acute settings like CPEP and ED, and it has multiple allergies, can be due to substance, can be due to paranoia, or it can also be due to mania. We need to manage agitation for patient safety. When they get agitated, they can hurt themselves, they can hurt other patients, and of course for staff safety, too. Substance-induced agitation can happen in two ways. First, intoxication due to neurochemical impact on the brain. Which can cause, as we said, paranoia, sometimes hallucination, visual hallucination. Also, it can cause false belief of superpower, and it can happen due to withdrawal, which patients feel uncomfortable feeling, irritability, and they have poor affect regulation. Some of substances that can induce agitation, as we know, cocaine, which can cause euphoria, confusion, aggressive behavior, and emotional liability, and methamphetamine, which can cause delirium, agitation. Also, we have cannabis, which can cause sometimes psychosis, and impaired judgment, K2, confusion, depersonalization, and psychosis. Hallucinogens can cause, of course, hallucinations, depersonalization, distorted sense of time. Also, they can get panic reaction, and they get agitated. This is the study we did. It was a retrospective study, cohort study. We did it over a period of six months in 2019, late of 2019. We reviewed all the charts from CPAP admission charts. We looked at substance at that time used, history of substance use, neurotoxicology, and the history. Agitation data, we collected episodes of restraint, intramuscular medication used, past diagnosis, time to emergent medication. Number of cases was identified. 70 cases were restrained, and we got IRB approval at that time. This is the data we got. UTOX, at that time, 35 patient has positive UTOX. 24 patients, they have negative UTOX, and 11 patients, they didn't have a UTOX. It wasn't ordered at the time of admission. Age distribution, we had the most, was between 25 and 34 years old. The second most was 35 to 44 year old. Gender distribution, the most was male. 50 case was identified, and 20 case were female. Racial distribution, the most was African American, second most was Hispanic. Also, we have Caucasian too, and 35 were others. Outcome, so I should have started with this. Our study question was, what is the best IM medication for intoxicated patient, and we reviewed all those data. After that, we put some criteria. Our criteria were, we considered the IM was successful if we didn't need to repeat the IM in 24 hours, or if it was repeated after 24 hours. IM medication wasn't successful if we had to repeat the IM within 24 hours after a patient got medicated. So this is what we found. The medication was used. It was Haldol, Thorazine. We have Ativan, Olanzapine. The most IM medication, it was Haldol. It was like 41 patients got medicated with Haldol, and 21 case wasn't repeated. 15 was repeated after 24 hours, and 5 cases, it was like, we repeated, it was repeated less than 24 hours. The second most medication was used is Thorazine, which was 24 cases in total. 8 cases, we didn't need to repeat the IM medication. 5 cases, it was repeated after 24 hours, and 11 cases was repeated less than 24 hours. So we have Ativan, it wasn't that much. Olanzapine, it's like in our hospital, we don't use it that much. So, result and conclusion. So we found that the most substance of intoxication, it was cannabis. Most IM medication was used was Haldol. When we calculated B-value, it was 0.1, it was insignificant. But T-value was like 1.3, which tells us like if we use Thorazine, it's less likely to be, patient is less likely to be medicated again. Here I added the consensus statement of Emergency Psychiatry Project Beta, how to medicate agitated patient. So as we see here, if we have agitated patient, we try like verbally to de-escalate and we do the assessment. If we feel the patient is delirious, our assessment, if the patient is delirious, we need to stay away from benzos. We can't medicate the patient with first-generation and second-generation antipsychotics. If the agitation is due substance, if it's intoxication by ethanol or its CNS depressant, we should stay away from benzos. We can't medicate with first-generation, second-generation. If it's due withdrawal from alcohol or benzos, we can't medicate the patient with benzodiazepine. Finally, if it's like due to psychosis, we can use first-generation, second-generation plus minus benzodiazepine. That's it for me. Thank you. So I just go ahead. All right, so hey everyone, my name is Saral. I'm a PGY-1 psychiatry resident at Tower Health Phoenixville Hospital in partnership with Drexel University. Today, I'm going to talk about this topic of depression, suicidality, and co-occurring substance use among adolescent hallucinogen users. So I don't have any financial disclosures to any conflicts to disclose. So at the end of it, you all will be able to identify prevalence and trend of hallucinogen use in American adolescents. We'll be able to identify prevalence of depression, suicidality, and co-occurring substance use among them, and identify the association between depression, suicidality, and hallucinogen use. So recently, as you all know, hallucinogens are gaining popularity as a new intervention for different kind of treatment of psychiatric disorders. For example, in supervised settings, psilocybin shows promise in treatment of anxiety, depression, PTSD. At the same time, there is a rise in microdosing of illicitly purchased hallucinogens as a form of self-medication to help them with anxiety and depression. So illicit hallucinogen use without supervision. We don't have data. It may lead to bad outcomes. There are reports out there suggesting bad trips, worsening of pre-existing psychiatric conditions. So and there isn't much data about this effects on adolescent, you know, developing brain of adolescents. So there isn't much research in this area yet. So in order to do our study, we used CDC's Youth Risk Behavior Surveillance System 2001 to 2019 data, which represents, so this is school going adolescents. I need to clarify that. And in database, they ask about questions related to hallucinogen use, which includes LSD, PCP, and mushrooms. They have also used street names because these are questionnaires. So we did statistical analysis, Ralph Scott chi-square test for categorical variables, and unpaired t-tests for continuous ones. And then we did multivariable logistic regression analysis to establish the association between