Hello and welcome. I'm Dr. Benjamin Druss, Professor and Rosalynn Carter Chair in Mental Health at the Rollins School of Public Health at Emory University and Health Systems Expert for SMI Advisor. I'm pleased that you are joining us today for the SMI Advisor webinar, Expanding the Evidence Base for the Crisis Care Continuum, Call Centers, Mobile Teams, and Stabilization Units. SMI Advisor, also known as the Clinical Support System for Serious Mental Illness, is an APA and SAMHSA initiative devoted to helping clinicians implement evidence-based care for those living with serious mental illness. Working with experts from across the SMI clinician community, our interdisciplinary effort has been designed to help you get the answers you need to care for your patients. Today's webinar has been designated for AMA PRA Category 1 Credit for Physicians, 1 Continuing Education Credit for Psychologists, 1 Continuing Education Credit for Social Workers, and 1 Nursing Continuing Professional Development Contact Hour. Credit for participating in today's webinar will be available until December 18, 2021. Slides from the presentation today are available in the handouts area found in the lower portion of your control panel. Select the link to download the PDF. Please feel free to submit your questions throughout the presentation by typing them into the question area found in the lower portion of your control panel. We'll reserve 10 to 15 minutes at the end of presentation for Q&A. Now, I'm very happy to introduce you to the faculty for today's webinar, Dr. Matthew Goldman. Dr. Matthew Goldman is the Medical Director for Comprehensive Crisis Services in the San Francisco Department of Public Health and is a Clinical Assistant Professor in the UCSF Department of Psychiatry and Behavioral Sciences. Dr. Goldman's research regards best practices in mobile crisis team settings for suicide prevention. Dr. Goldman, thank you so much for leading today's webinar. Well, thank you so much, Dr. Druss, for the kind introduction and for the opportunity to join you all today. I have no disclosures or conflicts of interest for the material today. And so as we're meeting today, I've got a few learning objectives that hopefully will cover. So upon completion of this session, participants should be able to articulate the current state of knowledge on what constitutes high quality and equitable mental health crisis services, to evaluate the unique challenges and opportunities of crisis services using implementation science principles and select research methods, and to advocate for meaningful data collection and program evaluation within mental health systems. So I'm going to start off today by framing why are we talking about crisis services and why are we talking about research and evaluation in particular in crisis services? And the broad message is not necessarily the most optimistic one. It's that in many ways, our behavioral health systems are failing. There are year over year increasing suicide and overdose deaths. I'm sure many of you saw the headlines and many of the major news outlets just in the last couple days highlighting the hundred thousand people who died from overdose, the vast majority of whom died from opioid overdoses in the year following the beginning of the COVID pandemic. And suicide rates continue to increase as well. Extensive criminal legal carceral involvement of people with mental illness. People with mental illness are highly overrepresented in criminal legal settings. They tend to spend much longer in jail compared to people who don't have mental illness, and they often do not have access to treatment when they're in those settings. And so this is a major problem. Inadequate access and capacity, and that's at every level of our system. Difficulty accessing outpatient treatment, difficulty getting into inpatient, which results in things like emergency department boarding, which is the last point raising here, which is particularly relevant for crisis services, because where there is an absence of crisis services, often people end up in a medical emergency room. And studies have shown that people who are in an emergency room with mental health presentation spend much longer waiting for evaluation and then waiting for ultimate referral to care than others who are there for non-psychiatric presentations. And so these are all major issues that are straining our systems across the board. And at the center there, I've got disparities, because disparities are really important factor across all of these issues. We know that it just, again, since COVID started, there was a study out of Maryland that showed that the rate of suicide deaths among the black population had doubled, whereas in the white population it had decreased. I think criminal legal carceral involvement of people of color goes without saying, but there have been additional studies that show that people of color who are in carceral settings have less access to mental health treatment compared to others who are in those same types of environments. Inadequate access and capacity, there again have been studies that have shown that black, Latinx, and Asian populations have much less access to outpatient services than the general population. And emergency department boarding, also there have been studies that have shown that people of color wait longer in emergency rooms when they have psychiatric presentations as compared to other populations. So there are disparities across the board. These issues are huge for the entire population, and they are particularly problematic, especially for populations of color. I added here, plus COVID, because of course COVID complicates everything. All of these issues are strained with the presence of COVID. And of course, this is something that we're still all facing every day. And so seeing all of this, so wow, a lot of challenges that we're facing, and all of them intersect with mental health crisis services in some ways. The conclusion is that, well, we need better. And part of the argument for that is that we need parity in mental health crisis services, the same as for general health emergencies. Mental health conditions should be treated no different from general health conditions. And there's a clear justification for having high quality mental health crisis system response be available the same way that there are ambulances and emergency rooms for every person who's having a heart attack in the country. So with that, I'm going to spend today talking about the mental health crisis continuum, which includes a few different components that I'll get into describing. Before I get into that specific components, though, I want to give a little of the history of crisis services in the United States, because we're not starting from scratch here. We're building on quite a few years of history, and that history is relevant. So in 1968, that's when the first 911 call was placed. 