Hello, and welcome. My name is Rachel Talley, and I'm an Assistant Professor of Clinical Psychiatry at the University of Pennsylvania, and one of your three presenters for today. And I'm pleased that you're joining us today for our Striving for Excellence series, Equity and Access, the Self-Assessment for Modification of Anti-Racism Tool, or SMART, and the Level of Care Utilization System, or LOCUS. This funding for the Striving for Excellence series was made possible by this grant from SAMHSA of the U.S. Department of Health and Human Services, and the content presented here are those of the authors and do not necessarily represent the official views of nor an endorsement by SAMHSA, HHS, or the U.S. government. Today's webinar has been designated for one AMA PRA Category 1 credit for physicians, and credit for participating in today's webinar will be available for 60 days. A PDF of the slides is also available in the chat tab, and captioning for today's presentation is available. To enable the captions, click Show Captions at the bottom of the screen, click the arrow, and select View Full Transcript to open the captions in a side window. Please also feel free to submit your questions throughout the presentation by typing them into the question area, which is found in the Attendee Control Panel, and we'll reserve 10 to 15 minutes at the end of the presentation for Q&A. With that introductory portion, again, my name is Rachel Talley, Assistant Professor of Clinical Psychiatry at the University of Pennsylvania, and a board member of the American Association for Community Psychiatry, and I am delighted to be here with my co-panelists, Dr. Sosin Maluch-Oyenka, who is Chief Medical Officer of the Department of Behavioral Health and Intellectual Disability Services in Philadelphia, and also a Clinical Associate Professor of Psychiatry at the University of Pennsylvania, also a board member of the American Association for Community Psychiatry, and Dr. Ken Minkoff, who is Vice President and Chief Operating Officer of Zia Partners in Tucson, Arizona, and also a board member of the American Association for Community Psychiatry. The three of us are so pleased to be here to present a little bit of content on the American Association for Community Psychiatry's family of quality improvement tools and ways in which these tools can assist with assessing equity and thinking about access to care through an equity-focused lens. First, a couple of disclosures. I'm a medical advisor for Ivana Health, and Dr. Sosin Maluch-Oyenka is a founder of Cyntia Health. The content of this presentation is not related to those roles, and so with that, I'm going to turn it over to my colleague, Dr. Ken Minkoff, to kick us off. Okay, and are you changing the slide because you can move right into the next one? Well, hi, everybody. I'm Ken Minkoff, and we're all excited to be presenting this webinar on behalf of the APA, but also on behalf of the work that we do at the American Association for Community Psychiatry. Because we're psychiatrists who are particularly interested in systems of care and improving those systems of care for the benefit of the people and communities that we serve, one of our activities has been over the years to look to develop tools that allow all kinds of professionals, including psychiatrists, of course, to figure out how to make progress around issues that are significantly challenging in the field. And one of those tools historically is the LOCUS, or Level of Care Utilization System, which I'll be talking about a little bit later, but the job of which is to provide a clinically driven framework for making service intensity or level of care decisions that can be valuable at both an individual clinical level and also, as we'll see here, in looking at data that informs system improvement for how the system of care operates in general for promoting access to services and appropriate level of care decision making for people who may come from marginalized communities or minorities who may receive inequitable services. And more recently, the SMART tool that Rachel will be talking about, and Sosa Molo will be talking about shortly, was created with the same idea in mind, that we recognize the importance of helping systems and organizations in the behavioral health space specifically to have tools for improving the effects and impacts of structural racism on the internal organization and, very importantly, on the people served and the communities served. And together, these tools in this presentation, we're going to illustrate how to use both of these tools in a combined way to look at particular targeted inequities that may impact people who are coming into crisis services and, in particular, may experience, as is often the case, inequitable placement in involuntary commitment. Part of my role on the board is I'm the co-chair of what we call the Products and Services Committee. So I've been directly involved in the implementation and development and dissemination of both of these tools. And it's really exciting for me to work on this and present with Rachel and Sosa Molo on this issue. I'm going to hand it back to Rachel and she's going to walk us through some more detail about SMART. All right. Thank you, Ken. And so I'm going to kick us off talking a little bit about some of the background of how we, our process of developing the SMART, as well as a little bit about implementation. And Sosa Molo will be chiming in here as well to help me in presenting this overview. So starting out, first just taking us back at this point about three years back to 2020, basically the American Association for Community Psychiatry, like a lot of organizations, I think found themselves at sort of a crossroads in terms of thinking about what our role should be in addressing issues of systemic and structural racism. This is, of course, in the setting of what was sort of a revitalized national dialogue around these issues that occurred in 2020 in the spring and summer, primarily kicked off by the highly publicized killings of unarmed Black Americans at the hands of the police. And so posted here on the screen, I have four basically stated commitments