depression, suicidality, and hallucinogen use after adjusting for potential confounders, which were socio-demographic variables and co-occurring substance use. So what we found was that there were 125,550 participants. Out of them, 8.4% reported ever using hallucinogens in their lifetime. More number of male participants reported hallucinogen use compared to females. And what we found was that compared to adolescents who never used hallucinogen, those who used it had a higher prevalence of feeling sad and hopeless, considering suicide, and attempting suicide that required medical attention. And they also had a higher prevalence of co-occurring substance use, such as alcohol, cigarettes, e-cigarettes, marijuana, synthetic marijuana, inhalants, heroin, cocaine, methamphetamine, and ecstasy. So this is the trend of hallucinogen use. As you can see, that trend line is going down. So it went from 13.3% in 2001 to 7% in 2019. So it is going down. Once again, this is for school-going adolescents. It doesn't represent all U.S. adolescents. This is the prevalence of sadness and hopelessness, and other mental health variables. As you can see, that blue line represents those who had used hallucinogens, and the orange one represents those who have never used hallucinogens. And you can see the prevalence of poor mental health outcomes were higher than those who ever used hallucinogens. This is the prevalence of co-occurring substance use. Once again, you can see that those who had tried hallucinogens, the prevalence of all substances that we included in our study was higher in those individuals. Alcohol being the most common, cigarettes, e-cigarettes, marijuana, and then the prevalence decreases, but still way more higher than those who didn't try hallucinogens. So this is the result of multivariable logistic regression. So as you can see, the initial sad and hopelessness, we adjusted for age, sex, race, and co-occurring substance use. For suicide, we additionally adjusted for feeling sad and hopeless because that would increase your risk of suicide. Then for suicide plan, we adjusted for social demographics, co-occurring substance use, feeling sad and hopeless, and considering suicide to account for that as well. And we did the same for injurious suicide attempt. And as you can see, the results were statistically significant for sadness and hopelessness considering suicide, making a suicide plan, but we didn't find any association between injurious suicide attempt and hallucinogen use. So the conclusion is that there is overall decreasing trend of hallucinogen use among school-going adolescents. There is a high prevalence of depression, suicidality, and co-occurring substance use among them. They are at higher odds of feeling sad, hopeless, considering suicide, and planning suicide, but not associated with suicide attempts. So there are limitations to our study, the main one being it's a cross-sectional design. It only can tell us about association, not causation, so we don't know the direction of this association. It is possible that depressed individuals and suicidal adolescents may be more prone to experimenting with hallucinogens, or those who experiment may then develop depression and suicidality, so there's no way to know that. So yeah, that's our study, and the main point of our study is that we wanted to highlight this association and bring in light that in light of current attention that hallucinogens are getting, and some states have already decriminalized it, so this needs to be taken into consideration, that this is something that we don't know yet, and it's not like you go out and start popping mushrooms, and we don't know what are the effects in adults and population. We don't have much research, so yep, that's my research. Thank you so much. All right, I'm Shelby Powers. I'm a PGY-3 at Duke, and today I'll be presenting on some people who have a lot to say. So I'm going to go ahead and introduce myself, and then I'm going to turn it over to my co-presenters. So I'm going to go ahead and introduce myself, and then I'm going to turn it over to my co-presenters. All right, I'm Shelby Powers. I'm a PGY-3 at Duke, and today I'll be presenting on some data on our inpatient management of peripartum patients with opioid use disorder at a large academic medical center. I have a lot more to present than I have time to present, so I'll just be highlighting some of the main findings on these slides. We don't really have any significant disclosures, but this work is funded by the Duke Endowment. Our learning objectives today are to understand the growing burden of OUD in pregnancy across the U.S., to learn about inpatient management of peripartum women with OUD who are admitted to a large academic medical center, and to identify gaps and challenges in providing appropriate community transition and harm reduction support for these patients. If you were at the great presentation earlier today by Dr. Greenfield, she already did a great job at highlighting that opioid use disorder in pregnancy is a growing problem. So I'll just highlight that the highest burden of overdose deaths are in that reproductive age population, and along with other opioid trends, there's been a four-fold increase in the number of patients with OUD at labor and delivery. Additionally, one in 20 pregnancies among those with Medicaid in nine states, in the study identified here, were found to have a pregnant patient with an opioid use disorder diagnosis. The implications of this for both the pregnant patient and for the neonate are significant. An OUD diagnosis at delivery is associated with a 4.6 times increase in death. The rates of pregnancy-associated overdose deaths have doubled, and opioid overdose is now the second leading cause of postpartum death. Opioid use disorder in pregnancy is also linked to a number of poor outcomes for neonates. Pregnancy also could be a prime time for intervention. Pregnant patients who present for obstetrical care should have a screening at their first visit for substance use, which is a point of interaction which they can be identified for treatment. For those who continue to have obstetrical care, their increased contact with the medical system, once again, is an opportunity to get and continue treatment for these patients. And there is some evidence that there is high maternal motivation for positive behavioral changes during pregnancy. However, only about 50% of patients are receiving, oh, pardon me, I think, I forgot the slide was