911 was established because prior to 911, there was basically a patchwork of different emergency numbers in each different region in the country. And if you wanted to call the police in an emergency, you would have to look at your phone book and see what the number was there. So there's a long history of 911 implementation. I will say it took decades to get to the point that we're at today, where 911 is truly universal and nationwide. And even then, there are still actually a couple patches in the country where 911 access is insufficient. In 1987, that's when CIT, crisis intervention teams, were initially implemented. That was called the Memphis model. I'll talk a little more about that later. And these were basically mental health crisis training for law enforcement officers. In 2005 was the beginning of the National Suicide Prevention Lifeline. This is a network of call centers that are supported. Well, the network is supported by SAMHSA, the Substance Abuse Mental Health Services Administration. And the specific call centers, though, are each independent, small nonprofit organizations. But the Lifeline, as it initiated in 2005, was the first time that there was a real sort of standards of care for suicide hotlines. And that was a major moment for crisis services and also was the origin of some helpful research on call centers that I'll get into in a little bit. In 2011, the Zero Suicide Initiative was founded, which is basically describing best practices in suicide prevention. It described things like identifying people who are at risk of suicide through universal screening, engaging those people, helping treat them with evidence-based interventions to reduce suicide risk, and then transitioning them to appropriate care settings. The challenge that the folks who are involved in zero suicide came up against, though, was, okay, we have this great idea of top-to-bottom suicide prevention throughout a whole institution, but who actually responds when somebody is in a crisis? There was not a clear and robust response immediately available for a lot of people. And so on the heels of that work, where zero suicide was informed in many ways by the National Action Alliance for Suicide Prevention, that same group went on to create something called Crisis Now. And that's what really created this concept of having three levels of crisis service response. Call centers, mobile crisis teams, and crisis facilities. And you're going to hear me talk about that a lot in the coming slides. In 2019, the Federal Communications Commission made a ruling that they were going to designate a three-digit phone number to contact the National Suicide Prevention Lifeline. So, whereas right now, the Suicide Prevention Lifeline, NSPL, I'll refer to it sometimes by its acronym. NSPL is currently accessed by dialing 1-800-273-TALK, often a phone number that you might see at the end of a news article that mentions something about suicide. What the FCC decided was that, well, rather than 1-800-273-TALK, which is a 1-800 number, harder to remember, we need a 9-1-1 equivalent for mental health. So what some people have referred to as a 9-1-1 for the brain. And that's what 988 is. So 988 is going to be that three-digit phone number that's going to be accessible nationally for people who are in crisis or who are in need of mental health assistance in an urgent manner. In 2020, and so it was 2019 when FCC decided that. And it's in 2022, July of next year, when 988 is going to go live. A couple other things just to make note of here, and this should be available in the slides. In 2020, SAMHSA released its best practices toolkit for crisis services that in many ways was highly reflective of the Crisis Now framework of framing crisis services as a need for call centers, mobile teams, and also crisis facilities and having high functioning linkages between those three different settings. And then in 2021, the National Council for Mental Well-Being, formerly known as the National Council for Behavioral Health, put out a report that was authored by the Group for the Advancement of Psychiatry, which is a sort of top-to-bottom description of the ideal roadmap to crisis care, where it really gets into a lot of the broad systems, issues, and opportunities for systems to really expand and grow their crisis systems. And so I've mentioned these three different services, and this schematic I find extremely helpful for framing kind of what is the vision, what's the idea of how crisis services can work. And this slide, I have to give credit where credit's due. Margie Balfour, who's a psychiatrist based in Arizona, made this slide. I think of this as the Mona Lisa of crisis services slides because it's so good and clear the way that she framed this. Basically, moving from left to right, the idea of a crisis system, and this is a system that's functioning in Arizona. So they really have, they're national leaders in Arizona in having this system in place. The idea is basically there's a person in crisis. That person can call a crisis line. About 80% of those crises might be resolved on the phone. If needed, 20% of those calls need an in-person response, and so a mobile crisis team could respond to those, ideally dispatched by the crisis line. About 70% of those mobile crisis visits can be resolved in the field so that they don't have to go to an emergency room or any other setting, although there might be about 30% of people who are seen by mobile crisis who do then need to go onto a crisis facility. Then when people are seen in crisis facilities rather than in emergency departments or jail, they're able to have whatever their needs are resolved, and about 70% get discharged to the community without needing any kind of higher level of care. Then importantly, something that Margie pointed out in the Arizona system here is that they also have some capacity for post-crisis wraparound services. What they found is that, at least in the programs in Arizona, about 85% of people remain stable in community-based care. Also, you see some of those logos here pointing at the different services. What that's meant to indicate is that 911 can ideally transfer mental health-related calls to the crisis lines. Similarly, police or law enforcement who might encounter somebody who's in a mental health crisis would ideally be able to request a mobile crisis team to come and either co-respond or take over a crisis call. Also, police, if they do pick somebody up who's in a mental health crisis, should have direct access to be able to drop somebody off at a crisis facility. Those are really important coordination points for these systems. What that close collaboration with law enforcement allows for, of course, is pre-arrest diversion. What the system overall allows for is decreased use of jail, emergency department, and inpatient utilization, which is very much meant to address some of those key issues that I talked about at the top of this presentation. That big arrow at the bottom that you see is really meant to indicate that not only are these services less costly, as you're just talking on the phone or potentially even just having a mobile team visit, they're also less restrictive. This is one of the things that I love most about working in crisis services. My program in particular that I'm medical director for is a mobile crisis program. Getting to see people in their own home environments, it's so much more person-centered. It's so much more trauma-informed. It's