that were part of a statement released by the American Association for Community Psychiatry around this time. The AACP, like a lot of organizations, felt moved to make a statement condemning police brutality and making a stance of standing with those who were focused on fighting against structural racism. And our goal in including these commitments was basically to sort of articulate a framework for our members to think about not simply talking the talk, but really walking the walk in terms of what it was that our organization should be encouraging its members to actually do to address structural racism. To give a little bit of background to the audience, for those who may not be familiar, basically the American Association for Community Psychiatry is an organization that represents psychiatric providers in public and community spaces nationally. And so our thought process was really trying to start to consider, for those who are working in these public mental health spaces, how can we give guidance in terms of how to address structural racism as it occurs in our own backyard, if you will? Or in other words, how can we guide our members to think about how to address structural racism as it is specifically uniquely relevant to the community, the public and community mental health space? And so with that in mind, basically having released this statement, our thought was, well, how do we really make sure our members are walking that walk and living up to this commitment? And so with that in mind, our board, and basically myself, Sosomolu and Ken in particular, started to think about how might we create something, some form of tool or some form of support to assist our membership in addressing structural racism in the community mental health setting? And so here I have on the screen what I would say sort of are the three goals that we had in mind as we approached this process or this idea, ideas that ultimately became the smart tool. First, our goal was to ideally create something that would provide metrics for the community psychiatric provider to consider that were specifically relevant to the community mental health setting and specifically relevant to disparity and inequity issues that might disproportionately occur in public and community mental health settings. So in other words, as I'm sure many on the call know, structural racism is rampant in our society and impacts all manner of different organizations and institutions. Knowing that that could feel overwhelming to anybody, certainly to the public and community mental health provider as far as where to start, our goal was to create a tool that would really help our membership to zero in on a set of key target issues that might have the strongest relevance to their work settings. So to give a brief example, for instance, one issue that is covered in the SMART is the issue of access to clozapine, life-saving clinical treatment for individuals with treatment-resistant schizophrenia. We selected this among the array of possible items to include in SMART, in part because we felt it was uniquely strongly relevant to public and community mental health settings because these are the settings where we're primarily seeing individuals with more severe mental illnesses, and also because there's copious research literature that has consistently demonstrated that clozapine is underutilized in minoritized black and brown populations. And so this was our lens in terms of thinking about how to draw together a key set of target issues. Our second goal with SMART was to ideally create something that did not reinvent the wheel. Knowing that there are a range of wonderful inequity-focused sets of principles and guidelines and guidance already out there for healthcare providers to use, our hope was to ideally not replicate those efforts but really build upon them to create a tool that would sort of holistically address all facets of structural racism that might impact a community mental health setting, including thinking about its structure, thinking about its hiring practices, thinking about its organizational culture, really to think about this in a holistic manner, drawing upon some of the wonderful existing standards that are already out there to help guide inequity work. And lastly, what I would say really boils down, what is the SMART in terms of what sort of item or tool that we want to create? Our goal was to create a tool that would promote a stepwise, concrete quality improvement process that could be adapted for self-directed use in the community mental health setting. I emphasize self-directed because an experience that the three of us had as we were thinking through this tool, as our board as a whole was thinking through the creation of this tool, was that a common sort of reaction to this dialogue around structural racism in 2020, I think amongst many of us, certainly amongst many of us in the community mental healthcare space, was again, these are big, overwhelming issues. Where might we begin in trying to address these issues? And so a common next step was to think, well, let's bring in an outside consultant to try to address these issues in our organizational space. Our mindset was that that's certainly not the wrong move, that that is certainly a fair approach in terms of addressing structural racism. But recognizing that for community mental health settings that are often lower resource, where there may not necessarily be the financial option to bring in an outside consultant, our goal was to ideally create something that would empower community mental health providers to draw upon their own knowledge of their organization. Basically, the folks who know it best and the stakeholders who know the organization best might be able to draw on their own knowledge of the processes and the data available in their organization to at least start to take some self-directed quality improvement steps in terms of addressing inequity issues without the support necessarily of an outside consultant. This is, I'll be brief here to be sure we can sort of move towards some of our more sort of real-world application content of this tool. But in brief, our development of the