This session explores the impact of providing housing with wrap-around case management services to unhoused patients with a high-utilization emergency department services, with a history of mental health and/or substance use disorders.
All right, um, before I bring up our speaker, I'd like to introduce Doctor Jatty, MD. He's a professor of emergency medicine at Washington University in Saint Louis. He also serves as a physician adviser to Barnes Jews Hospital and a physician consultant to BJC Healthcare, excuse me, to BJC Behavioral Health and has served as the president of the Barnes Jewish Hospital Medical Staff Association. As a member of the Barnes Shears Hospital Board of Trustees and the Washington University Faculty Practice Plan Board of Trustees. He's a native of Saint Louis and I found out, and he lives in Webster Groves. Please join me in welcoming Doctor Gotti. Thank you very much and it's, it's a pleasure and honor to be here, especially with such organized people. I want you all to organize my life and I can, I can give you the schedule. So we're gonna talk about the hospital housing program that I've been involved with for, um, several years now, actually since 2019. We have about 30, 31, 30 slides, 30 minutes. I'm gonna try to do that less. Some of the slides are pretty basic, uh, pretty self-evident. Some of them have a whole lot on them. We're not gonna go into a whole lot, but, uh, it's available there and there's some of you, some of these things if you wanna ask questions, I'm, I have, I have some contact information you can reach out to me. So let's just, let's let's just start. First off, disclosures. I have no financial disclosures, extensive attributive disclosures. I am one part of this program, one small part, even this presentation, this presentation which unfolds, others helped put this together and um so I do the medical part. I'm an emergency physician, but I. I know the patients inside and out. I know the process of what we do with them, especially clinically. Cencia Mitchell is a, a senior manager of community health initiatives of BJC. She helps run the pilot, the program is beyond the pilot. We've been going on for years now, but she manages all that. And one of these sets of the presentation are things that she's done, so you know that. And then Matt Reedhead, he is our data guy. He's uh vice president of research and analytics of the Missouri Hospital Association. The third part of the presentation is a lot of what he said, say, well, what does this have to apply beyond these patients in the emergency department of Barnes, whether you go to Saint Louis or if you go to all of Missouri. In terms of objectives, objectives today are to recognize these pathways. We have to identify groups with unique needs among hospital housing clients. What did we do? What were the interventions that improved their care? And how do we measure it, especially how do you measure it so that you find out if it's worth the effort? Uh, tomorrow, and technically I already have a bad slide because tomorrow is Saturday, so I wanna say Monday. So Monday we want you to apply this, you're gonna have the same cohorts of patients in your groups. The in in the areas you provide care trying to figure out how do we help this subset of people, what do we do and how do we find out if we're being impactful in their lives as well as in the larger health system. So this is the presentation framework. This is me when in terms I know in depth what we did, picking the patients, uh, who's eligible, how do we enroll them. Clintony Mitchell, and I know this part well, how we did it, but I have her slides here. I wanna attribute that that in terms of the structure, the partners we have, the roles, and then you'll see that it's not a huge cohort of patients. There's a lot of interventions that happen. We have about 80 that we've actually had in the program. Um, Matt's approach is to, are there others like this in Saint Louis? Are there others like this beyond? What would it be like in terms of we expanded this? So let's jump right into my part, what I know deeply, the program. So we've been in place since 2019, May, just pre-COVID, uh, went through COVID pretty well. We've had about, we've had 80 patients that have been enrolled, pretty strict enrollment criteria, making terms sure that they've been in the program and what their eligibility is. So they have to be adults. At least 10 visits to the emergency department in the last 3 months. Um, it can be a rolling average, but we want our highest utilizer, and we also want a clear diagnosis of a mental health disorder as documented by a psychiatrists, or a positive urine drug screen or serum ethanol over 80. So that has to be documented. You have to know it's that's that's one of their challenges because these are some of the people, the hardest ones to assist in the larger setting, and they have to be unhoused and that's usually uh there's a chart reported that with social workers are sending them to a shelter. I'm here. I need food. I need some time, but then we actually approach them and confirm that they tend to be male, uh, males about age 50, Medicaid or uncovered. And the things we do are, it's, it's we call it relentless case management. We