so much more recovery-oriented. All of these concepts that we really strive for in mental health treatment, mobile crisis work is able to achieve a lot of that because you're meeting somebody where they're at. It is always so helpful to see somebody in their home environment, to understand what their lives are actually like in that moment, and to meet them where they're at as opposed to bringing them into the whitewashed, fluorescent-lit halls of a crisis unit or a hospital. That least restrictive principle is very important here. I'm going to now go through each of these different components of the system and talk a bit about the evidence base that supports each of these different components. I mentioned a little bit about crisis call centers. NSPL, or 1-800-273-TALK, which will become 988, launched in January of 2005. They answered about 2 million calls in 2019. Now it's estimated they're more around 3 million calls. Call volume definitely increased during COVID. Again, this is a network of about 170 call centers. They're overseen by a nonprofit called Vibrant that's based in New York, and they're supported by federal grants from SAMHSA. There's also a veterans crisis line, which has been highly utilized, especially by male veterans. You can see there's some information that between 2007 and 2010, about 170,000 calls were made to the veteran crisis line, and 16,000 referrals to care were made. There's also a crisis chat and text component to the crisis call line to NSPL. I think actually it might be today or yesterday that the FCC made a determination about whether 988 is also going to be functional with chat and text, and so that's a big deal forthcoming. There's a good evidence base for crisis call centers. NSPL has been studied to show that calling into the line is associated with a decrease in suicidality during calls. About a third to half of callers were connected with mental health referrals. Callers said that a follow-up initiative that was implemented across NSPL helped stop 80% of callers from killing themselves and kept 90% safe. The chat function was studied in just a recent report that came out in the last few months, which showed that two-thirds of callers reported that the chat had been helpful and about half reported feeling less suicidal. An initial study of third-party callers to NSPL showed higher rates of requested transport to the ED, so higher ED transport rates among third-party callers. That is a really important finding because there are big changes ahead for crisis call centers because of 988. Transitioning to 988, yes, it is likely to increase call volume, and you can see this image here, that green box. There's three boxes there, so about 3 million lifeline calls happen a year. There's an additional 13 million calls that go to local, county, and state crisis lines that are not affiliated with NSPL. And then there's an additional 24 million calls that go to 911. That's estimated to be about 10% of the 240 million annual calls that go to 911. And so you put that together, you're looking at 40 million-plus annual calls that could go to 988, whereas currently there's only 3 million calls going to NSPL. It's 1-800-273-TALK. And that increase in volume is not expected just to be because a lot more people are going to be able to remember a phone number and are going to be able to access a suicide hotline. That much higher number is because the goal here is to expand 988 and suicide call centers to be more than, and I shouldn't, I didn't mean suicide call centers, to expand mental health crisis call centers, to go beyond just the suicide hotline functionality and be able to really serve all people who are in mental health crisis, including those who might be calling 911. And so that change in the characteristic of calls, you know, getting more people who are observing somebody who appears to be in some sort of psychosis, or, you know, somebody who has a friend who just had a recent relapse on substances, or, you know, all kinds of range of behavioral health crisis that might present to somebody, that if these are all coming into the 988 system, that's going to be a big change for that system overall. And importantly, although NSPL and its suicide hotline iteration has been studied, there's been no work yet on these kind of broader questions around 988-type all-encompassing crisis call centers, and so there's a lot of work to be done in that space. There's also a vision of 988 to be able to divert calls from 911, so, again, transferring calls from 911 to NSPL call centers, dispatching mobile crisis teams, ideally, you know, from a 988 call center or potentially in collaboration with other mobile providers. There's a vision for NSPL 988 sites to be able to monitor and maintain bed registries that have direct view into where crisis facilities might have availability, so that if somebody does need to self-present to a facility, they would be able to direct them where to go. Also being able to help facilitate referrals and provide follow-up and wraparound services, so this is the vision of 988, and there's a lot of work for us to get there. Now, mobile crisis, so the first mobile crisis teams, they were established as early as the 1930s in Amsterdam. As of June 2020, we've confirmed there's at least 34 states in the US that have mobile teams, although few operate statewide. By now, it's almost definitely more than 34 states. That's just what's been recorded, and we're currently working on a national survey of mobile crisis teams in partnership with Vibrant and the National Association of State Mental Health Program Directors to get a better count of how many mobile crisis teams are actually out there. Various compositions of these teams are possible. Typically, it's one or two clinicians. Might be a behavioral health clinician plus minus a psychiatric technician. There are plenty of correspondent and ride-along models, you know, behavioral health clinicians, and a police officer is one model. There's the CAHOOTS model, which is very popularly known, where a behavioral health clinician is paired with an EMT or a paramedic. Some models also include peers, including one that was recently implemented in San Francisco. And I mentioned correspondent. I want to take a minute in particular to talk about CIT because this is a really widely used model. CIT began in the 1980s in Memphis in response to a police shooting involving a person who had mental illness. And so, you know, now there's obviously mobile crisis is really growing in interest nationally as there's increasing attention to the reality that law enforcement are often not ideally positioned to be responding to people who are in a mental health crisis. And CIT was one of the earlier responses to that. So rather than saying, well, let's take law enforcement totally out of the game, they said, well, let's help train officers to be better at this because they're still the ones who are on the front lines and they're the ones who are making these calls. So about 3000 jurisdictions across almost every state in the country has implemented a CIT program, which typically includes a 40 hour training for self-selecting officers. It's associated with