SMART, and we do have a publication in the Community Mental Health Journal that dives into this process in more detail, but it was basically a mix of drawing on the experience and perspectives of the American Association for Community Psychiatry's National Membership, a discussion amongst some of our board subcommittees to start to boil down a key set of target issues, and also a review of the literature, both to review previously existing, previously published inequity-focused healthcare frameworks and sets of principles, but also reviewing the literature to identify what our key target issues would be, namely those issues that both had high relevance to the public and community mental health setting, and for which there was consistent literature evidence suggesting a likelihood of disparity and inequity by race, ethnicity. With that, I'm going to turn it over to my colleague, Dr. Shoyanka, to continue to speak a little bit about this process of how we developed and the implementation of this tool. Good afternoon, everybody, and thank you, Dr. Talley and Dr. Minkoff. It's a pleasure to be here with you all this afternoon. So, picking right up from where Dr. Talley left off, there are a number of common pitfalls in this kind of change work, and I say change work broadly, including, of course, the DEI efforts, but also, more broadly, looking at any sort of system-level transformation efforts. And particularly, the challenges tend to be, on one hand, particularly with structural racism, that sense of being overwhelmed, that sense that this is so pervasive that we really struggle to know where to even start, and what the likelihood of success might be. That's one pitfall, potentially. Another potential pitfall is the idea that this happens in all organizations. We know this happens elsewhere, but we are somehow magically exempt from this in our organization. We are good people, and this doesn't happen here, which is certainly the case, often, that people are well-meaning, but the very nature of structural racism is that it hides itself, embeds itself, if you will, in processes that are hard to even notice on a day-to-day basis. And that's actually what drives the pernicious nature of structural racism. And so, SMART and LOCUS both focus on looking at identifying processes, processes that require change and transformation. Now, to get to move organizations beyond that first challenge, that first potential pitfall of the sense of being overwhelmed, the Serenity Prayer System change is actually an answer to that, the understanding that we may not be able to change what happens with everybody else, or anybody else, really, but we actually can change what is directly within our own wheelhouse, what is within our own purview, our own program, perhaps even on the micro level, our own caseload. And if opportunity permits, I will share the story of one of our colleagues who, when we began developing SMART, took a look at the caseloads of herself and other people in her, that she was supervising, and made some changes. LOCUS PDC is that acronym that gives us a framework for system change. The first is, in this instance, of course, is finding a process to change, which is structural racism, then organizing a team to use that tool, the smart tool in this case, clarifying the baseline using the tool, which helps us understand, begin to answer the question of, well, why is that baseline, you know, the way it is? The simple fact of measuring how our processes operate actually points out many opportunities for transformation, just in and of itself. The other pieces are selecting specific issues to begin to address, and then using the PDCA cycle, which is a well-known rapid cycle transformation framework to begin to address them. Focus PDCA. Next slide. All right. Very good. So, how does this work? So, oftentimes, one of the more difficult initial challenges is to even begin to have these conversations. People often shy away, because, obviously, these are oftentimes emotional-leading topics, and, you know, people want to stay away from the possibility of that sort of back and forth. And so, what one of the geniuses of SMART in particular is to replace what is often or can be interpreted as subjective interactions or impressions with information, again, changes the focus from the individual to the organization, and, very importantly, provides data, which then allows targeting or targeted interventions, and simultaneously, oftentimes, will identify areas where more data is needed. And the requirements for a diverse group of staff doing this work, I mean, of necessity, any kind of system transmission is really teamwork, but, in particular, DEI work requires having BINAP at multiple levels, and the requirements for both creating that diverse group as well as the requirement for reaching consensus on scoring actually tends to drive the conversation to a much deeper and granular level. And then the last benefit is that this doesn't require any extensive training. Simply using the tool will train the staff. Next slide. All right. So, these are the five domains that are covered under SMART. Because it is maybe the primary driver, it's important to look at hiring, recruitment, retention, and promotion, and these are, I should say, all areas where the literature is very clear that there are disparities. I'll give you just one example. This is from management literature broadly. Hiring and the representation of top-level executive leadership for the representation of minorities at that level is something like 6% nationally, and compared with the representation of minorities in the overall population, which is something like 30%. So, there are clearly disparities in that arena. But you can see the same thing in promotion, certainly in academic settings, other settings, as well as things like discipline, the application of discipline to staff. So, clearly, disparity. Then clinical care, Dr. Talley has talked already about. Close the rule, but there are multiple other areas in which you can see this pain management, obstetric outcomes, use of application within mental health of coercive treatment. These are all areas where there are