will call you um if you don't answer the phone we're gonna knock on your door if you don't open your door we have a key. It's not a landlord tenant relationship it's a programmatic component so we you've got to be a participant in the program, yeah, and, and in the processes that are implemented. And we give housing sometimes that um same day, sometimes it can be months um depending on if they if we have uh what our availability is. And then we follow, particularly clinical utilization. There's other things that we follow, but clinical utilization fiscal impact is what I'm gonna present here. And how do we identify these patients? Um, Epic. Epic reports. And does anybody here use Epic reports for some things, some of you are familiar with that you can ask. So, I asked, who are the highest utilizers in the next period of time. Initially just the BJH. Now throughout all BJC, BJH is still the, the principal site. And then I look in the chart to make sure they have those psychiatric consults, uh, that with a diagnosis or positive, uh, lab studies. And at that point, next time they're their social work will, uh, speak with them in the emergency department next time they triage in and tell them about the program they're eligible, uh, confirm that they're unhoused or homeless, and do you, would you be interested, uh, maybe about 50% say yes, 50%, you know, we're it's, it's surprising not everybody trusts the program initially. It takes a long time to get to that point. They say yes, they sign a consent so then we can share their health, personal health information. Technically, to, um, our community partners, uh, Saint Patrick's Center or Live Sober Living, they get placed start case management, and then we start following how is it all going. So this is basically, um, this is just an epic drawdown, uh, these dark blue lines are the, um. I guess, are, are the ED visits, emergency department visits one year prior to the program, so the prior to when they got their key, um, when they had their could open their door and they have a place to stay. The, I guess those are like, I don't know what your color is. I'm gonna say gray. My wife would say it's, I, it's not, I don't know what we have a hard time with colors. Um, it's different. This is the one year after and it's not the whole year they haven't necessarily had a whole year, uh, with us. Sometimes things happen, but we just say a year after, has it changed them. And this is BJH. and then we say BJC too. Are we, are they going to Christian Hospital? They're going to MBAP elsewhere? And then we actually look regionally. Are they going to SSM, Mercy, things like that. And, um, so the these are number of visits 36 to 19 to 20. So the ED visits at Barnes dropped 45% after enrollment in the program being housed, 45% at BJC, and 41% regionally. And then we also have a cohort that's not housed those who said no that were fully eligible and they weren't interested or they're interested but they didn't uh show up at Saint Patrick's Center or live or they showed up but it's, you know, we have a limited um inventory of apartments, um, so it just depends if they go there when it's available so we follow them as well and these in terms of their ED visits. They increased one year before versus one year after and that turning point was when we, since they didn't get a key it was when we offered the program and they said yes or no but nevertheless that opportunity point so it rose 13%, 20%, and 25% if they were not housed so very different outcomes we looked at inpatient days. So inpatient days at BJH dropped 42, 43%, 47% at BJC, and 47% regionally. For the not housed group, they rose 6.7, 7.6% for BJC. Regionally they dropped 6.7. I'm not really sure why, but a little different on that. And this is just one individual, um, just to reflect that, you know, this is what happens. I mean, uh, this starts February 23. Each of these is a number of ED visits per month 123, 11, 59, 14. You can see that in the winter months of November 23, December, they really go up. February 5th, he was referred of 24. February 8th, we actually got housed, and then you see 0 thereafter. And this in July, he went into a permanent supportive housing with Saint Patrick's. So that's just. As we have the, we sit our, you know, we didn't take it personally. We have these turkey sandwiches and all that. They're not there for that really after all, we always joke about it, but that's not why they're there. And this also is one of the things that drives me. So as an emergency physician down there we work with all the uh tran the services that say we wanna transfer a patient in from Springfield, Missouri, Springfield, Illinois, Western Ohio, uh, BJ West County, things like that, and this is just in that time period. This is the morning of August 14th, 2024 in the middle of that period. These are all the locations that want to send them to us. Some of them are here in our, you know, metropolitan region, variable size hospitals. Many of them are further out like, uh, Joplin or, you know, actually in Kansas, um, whatever, uh, Iowa, all over Illinois, and these are the kind of things they have in these small hospitals, things like a blast crisis with new leukemia, uh, ventricular failure, ventricular