increased use of verbal redirection, higher likelihood of referral to treatment and lower likelihood of arrest. So there is good evidence to support CIT programs. Importantly, the full CIT model recommends a crisis system that's ready to receive individuals from law enforcement with quick and easy access and 24 seven availability. There's no wrong door idea. And often CIT programs don't have that. They often are more focused on training officers, but without the actual full system to support a CIT type model. And so that is even more justification for expanding crisis facilities more nationwide. In terms of the evidence for mobile crisis, so there have been a handful of single site quasi-experimental studies that focus on post-crisis service utilization primarily. They've looked at decreased hospitalization, which has been seen, increased community-based mental health services. So linking people to routine care post-crisis episode and also reduced ED use among youth, emergency department use among youth. Although another more recent study actually recently showed that there was more emergency department utilization and a different study showed that there was higher utilization of emergency medical services on a co-response team. So some equivocal findings, but I will say that most of these studies have really just looked at services outcomes. There's been very little that's been done in the mobile crisis context to look at best practices, to look at the composition of the teams that might be ideal. You know, should it be two clinicians? Should there be a peer? Should there be a co-responder with a police officer with a paramedic? That has not been studied in any kind of head-to-head way. Some evidence has shown that mobile crisis can reduce short-term arrests, although not long-term jail bookings. And studies have shown that there is significant cost savings due to diversion from inpatient admission. So sort of what you would expect that if you're helping people resolve their crises in the field and helping get them care that way, that then that would reduce costs because they're not accessing those higher acuity, more expensive services. Okay, the next component of these systems are crisis facilities. And there are many flavors of crisis facilities. If you hear crisis stabilization unit, you know, no two states have the same crisis stabilization unit definition. There's a real challenge in terms of nomenclature and licensing variability across the country. But broadly speaking, there's a few main models that are prominent as crisis facilities. There's 23-hour observation units, subacute living rooms, sobering centers and social detox, and then crisis residential. And these vary, I'm not gonna go into every single box on this slide, but these vary based on whether there's medical coverage with onsite nursing. It varies based on length of stay, 23-hour observation as it's called is less than 24 hours, whereas subacute tends to be more short term. There's variability in terms of whether people can be on an involuntary hold or not. Typically 23-hour observation and subacute are locked settings that can accept people who are on involuntary holds. Whereas living rooms, sobering centers and crisis residential tend to be unlocked. Police drop off then varies accordingly. What is important is that peers can and should be present in all of these different crisis facilities top to bottom, such an important role for these types of settings. And importantly, people can move between these different types of levels. So an episode of care might start at one level and then either step down to a lower level acuity setting or potentially even step up to a higher level acuity setting if they find that they need that additional support. There is some evidence for crisis facilities. The most commonly defined setting that crisis stabilization unit at 23-hour observation has actually been really minimally studied. Psychiatric emergency services, that's been extensively studied. That's a whole field of emergency psychiatry. There are a couple specific types of crisis facilities that have been studied more. So empath, the emergency psychiatric assessment treatment healing type settings have shown doing pre-post analyses. They were shown to have a reduction in admissions, a reduction in 30-day return to the emergency department and a reduction in ED boarding time by as much as two thirds and also had a 60% increase in 30-day follow-up care. So that's a quite well-known model empath and does have some good evidence to support it. Also, there's an extensive crisis residential literature that shows greater client satisfaction, lower costs. Although there is some need for more studies of peer respite. And now finally here, so there's the three components, but then there's a lot of topics that crosscut that are important to consider, whether we're talking about call centers, mobile teams or facilities. Post-crisis care is one, it's a really important one. And there have been some studies that have looked at post-crisis wraparound services. Link to patients or these types of services have been able to link patients to longer term treatments. They've been able to avoid reutilization of crisis and other acute services. They've been able to prevent future encounters with law enforcement and also some have even been able to address some social determinants of health like housing, transportation and food. And these types of post-crisis care can be provided by behavioral health programs, of course, including peer navigators, but also potentially law enforcement based case management and even community paramedics. I mentioned peers a couple of times and I do wanna highlight that there has been some actual research work that shows the effectiveness of peers. Peer support specialists working in EDs have shown promising recovery related outcomes. They've been shown to be effective at communicating with people who are experiencing a psychiatric crisis. They're effective at intervening in the cyclical nature of emergency department admissions for people who are in mental health crisis. It's all too common that people are cycling through the ED multiple times. And peers can be really good at sort of breaking that cycle and figuring out, well, wait a minute, how do we actually help you and try to make this cycle not keep going? And then there's also a real potential to decrease emergency department boarding by people with mental illness and decrease at first outcomes secondary to a crisis, things like restraints, emergency medications, IMs, things like that, intramuscular injections, things like that. Also a relevant issue, whether we're talking about calls such as mobile teams or crisis facilities is the issue of involuntary treatment in crisis services. And this is another area that's really not well studied. So there have been some studies that looked at, who are people who are at greatest risk for involuntary hospitalization? Might come as no surprise that one of the greatest risks for being on an involuntary hold is having previously been on an involuntary hold. Also psychotic disorders and also economic deprivation, both on an individual and population