Today we're going to be talking about, so our presentation is about changing the trajectory, innovations in first episode psychosis to reduce the risk of suicide, violence, and legal involvement. I'd like to start with some introductions. So my name's Deirdre Caffery. I'm a psychiatry resident, a PGY3 resident at Columbia. I'm joined today by Dr. Nossel. She currently serves as the co-director and medical director of OnTrack New York. OnTrack New York is a coordinated specialty care program for early psychosis. Dr. Nossel is an associate professor of clinical psychiatry at Columbia University. Joining us actually virtually is going to be Dr. Stephanie Rowland. She is a forensic psychiatrist and physician scientist at Columbia University where she studies issues related to violence during the early period of psychosis. And Dr. Michael Compton is triple boarded in psychiatry, preventative medicine, and lifestyle medicine, and is a professor of psychiatry at Columbia University and a research psychiatrist at the New York State Psychiatric Institute. We have no relevant disclosures. And just to orient us all together, today our objectives are formulating the risk of violence, suicide, and legal involvement in the prodromal and early stage of psychosis, apply novel approaches to identification, risk assessment, and timely intervention for patients experiencing early psychosis, consider how innovations in practice setting can better meet patients' needs, and review cases of early psychosis together and recommend new ways to reduce high risk behaviors. So this morning I'm wondering if to start, each of you could think about a patient that you have that you're worried about that has, that's in the early stages of psychosis and you're worried about their risk. In thinking about that patient, either a patient you have right now or one you've worked with in the past, what were you most worried about? Were they at risk of suicide? Were they at risk of violence? Were they at risk of being arrested? How did their symptoms impact their risk? What were their protective factors? And were any of their risk factors modifiable? So we can all think together. I'm gonna present a fictional case that's based on a lot of clinical experiences about this patient population. About an 18-year-old male who previously was living with his mother and one younger sibling. He completed the 11th grade of high school but stopped attending school over the past few months. And he was diagnosed with schizophrenia after his first hospitalization a year ago at age 17. He has a history of suicide ideation over the past year and two past suicide attempts, both while experiencing paranoia and auditory hallucinations and both leading to hospitalizations. He also has a history of self-injury by cutting. He has a history of aggression, primarily throwing objects and punching walls, but recently had also had episodes of violent behavior towards his mother, including pushing and choking. And in the most recent presentation to the hospital, the patient was brought to the emergency room under arrest after threatening to kill himself and choking his mother. His mother pressed charges against him, assault in the third degree, and his mother now has an active restraining order against him. After this incident, he was hospitalized. He was restarted on a long-acting injectable antipsychotic and on discharge from the hospital, he was connected to OnTrack New York, an early intervention service site. He is now living in the shelter system and he has very few current social supports, but he does have a grandmother that sees him occasionally. At his intake appointment at OnTrack New York, which was immediately following his hospitalization, he states that he has schizophrenia. He knows he was started on a shot for medication and he thinks this might be helpful. He recalls the events leading up to his hospitalization, recalls being very worried that his family was in danger and that his neighbors were bothering their family. He reported he didn't mean to hurt his mother, but he was really mad that she wasn't listening to him and that no one believed him about the neighbors. He denies current auditory hallucinations, denies paranoia, denies suicide ideation, and on mental status does not appear internally preoccupied. He's future-oriented. His goals are to find a job and to find his own apartment, and he's pending a court date regarding his legal charges. Reading this case as a resident, I can say that this would be somebody that I would be very worried about, and I'd be very worried about taking care of them in the outpatient setting. I would be worried about the risk of suicide, risk of another violent episode, and risk of further legal involvement. And I wonder if any of the cases that you all had in your minds regarding patients that you're working with had some overlapping risk factors. So at this point, I'm gonna turn it over to Dr. Nossel to go over first episode psychosis and suicide risk. Thanks very much. It's great to be with you all today. I'm struck by how many people are here at 8 a.m. to talk about suicide and violence and legal involvement with young people with early psychosis. These are some of the most challenging things we face, and I know my heart rate kinda went up just hearing that example and thinking about my own experiences working with young people and those of others. And I think one thing that I hope we can all carry with us is that there's a lot of hope in this work. We need to think about risk and think about how to mitigate risk, and that's in the service of supporting people in living lives that are meaningful and pursuing what they wanna pursue. And so I think we always have to keep that hope in mind even when we're grappling with risk. So let's start a little bit with the statistics. We know that schizophrenia is, those with schizophrenia are at elevated risk for suicide, and the early psychosis period is a period of particular risk. And when I see these statistics, I'm reminded of a young man I saw early in my training when I was working on an inpatient unit that specialized in working with people with schizophrenia. It was a 22-bed unit, and I was working with a young man who had recently been diagnosed with schizophrenia for the first time. He was in the hospital for the first time after experiencing his first episode of psychosis, and he was learning about schizophrenia and reading about it, and he came to me one morning on the unit, and he said, I've read that 10% of people with schizophrenia die by suicide. And I said, yes. And he said, that's two of us on this unit. And that has always stuck with me when we think about sort of the percentages that it's really the people that we work with that that really affects. And so I carry that with me. I wanted to go through a framework of understanding and thinking about the risk factors for suicide and risk factors of violence because there is so much overlap. I know I was trained in the biopsychosocial model, which I think is a very useful model, and we've needed to since expand that model in terms of thinking about the contributors to a presentation. And I'll walk through it a little bit, commenting on the example we just heard about. So there's psychopathology, psychiatric symptoms of psychiatric disorder. And we heard in this young man about his paranoia and fear. We don't know a lot about that, why he was afraid of the neighbors, what his fears were about them. And we know he was diagnosed with schizophrenia. We don't know enough to know if that diagnosis was an accurate diagnosis, if there's more. But we would need to understand more from him and from those who know him and the records. There's psychological factors, including coping skills, how one responds to different stressors, and how one feels about oneself. And here, we know some things from this young man. We know that he has a history of cutting, and he has a history of suicide attempts. So when the stress gets very high, we know there's a history of acting on those, also a history of acting in aggressive ways. We don't know about self-esteem, but we might really sort of suspect some things and want to probe further to understand how is he feeling now. We know he said he didn't mean to hurt his mother. How is he feeling about the fact that that happened and his sense of self around his own experience, around this diagnosis, around having left school, around being homeless? So there'll be a need for a lot of attention there and a lot of concern there in terms of how these things could impact him. Social factors, we know he has limited social contacts. He's now estranged from his mother. We wonder about loneliness. And we don't know about adverse childhood experiences. We know very, almost nothing, I think, about his childhood. So that would be really important to understand. What happened sort of before all of this? Who is he? What has his experience been? And has he had adverse childhood experiences? We don't know. For example, if he had a father involved and if there were losses, we don't know anything about sort of his earlier history. So that would be important as well. Cognitive factors. Again, we don't know much here, but we have a few things to go on. We know he was in school until 11th grade and stopped going. We don't know, was he in regular education, special education? Does he have a learning disorder, either identified or not identified? And also, what cognitive factors might relate to the onset of the psychotic illness? And then biologic factors, we don't know about that as well as their genetic risk. And of course, structural factors. We know some of the current. We know he has homelessness. We know he has legal involvement. We know he's unemployed, not in school. But we don't know about