in here. We're also, we have medications with which to manage these patients. Historically, it's been methadone or butanol product. However, there's evidence that butanoloxone is also likely safe for our patients in pregnancy. And we know that the benefits of MOUD during pregnancy are significant both for the pregnant patient and for the infant. Yet only about 50% of pregnant people will receive evidence-based treatment during their pregnancy. So, once again, with each contact with the medical system is an opportunity for patients to initiate treatment. And inpatient hospitalization represents a critical opportunity for us to also offer evidence-based OUD treatment to pregnant and postpartum people. And this has not yet been described. So, our aim was to describe the population of pregnant and peripartum patients at Duke University Hospital with OUD, who were evaluated and treated by our OUD consult service. Our consult service is called Comet. It's a great, great name. It is unique in that the hospitalists who staff it, none of them are addiction trained or addiction boarded. They come from internal medicine, med psych, and med peds. They also have a dedicated social worker Monday through Friday. And it's an elective on which psychiatry residents can rotate. With the Comet service, we evaluate for OUD. We treat acute withdrawal. We provide treatment by initiating or continuing MOUD for our patients. We educate patients, and we try to link them to community resources like MOUD services on discharge. Our methods for this work was it was a retrospective chart review. We looked at all pregnant or postpartum patients who were admitted to Duke between May 2020 and May 2022 who had a consult completed by this service. And we looked at a number of factors for them. There were 27 total admissions, which met our criteria. 26 were during pregnancy. One was postpartum. And there were 25 patients involved. The average age for these patients was 31. It was a range from 22 to 40 years old, all identified as female. And here's our racial, ethnic, and payer data for these patients. Not only at Duke, but also at sort of a nearby academic institution down the road. A lot of us in addiction have been talking about equity in these programs which we've started. Like, are we serving the communities in our area with these programs? And I'm still not sure about that, but I do find it interesting that at least as far as the racial breakdown of patients who are seen on this service, it very closely approximates the racial diversity that we have in Durham County. As far as pre-pregnancy comorbidities for these patients, the most common were chronic pain, hypertension, and hepatitis C. Psychiatric comorbidities most commonly were anxiety and depression. Notably, 28% said that they had experienced intimate partner violence. Urine drug screens were not ordered by this consult service. However, they were done on 82% of admissions, usually by the admitting team or in the emergency department. The opioids found on UDS were mostly opiates. However, fentanyl is not yet included on our UDS. As far as non-opioids on UDS, THC and cocaine were the most common. When we went and evaluated these patients as part of our consult, we found that 25 of the 27 admissions met criteria for opioid use disorder, and 88% of these admissions had active opioid use disorder. 72% of the patients had previously used MOD at some point, and then 36% had used MOUD during their current pregnancy. Patients reported their substance use prior to admission, with heroin being the most common opioid misused, or what the patients believed to be heroin, followed by opioid pills and fentanyl. A concurrent substance use that they reported was mostly nicotine, cannabis, and cocaine. 70% of the admissions had MOUD provided. This was mostly bupenaloxone, followed by methadone, and then bupenol product. Reasons that patients might not have received MOUD was it was not indicated, they had self-directed their discharge and had not yet been able to receive MOUD, or that they had declined to have it started. As far as the pregnancy outcomes or characteristics for these patients, 48% had limited prenatal care. 68% of them delivered during an admission in which a consult was completed. And then most of these deliveries were via C-section, and that was 70% of them, which maybe speaks to the acute nature of these patient presentations. There were also many complications. 59% of the patients were receiving MOUD at delivery, so they'd already had a consult completed and MOUD had either been initiated or restarted for them. As far as infant outcomes are concerned, AFGAR scores were very average for these neonates. However, 65% of them were admitted to the NICU. Because we only looked at maternal charts, it was very difficult to try to discern what was going on with the health status of these neonates. 36% discharged with their mother, and 50% of cases had CPS involvement. Meds provided at discharge included MOUD for 60% of the admissions. This was bupenaloxone followed by having methadone arranged in the community. We prescribed Narcan for about half of our patients. Some already had a prescription at home. Some had self-directed their discharge before we could provide it. But a full 26% of them we did not provide Narcan, and it's unclear why in their chart. As far as post-discharge linkages to care, 72% of the patients were connected with MOUD treatment in the community. One patient declined, but as far as the others, they either had self-directed their discharge before this could be arranged, or they weren't linked to this care for unclear reasons. So we found that offering MOUD during inpatient admissions with a dedicated hospitalist-led consult team resulted in MOUD initiations and restarts for pregnant and peripartum patients. Just to recap, 36% of them had been on MOUD at admission or shortly before admission. 