level. So there's clearly something structural happening here in terms of the use of involuntary holds as a tool. There are usable counts for involuntary hold data in only 25 states. So it's not just that these haven't been studies, that the data to be able to study these types of involuntary holds is really poor across the US. There was one study that was done just recently published in psychiatric services that showed that in 24 states, about 600,000 emergency involuntary detentions happened in 2014. So clearly a high rate in what was equivalent to about half the population of the US. Emergency detentions range drastically. In Connecticut, it's about 29 out of 100,000. In Florida, it's 966 out of 100,000. That is a over 30 fold difference between Florida and Connecticut in terms of the rate of involuntary holds that are put in place. And so that's meant to just really highlight that, given that mobile teams are often in a position of needing to decide about voluntary or involuntary treatment, there's a real variability in terms of how that tool is used. And it does appear that the rate of these involuntary holds being placed is increasing. So from 2012 to 2016, there was an increase from 273 to 309 per 100,000 on average across the states that were studyable in this data set. Finally, public tracking of civil commitment is needed for oversight, but it is complicated. Privacy concerns abound. There's decentralized systems and mental health care. So the mental health care systems aren't making it any easier on the data reporting side. And there's variable commitment criteria across jurisdictions. An involuntary hold in one state might be 72 hours, but in another state, it might be much longer. And so all of these are real challenges for improving our data quality in this space. Lastly, in terms of cross-cutting issues, I really want to focus here for a second on equity. You know, if we're talking about crisis services as being an opportunity to address disparities, as being a way to reduce the role that law enforcement might have a mental health crisis response, especially in terms of how that disparately impacts communities of color, we have to have the data to back up those claims. So Dr. Sarah Vinson and Dr. Dennis in an article in Psychiatric Services, and it was also one of the, I think, bigger Do Good series by the Scattergood Foundation, wrote an article on this topic. And I want to read this quote, because it's really powerful. Understanding a problem is a prerequisite to addressing it. For the mental health care system to play its role in remedying the incarceration of people, of a population that is disproportionately Black and Latinx, the extent of racial inequities in this population's mental health treatment must be fully characterized. However, the system's current functioning does not support such understanding. So as they suggest, the literature on crisis services has very few examples of equity analyses. And there's one study recently that demonstrates just how powerful this type of equity analysis can be. So there's a study of a correspondent team where it was a clinician and a police officer, and they found that there was no change in overall long-term risk of justice involvement. Although, they did find that incarceration was significantly reduced among Black correspondent team recipients. So again, so that means if in the overall population, so thinking about it, looking at it race-blind, which of course is problematic, they saw, okay, well, there's no overall term change in long-term justice involvement. When they did that sub-analysis by looking at the Black subgroup, they found that actually there was a significant decrease in longer-term incarceration risk. And so it might help us then interpret, well, wait a minute, so that correspondent team, although maybe it wasn't as effective for the overall population, as an anti-racism measure, as a measure to try to reduce the disparity of people of color who are justice involved, maybe this type of correspondent team could actually be effective in that sense. Also, they found that it was recipients of the service were more likely to have any emergency medical services contact at six and 12 months. And they found that these trends were actually driven by the white recipients of the correspondent team and was not observed in the Black population. So what does that say? People who got the correspondent team were more likely to have EMS contact at six and 12 months. What is it about the system that's making it so that the white population that received this service are accessing EMS more and not the Black population? Is it something about how EMS was made available as an option to clients? Is it experiences of discrimination previously that helped made service recipients feel more or less comfortable and that varied by race? Really important questions here to get a better, more nuanced understanding. So again, the general principle here is we need to be able to actually characterize equity as part of this work. Future directions. So we've got a lot of work to do as we know. There are some unique challenges of crisis services research and evaluation. Some of the questions here. So there is an enormous range of topics. I talked about a lot of things in the last 30 or so minutes saying there's not a lot of data on this. There's not a lot of data on this. And part of the issue is where do we start? So that is a real issue. There's so much to try to tackle here. There's poor data quality across the board and especially with race and ethnicity data that's needed for those equity studies that I was just describing. There are minimal linkages between relevant datasets, inconsistent billing and quality reporting, which means that there's limited administrative data. A lot of crisis services are not billable. And so that makes the data that we get from things like insurance claims not studyable if they're not being claimed. Limited data on non-hospital field-based settings. So emergency medical services, mobile crisis, often their data systems are much more kind of simplified because they're plugging things in the field and that limits the value of their data. Siloing between state and county health officials and academic partners, meaning that there's all too often the administrators or the operations folks who are right now across the country actively implementing these programs. I've met more than one system administrator who when you say the word research to them, they fall asleep, right? So there's a historical divide here between the implementers and the researchers. And that siloing then is a challenge, especially when we're thinking about something that's growing so quickly. Difficulty of generalizing results and implementation research and fidelity to evidence-based models. These are challenges just across the board and in mental health services research. And then also a lack of mental health services research funding from the National Institute of Mental Health and others. These are all challenges. And so where that's left us is there's a lot of gaps in crisis research. The way that I've organized this, and again, I'm not gonna get into all the details