the structural factors that may have kind of been part of his life prior to the onset. This is some data from OnTrack New York. So as Deirdre mentioned, OnTrack New York is New York State's coordinated specialty care program for early psychosis. Just a word about that in case people may not be familiar. It's a multidisciplinary, team-based approach to work with young people. In our case, within the first two years of a non-affective psychotic disorder. The OnTrack New York program has been in place as a clinical program for the past 10 years. It's grown from four programs to now 27 programs across New York State and growing. We collect data for the purposes of quality improvement and program evaluation. And so this is some of our data related to suicidal ideation. This shows that a quarter of people coming into the program have had suicidal ideation in the three months prior to enrollment. That drops after enrollment about 14% at the three month point and then levels out around 10 or just under 10% every three months following there. And we'll say a colleague of mine recently published looking at trajectories of suicidal ideation. And we know there's sort of heterogeneous trajectories. There are people who have suicidal ideation that resolves after they come to treatment. And that's sort of hinted at in this graph. There are the people who have persistent suicidal ideation and we see that too. And then there's a group that does not present with suicidal ideation and the onset of that is following enrollment. So, and that you sort of miss in this data but important to be aware of. This is our data regarding suicide attempts. You can see in our program 6% of young people have had a suicide attempt in the three months prior to enrollment. That drops to about 1% per quarter after enrollment. Since On Track New York began, we've enrolled more than 3,000 people. We have about 800, upwards of 800 enrolled now statewide. And so you can imagine that 1% of that we see suicide attempts. I wanna bring up a concept that we've used quite a lot in On Track New York that I think is really helpful to our understanding which we refer to as unintended self-harm. This is harm that occurs in the context of psychosis but that is not intended to cause harm per se. So this is different than non-suicidal self-injurious behavior where someone might be cutting to feel something or to relieve pain or where the intention is that they're acting to harm themselves with that as the purpose. I'll give some examples in the context of psychosis. Cutting oneself due to believing that someone has a recording device implanted and trying to get it out. Drinking bleach to cleanse oneself due to command hallucinations to do so. Jumping off of a building to make the voices stop or walking into traffic due to disorganization. And I give these examples because if you ask about suicide if someone has had thoughts about harming themselves, thoughts about not wanting to live, thoughts about suicide, someone experiencing any one of these might say no to that. Including if they'd acted on these and you ask them about a history of suicide attempts they may say no about those. And so you wanna make sure when you're asking that you're doing it in a really comprehensive way that asks about any thoughts or behavior that could lead to harm or that someone else might have been afraid might lead to harm. And to really try to find a language with a person you're working with to refer to their thoughts that resonates for them or that resonates for them in terms of their past behavior. An example of this in my clinical work was a young woman who came into our program following jumping from a fifth story fire escape in the context of psychosis. When I had asked about suicidal ideation, suicide behavior she said no, she'd never thought about suicide or tried to take her life. When I learned about this jump she said and I asked about what she was thinking at the time she said I just thought it was the best next thing to do. It was not done with the idea that she would die and she's still, I think this is some 10 more years later, does not refer to it as a suicide attempt she refers to it as an accident although she did not slip. So that's the language we use in talking about it together because that's the language that resonates for her even though clinically I understand it as a suicide attempt and think about it in those ways. So I want to talk about the model we use within On Track New York related to assessing and intervening, the AIM model. This is a model Barbara Stanley developed and trains on. Barbara Stanley, for anyone who does not know, is a professor of psychiatry at Columbia, suicide prevention expert. And unfortunately we lost Barbara Stanley earlier this year when she died too early. She just had a tremendous effect on the field in terms of her work in suicide prevention and has had a huge impact on our work in On Track New York. So this AIM model stands for, it's very simple, it's a framework to support sort of making sure you have a systematic way of approaching assessment intervention and monitoring. So I'll talk about the pieces. Within On Track New York we broke assessment down into three components. Doing structured screening, which I'll talk more about in a moment, doing a comprehensive assessment and then ultimately summarizing that. Then intervening, which is planning to mitigate risk, and monitoring over time. And I'll go into all of these pieces. So structured screening, this is the use of a structured screening tool. We use the Columbia Suicide Severity Rating Scale. You may use that one, you may use another. The idea is to do universal screening with all participants in the program, both initially and over time. When you're doing that initial screening, ensuring you're asking both about lifetime as well as recent, and at follow-up asking about since last contact. This should be done as frequently as clinically indicated, so that might be weekly for people at elevated risk and at a minimum quarterly for those who might not be at higher risk, although I would say people who come in early on are at elevated risk, and so it should be done more often in the beginning. In this group of young people with early psychosis, screening is not enough. When you get a negative screen, that is not sufficient as an assessment in terms of suicide risk. So you really need to do a clinical risk assessment with everyone. That's because there's risk factors that are specific to early psychosis, and early psychosis itself is a very significant risk factor. So screening is useful but not sufficient. The goal of a comprehensive assessment is to identify risk factors that are both specific to the individual and environmental, both recent and distal, and the same with regard to protective factors. When you're doing that assessment, certainly include the input from the participant in terms of your interview, but also any other team members working with the person as well as family may be able to share really important things. An example in our work is the peer specialist may have a very different relationship and hear about different things than the psychiatrist may, and so really important to make sure you're getting input about risk and protective factors for anyone who might have information to share. I won't go through all of these risk factors, but I'll walk through it a little bit with the example that we were just talking about, and we have many here. He's someone who has a recent suicide attempt and self-injury and recent suicidal ideation. He was recently discharged from the hospital. He's within the first three months of outpatient treatment. We don't know about whether he has a history of engagement or disengagement in prior outpatient treatment. We know he has recent aggressive behavior. We don't know about hallucinations, and we know he had recent positive symptoms, but those have improved with the LAI. We don't know about most of the remaining risk factors listed here that are individual-specific proximal risk factors. We know he's socially isolated. We don't know about his sense of hope, although he's planning for the future, so that gives some sense, and we don't know about substance use. We know he has prior suicide attempts, and there are many things here we don't know. We know he had an early onset of psychosis prior to 18. We don't know about his sexual orientation. We don't know about his functioning prior to or following the onset of psychosis other than dropping out of school. We don't know about trauma history or family history or mood symptoms. In terms of environmental risk factors, those are high as well. With this young man, he is homeless, and we know about his legal charges. We don't know about access to lethal means. That would be very important to ask about. We don't know a lot about his protective factors. That will be really critical to explore, and we also don't know about his strengths. What are his reasons for living? We know he wants a job and an apartment. We don't know what's important to him, what matters most to him. That would be critical for us to understand. We know he finds the medication helpful. It would be really good to understand more about that. How does it help, what is he noticing about that? We know he has some contact with his grandmother, so we'd want to explore, is that a protective factor for him? So once we gather all of that information, we just want to make a summary statement, kind of putting it all together, how do we think about risk? And so for this young man, I would say he's an 18-year-old man with a history of prior suicide attempts and self-injurious behavior, history of aggressive behavior toward his mother, recent arrest, his mother has a restraining order, he is currently