disparities. And then workplace culture is included as a key driver and determinant of behavior within the workplace, oftentimes, many times assumed or unspoken, but very powerful and oftentimes stably transmitted across generations. And then community advocacy, because of the rule of mental health in the community and the significant overlap with many other social service organizations, such as, for example, the criminal justice system, there's a representation in that arena. And then population health outcomes and evaluation, because that is really at the bottom line is what happened, what are the outcomes at the population level. Next slide. So, what is the recommended process? Well, it really starts with convening a group. It starts with, and I should say that organizations that have been successful in using SMART have really been, it's really been driven really from across the organization, but very importantly with buying from the senior leadership levels. So, you need to recruit a group of staff, diverse, obviously, in terms of ethnicity, but also in other ways, diverse in terms of seniority, diverse in terms of clinical background and certification levels, as well as in terms of age is also important. And then we're looking at this as a catalyst for really for a broader conversation. The idea is not that this is an end in itself because this type of work, really truly doesn't really ever end. And it really oftentimes is a catalyst for other conversations. And then, as I mentioned, the consensus core requirements is one of the key elements and is intended to generate an action plan, which based on low hanging fruit, and then the ideas to reassess periodically, ideally every six to 12 months. Next slide. All right. So, this is just one example of how to apply the SMART tool. So, involuntary treatment. So, again, this is an arena in which there is abundant evidence that there is disparity, whether that's involuntary treatment, as you will see at the individual sort of point of care, or whether that is the more sort of sustained outpatient treatment orders. There's evidence of disparity. So, an example of how to apply this tool would be to look at the potential disparities in the imposition of involuntary commitment. So, to what extent after in this instance of this scenario, organizing that group would be then posing this question to the group. To what extent does the organization track and address potential racial disparities in the imposition of involuntary commitments, either emergency commitments or AOT? Score one is we don't track this. We haven't addressed it. Score two is we do track this, have identified disparities, but have made no progress. Score three is we do track this fairly well, have identified disparities and have made some progress. Score five is we track this consistently and have processes in place. In other words, we have solved this problem and have no further issue with it. Next slide. So, I'll turn this back over to Dr. Minkoff. Thank you. Yes, thanks, Dr. Shoyinga. And what I'm going to do is spend a little time introducing a locus for those of you who are not familiar with it. And to make a particular point of how it's relevant to the issue in quality of care generally of a potential area of inequity concerning imposition of involuntary treatment orders. Which is one of the things that a service intensity assessment may be addressing. So, next slide, please. So, the purpose of the locus tool, this is a very important part of the message, is the recognition that service intensity assessment and planning, which is our preferred term, shorthanded into, quote, unquote, level of care placement, is fundamentally a clinical, organized clinical set of decisions that is orthogonal to diagnosis. And that there are clinically understood best practice approaches to making those decisions that all clinicians should use. But not only should all clinicians use it, there should be a common language for persons served, for people paying for service, and for folks who are managing services at the system level. And we need to get away from the idea that these decisions are made by managed care companies for payer-driven reasons, so much as the same as we do with diagnosis, there's a common set of approaches and criteria and a structured, measurable process for doing these things with reliability and validity. This can be applied on the individual level, for an individual person, but also looking at the data aggregately, we can use this to see whether there are patterns or disparities in the way these are applied, as well as looking at broad system gaps. Next slide. So, the essence of the locus was originally developed, the first version, in 1996, and it's been updated a couple of times since then. It uses a framework similar to the ASAM criteria, multidimensional assessment criteria, it maps those using a computerized decision scoring algorithm into a set of service intensity bundles that are shorthanded as levels of care, which incorporate many different possible services that can be put together to meet the service intensity needs of that individual. And as much as possible, it tries to approach this in a sense of what is the best way to engage the person in a recovery-oriented process, and so are we looking at the person's strengths and capacities, as well as areas where there are challenges that may affect the right intensity of services for that individual. Next slide. There is a, we now have what we call the locus family of tools, through our partnership with the American Academy for Child and Adolescent Psychiatry. We derived the CALOCUS from locus, and now we have a new sort of evolved partner tool with ACAP called the CALOCUS-CASI. And then ACAP developed an early childhood, zero to five version, called the Early Childhood Service Intensity Assessment in 2009. Next. So, all of these have six dimensions of assessment, multidimensional, and a six-level resource continuum. As I said, they're transdiagnostic with a scoring algorithm. And increasingly, these have been recognized in both state and currently in proposed national legislation as kind of the standard of practice for these kind of service