tachycardia. They had a C-spine fusion, the wounds open, or they have a, um, a liver transplant, they're not going to make it. You know that. We don't have space to put them here necessarily, but we know with these kind of diagnoses, you're probably not going to make it. So, trying to be efficient and saying if we can get these people that are here for turkey sandwiches, get there somewhere else like that one picture we have where they just stopped coming. It allows us, you know, for our extensive resources we have here to apply them to them. So how did we do it? This is, um, and I know this well, but uh, Clancy, I wanna uh attribute this to her. She's the one that, uh, really helped see the program. Puts it in place and I'm just gonna go on these in the big scale. So this is the hospital staffing support structure complex care social workers, BJH Finance was uh Mark Krieger, Paula Rovik. They've been Madeleine Thieler, Madeleine Skobe, and Ruth. They, we've been, they've been fantastic. BJH has been BJC emergency room physicians and Community health support data analysts in terms of within Epic drawing down what are we doing? We also have substantial support in our community partners. We have case managers out there that will go in and they don't open that door and they'll open it and they'll find out what's going on. We've had them ask to evict, uh, drug dealers from their apartments, things like that which I don't do, I hope, but. Um, all kinds of requests. Our landlords have been really good. We're good, and, and actually Saint Patrick's Center manages the, uh, the leases, but we're good tenants. If there's damage, you know, it's gonna be paid. There's, you know, there's never gonna be a late rent, things like that, but they're very tolerant sometimes when you have someone who has schizophrenia, and they're gonna need a 96 hour hold in a day or two because they're just not. Um, they're not, uh, on their medications. The staff, peer support, just, you know, great assistance with many broad parameters, and this is what those apartments look like. This is McFarland Place. It used to be a school, I think an old school down here, but you can see pretty utilitarian. There's a, the, uh, stove, the kitchen, you know, the refrigerator, simple table, one chair, just, but it works. It's very different from the streets. Um, this is live sober living, much more structured. Uh, they have classes. It's like they have, uh, 4 or 6 people per, um, per room with like an RA. It's sort of like the college experience with like a lot of rule, more, more restrictions in college. You can't drink. You can't, yeah, so not that, you know. And then, uh, uh, Clancy, you know, the BJ HBJC, they manage all these contracts like with the Behavioral Health Network. They had a contract for me for some of my FTE time when I was, uh, when we had a grant from SAMSA, um, Saint Patrick's Center, Live Sober Living, all of that, um. The things that we followed financially, and we're going to start going into a little bit of the finances are the program operating costs, variable costs which are different per patient depending on what you do, and then fixed costs, which are the costs of just having the building open and public safety and all this stuff, having the stretchers in the room waiting for patients, things like that, and we follow the program financial impact. So this is just BJC, so it's a continuous positive impact. So it took a 90 cost savings. So these are patients that they don't reimburse that well when they come here. Some of them are completely uninsured and we provide care for them. So what is it like for them to be patients on our campus before the program or after the program financially? So we actually saved almost a quarter million dollars in 90 day cost savings for these patients. And then the return on investment, so. I'm not a finance person, so like if you, we, we have, I, I'm not a, we don't have Capital One, so we have, but I just pulled that up. So Capital One, your return on investment will be 3.4%. So if you give them $100 right now, a year from now you get $103.40 for this program. If BJC in terms of cost savings, if BJC invests $100 a year from now they'll have $203. So it's very. Substantial in terms of what it does. It does not include, I'm a WashU employee. It doesn't include the impact on WashU. It doesn't include the impact of SSM, Mercy, RMO Health Net, or Centene or who'sever paying the bills. And then this is now what we're gonna do. So we like, we, it's not bad what we're seeing. We've seen what it's done with BJH in our house, on our, you know, in our system. What about other places do we know? How does it impact things like UC or SSM or others and Mo Health Net and everybody is it, and are there more people like this out there? So we actually, Matt, this is what Matt Reed had, who's that data guy out there, so. This is a little bit, you know, the whole huge AI, and I, it's not AI, it's these huge data sets because Matt directed this. So how would I compare, so Matt, he works for the Hospital Industry Data Institute, which is part of Missouri Hospital Association, but it's an independent. So they're they're called a neutral conveyor of data. So let's say I'm a patient and I go to