here, but you can kind of think about it in terms of, okay, so we've got call centers, mobile teams, stabilization-type crisis facilities, and then post-crisis follow-up and linkage-type services. And there's gaps in evidence base across those different settings in different important areas of research. So things like accountability and finance, the types of capacity and services, clinical best practices, which I think is actually the most important. For example, what are the clinical best practices for non-suicide crisis call triage? What are the important clinical best practices and clinical pathways for various presentations of populations who are being seen by mobile crisis or outreach teams? And facilities, what are the ideal roles for peers or for short-term medications? A lot of questions here not yet answered. And then there's cross-cutting topics across all of these different types of questions. So again, disparities and equity analyses, staffing and workforce questions. What are the role for peers? Rural versus urban settings being so different. What's the role for technology? How has COVID disrupted these types of settings? In particular, looking at high utilizers of these settings as a special population. And then other special populations, LGBTQ, Native American, American Indian, veterans, people who are hard of hearing, people who've experienced homelessness, et cetera. Looking at client satisfaction, looking at involuntary versus voluntary treatment as discussed. The impact of specific policies on how these services are effective. Lots of gaps in the evidence base here. So some solutions. So one is there's a need to invest locally in data infrastructure and linkages. So I've got this slide here with cobwebs on it because our data infrastructure is so old and poor. There's so much fragmentation between these different types of data. So at the call center level, we've got NSPL call centers, non-NSPL call centers. We've got 911, what are called public safety answering points or dispatch. Other types of crisis call lines. There's mobile crisis, electronic records, and dispatch logs, emergency medical services logs, police reports. There are crisis facility electronic records, emergency department and hospital electronic records, all of these different types of data that are most often not at all linked. And then there's data that cross cuts. So benefits eligibility, insurance claims, social determinants relevant data like housing, education, and employment, death data, so vital statistics, those sorts of things. And then importantly, collecting race and ethnicity data throughout, again, to inform these equity analyses. So there's a lot of gaps here. And, and, you know, the vision is, can we turn these cobwebs into an actual web, you know, into a deliberate network that does link these different types of data, while also taking into account data security and privacy and ensuring that you're, you know, you're, you're really making sure to, to address those important ethical considerations. There's also a need for more sophisticated approaches to measurement. So what I'm showing here is there's a few different types of measures that I think about in crisis services. I think about descriptive measures, which are things like, you know, number of teams in beds, number of calls and visits. There are performance measures, so call and visit response times or duration. Then some more meaningful process measures where we might track referrals to routine care, 911 transfers, emergency department drop-off rates, involuntary hold rates. And then also outcome measures. So these are real outcomes, things like suicide rates, overdose rates, symptom reduction, client satisfaction, including patient reported outcome measures. Now the challenge here, of course, is that as you move further down this chart, they're more meaningful, but they're also more difficult to measure. So to get to that point of having a more sophisticated measurement approach really requires a process of building consensus. Again, we need equity analyses with the way that that can be done in this context is by stratifying measures by race and ethnicity. So you deliberately look at how programs are either impacting disparities, if they're perpetuating disparities, or if they're dismantling some disparities in a system. And then I've got here on the side, my cheesy logo here for benchmarks. So the benchmarks are important because we need to understand, well, what's the meaning of these rates? So take, for example, a process measure on reutilization of crisis services. The idea here is, well, you've got a crisis service, and if somebody comes in and is seen by a mobile team, that that mobile team should be able to help that person get what they need. And then if that person represents to another crisis service within a couple days, that that might somehow be seen as a failure of that mobile team for not getting them an outpatient linkage or for not somehow responding to their need. And that might be true in some cases, but then we also know that when you see somebody who's experiencing a suicidal crisis, what's like the first and last thing that you say to them? Thank you so much for calling. Please, if you experience this again, call us back. Right? Please let us know so that we can come see you again. Please come to the crisis facility to seek care and keep yourself safe. So there's some amount of reutilization that's actually going to be really clinically appropriate. And so that idea of benchmarking is figuring out, well, what's the correct rate? You know, what's the rate that is, you know, it's the rate that we would hope for and expect, but it's not too high because the systems and programs aren't actually doing what they're supposed to be doing. So that's a whole process and is in real need here in crisis services. There's also a real value in the idea of aligning quality improvement and research activities. So, you know, there's a lot of people who are doing QI out there and QI and research are actually a lot closer than they might seem. So, you know, quality improvement, when you think about stakeholders, usually QI folks are administrators, program directors, quality officers. On the research and evaluation side, it might be academics or payers who are asking for evaluation data, you know, to support their reimbursing for these services. But there's a lot of alignment here too, you know, patients, payers, and policymakers all want to see good quality research. Funding, you know, QI is often funded by grants or unfunded, whereas research and evaluation kind of relies on federal grants or foundations. But there are real opportunities for alignment here. So research funds can definitely help build infrastructure to support quality improvement and vice versa. And I've really found that myself being in a Department of Public Health, but also being a researcher on grant funded projects that our research can really help improve the quality improvement function of the department. And also all the really smart quality folks are able to contribute to the research in really helpful ways, because they understand the data super well and are able to make sure that