70% of them who received a consult received MOUD when this was offered. And then 60% of admissions with OUD were discharged with MOUD either prescribed or arranged. So a major challenge we identified here was self-directed discharges. About a third of our admissions had self-directed discharges, including several elopements. Only one of those nine self-directed discharges was on MOUD at some point during their pregnancy. Only three of the nine received MOUD in the hospital, and six discharged without an outpatient MOUD plan in place. So we're wondering, are these patients different? And although we didn't do any statistical analyses of these small samples, we did want to look and see what differences did we notice. We found that they had lower cow scores, although again haven't run a T-test to see if that's significant. They tended to have increased cocaine use. They were less likely to have delivered a living infant. They had increased CPS involvement and an increased history of intimate partner violence. And so as we look at each of these factors, particularly maybe after being just in the induction without withdrawal presentation that was so wonderful just before this, it makes me wonder, you know, in what areas can we be improving our care for these patients? Some of the remaining gaps in care, which we found, we're kind of wondering if we're doing a good job of identifying these patients in pregnancy. This data is just based off of consults we received. For all we know, there are a lot of patients who are not getting consults who we could be providing care for. Another gap is identifying and addressing factors related to MOUD refusal, self-directed discharges, and misconnections to care, improving our Narcan provision at discharge, enhancing linkage to other postpartum care services, and then assessing if patients are successfully receiving care in the community, which I don't cover here but is something we have been attempting. For our next steps, we are examining all pregnant and peripartum patients with OUD across our health system to assess if there's a difference in care between the service we're providing through Comet and at the hospital where there is no OUD service. Like, can and should we expand this consult team? And then we're looking to identify factors related to MOUD refusal and connections to care. So here are my acknowledgments for the Comet team and my references. Thank you. Okay. I am Brian Furline. I'm the Director of the Psychiatric Emergency Room at VA Connecticut and on the faculty at Yale. I'm going to be discussing a study we did on the highest utilizers of psychiatric emergency room services at VA Connecticut. The majority of this work was done by a senior medical student who was unable to attend because she's applying for psychiatry residency and doing her interviews now, which apparently psychiatry residency has become quite competitive, which I guess some of the residents up here could attest to, but it's a little bit more competitive than when I was a psychiatry resident a few years ago. No disclosures. So a few learning objectives. So we want to know what the characteristics are of the patients who really are the highest utilizers of emergency services, at least at VA Connecticut, which we do have a distinct population of patients. What are the baseline characteristics of these patients, and what are the longitudinal outcomes of this group? So first I'll tell you a little bit about the Psychiatric Emergency Room at VA Connecticut. In the community, psychiatric emergency services is not an unusual thing, but at the VA it is quite unusual. There are very few psychiatric, true dedicated, freestanding locked psychiatric emergency rooms at VA hospitals. In fact, we really don't know of very many others outside of ours. There are small areas within medical emergency rooms where the psychiatric patients are often housed, but there's rarely a completely locked, dedicated, separated unit the way we have that staffed 24-7. We have a capacity currently of 12. We are expanding, and our patients are almost all voluntary, unlike our community partners where police will often bring patients into psychiatric emergency rooms involuntarily at VA Connecticut because of the nature of it being a federal building. The vast majority of our patients are voluntary. If the police encounter someone outside of the hospital, even if they're a veteran, they will typically bring them to Yale Psychiatric Emergency Room and not to VA Connecticut because we don't really adhere to state laws because by nature of us being a federal building. So we also obviously, being a VA hospital, we serve almost all men. It's about 90 percent men. It's pretty much all veterans unless there happens to be a humanitarian case, and the average age of our veterans is a bit older than the general public, and they're, like I said, almost all voluntary. So we have a very different population of patients than our community psychiatric emergency room at Yale, which is just a few miles away and serves the greater New Haven community. We serve a much more distinct population of patients. We have approximately 30 doctors, attending-level psychiatrists that staff the psych ER at some point throughout the month. We have about 30 who do shifts. Some are VA faculty, such as myself, who do extra shifts, our moonlighting physicians who have other jobs in the community and just work occasional shifts in the emergency room for extra pay. We get between 100 and 120 new patients per month. So it works out to be about four new presentations per day, and our average length of stay is about two days, so we're not really a true emergency room where people come and go quickly with a high volume. We have a lower volume and a longer length of stay than a typical emergency room, but definitely more than a typical medical emergency room and even more typical than other psychiatric emergency rooms. And you can see what our patients present with. The most common presenting complaint is alcohol use disorder, and it adds to more than 100 because our patients have more than one thing. So the top 20 utilizers, the way we defined them is we looked at the patients who presented between 2010 and 2014. We went a few years back because we wanted to look at some longitudinal outcomes of this group. So we looked at all of the patients who presented between 2010 and 2014, number of presentations, and we got obviously other information in the spreadsheet, but number of presentations, and then we sorted it by number of presentations, and we took the top 20 who had the most presentations during that five-year period. And the most amazing part of this is that the day I got the spreadsheet from medical informatics, the first thing I do is sort the data and look at the top 20 people, and I looked at the top five people because I was curious, and of the top five, three of the five were currently in the emergency room that day that I got the spreadsheet. There were only like five patients in the ER, and three of them were the top three from this five-year period more than 10 years