intensity decisions. So that in California and Oregon right now, active legislation requires that these are the tools, along with ASAM, for mapping into the addiction continuum by insurers and others for determining appropriate service intensity placements in the utilization management framework, as well as in the clinical framework. Next. The six dimensions are risk of harm, how dangerous or potentially harmful to self or others, how dangerous or potentially harmful to self or others, functional status, the person's ability based on their mental health and or substance challenge, to function, you know, to function in their life, to do normative activities, the presence of comorbidity or complexity in relationship to the presenting issue, to what extent is that confounded by active medical or co-occurring mental health and or substance issues, what's happening in their environment, what's the degree of stress, and conversely, what's the degree of support, what's their history, to what extent have they responded well to certain kinds of interventions previously, and to what extent are they actively engaged as a recovery partner in the work, or to what extent are they disengaged and therefore needing other kinds of interventions to help them be safe. Next. So, the way, it's a very easy to use tool. It's intended to be used, you know, by individuals, evaluating individuals. Each dimension has a five-point rating scale. You can rate from one to five. Within for each rating, there are set of anchors that describe characteristics that might put somebody in that. You pick the highest one that fits for any particular dimension. And then, there's some summation and then a scoring algorithm to help determine the quote-unquote composite rating that leads to a recommendation for which quote-unquote level of care is needed. Next. These are what are called independent criteria. And what that means is, if the person scores high, let's just say risk of harm, then no matter what else is going on, it requires a certain level of intervention. And what we're going to show you a little bit later apropos of the involuntary commitment is subtle perceptions of how you rate can create some bias in the structural racism sense that may lead to individuals who are Black, say, being viewed as more dangerous on the average even in similar circumstances than people who are not Black. And that could lead to disparities in how involuntary commitment is applied. And that kind of data, as you'll be seeing later, can really be informative to the quality improvement process for addressing anti-racism around that issue that the SMART tool is intended to address. Next. So, there are six quote-unquote levels of care. The basic level is what's called recovery maintenance, which is kind of an ongoing support level of care for folks who have had previous active episodes but are now essentially stable and just need ongoing support to maintain themselves. Level two is the lowest level of active care, which is basically like outpatient. Level three, high-intensity community-based, is some version of intensive outpatient, intensive case management, that kind of thing. Level four, medically monitored non-residential, is a much more highly structured and intensive community-based service, like an assertive community treatment team or an integrated dual disorder treatment team or a partial hospital program. Level five is a medically monitored residential setting, which can have varying degrees of short, medium or long-term treatment capacity and rehabilitative capacity. And level six medically managed is essentially equivalent to a hospital, although in many states, there are hospital alternatives that fit into the level six criteria as well, like a psychiatric health facility in California, for example. Next. So what we're gonna do now, this is just the basic introduction. As we posted in the chat, you can go to the communitypsychiatry.org website for AACP and get lots more information on all these tools and take them home and have fun with them and play around with what you got. But for right now, we're gonna illustrate the application of this in a quote-unquote real person, but in a definitely real system. So I'm gonna hand it back to Dr. Talley. All right, thank you, Dr. Binkoff. So I'm going to kick us off just sharing a little bit of a hypothetical case, but a case that I imagine may resonate with many in the audience as far as being drawn from sort of real life, real world situations of those who interface with our crisis services to start to dive into thinking about how we might apply these tools and utilize them to think about inequity and how to address inequity, identify and address inequity on a structural level. So this is basically to remind from Dr. Czajnka's portion earlier, this is the smart item that we will be sort of thinking about and focusing on with this case, specifically the issue of involuntary treatment orders. So this case, again, hypothetical case, we have a 27-year-old male brought in on an emergency involuntary evaluation order in the city of Philadelphia where Dr. Czajnka and I work and practice. This is termed as basically a 302 based on basically the state code terminology due to bizarre behavior, making vague threats. This order was petitioned by a family member after the patient expressed a wish to, wish to, quote, unquote, get back at them. And this patient at the time of interview expresses several concerning symptoms, anger, paranoid thoughts, hopelessness, also reports when prompted several symptoms concerning potentially for depression, appetite, weight loss issues, insomnia, and fatigue, and also reports struggling with some issues with substance use, has a history of some prior interface with the acute treatment setting, as well as some challenges adhering to traditional outpatient level of care. This young gentleman does not have many significant active chronic problems, although some prior surgeries, has a history of use of several substances for the past year, primarily cocaine in a binge pattern, also some issues with drinking alcohol, five to six beers every few days. With his last drink yesterday, he reports