um BJ West County because I break my ankle. This is where my data is stored out there in West County. So BJ West County will de-identify me, send it out there, and they've got it, but it's de-identified. And then I go to SSM for um pneumonia. This is their category over here. They know Randy Jadi went to SSM for pneumonia in December of whatever, and they have a uh uh just they have me de-identified, but I'm there and then say if I went to Mercy for um. Uh, I don't know, behavioral health. I'm stressed. I need, you know, I need to talk with people, so that's over there, but they, all this stuff gets sent there. They compile it all together and they said this is Randy Johnty. This is what was spent on him. His number is 10745389. No one knows that, but they, you can compile all that together and that's how things like in the middle of COVID, you say we're, we're starting to see a lot of people with. You know, certain conditions going on or Canada RS or things you get a sense public health wise without everybody knowing who those patients are so Matt does that that's and it's an incredible. So first off, where do these patients go? They just come to Barnes? No, these, uh, these H2H patients, they go, they get around. This is Las Vegas. This is up, um, Minnesota, uh, Rhode Island, uh, a lot of Missouri. They mostly come to BJC, like 60%, 54% to Barnes, and uh we're gonna go a little bit greater detail here. Um, this is where they come, like BJH, uh, and this is for the whole time period. Matt just looked to say, where are these patients going 2018 to 2024. 2900 visits to BJH, um, they go to Slough a lot as you can imagine as well. Um, they also we got DePaul here is a pretty heavy hit, uh, Mercy South Saint Clair, they, they go around quite a few places, um. And then this is, this is a slide that probably reflects what is the impact. So what Matt did was this is the day, this is the week before anybody was housed. This is and this is our clinical utilization. The, I'll say that's black, that's, um, uh, other hospitals, the SSM Mercy everywhere else. The blue is BJC. So this is their utilization the week before and the utilization the week after in terms of visits. Uh, 3,413 visits pre-placement and 1700 visits post placement. They just stopped going everywhere. They don't, they don't, you know, they don't need to be there. Um, and then cost, um, $4 million in total cost beforehand, and $2.3 million total cost afterwards. Per client 59,000 per client beforehand, 32,000 per client afterhand. So it's just very inefficient utilization of funds beforehand and we get a lot better at that. So the question that comes up is, so these are 80 folks, there's certainly another 80 out there. How many are there out there like this? Is it just this one individual there? Is that the only one we found them by chance? And there's no one else in the community of Saint Louis, Missouri, the nation that needs us or not, so. So, um, there's a means of finding out how many people are just like this, and you're you'd be amazed, I think you get a sense of how much data about about us is out there when you look at this. So it's called complexity cohort complexity profiles. So, all the stuff that's known about patients, that's been de-identified, there'll be, and I'm gonna show you some of these. There's social factors, behavioral health factors, physical health factors, like, you know, congestive heart failure, or whether you have COPD or diabetes, hemoglobin A1C, risk behaviors, and I'll show you what those are too. And then we say, well, is it different for hospital housing clients versus other unhoused patients and everyone else? Do they go into different baskets and are they identifiable as unique? And they are. So I'm gonna step back just so I can see this best and I apologize for everybody you can see. So this is, um, the dark blue is hospital housing and we know those patients deeply because I shared, you know, we're able to share with them the, the patients they can say who are these individuals, what makes them unique. And then the, um, the orange or other unhoused, other Saint Louis unhoused, there's something called Z codes. So, there is a very reliable ICD 10 diagnosis that goes Z codes if you are really not, and it's pretty reliable. Um, it's been around for, since whenever it went in. 2015 or something like that identifiable whether they're unhoused um and then everybody else so any behavioral health diagnosis 95.6% of them had that um or are present in our hospital housing we have some who are strictly substance use disorder, about 75% are present in um the other end house 28% in the general community. And then let's get a little bit deeper into this, um, schizoaffective disorder, 76% in the hospital housing, only 2.3%, so 35 or whatever more than that times more likely in the hospital housing, and then um double what it is in the other house. And uh personality disorders, we see a lot of people that have that and uh the difference is um over about nearly 6050 to 60 times more likely in the hospital housing. Than in the RXU 50 times more in the hospital housing than the general population, and still 14 times more likely in the unhoused population to have a personality disorder. How about other things? These