ago. It was quite remarkable that they happened to be in the ER that particular day. So it just speaks to the fact that they're still very high utilizers of our services. The mean number of visits they had was 61.5 during the five-year period, and you can see a little bit about the patients, right? They were 51 years old on average. Most of them were white. Nineteen of 20 were homeless at some point during the period, either before or during the period. Seventeen of the 20 had legal problems. And the stat that I highlighted, which is pretty remarkable, is if you take all of the patients who presented, these 20 account for 0.38% of all patients, right, not even half of 1%. But during the 11-year period from 2010 to present, they accounted for 15% of all visits to the psych ER. So one out of every six presentations over the last 11 years was one of these 20 patients, which is quite remarkable. A little bit about the patients at the index visit. The majority of them, they presented that day for substance use. Most was alcohol. Three presented for psychosis, one for mood, one for anxiety related to PTSD. And you can see a little bit more about them, right? The majority had a history of substance use. One actually had no history of substance use disorder at all. And you can see 16 with alcohol use disorder. Fourteen of the 20 had one other mental health diagnosis other than substance use. They had on average almost four chronic non-psychiatric medical conditions, and they were on a little over five medications. And 11 of 20 had an alcohol-related medical condition, so we coded those a little separately. So pancreatitis, GI bleed, things like that. So more than half had a serious alcohol-related medical comorbidity. So looking ahead now, this group and then moving ahead, what happened to them? Over the 11-year total period of the study, they did have a significant number of primary care visits. They were very plugged into care. So this is not a group who only came to the ER. They went to the primary care. They averaged 43 primary care visits per person. They also had an average of 53 hospitalizations. So they're hospitalized very – that's not emergency room visits. That's hospitalizations either in psych or on medicine. And then you can see the stats. You know, 18 of the 20 had an ongoing alcohol-related visit. So all of them used social work services. Again, they were plugged into primary care. They're hospitalized frequently, and they're plugged into social work services. Half of them received homelessness services during the course of the study. So what happened to this group at the end of our study period at the end of 2020? Five of the 20 had passed away. We looked through the chart to see how. Two, it said, were natural causes. Two were cancer, and one was related to alcohol. Three of the 20 moved, so they were no longer receiving care at VA Connecticut. They were alive. They were just living in a different state. So that left us with 12 out of 20 that were still alive and still in Connecticut and still receiving care at VA Connecticut. Eleven of those 12 continued to come to the psych ER routinely. So one didn't, and the one who didn't was because he was living in a nursing home due to having a severe stroke. So he was still alive, still in Connecticut, but living in a nursing home and had a stroke. So of those who were alive in Connecticut and not confined to a nursing home, all of them continued to use psychiatric emergency room services. The majority of them had four or more visits last year, which the literature usually defines as a high utilizer. And now they had an average of seven chronic medical conditions as opposed to four, and now they're taking an average of 11 medications as opposed to the five that we saw earlier. So it was an interesting study. We just wanted to see sort of some of the outcomes of this group and sort of some of the conclusions are that, at least, again, in the VA group, overwhelmingly it's substance use disorder is the primary contributor to patients utilizing psychiatric emergency room services. Often, as you can look in the literature and see this is true across the board, a small percentage of patients often makes up a very large percentage of the cost or a very large percentage of the utilization of the health care service, and in our case that was very much true. An overwhelming majority of our patients had homelessness services, and they had legal involvement. So that is another thing to really pay attention to in this group. They all engaged in primary care and social work, and despite being very connected to the VA and very connected to a wide array of services already, they were still very high utilizers of the emergency room. So looking forward, how do you resolve this? It's unclear, right? If we would have shown that this group only comes to the ER and doesn't do anything else, okay, now let's get better wraparound services. But this group had very high services already but yet continued to utilize our services at very high rates. So there needs to be a little bit more research that goes into trying to figure out how we can prevent some of these high utilizers from coming in so frequently. And with that, I will stop there and turn it over. All right. So it's interesting when you're the last person to present and you don't know if people are still here, but hopefully you can hang in here as I go through the presentation. My name is Funchal Oladunjuye, and I am a PGY-2 psychiatry resident at Baylor College of Medicine. And I will be presenting on behalf of now she is interviewing a medical student who owns this paper, but one of the co-authors on the paper, so I'll try my best to present the case. I'm having troubles with the, okay. Looks like it's frozen or something. Yeah. Oh, my, and it's my turn, right? Okay. All right. Well, I assure you it's going to be a very short presentation, but here we're just trying to show the prevalence of substance use disorder in obstetrics patients, pediatric patients with eating disorders. And we don't have any disclosures here and no conflict of interest. So the objectives of this study is actually to show the prevalence of various types of substance use disorder in obstetrics pediatric patients with and without eating disorders, and also to show baseline characteristics of this patient in the inpatient service. So generally what we aim to do here is what I call more like a sensitization. It's not like in-depth analysis, but just a general