some occasional use of marijuana as well, reports a few episodes of outpatient care to address some of these challenges, but reports no significant periods of abstinence in the past 10 years. And then lastly, to review some of the social stressors as well as this gentleman's presentation. And so he was, until recently, living with his wife. However, after a dispute, his wife has recently left him, certainly a major stressor. Also recently lost his employment, and there's a concern around an issue of his employer potentially pressing charges due to an accusation of this gentleman stealing at work. His wife has told him that she will not return until he gets some help with some of the, addressing some of the challenges that he is dealing with. Also is estranged from what we might imagine could be some primary social support, his other family in the area, due to issues with unpredictable behavior. And then we have his mental status exam. This gentleman presents disheveled, dysphoric. He does not basically report any prior suicidal or assaultive behavior, shows some limitations and insight into his situation, expresses hopelessness, but makes no specific active threats in terms of desire to harm himself or to harm others. Recognizes that he struggled with treatment success in the past, but has some ambivalence towards reengaging with treatment, and then cites his key goals as reuniting with his wife and obtaining another job. And so if we were to think about this case through the frame of the locus, I have presented here basically, if we were to apply the locus dimensions to this case, what a hypothetical score might be. And we see with the score, I want to focusing specifically on the risk of harm dimension that a case like this should ideally have a score per basically that score is no active suicidal or homicidal ideation, but extreme distress and a history of prior behaviors in this vein exists. And so that might objectively fit this case well. However, what my colleague, Dr. Sriyanka is going to illustrate with some data from Philadelphia, basically looking at this process of involuntary commitments and involuntary treatments, is that our systems are often subject to certain implicit biases, where despite having a certain array of sort of objective information, as far as a presentation, we find sometimes in terms of the end result, in terms of how a person is assessed, how risk and danger is assessed and sort of the subsequent steps in terms of how that person is treated, that implicit bias can absolutely come into play. And what leads us to that conclusion, potentially it can be these noting the existence of disparities in terms of how risk assessments and sort of the following involuntary treatment commitments are applied depending on a person's racial or ethnic identity. And so with that, I'll turn it over to Dr. Sriyanka. Thank you, Dr. Talley. So this data that I'll be sharing is based on work that has been done in Philadelphia. For context, Philadelphia began a system level soup to nuts crisis transformation effort back in 2019. That began with the mapping of our system data, what data was available. And very interestingly and appropriate to this conversation was that one of the key data points that actually triggered that transformation effort was looking at the emergency involuntary commitment orders in Philadelphia. And obviously we are fortunate that we have a pretty robust database and we're able to interrogate that database. I will come back to that particular item. It's the 302 form that Dr. Talley mentioned earlier. Now, 988 is a national transformation effort involving crisis systems across the country. One of the pieces of data that we track in Philadelphia is who is calling our 988 line, essentially a call center. And it's very interesting when you look at the data from a systemic level. Now, this is a little blurry, but if you look at the pie chart on your left, on the left, it's essentially crisis call data broken down by race ethnicity. So you can see that the bulk, almost 68% of persons calling the crisis line are black, identify as black. Now, for additional context, Philadelphia is a city that is challenged by many of the same socio-economic, socio-cultural factors that many of the large cities have to deal with, including the fallout of decades of economic instability and disinvestment in certain parts of the city. So that's important for context. Essentially, you can map on to the map of the city and indices, what I call indices of distress. You can actually see that they map onto areas where historically they had been redlining. And those areas tend to have a disproportionate representation of black and brown people. So when I look at this call volume, this call data, about 40, 44% of Philadelphians are black, identify as black. Compare that with 67% of callers to that call, to those call lines. Now, this raises questions for me by contrast, 21% are white and you can see the breakdown by other race ethnicity groups. Well, what this raises for me as an administrator is a number of questions. Number one is why do we have so many black identified, self-identified black persons calling our call line? On the one hand, it's a good thing that they're calling our call line because it means they're getting some level of assistance. On the other hand, it begs the question, do people know that other services exist that they can access? Is there a question of trust in the treatments that they expect to receive from those services? Is that why they prefer to call the call line rather than attempting to access the other services? Or is it some other reason? Is it that this is simply easier to do rather than trying to go to a crisis center or crisis response center or some other level of care? So this raises a number of questions. One of our responses in Philadelphia has been to commission a whole scale and that study of Philadelphia by neighborhood and really assess what people knew about services, mental health services in general. And what we found is that one of those hypotheses actually rings true. People do not know what services are available to them in Philadelphia. And what that has