we're actually measuring the right things. The rationale for this, so, you know, quality improvement is often focused on mandated reporting, regulatory compliance, you know, just quality assurance, whereas research and evaluation is focused on expanding the evidence base, understanding fidelity, feasibility, acceptability. But again, all of these things can help justify payment and also respond to community needs, including, you know, in a research sense through methods such as community-based participatory research, which can engage communities directly in the development of the evidence itself. And then the methods. So, you know, QI often is descriptive, looking at a pre-post comparison, doing a plan-do-see-act type QI cycle. Research and evaluation might use more rigorous methods like implementation science, mixed methods, interrupted time series, propensity score matching, step wedge, all kinds of different implementation science methods. But the alignment here is that you can share data sets and you can align your metrics. And, you know, research does not require cumbersome research designs like randomized control trials necessarily. Yes, there is a time and a place for that type of rigorous gold standard analysis. But there is also a lot of information that we can get from implementation science type techniques. And when you have quality improvement metrics, those same metrics and that same data can be used in implementation science type techniques that can contribute to the evidence base in meaningful ways. So, you know, in summary here, there's a lot of strategies potentially to advance crisis services research. There's opportunities to develop a crisis research agenda, which might be based on that, you know, that matrix of research gaps that I went through earlier. Opportunities to advocate for mandates and funding for evaluation and federal and state bills. Identifying strategies to engage academic partners and seek grant funding. So trying to kind of bridge those silos. Sharing best practices for implementation science methods to study real world models. So sharing some of the ways that that QI and implementation science methods can overlap. Sharing best practices for obtaining, linking, and analyzing complex multi-sector data. So again, back to the data linkage issue. Aligning research with quality improvement. Defining consensus, crisis metrics, and benchmarks, as I discussed. Leveraging big data for informatics and population-based approaches. I didn't talk about that so much, but as we get more and more clinical and also hopefully claims information on crisis, those will present opportunities for those types of informatics approaches. And then promoting research opportunities. So when there are grants or requests for proposals or conferences, journals that come out, these are all opportunities to advance crisis services research. And so some takeaways, especially, I know I'm mostly talking to a clinician audience here. And so the research is hopefully interesting and will, I think, resonate with everybody insofar as we all care about evidence-based practices. And we all want to make sure that as we're moving these different crisis services forward, that we're doing so in a way that's sound and is really in the best interest of our clients. There are some important takeaways that clinicians can really focus on. So first of all, building programs based on evidence, based wherever possible. So that's part of what we can do as clinicians is we can think about, all right, where is there room for applying the evidence that at least exists? And then also recognizing where the evidence doesn't exist and needing to really go more on clinician or expert consensus. Drawing on clinical best practices from other settings to inform protocols and procedures. So that's another strategy. Tracking data on quality of care, including by race and ethnicity. So even if not participating directly in research, that equity analysis piece is so important. That can be applied in QI just as much as it should be applied in research. And so that's definitely a way that clinicians, as advocates for our clients, can really put that forward. Partnering with academic colleagues to evaluate implementation of new crisis programs. So we can be the change makers. We can be the folks who are reaching out to people at the local university or some research postdoc who you somehow have a connection to who wants to do research in this space. Trying to build those bridges to invite opportunities for evaluation of implementation. And then engaging with communities served to understand needs and inform quality goals. So again, we can't do anything without the community. We got to be looking for every opportunity possible, especially around specific community engagement plans around how to structure a crisis system, community engagement plans around how to address racism in a mental health system. These are all important opportunities for engaging the community and also specifically people with lived experience of mental health crisis to really make sure that they're having a direct voice to inform some of these quality goals. So here are some sources which are in the slides for you. And with that, thank you so much for your attention and looking forward to the conversation with Dr. Dress with any questions. Thanks so much, Dr. Goldman. That was, you know, really clear and just fascinating presentation. Just a reminder to attendees that if you have questions, you can submit them by typing them into the question area, which is in the lower portion of your control panel. But before we shift into Q&A, I just want to take a moment and let you know that SMI Advisor is available from your mobile device. Using the SMI Advisor app, you can access resources, education and upcoming events. You can complete mental health rating scales and even submit questions directly to our team of SMI experts. You can download the app now at smiadvisor.org forward slash app. So, Dr. Goldman, you know, when I first heard about 9-8-8, I had thought of it primarily as you had discussed about kind of the focus would be on suicide. But it sounds like there'll be a large range of the sorts of crises that will be fielded. And I'm wondering if there are any data or any ideas of what will be the mix between either existing patients or new ones or suicides versus suicidal ideation versus overdoses versus other sorts of crises. Are there any projections about that? There's definitely projections. I know that Vibrant, which is, again, that's the organization that operates the National Suicide Prevention Lifeline with support from SAMHSA. I know that they've done projections on volumes and also potentially on caller type. I mean, at first, just to reinforce, that's correct. That's my understanding as well, is that the focus of 9-8-8 is not just going to be on, hey, here's an easier way to remember your suicide hotline. It really is meant to be broadly. Here is the new mental health crisis line. Call us instead of 9-1-1 if you need anything urgent. And so that, you know, it really is a fundamental change. There's not a lot of study. I am working on a couple analyses that might get at