overview, something to provoke us to take things to the next level. So we aim to study the prevalence of substance use in obstetrics patients with eating disorder in the United States. As we all are familiar with, most of the times the conversation has been around like mood disorders, comparing things like bipolar disorder, anxiety, depression, strongly linked with substance use disorders. But it's been in the last couple of years now, green literature showing that there's some association of eating disorders with substance use disorder. And what we did here was we went using what we call the KIT database. It's the KIT Inpatient Database 2016. We have up to like 2019, the latest, but it takes time before these things are updated. And as you can tell, it's kind of a large database. It's described as the largest inpatient or peer public pediatric database in the United States, and it's publicly available. You just have to pay a fee for a subscription, but then you have the opportunity to analyze the data, and you have, whether it's using data analysis or SPSS, there are different statistical softwares available to do your analysis. The database we have for the KIT database shows ages less than 21, so the analysis here will be kids less than 21 years of age, and it's nationally, it's a national database, typically representative of about 48 states in the country. In our study, the method basically was we analyzed looking at the diagnosis of eating disorders from all the ranges of eating disorders using the ICD code F50, and we used Chi-square to compare analysis looking at the demographic characteristics and all the associations related to eating disorders and substance use. And eating disorders patients were included anorexia nervosa, bulimia nervosa, binge eating disorder, and all the other types based on the ICD codes we drew from F50 IC codes, and we used the P value as a limit for 0.05, and we didn't need IRB approval because it's de-identified and publicly available, and we didn't have to go through, as we all know, the hassle of getting IRB approval. So in our results, what we found was that about 0.2 percent of patients in the 2016 database, totaling about 13,000, had eating disorders, and most of them were females, about 87.2 percent, and the mean age was 15.6. And in the prevalence, we found that many of our patients had, like, nicotine use disorder, comparing those who had this eating disorder with those who didn't have. There were higher prevalence of those who had eating disorders with 7.7 percent compared to 1.6 percent with those who did not have eating disorders. Another very high number came with those who had cannabis use disorder. They were, like, 10.8 percent compared to those who did not have eating disorders, which was, like, 1.4 percent. Also strikingly noticed here was also alcohol-related disorders, which was 4.8, compared to those who didn't have eating disorders, 0.4 percent. And this actually highlighted to us that, you know, generally across board, looking at some of the other substances, opioid use disorder, sedatives, amphetamine, and all that stuff, looks like across board they were higher in the folks who had eating disorders compared to those who did not have. And that, to us, shines a light to the need to begin to walk in the direction of finding out what's the association here, what's responsible for this. And so far, what we have is that we have a scarcity of data available highlighting this need. And just like I said earlier on, we have, you know, we've talked about some other associations. We've talked about mood disorder, related to eating disorder. But maybe it's now time to begin to, you know, talk about substance use and see what we can do to help this population. You would imagine, and because right now I'm on the CL service, you imagine how these kiddos get these drugs. I mean, from Instagram, they're more exposed to opportunities to be able to purchase and, you know, have access to these substances. And it's mind-blowing to see the rate at which these things are at the moment. So the conclusion we have here is that there's a higher prevalence of substance use disorder in patients with eating disorder. And the conclusion would be that we should probably start to screen patients with substance use disorder and see ways in which we can help. And with those who have positive screening, provide things like counseling and resources to help them. And sometimes we may probably think because they're little, they're naive and things like that. But they're also a very important part of our population, which we need to pay attention to. And, of course, doing this would require a multidisciplinary approach. Mental health providers, social workers, just the teamwork thing to get this population to the safe haven. Thank you for this, for listening to my very brief presentation. Thank you. I have one, and that's your VA program. So often, you know, you get the message that if you link people to the right services, it's going to help them and save money and all that kind of thing. But that didn't look like that necessarily happened with you. And I'm wondering what your theories are about what kept these people doing what they were doing. Yeah, it's a good question. You know, I sort of wish I knew. And I know these patients really well. I've seen, I mean, I've personally seen most of these patients now dozens and dozens of times. And they do have different problems. But most of them, it comes down to alcohol use disorder. That's what most of them, that's what it comes down to. And despite the programs, at least several of them have been committed to substance use treatment, like formally committed for long-term treatment. Despite all of our efforts in a variety of CTI programs, MECM programs, wraparound services, visiting them daily, despite that, they continue to drink and continue to have symptoms of mental health and continue to come to the ER. And I think that in out ‑‑ I mean, my take on it is in the absence of all of the stuff that the VA has to offer, because I'm in such a rich system, I think a lot of those, I think more of those patients would have passed away by now. And I think the fact that they continue to come in to any of the services is really the reason they're staying alive. And the fact that we can offer so much is one of those reasons. But why we can't break that cycle with the current resources we have, we're just unable to do it with a certain group. First of all, thank you all for