resulted into too is a essentially social marketing campaign around specifically 988, but also around other mental health services that are available in Philadelphia. So that's one example of how this kind of data is useful, how it can be useful to planning and targeted interventions. Next slide. Now, this gets us to the data that we've mentioned several times. And this is essentially a breakdown of persons to whom 302s have been applied. The 302 is, again, it's the Emergency and Voluntary Commitment Order. And this breaks down how that tool has been used across Philadelphia. Now, let's start with the right, the column on the right, on the far right here. And you can already see that from the beginning when you look at Medicaid eligible individuals, there's a disproportionate representation of black individuals. This is 50% of black people on Medicaid in Philadelphia are black, identify as black, as compared with the representation in the overall population, which is about 40, between 40 and 44%. So there's already disparity there. When you look at the middle column, middle two columns, that is the individuals who have been served a 302 or have been treated under an involuntary commitment order, involuntary emergency commitment order, that rises to 57%, almost 58%. So more black people, relative to their representation in the population, relative to other race, ethnicity groups, are experiencing being served 302s, which is a cause for concern. And I'll tell you one of the reasons why it's a cause for concern. Philadelphia, in some of our other research, we've identified 302s as associated with worse outcomes clinically. Specifically, one of those worse outcomes for certain populations is suicide, death by suicide. And so this is a concern. The other concern is that people, when they get 302ed in Philadelphia, the process includes that 302 being served by a police officer. So if a police officer, and we'll see this in the next slide, gets involved, even though we have CIT-trained officers and police do obviously the best they can to process these in as gentle a manner as possible, the risk goes up. And case in point is Walter Wallace, who many of you may be aware of, familiar with the story, young black man. In West Philadelphia, one of those communities that was historically subjected to redlining was in crisis. His family called for assistance several times that morning. They eventually got a police response. He was killed by police in front of his entire family within a few minutes of their arrival. So 302s are, in our system, in our transformation, one of our values is to reduce the use of 302s because they are experienced as traumatizing by the population. And this has historically been applied disproportionately to black people as this data shows. Next slide. Let's delve a little further into this. So when you look at the reason, the justification for 302s, and let's think about the case study that Dr. Talley just read out to us. There's some question, concern about potential threats to itself or others in there. But when you look at the overall data and the reason people with 302, when you look at this column over here, risk of harm to others, which is what kind of gets everybody concerned. You see that the representation is 67%. So of every hundred 302s, when there's a concern about potential harm to others, 67% of those persons are black. Let me put that into context, further context for you. When a police officer receives a dispatch and they are told this person that you're, this person 302 has been completed, you need to go serve this person. When you talk to officers, it's a very, and to the dispatch team, there's a very different, the tone of the dispatch is important. If the stated reason for the dispatch is we're going to protect the community from this person who is potentially danger to other people, then that already sets the tone for a different kind of encounter than, well, we're going to help somebody who may not be able to care for themselves or may be at risk for death by suicide. So this is important stuff. I'll just wrap with one very quick scenario. We had a young individual in Philadelphia, a nine-year-old who had already been traumatized, just kind of making, really driving this home. This experience, they were already traumatized, they were in a treatment setting. And at the time, this is before Philadelphia's transformation efforts began, we didn't have very many options. And so the only way to get this child into treatment was to 302 them because their parent wasn't available and so on. And this young person ended up in cuffs at the back of a police car and was traumatizing to everybody involved, to officers of settling the child and our crisis call center manager who had to try to figure out that situation and certainly to the clinic staff that were involved. And that's one of the reasons why this has been a focus. So I'll stop there and I'll hand this back to Dr. Talley. Thank you. All right, so at this point, we just a reminder to those on the call to submit your questions via the Q&A area of the attendee control panel. And we are happy to take questions as well as hopefully have some discussion amongst the three of us. So, could I toss a question to Dr. Shoyinka? Absolutely, absolutely. Yes. Yeah. So, in the spirit of this presentation linking the objective use of the locus to trying to improve the potential, the disparity that you've demonstrated showing your data, how might you be able to use the locus as part of an objective quality improvement effort to reduce that disparity in terms of the perception of risk of harm to others for black people among your system evaluators? I think that's an excellent question. So, one way that that could be done is would be in a similar fashion to how this data around 302s has been analyzed and is being used now, is to take a sampling, a blinded sample of persons who have been served in our system at whatever level, crisis center level, outpatient level, and apply the locus to obviously create a panel of assessors and apply the locus to those cases, have the panel apply the locus to each of those cases or case scenarios, and then unblind the