are risk behaviors. So these are big ones. So, in terms of, and what's the finest risk behavior? Well, it's like substance use disorder, tobacco use. It's a thing I, I'm not sure why it's called risk behavior. It's maybe something that isn't gonna happen to you necessarily, I don't know, genetically or is it just some to some point, it's, um. It's a behavior, you know what it is. But malingering, which, as Matt said, is, is kind of an offensive term, but it's applied. People get that diagnosis. What you're telling me, you're here for chest pain, but you're here for chest pain yesterday and the day before and the day before. You're here because you, you know, you're not really here for clinic, uh, clinical chest pain needs, and people get that diagnosis malingering. I'm, I'm quite confident nobody at this conference has that diagnosis, but there are those who have a, um, so it's very rare in the general population, 0.2%, 80% of our hospital housing patients have it, so that's an 800% increased likelihood to have it. In the, in the other unhoused, it's still, what's that that's over 20 times more likely or what's whatever the math is, um, 5 times 50 times more likely in the uh in the other unhoused population. Restraint use is also out there as one of those little data draws if you've, if someone, I, I don't know, I guess I could say I've been restrained, but I, I should have been restrained probably at times. But um you can see much more likely in the hospital housing as well as in the other unhoused, you know, so just in terms of things, who are these folks. So what it is, is when you're trying to figure out, are these patients out there. So I guess uh, if we look at the big population of Saint Louis, it's, you know, apples and oranges. Most of Saint Louis, most of Missouri, most of the nation is apples. Our hospital housing patients are oranges, and we're trying to say how many oranges are in this big eighteen-wheeler bed of fruit. And how do we identify them? Well, it turns out by things, looking at things like schizoaffective disorder, restraint use, malingering diagnoses, other things like that, you can know exactly what an orange looks like, and exactly what an apple looks like for this population. And that's it right here. So this is something. It's, uh, let's see how much I've got a couple minutes. I, I just give myself an extra 5. So we're close to the end. So this is like if it, this, oops, I'm gonna go back. So it's called, it's or actually receiver, so I'm not statistics. This is Matt, but what you're looking at is is saying is this a flip of the coin or is this in terms of the data you're looking at. So it's called a receiver operating characteristic curve. So if it's a flip of the coin, it's this line. The higher up you hug over here, the better it is. Anything over 90% is considered to be excellent. So this is 97.6%. So Matt said he's never seen anything like this. And what you're looking at is trying to define those apples versus oranges. How well have you defined an orange? Is there a chance you're slipping an apple in there? And it's things like malingering. It just doesn't happen. The P value is less than, you know, you want it to be less than 0.05, so it's less than 0.0001. It's just people don't get that diagnosis unless they have these other challenges in their life, and they need to get out of the cold, and they need to be, they need that turkey sandwich or whatever, otherwise they're going to go hungry. So it's, it's unique. So, and these are just, you know, the, the differences are so this says that there's his Matt's data is 32,000 other Saint Louis unhoused individuals out of this is a database of 4.5 million. These that's how many individual. Patients he has in the Heidi system and her 70 at that time in the hospital housing on 140 versus 13 versus 3 ED visits, um, this the middle one I'm saying is the other unhoused individuals, uh, inpatient days, uh, or, um, inpatient hospitalization 14 versus 5 versus 1 and about 200,000 versus 72,000 versus 15,000. So, and there's certainly in this group here, there's some who are from here and we've looked at that. And then this is just, and, and they come in for things uh for psychiatric admissions primarily. 31% versus 2% in the house, uh, suicidal ideation, 30% versus 2%, and things like substance use disorders. So they, you, you can, we know who they are and why they're there. And this, this is my last slide, I'm, and I'm gonna, we will have time for the, um, uh, video or just a few minute video. So Matt looked at this and he said out of these other oranges in the, in the, uh, in the tractor trailer, how many, let's say how much you were spending on them before and after type of comparison and the savings again, um. Not knowing how much we're saving, how much we're spending on them and the per patient per year in the community, saying they're better and they're going to Affinia and they're going to Preferred Family, you know, just somewhere we're not at Barnes Hospital, variable amounts there, uh, return on investment is anywhere from $1.59 for every $1 put in up to 326 cents. That applies to the state of Missouri. So this is what and actually finances is not was not part of our initial model in May of 2019, but