excellent presentations. Each discussion was very enriching, and I appreciate that. I wanted to speak to you regarding this VA case. This may not apply at all, but I've run across a few people. There was one lady who's had no fewer than 500 crisis visits since 2008. And she ‑‑ I saw her last go‑around, and she was diagnosed with schizoaffective disorder, but it actually turned out when I actually sat down and talked with her, she had a substance use disorder, of course, but she also had Borrelian Purslane disorder, which is kind of overlooked. I'm making a gross generalization here, but I think there's a ‑‑ I'm not applying this to you whatsoever. I think there's sort of a hesitancy to look for personality disorders and factor them in. Some people feel it adds stigma, maybe. Maybe some people feel uncomfortable with it. I was just wondering what your thought is about the possibility of personality disorders factoring into some of these cases. Can you also say your name and where you're from? Oh, I apologize. David Weinstein, Bakersfield, California. Yeah, David, I agree. And, you know, we don't ‑‑ so at least in the emergence room, we're not typically assigning the personality disorder diagnosis, at least not in the context of the substance use and the multiple presentations, but there's certainly a subset of this population that that would apply to. And, in fact, there's one patient now who's coming in almost every day. He's not in this group because he's sort of new to coming in every day, and he would be on the group now if we looked five years later. He's very similar to one of the patients that is in this group who they don't have substance use disorders, but they have very, very poor coping skills. They both have a chronic psychotic disorder, very poor coping skills, and a complete inability to adapt to a home that they're living in now. And they basically come to the ER as comforting for them, and they'd almost rather, if you offered them to just live in the emergence room for the rest of their life, they would jump at that opportunity, and that would probably be the happiest day of their life, to be told they could never have to leave the ER. And there is a subset that that applies to, and I don't know if that's a dependency, if it's a borderline, but there's certainly a subset of this group that has a personality disorder diagnosis for sure. Thank you all for the help that you provide for patients. Yeah. Hey, Mary Ann Cole, used to be in Milwaukee VA, now Tulsa, Oklahoma. So I feel like we share the same patients. So Milwaukee had this thing where years ago was struggling with the same thing. They created a treatment alternative group. I called it the lost man group. We had usually one or two or three women on this group. But what it was, people would be allowed to come starting 7 a.m. They could leave 4 p.m. And interestingly enough, there was some programming. They had food, they had shelter, they could wash their clothes, whatnot, access other treatment services if they desired to do so. And over the years, there were some patients who collected about eight years of sobriety in the program and refused any other treatment. They were like, this is my group, this is where I'm going to stay. They didn't have a lot of filter, could be very loud group. But the only requirement was like they couldn't be too rambunctious, but they could even come if they were intoxicated. So this was like one solution to cutting down on a lot of the high utilization. Similarly, we had like community engagement from the EMS services. We really tried to engage them from the VA into the community and vice versa. I'm not sure if that really cut it down. I think the other part is that sometimes we forget that it could be a palliative service. There are some people where this is truly their home and their family, and so I'm not sure if we should always phrase it in such a negative light, even though it's treatment, like cost intensive, and you wonder where can you cut down on the cost, because they will come, which is a good thing. Yeah, and I agree. And what you're saying with it's a good thing, like I said earlier, I truly believe if it wasn't for this rich array of services, these patients would not have lived this long. And I tell some of them, if you weren't a veteran and couldn't come in all the time, there's no way you would have survived this long. And I truly believe that. So I think we don't worry so much about the cost. We're trying to figure out how can we get them sober and how can we get them living a more productive life. But the patients don't mind. These are the patients that want to be in the ER and want to be there. To them, it's not a problem. The problem is why are you kicking me out again? I'm just going to come back tomorrow. It's not a problem for them the way we see it as a high utilization. Yeah, Mike does. I'm VA Boston, so I'll share this similar dilemma. I think one thing that started the past five or six years at VA Boston is we have a high-risk, high-utilizer team of specific case management, and they carry anywhere up to 30 to 40 patients, and probably over half of them are very much what you described. They've been modifying what they're doing, and they're getting better trained in terms of motivational interviewing and engagement and connecting with our addiction services. We've, I think, made an inflection point of like 10%, 15% better, but when you actually start looking at the cost data that's saved even with that amount, when you cost ER visit versus emergency room versus inpatient, which is the common theme of what we're doing, it really adds up. So that's just a thought. But I do have a question for the panel because each of you are in, well, either training or trainees, and I'm just wondering what kind of QI projects and next steps you would take based on the data that you've presented. Yeah, in my case I would focus on more precise history taking because we don't have a standard urine drug test to detect all of hallucinogens. We can detect PCP and some of the others, but nowadays what I've observed is that a lot of people are using psilocybin, and there's no way to detect that with standard urine tests. So I would focus on QI projects that ask for this question specifically. So I am working on a couple of QI projects. One is sort of on prevention of this problem of pregnancy with an opioid use diagnosis. A lot of these patients have unintended pregnancies based on some of the chaos in their lives and