homeless, not employed or in school, socially isolated, and so he has many risk factors and is at elevated risk. He is hopeful about the future, psychotic symptoms have improved, and he feels medication has been helpful, we have to do much more to understand his protective factors, and so he would certainly be on an elevated risk list. This is where you really want to be collaborating with your team if you have one, with those who are at elevated risk to discuss how to develop a plan to mitigate risk. So there are some things that should be done with all young people in an early psychosis program or all young people with early psychosis in care, and then certain things that should be done particularly for those deemed to be at elevated risk. Psychoeducation about risk for the young person and for the family or those involved should be provided to everyone. As I mentioned, there's that trajectory of people who haven't had history of suicidal ideation or behavior, and then there is the onset following entry into treatment. So everyone needs to know there is elevated risk with this condition. You want to be talking about access to lethal means and taking steps to restrict access for everyone in the program and providing information about crisis and how to respond to crisis. And then there are specific interventions you want to engage in for those who are particularly at risk. And then I'm going to talk in a moment about the safety planning intervention. Again, we use the Stanley Brown safety plan, which you may be familiar. That is named for Barbara Stanley and her colleague. Developing a crisis narrative. This is something that's important to do, and we really want to understand for this young man what were the circumstances of his prior suicide attempts, what was going on for him at that time, and also about this most recent incident that involved the aggression toward his mother as well as him thinking about suicide again, understanding what was contributing, what led him to act. And this is something that often needs to be done, developed, revised over time. I understood much more about the narrative that led to that participant I mentioned earlier who had jumped from the fire escape. That was something, a crisis narrative that developed over time, over many months. So this is not a one-and-done situation, but something to revisit over time as you develop trust, you get to know the person, they have a sense to reflect, they may have different perspective. You want to provide education about what a safety plan is and how to use it and complete the safety plan together, including means reduction counseling. I'll show it in a moment. One of the most important things is the safety plan needs to be in the hands of the young person. So it's great to have a copy or a reflection about it in the chart, but the most important thing is that it be in the hands of the young person, whether that's something in their wallet, whether that's using their phone, there is an app for safety plan which you can use and the person can download and enter indirectly. If someone has supports, they're open to involving in safety planning, that can be incredibly helpful and an important part is identifying what are their early warning signs. And again, that's often something you're doing over time and in your first sort of efforts on developing the safety plan. There may be less and then that develops further over time as they come to identify those. The URL for this is on the bottom. If this is not something you have, this is the Stanley Brown safety plan. It's a stepwise intervention, but the person can go to any step as they need to. So the idea of step one is identifying their early warning signs or triggers and identifying what internal coping strategies can they use to provide distraction. And so this might be a really important thing to develop with this young man. What are his internal coping strategies? How can you work on further developing those together? The next place is identifying people and places that provide distraction. So if you're feeling upset, distressed, thinking about suicide, what can help with distraction? These may not be people you're turning to for help, but they may be people that you want to hang out with or spend time with or can distract you. Same with places, places that provide some distraction. Step four is who can you turn to and ask for help? And there may be people, for example, that you're okay to spend time with, but you don't want to ask for help. This is who would you ask for help with? And again, sometimes in the beginning, there's not a lot to write here, and this may be something you're developing over time, helping someone strengthen their support system. Places or agencies they can contact. I think this needs to be updated now with the 988 number for suicide prevention hotline. And step six is critical, which is making the environment safer. What are the plans for lethal means restriction? Whether that's access to a firearm or knives or medication or other strategies, what steps can be taken? And again, really helpful if someone is open to it, to involving family or other supports in this step as in many of the others. Other pieces in addition to doing the safety plan are increasing the frequency of visits, doing assertive community outreach for those who have disengaged. If this young man wasn't coming in, this is certainly someone we would want to go into the community to try to find and try to reengage in care. Increasing the engagement of family supports. With this young man, we might start with the grandmother. Right now, mom has an order of protection, but we may explore over time, you know, what is possible there. And then thinking about interventions to address modifiable risk factors, including optimizing medication, considering clozapine for suicide risk, and the other psychosocial interventions that can support risk. Psychotherapy, again, focused on developing coping skills, responding to suicidal thoughts or responding to psychotic symptoms that may be a trigger. Case management, that's critical, right? Helping this young man attain housing can be an incredibly powerful suicide prevention step. Peer support, if relevant, substance use treatment. Enhancing protective factors. And then you always have to be thinking about, you know, is the level of care right here? Can we navigate this risk in the current level of care that we have? Is there any need for consideration of a higher level of care when it's not possible to create a safety plan, a safe plan for outpatient care? This is, of course, an ongoing process. We need to monitor risk over time. As a colleague of mine has spoken about, sometimes people are one stressor away from a suicidal crisis, and so you want to really be mindful of that, educating them and their supports around that, and be aware that, you know, having a plan for if the shelter should fall through, if something with housing, something with schools, a relationship breakup, whatever it might be, how do you increase the supports to intervene in that time? And monitor risk frequently. Obviously, that depends on the acuity. Sometimes it's every session, and just having a low threshold for reconsidering risk. I have a few moments, so I do want to also talk briefly about after a suicide. I don't think we talk about this enough in psychiatry, that if you do this work, you are likely to have someone that you lose to suicide, and I think it's really critical for us to talk about that, and so I wanted to speak about that a little bit. Being able to support the clinical team, the psychiatrist, the therapist, the peer specialist, whoever is working with the person, our employment education specialists, individually, as a group, immediately and over time, again, not a one-and-done kind of support. Some months ago, I had a clinician contact me about the loss of a participant that he worked with some six years prior, and the way that had stayed with him, the impact that it had on his sense of professional identity, his work with other patients. So really to think about that support as something you want to be providing in an ongoing way. Providing space for reaction, reflection, processing, and thinking about the impact on ongoing work, which it undoubtedly has. Providing an opportunity for the team to debrief together, and normalizing a variety of reactions. I would say often psychiatrists are in the role of themselves coping with loss and having to support colleagues through loss, and that's a sort of unique and special important burden, so the need to get support while you are experiencing your own grief and supporting others. There can be a variety of reactions, and important to name those and understand that different people may experience reactions differently, and it may differ over time, including sadness, grief, guilt, fear, anger, numbness, helplessness, despair. I often think about that given the elevated risk in early psychosis, suicide is not unexpected, even though we work so hard to prevent it, and that we often don't know about the ways in which our work saves lives, though sometimes not everyone. I also think it's important to convey that suicide is not the team's fault, and suicides can happen even in the context of excellent care. Important to connect with other clinicians who have experienced a suicide, or to share one's own personal experience, and provide supervision, sometimes with someone outside of the sort of clinical hierarchy is really helpful, strengthening, supporting the team. And then there's considerations of whether and how to support the family. They might include thoughts about whether to go to the funeral, whether to convey some of the same information and messages as I just talked about, offer to connect to