in the middle of COVID someone said, hey, these people can, you know, we met some people from our finance you wanna follow this program and see what it does and this is really what's BJH Foundation's been fantastic. John Lynch Barnes have been fantastic in keeping this going. But the video that follows is the reason why we, you know, drives us to do it, and I'm just gonna let it speak for itself. We are here really in the heart of North Saint Louis around one of the homeless encampments. What you see here are just some of the living conditions that people face. I was outside. I mean, I mean, literally outside sleeping in the rain, sleeping in parks, sleeping in airports, but during the course of being homeless, I wind up in hospitals. I lived in the emergency room. If it wasn't for Barnes, if it weren't for Barnes Hospital, I don't know where I'd be right now. The, the problem that we first observed was that we had our familiar faces, our, our patients that were using the, the emergency room to solve their daily issues, and we know that that it was not good for their long term health. We could only put a band-aid on it. And then expect that they would return. It was also not good for an overcrowded emergency room, which many urban emergency rooms are experiencing. In my role as a as a physician, as an emergency physician, pretty much everybody throughout the greater Saint Louis region, the Midwest, they're potentially patients, and I always want to be prepared to understand their needs, to be able to address their needs and help them. And the hospital housing program is giving me insight how to help those in sometimes the furthest margins of society and pull them back into the fold a little bit. We really started hospital to housing 5 years ago to really help people just like this who needed to be connected to housing and to other services. Partnership like this is really new for the Saint Louis area where we identified individuals who were using the emergency department for not. Emergency medical care, but they were there because they had no place to go. They need a social service type care. To take somebody from that setting and putting them into housing is so important. And every single person we move out of homelessness into permanent housing is a huge win for everybody. That's so great. When you start to get success with our community, other community partners step up and want to be part of that success story as well. Another key partner that has stepped up is live sober living. And they help us with a special population of patients that may be dealing with substance abuse disorder as well as being unhoused, as well as maybe dealing with other psychosocial issues. Well, I was having what I would call a nervous breakdown. There's something in that night that I was somehow I was led to Barnes Hospital by the grace of God and they're good people over at Barnes. And they saw me and and and held me for a while and they fed me and they, they made arrangements to bring me here to meet, to meet you guys. When I think about the work that you do, it moved me so much. I'm so happy that your organization is part of the hospital, the housing program, getting them access to services, you guys recognizing that. The hospitals can't always serve the substance use and mental health population. And how do you feel like that fight for your recovery is going the last week since you've been here? Oh, it's been great. I mean, these past few days has really been good. I mean, I get to get out and. I'm actually having a conversation with people. The concern you guys bring to the table and everything and the help that you guys bring is really amazing. The patients in the hospital housing program, they don't need to come to the hospital for their non-clinical needs. So consequently that frees up space. They're off our clinical grid. They're well, so it's, it's very inspiring to know that there's, there's an opportunity for everybody and you just, you don't give up. It's clear that by reducing utilization by 50% in some cases. The patient benefited, but the system of care also benefited patients that you guys are referring rather than going back to the ED, they're calling us. I could see where a reduction is happening because instead of, you know, oh, I gotta go to ED, that's, you know, because that some people get caught up in their norm. So once you become homeless, that's a pattern within itself. Within, within your 1st 30 to 90 days, you don't know how to live. You don't know how to do anything. Well, this program teaching me all that all over again. I mean being able to shower, being able to cook, being able to stick the key in the door and come in as you want when, when you get ready, those are things that are important. They're training me on how to live again. These are people and these are real lives that we're able to rebuild. We are actually shifting the narrative and supporting programs that really have a meaningful impact to do community good. And when we succeed together, we build trust. We fulfill each other's missions and we help our community improve their health status, but the most important part, it helps someone like Perry get from the hospital. In the housing and beyond. Right. Yeah.