their substance use, and the patients we currently treat are seen by a hospitalist-led team who frequently are not prescribing contraception. So I think a lot of the patients that we see who are capable of becoming pregnant are maybe seen in the hospital, maybe started on MOUD, but are not started on contraception. And so that's a project I'm working on, on educating the hospitalist-led team on the importance of that work and seeing if we can make a difference in that light. I'm also intending to do something about our Narcan prescribing because I think getting naloxone to these patients, although they could walk into a pharmacy and pick it up, I think if we can help them with that step, I think that that's an important piece as well. All right. So in my own case, we have kind of a lean inpatient service for children, and where we rotate currently, Texas Children's Hospital, the inpatient is not as robust as some other places. If anything, we're just starting in February next year a 10-bed situation. But on the CL service, we might have some more patients to kind of do a QI project on, and that's kind of where the direction is going with this. Of course, this is a large database. Some of the things we have talked about, like logistic regression, which we didn't go to in this analysis. I think there are some things we can explore to kind of, like, dig into some of the reasons and possibilities here, which I think is a huge thing. But just like I said, it's like sensitization in the beginning, and I think there are a lot of opportunities here, which we foresee will happen in the next couple of months. Hi. Hello. My name is Paige Marnell. I'm from Savannah, Georgia. I had a question about the peripartum talk. It was an excellent talk. Thank you so much. I was a bit struck by the fact that opioid overdose deaths are the second leading cause of deaths in the postpartum period, and I'm wondering about that population. Are these patients that were started on MOUD during pregnancy and then fell off after the postpartum period, or are these more likely to be patients that were never started on medication? Yeah, so the statistic I shared about it being the second leading cause of death in the postpartum period isn't from our work, so I cannot speak to what those causes or trends might be as far as MOUD. I can say that since submitting these slides, looking back, we have found that of the 23 patients who did have OUD, who we saw that three of them are now deceased, so that's about 13%. It's a small sample, so who's to say, but it's still a pretty shocking number for us. I haven't gone back and kind of drilled down to see what was going on with the treatment of these patients. In the induction without withdrawal talk earlier, that really has me thinking just about our data too, and we did see sort of as time passed, and there likely is more contamination of the supply with fentanyl and people using what they think is likely heroin or is heroin but is likely contaminated, we did see more self-directed discharges, elopements in the latter half of that period. And I can't help but think now as we're talking about these patients who are more likely to self-direct their discharges who have a lower cow score when we go to see them, do we even need to be letting these cow scores get higher before we see these patients? Can we make a difference for these patients by using some of the practices I learned about earlier? But I'm with you. I would also like to know how we can help prevent those deaths. Hello. Thank you all for your presentations. I came in a bit late. Yes, I'm Zola. I am a fifth year family medicine and psychiatry combined resident at UC Davis in Sacramento. Interested in addiction medicine, perinatal psychiatry, a lot of things. And so I have a couple of questions. Let me try to organize myself. I guess the first one for the peripartum MOUD discussion. And so this might be outside of the scope of your work, but in terms of the mean gestational age of the patients being around in the third trimester and then people presenting to the hospital either near labor, in labor, did you note, I guess, barriers to earlier detection and earlier use? Was this something that providers that presumably if they're delivering patients on labor and delivery, they're also doing perinatal care throughout that? Is there an education from that side of things to get people to do more screening? Or are there different barriers in terms of not wanting to get CPS involvement or things of that nature that might have delayed treatment? Yeah, that's a really good question, and I don't know if I have the answer for that at this time. I think that that's something that we would benefit from looking into and trying to find out why these patients aren't connected to care earlier, as you know. The sooner is usually better, especially with the growing baby. And next question is about the life stages of these patients with eating disorders and substance use disorder. I don't know if there was a certain length of time that the patients had qualified for having a substance use disorder, and would you see any utility to noting any family history or either past or current use of substance use contributing to a fairly young age of developing a substance use disorder? Yeah, I think that's a good question. So with this data, for instance, it's a cross-sectional study, and with the available data points we have there, there is kind of no way for us to know how long they've had this disorder or how early it was. I mean, the only thing is, because it's an administrative database, we have maybe they were 15 when they were diagnosed, and it's dependent on the provider at the time the diagnosis were made. But I think it's a very interesting question because one of the things we're profiling in the CL service, for instance, kind of like knowing how long they've been having this disorder can actually help prognosis and outcomes, whether it's like worst outcomes or positive outcomes. So I think it's something we can explore. But then the dynamics and logistics of getting this information to is also going to be like very, you know, it's going to be kind of like long and prolonged and very tedious. But I think it's something we can look into when doing some QI projects.

QI projects

substance use disorder

obstetric pediatric patients

eating disorders

history-taking

high utilizers

psychiatric emergency room services

care and support

peripartum management

opioid use disorder