resources such as a suicide survivor support group. Often that involves discussions with the agency leadership about whether and how to have contact with the family. Also other participants or families may know the participant, for example, if they participated in a group, or sometimes there's news coverage of what happened, and so there may need to be a consideration of whether and what information to share with other participants in the program, and that's important to be very thoughtful about and to seek consultation about. Emergency review processes should be conducted sensitively and not in a culture of finger pointing or blame. That's critical. And lastly, we do want to think about whether there is something to learn, and do we have systems improvements we can make. So I'm now going to jump to Dr. Rowland, who actually pre-recorded her session because she has a newborn at home. Hi. My name is Stephanie Rowland, and I'm a forensic psychiatrist and researcher at Columbia University I'm presenting remotely today because I am in New York City with my three-month-old daughter Marvell on parental leave right now, but I look forward to seeing everyone at the annual meeting in New York next year. What I'm talking about today is first episode psychosis and violence risk, which is my area of research, and our objectives for today are to discuss violence risk factors, try to understand how violent behavior changes during the course of psychosis, and then we'll touch on some violence risk assessment strategies as well. So violence risk. Why are we talking about violence at this conference and during this session? So individuals with serious mental illnesses have a small increased risk of engaging in violence compared to the general population. The relative risk of violence is about four times that of the general population, although the absolute risk of violence is very small. Violence does appear highest earlier in the course of a psychotic illness and then decreases over time, with current studies suggesting that up to a third of people engage in violence prior to initial presentation for psychiatric care when they're being diagnosed with early psychosis. So past research has looked at risk factors for violence, and that's included being in treatment involuntarily, having criminal legal involvement, feelings of anger and hostility, and research has also found weaker associations for risk factors like being younger, male gender, longer duration of untreated symptoms, and substance use. These risk factors are pretty similar to risk factors for violence for people without psychosis, and so it can be hard to use them to differentiate who is and isn't at risk. I'm going to take a moment now to touch on mass shootings, and the reason that I'm talking about mass shootings is that it's hard to have a presentation about violence and psychosis and not have this come up as a question that people have. And although mass shootings are increasingly common in the United States, and we see headlines about them nearly every day, mass shootings do account for a small fraction of murders. So a lot of headlines, but a small amount of murders that are occurring, and my colleagues did an analysis of over 1,300 mass murders that occurred between 1900 and 2019, and it found that about 90% of people who engaged in mass shootings had no lifetime history of any psychosis. And specifically in the United States, mass shooters are more likely to have legal histories, to use recreational drugs or misuse alcohol, and have a history of non-psychotic psychiatric or neurologic symptoms. So when I'm talking about violence during the period of early psychosis, I'm not talking about mass shootings, because for the most part, people who engage in mass shootings are people who have no history of psychosis. So what kind of violence am I talking about? The violence that I focus on that's more common among people with psychosis are small, repeated episodes of interpersonal violence, that each individually may seem more minor in terms of the amount of injuries that occur, but over time have a cumulative impact on that person's life, on their family and their community, and then on society at large. So when we think about caregivers to individuals with psychosis, between one-third to two-thirds of caregivers have reported at least one incident of physical violence or aggression within the past year. And for caregivers who experience physical violence or aggression in their caregiving duties, they tend to appraise caregiving more negatively, they tend to have feelings of hostility towards the person that they're caring for, and they tend to have worse mental health themselves. There's also a challenge in being a caregiver to someone when a crisis does occur, because what we know is that if you call emergency services for assistance, a person with mental illness is at increased risk of being arrested, of being charged with a crime, and of having a criminal record. We also know that there's been a number of incidents in the United States where a caregiver calls police for assistance, and the police responds and then shoots the person with mental illness in their own home. That's something that we have heard has happened. And so caregivers are often very challenged to ask for help. Although non-violent crimes are more common than violent crimes, adults who have schizophrenia are more likely to be convicted of a violent crime than their age-matched peers. And what we know about violence is that it contributes to intergenerational trauma, that a violent episode or history of violence in one family member can really have an impact on a family for generations. So one thing I started by talking about was that violence risk is highest earlier in the course of psychosis, with up to one-third of people engaging in violence prior to presentation, and then we know that it decreases over time. Well, how does it change? How does it decrease? And in order to understand this better, I conducted a secondary data analysis of young adults who were entering care at OnTrack New York in order to identify the prevalence of violent behavior at time of entry, when people first come into care, and then every three months after that. This analysis included over 1,500 young adults who were entering care in the OnTrack New York network, which has more than 20 clinic locations distributed throughout New York State. And the data that I'm showing today was collected between October of 2013 and June of 2021. And at OnTrack New York, there's a number of standardized forms that the teams fill out, and then these forms are the same throughout all these clinics, so we're able to look at this data as a whole. And the question that I'm focusing on is this question 30, looking at behavioral patterns reported from any source occurring the 90 days prior to OnTrack New York. And I'm specifically focusing on part C, which is violent or aggressive behavior. So a clinician sees this form, they do this every 90 days, and they say, based on my personal interactions with this participant, based on what I heard from their families, any medical records that I reviewed, everything known to me, is this a yes or a no, or do I not know? And so people go through and check that out every three months. So looking at the results of this, of these 1,500 young people who were entering care, about over 70% had no violent behavior at any time point. So not at entry and not at any time point during treatment. For people who had violent behavior reported by their clinical team during treatment, about 14% had violent behavior only at baseline, and then not subsequently during treatment. And then about 13% had violent behavior during treatment. And this includes people who had violent behavior at baseline or didn't, and the reason that we combined them was because this was a group that was thought to be at increased risk of violence during treatment as they're being reported by their clinical team to have violence during their treatment. So here we're comparing who these three groups are. So we have, all the way on the left, the group that has no violent behavior at any time point, this is the 70%. And what stands out here is that this group has less men, more women in it. In the center, we have the group that has violent behavior at baseline only, and this group seems to be more likely to live with their families and less likely to be working. And then we've got the third group, which is people who continue to have violent behavior during treatment. And again, these are as reported by their clinical team. And this group seems to be more likely to be homeless, more likely to be using marijuana. So what, taking away from this, most individuals in treatment at OnTrack New York never have clinician-reported violent behavior. So at no point in time is there any violent behavior that is reported. And of those who do have clinician-reported violent behavior at any time point, about half have it at entry and then not during treatment, and then about half have violent behavior that continues during treatment. Overall, women are less likely to have clinician-reported violent behavior than men in this sample. So compared to people who never have any clinician-reported violent behavior, those who have violent behavior at baseline only are more likely to live with their parents, less likely to be working. And those who continue to have violent behavior during treatment appear more likely to be homeless and more likely to be using marijuana. However, this is a secondary data analysis of data that was collected for a clinical program OnTrack New York. There's very limited use of diagnostic research tools, and violent behavior is this binary non-standardized variable. And what we know is as individuals, as a society, we