sample with regards to race, ethnicity, and then take a look at what the scores show. So, basically, let me tie that to an actual example. So, it could be that we take a look at persons who ended up going to the crisis center, people who may have called from the community or people who may have been seen in the outpatient clinic. Let's say we selected 100 of those people or 50 of those people and then blinded those cases, those records, by ethnicity, and then we would potentially, in this case, convene a group of assessors who are trained to use locus and then look at what happened to those individuals. What was the decision made as far as to the next level of care? And that would give us, I think, some very useful data about, well, how do we make decisions about the next level of services for people? So, that would be an example of how to use the locus on a system level between levels of care. Yeah, I think that's really very interesting. And part of the objectivity of the locus, of course, is that you can evaluate objectively both the level of care that's being recommended as well as if you do a deeper dive, you can look at which anchors are selected that contribute to the scores that lead to that level. And so, there are some anchors that are more in the harm to others domain and some that are more in the self-harm domain. And it can kind of give you a deeper dive of how people are selecting and then that can also affect how people could be trained to use the tool in a way to counter those potential biases. Another issue that comes up is if you're using case presentations, and you know this, and it's just this is a softball to either of you, but it's like even how the thing is written up can generate the bias even before the person does the rating. And then you kind of build that in, you know, in terms of how the police, the radio call goes out to police. But in the case presentation, you know, 27-year-old African American male who's behaving in a threatening manner versus 27-year-old white male who's behaving in an agitated and anxious manner. They may be doing exactly the same thing. But the presentation actually starts right from the beginning. So, it becomes an interesting exercise for all kinds of people. How do you describe what you see? I'll jump in. I think it's such a great point. And I think what I see is almost tying into what I like sort of one of the take-homes of this presentation of how objectivity and data-driven processes, while they aren't necessarily sort of the cure-all for inequity, I think they really can sort of push an objective lens that starts to address implicit bias. Because exactly to your point, you know, acting in a threatening manner, what does that mean? Could this case have been reshaped to sort of state exactly what was it that this person was doing? What was the action? That already when we say threatening, that's a subjective interpretation of what somebody has observed and taken in. And I think particularly in psychiatry, where in part because ours is a field where we are still formulating a sense of our diagnoses of the behaviors and of the observations, we're so dependent on that subjective observation of the provider taking in, you know, appearance, behavior, these different facets of a person to come to an assessment and describing those things, that it's so critical that we have these sorts of, I think, tools like the LOCUS, like the SMART, we're really talking about. But would it objectively, what exactly is going on here? What is the exact behavior? How do we delineate this behavior versus that behavior to start to try to fight against some of the implicit biases that can come in with a subjective provider impression? And I'll just piggyback onto that within the child domain. One of the areas that the SMART tool talks about is the school-to-prison pipeline. Well, one of our other board members, Dr. Sarah Vincent, has done a tremendous amount of work. She's a child and forensic trained psychiatrist, practices in Atlanta at Morehouse, and just became the chair. Shout out to Sarah. Well, she has oftentimes remarked on this very thing that Dr. Minkoff just mentioned, which is how the same behavior oftentimes in kids who end up in the juvenile justice system is described in very, very different terms. So we know that trauma is highly prevalent in that population, but very often it's never even considered as a potential driver of behavior until somebody goes and asks. And then those kids end up really being punished when they should be getting treated. Let me mention one other sort of related anecdote here, which is what I mentioned early on in my portion of the presentation. One of our colleagues within the NAACP, Dr. Blackman, again to this very point, talked about early on when SMART was new and had just been put together, talked about how she just took a step back one day, and at that time she was supervising a panel of nurse practitioners. And she noticed that, very interestingly, all the young Black men were diagnosed with bipolar disorder, and they were, excuse me, excuse me, they were diagnosed with bipolar disorder, and every one of them was treated with Seroquel. And then when she took a step back and queried those diagnoses with the nurse practitioners under her leadership, they found the diagnosis was not infrequently altered to something else, and the treatment plan was not altered. And when you think about the long-term potential complications of unnecessary antipsychotic treatment, that is not insignificant. Absolutely. Absolutely. Appreciate you sharing that last example, sort of how does the individual provider hopefully can be empowered by this tool to identify inequities even at that level of individual practice, right? And lastly, within this Striving for Equity series, the next presentation will be on August 10th, 1 p.m. Eastern Standard Time, Dr. Gilbert Bastien presenting Between Two Worlds, Promoting Mental Health Among Migrating Populations. And thank you for your attention this afternoon.

SMART

LOCUS

equity

access

mental health care

structural racism

disparities

involuntary treatment orders

diverse teams

quality improvement efforts