some of this. And so one way that we could look at that is by looking at a setting that already operates a kind of universal crisis line. And one particular example where that's very well developed is Georgia. So Georgia has something called the Georgia Crisis and Access Line that operates statewide. It was developed actually in the aftermath of Hurricane Katrina, because after Katrina, a lot of people from New Orleans were seeking refuge in Georgia. And the state was getting a lot of requests for assistance. And somebody in the state was like, well, so what's the crisis line for each of these different areas? And they realized, my gosh, there is no crisis line in a lot of these areas. It was going to some individual's voicemail. It was a really fragmented system. And so they developed a really impressive statewide infrastructure for crisis calls called the Georgia Crisis and Access Line. Of course, Dr. Dress, you know about this because you're based in Georgia at Emory. But what this is able to do is they have a statewide clinical database for over 10 years with crisis line calls. And they are billed to Georgians, not just as the suicide prevention hotline, although they are a National Suicide Prevention Lifeline member and will get 988 calls. They also are billed for all types of crises. And so in a way, they're kind of like a 988-like crisis line before its time in a statewide context. And so I've been working with people in the Department of Behavioral Health and Developmental Disabilities in Georgia to utilize that call center data. And there's a couple interesting analyses that we could do. And one of the things that we're going to do is something called a cluster analysis, or more technically, it's a latent class analysis, where you basically look for what are the different factors among people who are calling the Georgia Crisis and Access Line, GCAL they call it. And so you can basically plug in like looking at age, gender, race, ethnicity, clinical presentation, region, all kinds of different variables. And then this analysis will tell us, well, what are the most common types of clusters that come out of that? Are people who are calling mostly young people with suicidality, middle-aged white men with suicidality, which of course we know is one of the demographics that has the highest suicide death rate in the country. Are there young people with psychosis who are actually calling a lot or people with overdoses? And so there are data-driven approaches to be able to answer the question of who is accessing a 988-like call line. And so that's an analysis that it's forthcoming, but I think that kind of approach, and it doesn't just have to happen in Georgia. I'm sure there are other settings as well that could do something like that. That's really what's going to, I think, help inform what we might predict with the arrival of 988. Yeah, that's really interesting and we'll all be excited to hear how that turns out. I imagine that will also have implications for how clinicians, particularly those who treat those with serious mental illness, will end up hearing from or interfacing with the crisis system. I expect or I guess I wonder how often that will involve if one of their patients ends up calling 988, how they'll be contacted in those cases and ways in which the patients will be referred back to them. Yeah, no, you're in such a good point. I think there's a couple of thoughts that come to mind about that. So first is that idea of clinical pathways to care. For somebody who's presenting with serious mental illness to a crisis call center, that's a very different situation to be able to help support, as opposed to somebody who's calling for themselves a suicidality. Especially it's different if it's not the person themselves who's in a crisis experiencing hallucinations or delusions, but rather somebody else, like a bystander, who's observing that person, who looks like they're in some sort of distress. Figuring out those clinical pathways for call centers for non-suicidal self-callers is a major task. That is definitely something that is being looked at closely as part of developing 988. The other issue that you're talking about is care coordination, which is so important as well. It's a point that I didn't raise in my slides, but it's really important, is that there needs to be really bi-directional support between the crisis care continuum and the traditional behavioral health care continuum. We've already got clinics and community mental health centers and CCBHCs and all kinds of different settings out there that are already doing this work in the community. We need to make sure that when those outpatient providers have a client who's in crisis, that they know exactly how to contact a crisis resource. We want to make sure that when a crisis call center or mobile team is seeing somebody who needs outpatient support, that they then have ways of linking those people to services. That kind of bi-directional support is so important. Just what you were talking about in terms of how do you contact somebody, that kind of information sharing is a key element of that bi-directional support between the crisis system and the traditional behavioral health type settings. Great. That's wonderful. I think we could ask you many more questions, but I think we're going to need to close with that. It's a great note to end on. Just want to let folks know that if you have any follow-up questions about this or any other topic related to evidence-based care for SMI, our clinical experts are now available for online consultations. Any mental health clinicians can submit a question and receive a response from one of our SMI experts. Consultations are free and confidential. SMI Advisor is just one of many SAMHSA initiatives that are designed to help clinicians implement evidence-based care. We would encourage you to explore the resources available on the Mental Health Addiction and Prevention TTCs, as well as the National Center for Excellence for Eating Disorders, and particularly relevant for today's talk, the Suicide Prevention Resource Center. These initiatives cover a broad range of topics from school-based mental health through the opioid epidemic. To claim credit for participating in today's webinar, you'll need to have met the requisite attendance threshold for your profession. Verification of attendance may take up to five minutes. You'll then be able to select Next to advance and complete the program evaluation before claiming your credit. Please join us in two weeks on December 3rd, 2021, as Guarava Agarwal with Northwestern University presents an organizational consultation model to advance a culture of well-being. Again, this free webinar will be December 3rd, 2021, from 12 to 1 p.m. Eastern Standard Time. Thanks again for joining us. Until next time, take care.

high-quality mental health crisis services

equitable mental health crisis services

challenges faced by behavioral health systems

crisis call centers

mobile crisis teams

data collection in mental health systems

program evaluation in mental health systems

evidence-based care

gaps in crisis services research

implementation of the 9-8-8 crisis services hotline