have biases in terms of how we perceive violence. And we may be seeing that reflected in this data as we are relying on clinician-report. For example, we're seeing that women have less violence than men in this sample. And there have been some studies that have shown that maybe that's not quite the case, but that is our perception of violence. And so I think taking this with a caveat, it points us in some directions, but there's definitely more work that needs to be done to better understand what's happening with violence during the course of psychosis. So now I'm going to talk about risk assessment. And the reason I'm talking about risk assessment is that current treatment standards endorse violence risk assessment for all individuals with schizophrenia. It's something that as clinics were recommended to do. But there's no real guidelines as to how to do violence risk assessment. And there's a lot of different ways to assess violence risk. So all of these different models exist. And the most common model that's used is unstructured professional judgment. And this is basically your clinical judgment. What do you think the violence risk is as a clinician based on what you know? And although this is the most common type of violence risk assessment that's done, it also has poor inter-rater reliability and poor predictive value. So people have come up with these other tools. One is an actuarial tool. And these are tools that you answer the questions and then the tools tell you what the risk is so that there's no clinical judgment or anything that comes into play. And two people using the tool should come to the same conclusion. Structured professional judgment is a combination of these where there's some actuarial component that has a checklist perhaps to go through, but then also a place for your professional judgment. So clinicians like to use their professional judgment. And so sometimes they like the structured professional judgment tools more. What's interesting about these risk assessment tools is that they've been developed in certain populations. And then when you try to use them in another populations, their ability to predict violence might vary. And what we know is that some studies have found that these tools have worse performance when used among people with schizophrenia than for other diagnoses like perhaps antisocial personality disorder. So although these tools exist and they're out there, it's hard to know how well do they perform for someone say with early psychosis. So in a pilot study that I had, I implemented multiple different approaches to violence risk assessment and then follow people longitudinally for one year to look up what actual violent behavior happened and then how well did these tools do at predicting it. I recruited 30 individuals through the OnTrack New York network starting in November of 2019. And then my recruitment ended in March of 2020, which is when New York City really started feeling the impacts of the COVID-19 pandemic. And my research project like other research projects was shut down at that time. So I anticipated having many more people, but this was the sample that I had. And then the tools that I was comparing, I'm gonna talk about each of these tools in detail, but I was comparing the HCR-20, which is a structured professional judgment tool. The COVER does actuarial assessment, a self-assessment, and then something called the UPS, which is a measure of impulsivity. So the HCR-20 is a structured professional judgment tool. It has the presence of 20 risk items that you go through and say whether they're present or not. And then you say how relevant you think they are to the violence risk. So there's 10 historical risk items, five clinical risk items, which is based on the person's current presentation, and then five risk management items, which is based on anticipated future problems they might have with housing or social support. And then you come up with summary risk scores, case prioritization, the risk of future violence, the risk of serious violence, the risk of imminent violence. You go through and using this, you come up with what you think the score is. The COVER is an actuarial program. It's a computer program. It asks you questions, you answer them. And then at the end, it gives you this report. The likelihood that sample A client will commit a violent act towards another is between 20 and 32%. Best estimate, 26%. And the thought is two people using this would come up with the same risk assessment. And then I was also very interested in data about recent studies looking at self-assessment. We do a lot of creating these tools, studying these tools, how well they're working, training people to use these tools. And then Dr. Scheme, among others, has suggested that we can ask people and maybe people know. And they found that asking people has comparable predictive value to many of these tools in the studies that they've done. And the question that they use is, we define violence as any act that causes physical harm to another or is intended to do so. Given a scale of zero to five, where zero is no concern and five is greatly concerned, how concerned should your therapist be that you might be violent in the next year? And then I also use the OPS, which is a measure of impulsivity. It produces five subscales, but I'm gonna focus on two today, which is positive urgency and negative urgency. Positive urgency is the tendency to act impulsively when something good happens, like it's your birthday or you just did really well on a test. Negative urgency is the opposite. It's the tendency to act impulsively when something bad happens. Somebody breaks up with you or you fail a test. And these are self-reported and they range from one, which is low urgency, low impulsivity, to four, which is high. After I assessed all of these at baseline, I followed up with people at four time points to measure actual violent behavior. Three months, six months, nine months, and 12 months. Follow-ups were initially in person and then transitioned to a virtual format, which was by phone in March of 2020. And individuals did have to remain enrolled in care at OnTrack New York for the duration of the participation in their study. And that was in case they told me something concerning, that I would be able to reach out to their team and help them, just to make sure that they were connected to who they needed to be connected to. And what I used to measure violence was the MacArthur Community Violence Interview, which gets into details about violence in a very granular way. And it asks, has someone thrown something at you? Have you thrown something at anyone? Has someone pushed you? Have you pushed anyone? And it goes through increasing severity of violence. And I asked this every three months. And at the end, what I found was that there were six episodes of violence by six people, which were three episodes of throwing something, two episodes of slapping someone, and two episodes of pushing, grabbing, or shoving someone. And that adds up to seven, because someone engaged in two of these actions in one episode of violence. And as I was talking about in the beginning, these are not severe episodes of violence in terms of the injuries that occurred. However, they really had an impact on people's lives. Overall, these episodes were more likely to happen when someone was not taking antipsychotic medication. And the impacts that it had was that two people reported they had to take time off work and school, and one person was taken to the hospital for a psychiatric evaluation and subsequently hospitalized. So although these might not be severe in terms of the injuries, no severe injuries were reported, however, they really did impact people's lives. So looking at this sample, 20% or six people engaged in violence, and this is a small pilot study. It's a pilot study, it was also during the pandemic, so it is a small sample, but we see that it seems like there are more women who may have engaged in violence in this sample. Now, here's how did these tools perform? Well, the first thing I ran was the HCR-20, which was the Structured Professional Judgment, my professional judgment, and it seemed like the HCR-20 really wasn't able to differentiate between people who were and weren't at risk of violence. And then I ran the computer program, the COVR, and the COVR also wasn't able to differentiate between people who were or weren't at risk of violence. And then I looked at self-assessment, and then participants themselves also weren't able to differentiate if they were or weren't at risk of violence based on the outcome for the study. And then I got into the OPS, which was the Impulsivity Measure, and what I found was that people who engaged in violence reported that they had higher levels of negative urgency, that they felt that they were more impulsive when they were already having negative emotions, and that that group of people, the people who told me that, were more likely to engage in violence. So overall, what I found in the study was that research related to violence is feasible and acceptable. In an on-track New York setting, in an early psychosis setting, people were comfortable talking about it with me. Both these studies show what we know, which is that most people with early psychosis, most people with mental illness are not at risk of violence. We're really talking about a small group. And violence itself was rare, with 20% of the sample engaging in at least one act of violence over 12 months, and none of these structured tools that were developed in these other settings really were able to identify individuals at risk of violence in the sample in this setting. In this study, I found that higher levels of negative urgency, that type of impulsivity, was related to future violence, and that was really interesting to me because higher urgency scores have been associated with increased aggression and reduced cortical thickness in notable brain areas, including the ventral prefrontal regions. Negative urgency overall has been theorized as a potential target mechanism for behavioral intervention to reduce violence, and that's the direction that I'm moving in now. So overall, the pilot work that I've described here and that I've done as well has really focused on trying to understand who is at risk of violence as a minority of people, and how can we identify someone who might be at risk of violence early in the course of schizophrenia so that we might be able to offer services that they'd like to engage in? And then the next steps will be to develop a behavioral intervention to reduce violence risk for young adults with early psychosis. Thank you very much, I appreciate it. So I'll be speaking about criminal justice involvement among young people with early psychosis. So why focus on first episode in criminal justice involvement? We know that the criminal justice system itself sort of represents an enriched setting, if you will, in terms of identifying young people with early psychosis. In other words, the prevalence of early psychosis within the jail setting is probably higher than the prevalence outside of the jail. And so we need to do things in the jail setting in order to identify young people with early psychosis who would then benefit from specialty care either within the jail or at least upon release. We know that there's a higher risk of criminal justice interaction during early psychosis. Some symptoms may lead to this criminal justice contact. But in many cases, it's not the symptoms, but rather the psychosocial adversities that come with the early illness that actually set the stage for some sort of disturbed behavior that results in a 911 call, which results in police contact and the subsequent arrest. We do know, as Stephanie was just pointing out, that there is a higher rate of violence during early psychosis, and so that can also contribute to criminal justice involvement. For a significant portion of young people with early psychosis, criminal justice contact is actually their first treatment contact. Police officers are sometimes critical to the pathway to care for many young people with early psychosis, whether that means in the criminal justice system itself or a transport to an emergency department, for example. We also know that criminal justice involvement lengthens the DUP. And as we all are aware, the DUP is one of the strongest predictors of outcome. The shorter, the better. So just a brief presentation of a case. This is an actual case of a young man that we worked with. It's highly disguised. Mr. A's a 24-year-old male who's been detained in jail for about five months. He graduated from high school, completed a year of college, attended trade school, but then he quit the trade school and his job to pursue being an artist. And he reports that the problems began around that time, which coincided with daily marijuana use. Over the course of about three months, there were two police transports to the ER or the crisis receiving facility. One of those resulted in a five-day inpatient stay, but he didn't really get connected to outpatient care after that hospitalization. Increasing paranoia had resulted in several confrontations and charges for criminal contempt. This is a charge that's really about violating a protective order, but also disorderly conduct, trespassing, and felony robbery. So these are some of the common charges that we hear about. Bail was set. He was diagnosed with schizoaffective disorder in the jail by Correctional Health Services. So how does jail detention complicate care in the case of early psychosis? He was eligible for bail, but couldn't afford it. His next court date was several months out and likely to be further postponed. If his defense attorney tries to secure an alternative to incarceration, it will likely involve mandated treatment. Which comes with, as we know, all sorts of complications. And then finally, discharge planning within the jail for the time of release. Services tend to refer to just a handful of behavioral health services known by the jail health staff and that are specifically for justice-involved people. So one of the complications is that within the jail setting, discharge planners, social workers, psychiatrists, et cetera, may not be familiar with early intervention services that are existing in the community. And so there's a gap that needs to be bridged there. This is a really nice paper on FVP and criminal justice system involvement that I would encourage you to take a look at. It suggests that one way of sort of wrapping our heads around this issue is to use the sequential intercept model as a framework for understanding how early intervention services, like coordinated specialty care, can liaise with the various aspects of the criminal justice system. So this is a diagram showing the sequential intercept model. The intercept one occurs at the time of the initial police involvement. Again, there's usually a 911 call. Police come out. The police have to figure out what are they gonna do to resolve the scene? Can they just calm things down and leave? Or is there a complainant, like Mom, we had heard about in the earlier case, in which case something needs to be done. There can either be a transport to the ED or there can be an arrest executed. And in most cases, because officers have so many misdemeanor charges at their disposal, in most cases there can be an arrest for something like disorderly conduct, criminal trespass, willful obstruction of a police officer, et cetera. Intercept two is really at the time of the first court appearance. This is when you go before the judge. Intercept three is within the confines of the jail. There may be a specialty court like mental health court or other dispositional courts. Intercept four is when the person is coming out of the carceral setting, whether it be jail or prison. And then finally, intercept five is also important. That's probation and parole or what we call community supervision. I should mention that since the original development of the sequential intercept model, which again is just a way for us to understand how our patients move through the system and some of the interventions and programs that can be developed along the five intercepts to help our patients hopefully transition out of the system, an intercept zero has been added and that's really the community crisis response services that can be put in place to hopefully eliminate law enforcement contact in the first place. So what do our patients get arrested for? Mostly minor misdemeanors in which the arresting officer has quite a bit of discretion. There are some charges where officers don't have a lot of discretion like DUI, like domestic violence cases or robbery, cases where there's a complainant saying, I want to press charges. But a lot of our patients get tangled up in the criminal justice system for charges where the officer can or cannot execute an arrest. The types of these charges and processing of the charges across the criminal justice system change over time based on both the formal and the informal policies within the criminal legal system, but probably also based on the changing life circumstances of our patients. So let me show you what I'm getting at here. This is from a study of individuals with serious mental illnesses in Southeast Georgia. These are not first episode patients. These are patients with SMI at about an average age of 35 or so. And as part of a study, we were able to pull their state rap sheets, which is their history of criminal justice involvement, their history of arrests. We looked at their, and this is among people who had had at least four arrests over the course of their lifetime, 99 patients. We looked at their earliest two charges and their most recent two charges, or I should say arrests, the charges embedded within those arrests, the top charge. And in terms of their earliest charges, we see things like marijuana, driving under the influence, shoplifting, criminal trespass. But in terms of their most recent two arrests, some 15 years into the course of their illness, we see a very different constellation of arrest charges. Specifically, we see marijuana possession is much less common later into the illness, disorderly conduct. And this sample became the second most common charge. And we also see other types of charges start to appear like willful obstruction of a law enforcement officer, which basically means doing anything to make a police officer's job more difficult, like not giving your name, or not giving your date of birth, or in any way resisting, whether physically or not, or attempting to run away, or pushing the officer back when being handcuffed, those sorts of things. That's willful obstruction of an officer. We see that those are very common charges later on in the course of the illness. And then we also see these new types of charges appear. The very top charge, the most common charge in this sample of individuals with SMI is probation violation. So they've already been entangled in the criminal justice system, released from jail, now they're on probation, and they miss an appointment, or they can't keep up with the requirements of probation. That was the number one charge that resulted in going back to jail. Other charges are things like failure to appear to court. So you have a court appointment, you miss your appointment, now you have a bench warrant for your arrest. The next time you have a police contact, the officer's gonna have to execute an arrest for failure to appear to court. So just a sampling of some of the types of charges that our patients with SMI tend to be involved with. In a study in Atlanta from a number of years ago, we looked at 191 first episode psychosis patients who were hospitalized at Grady Memorial Hospital, and we looked back at their own criminal justice involvement. 37% had been incarcerated during their duration of untreated psychosis. So during that period of time when they had early psychosis but were not yet engaged in any sort of treatment, 37%. And having had criminal justice involvement lengthens the DUP and is also associated with other types of worse outcomes. So I want to close by just telling you about a program that we put in place, thanks to an NIMH R34 Intervention Development Grant. and this is where we wanted to go into the jail setting and figure out how can we do a better job at early detection of young people with early psychosis who may or may not have been diagnosed yet, and then connecting them over to coordinated specialty care upon release. And so we looked at the TIPS study in Scandinavia and other studies and decided that we needed a two-prong approach. Just like in TIPS, we need a public information campaign and we need an early detection service, but within the confines of the jail. So we developed a targeted education campaign for correction officers, these are the guards in the jail, to help them better recognize specifically the early signs of psychosis, and then also a specialized early engagement process to work with those young people and their care staff, and then connect them to OnTrack New York upon release when possible. This is an example of some of the educational materials for the correction officers. This is a postcard, again an example of some of the materials. We surveyed a group of officers before the targeted education campaign, another group of officers, six months into the education campaign, and then yet another group 12 months in. This paper has been published in Early Intervention in Psychiatry. Bottom line is we found improvements, but then those improvements went away, which we're trying to understand. This study was happening all during the course of COVID, and so there was a lot going on within the jail for the correction officers to be dealing with, in addition to our targeted education campaign teaching them about early psychosis. We have a toolkit out there that you all can access if you're interested in learning more about what we did, and then importantly about what early intervention services, coordinated specialty care programs, etc., can do to better liaise with your own local criminal justice system, getting to know your own intercepts across the sequential intercept model, and then figuring out how to really work with jail staff to hopefully do a better job at not only early detection, but importantly also then connecting to specialty care upon release. So I will stop there and turn it back over to Deirdre. Thank you both. I'd like to now move into a Q&A session, so if anybody has questions and would mind using the mic. We do also have an online audience, so speaking directly into the mic would be helpful, and then I also have some online questions that have already been submitted. So maybe I'll start with an online question while people... So one question posted online was whether you have, whether you would recommend using DBT skills and mindfulness skills in interacting with patients with first episode psychosis while doing a safety plan? So yeah, I think they can be very useful. It depends often what the trigger is for suicidal thoughts and acting on those thoughts, and also different people, there may be need for adaptation of mindfulness skills for people with psychosis, and so working carefully with people to figure out what skills help to ground them, but anything that supports someone in developing skills for dealing with distressing emotions, distressing thoughts, and supporting, creating space between having those thoughts and acting on them, I think is very useful. So yeah, awesome. Great, thank you. Another question that was posed on, or actually let's go right here, go for it. I can wait. No, perfect. I was just curious about your thinking about starting with an injectable if you have fairly good evidence that this patient is not going to comply, and you know, how much you've used that modality. So the question's about starting with long-acting early, long-acting injectable medication from the beginning. So I would say in our program we really value the use of shared decision-making, and so I think that long-acting injectable medication should be offered from the beginning, and discussing with people what their preferences are, what their priorities are, what their goals are, the pros and cons and risks and benefits of both oral medication and long-acting injectable medication, and working to come up with a decision that works best for them in that moment. And so yes, absolutely offer from the beginning. Some people are open to it, want it. Some people do not, and so you really have to meet someone where they are and revisit over time. So thanks for the question. Go ahead. Just a couple of quick questions. From the presentation, you know, we know there's no foolproof method of accurately predicting suicide or a violent act, but sometimes we know society almost like puts an expectation on clinicians that if you don't get these rights, you know, you face like litigation, you know, problems from your colleges or the employers. So my question is about disclosure. You know, is there like a threshold whereby you've seen a patient and they disclose that they are planning to hurt themselves or to hurt someone else? Disclosure to third parties. Is there uniformity amongst clinicians? At what point do you disclose like to the police or to the third party, or is there legislation in place? And secondly, about treatment in the jail setting. You mentioned that DUP is affected if there is a treatment delay, and there's, you know, in your studies that you did, you did targeted education for the prison guard, but are there steps that could be taken to actually treat the inmates in jail, you know, especially those who are refusing treatment, you know, so that you can almost bring the hospital to them in jail rather than waiting for the legal processes to take its steps, and which could actually take a long time? Thank you. Yeah, I can take your second question, and yes, definitely. Jail health services, correctional health services, should treat early psychosis just as they're treating other mental illnesses. However, jails vary widely in terms of what that care looks like. Prisons are much more standardized, but jails, county jails, city jails, across the country are very, very different in terms of who's providing the care, what the care looks like, what the formulary looks like, etc., and so we would hope that anyone with early psychosis would be identified in jail, although it doesn't always happen, and we would hope that they would be started on treatment. It's usually only medication. There's not much in the form of psychosocial treatment that goes on in jail, and definitely not the types of psychosocial treatment that we would usually recommend in the context of early psychosis, like supported education, supported employment, family psychoeducation, etc. That's just not happening in the jail, and so it's important to bridge the person over to outpatient care when possible, continuing the medication that may have been started in jail. With regards to the first question, I think about the expectation that we be able to anticipate risk. I think that's exactly right. I think we have to educate others in the limitations of our capacity to predict the future. We have some tools that support risk assessment, and they are imperfect, and so I think we need to be humble about that, ourselves, with the people we work with, with the families, and with the institutions. I think there was also a question about duty to warn and how to navigate that issue. There's obviously state-by-state differences in terms of the laws there. When there's acute imminent risk, obviously you need to intervene. The issue comes when there's more chronic risk, and I think there, in terms of warning, you need to know about and follow what your state laws are, and often those are important times to seek consultation with others. I realize this is the first episode of data that you're talking about. Are you familiar with any data of patients with psychosis who continue with recurrent violence in the public, and what are your thoughts about that? Because it seems to me there should be a different way of handling those situations, which are much more precarious, and create problems with staff at the hospitals and everywhere else where there's recurrent violence, and there isn't a different way of handling that in public, other than through the forensic system, which is usually extreme violence. I'd like your comments on that. Exactly. We're just saying we need Dr. Rowland here to speak to this question, because I think it is, outside of our areas of expertise, though, a really critical question. You know, I don't know that I have the expertise to speak to it. I would say I think there's a little bit of evidence, and maybe some research going on right now, in terms of clozapine as a potential treatment that may target violence more than the other antipsychotics. And briefly, there was actually an online question regarding clozapine versus lithium and reducing suicide and violence risk. So you're speaking to part of that, that it sounds like there's some evidence for clozapine. Maybe to follow up, whether there's any evidence for lithium use in this first episode psychosis and suicide risk. I would say in the context of someone with schizoaffective disorder bipolar type, there may well be an indication for the use of lithium. In the absence of mood symptoms, then I think that, you know, that sort of argues for clozapine. Right. Thank you. Yeah, quick question. I work in a county jail in rural PA, and some CO corrections officers are very thoughtful and reflective, but other times they're either give too many privileges to certain patients that they favor, or inmates, and then other times they let people go on, they put them in solitary. And solitary seems to be used as a punitive measure if you're psychotic, or if you're just a behavioral problem, you run away from work release. I just wanted your thoughts on solitary confinement, and how do I, or how do you communicate with jail guards about the dangers of that? Because, you know, the majority of my inmates that have solitary confinement don't do well with it. They're, you know, they're in their thoughts, they're in their jail cell, and these are non, most of them are nonviolent, so I wanted maybe some tips on how to address corrections officers. Yeah, I would say that if you've seen one jail, you've seen one jail, like there, there's so much variability across county jails. I agree with you, though, that there's general consensus now that solitary is not good for your mental health, and if you have a mental illness, it's not good for your mental illness. And so, hopefully, there are efforts underway, either at the local level or at the state level, to reduce, if not eliminate, the use of solitary confinement. And I would hope that our young patients with early psychosis wouldn't have to endure that, you know, during their jail stay. Correction officers are variable, just like, you know, any other profession. Some are very good at talking to detainees, detecting signs of mental illness, and then referring to health staff. Others probably aren't so good at it, or don't view it as part of their work or a priority, so there's a lot of variability within the jail. Thank you. Sir, there's one person right behind you that was in line, if we want to grab that real quickly. Thank you. Thank you so much. I have kind of two different scenarios that I've experienced that are sort of in the opposite. It's the situations where you really, really, really, really wish you can get a patient in, but the police say nothing we can do, or the county jail assessor says nothing we can do. So I've had patients before who've made threats, for example, who they get brought in, and because they're well enough, they're still early enough in their psychosis that they're able to present well, the county jail psychiatrist or physician who assesses them says, no, they can go out on bail, and then they go, and they don't get treatment. Or the opposite situation, where I've got patients who are getting disorganized, you can tell they're in early psychosis, or they're having a psychotic relapse, and you send the police for a welfare check, and all of that, and nothing comes of it, despite the fact that you know where this is heading, because you've seen it before, either with this patient, or just because it's clinically something that you're used to. And so that's a continual frustration that I have, where the police sort of seem to say, they either go really hard, and then they incarcerate people unnecessarily, or they sort of say, patient is not actively trying to stab somebody, therefore nothing for me to do, goodbye, have a nice day. And I have spent hours on the phone with police officers, and all these things, trying to, you know, say this patient is not well, he's, you know, saying things that are very concerning to the family, his behavior is very strange, he's clearly entering psychosis, and I understand that this is how the law is written, but there does seem to be some discretion that different police departments seem to take, and I've worked in a couple of different places, and I've seen very different practices, so I'm wondering, it's not precisely the topic of what you all were talking about, but I'm wondering if you have any words of wisdom on dealing with that particular situation, or any thoughts on that. Yeah, I understand your frustrations, and I think the overarching problem is that the criminal justice system across the various intercepts is being used in an attempt to address a health problem, or even if it's not a health problem, disturbed behavior that's related to a health problem, and so there's no wonder that that's not a great solution, and it's going to lead to frustrations, and the only way to really address that overarching problem is for our field to take more responsibility along the lines of crisis response, welfare checks, those sorts of things, rather than relying on the criminal justice system actors to do that, so I think, you know, bottom line is there's a problem, a broken system, and so it's not surprising that it's going to lead to less than ideal, you know, outcomes and a lot of frustrations. Go ahead. Yes, I just want to offer a caveat, Alana, to one of the points that you made with regard to post-event quality improvements, conferencing, to figure out what went wrong and what could be done better, which is to say that it's important that that be done within the scope of your state's peer-review privilege statute, because those discussions are otherwise discoverable by plaintiff's attorneys, who, you know, suicide is one of the major causes of malpractice litigation in our field, and the presence of even medical students in some states will vitiate the privilege because they're considered outsiders who aren't actually employed by the facility, so figuring out how to keep that within the scope of the privilege is extremely important. Yeah, thank you for that. Go ahead. I think there's a public policy question which really overrides all of this. I find that, at least in Southern California, when someone gets arrested that I've been treating, it's the best thing that's happened to them, sadly enough, that you have a very disturbed patient who's running around naked, they get picked up by the police, they're taken to a psychiatric hospital, they're back in my office four or five days later, just as bad. The hospital can't keep them, either because there's no money or there's not insurance, and secondly, when they go to jail, they're often forced to take medication, and they're kept for a long time, much, much longer than a hospital would keep them. When I started practice, if someone was quite disturbed, we'd put them on a temporary conservatorship, they'd wind up at the state hospital, they'd get treated for six or nine months, and they'd be much better. That's not possible anymore, so the only thing we can do is change policies so that there are places for people to go and finance it, so that patients can stay long enough that they get treatment, because I think the biggest problem in violence and committing crimes is that there's no place for them to go, and compliance is critical. If they're not compliant, they're going to be back in the system over and over and over again, and the only way you can get compliance in many cases is to force them, and there are many positions in politics where people don't want to force people to do things, where they think that their rights are more important than their lives, and I think that's something that perhaps we should be working on. Thank you for that. I would also say that hopefully the work that's being done right now in Intercept Zero, which is the crisis response system that doesn't rely on law enforcement, will get us to a better place with more humane and non-police crisis response, and then more places for people to go for crisis evaluation, crisis stabilization, crisis respite, those sorts of things, so hopefully there's some progress being made in Intercept Zero to alleviate the problems in Intercept One through Five. I think there's one problem is that we have programs in Los Angeles where many of the criminal defendants are diverted out of the criminal justice system, which is a very noble effort. The problem is they walk away from these programs. They don't necessarily comply. You see them going into diversion programs two and three times. Unless you can force people to get the treatment that they need, they're not going to get it, and so all the best efforts are very short-lived, in my opinion. The only thing I would say, I very much hear what you're raising. We, I think of, you know, there's just such heterogeneity in how people present and what leads people to seek care and engage in care, and there are a lot of people who do engage in care in non-coercive, in non-coercive contexts, and many who are, you know, coercion can contribute to a disengagement in care, and so I think there's just a tension and challenge, I think, we as a field have around what is the, how do we maximize engagement using non-coercive strategies? What, you know, are, what are the needs for coercion at times, and sort of how to use it when it's needed and limit it, and really support engagement. I think my experience in coordinated specialty care has been that there is a huge amount of opportunity to engage people based on what they're seeking, that people engage in care in ways that in traditional mental health services have been typically disengaged from care, but, you know, it's absolutely something I think we're still grappling with as a field. All right, I think we're just about out of time. Thank you all so much for coming. Really appreciate you joining us on 8 a.m.

early psychosis

Columbia University

first episode psychosis

OnTrack New York

risk management

AIM Model

antipsychotic medication

clozapine

violence risk

criminal justice system

sequential intercept model

non-coercive engagement

crisis response