Good morning, everyone. I want to welcome you to today's webinar on Screening for Psychosis in Adolescents. I am Judith Dauberman, Program Manager for PepNet. With us today, we have Dr. Adelsheim, who is a Clinical Professor of Psychiatry and the Associate Chair for Community Partnerships in the Stanford University School of Medicine, Department of Psychiatry and Behavioral Sciences. Dr. Adelsheim is also the Director for the Stanford Center for Youth Mental Health and Well-Being and will be helping us with the Q&A. Today's webinar is presented in partnership with SMI Advisor, a SAMHSA-funded program of the American Psychiatric Association. Dr. Woodbury is a Clinical Social Worker, Licensed Clinical Psychologist, and Early Psychosis Researcher at the Maine Medical Center Research Institute. A graduate of Bowdoin College, she obtained her MSW from the Simons College School of Social Work and her PhD in Clinical Psychology from Harvard University. She is a Research Assistant Professor in the Psychiatry Department at the Tufts School of Medicine. And here's Dr. Woodbury. Thank you. Thank you very much. And thank you, everybody. I know that there's a lot now online for training, and I'm really excited about the interest in this topic in particular. My goal today is to focus on conceptual issues in screening for psychosis in adolescents. So I'm just going to put a warning out there that if you tuned in to know which screen to use, I'm going to apologize in advance. I absolutely share your desire to have a great, reliable go-to tool. I just don't think there's an easy answer yet. So the point of today's talk is to help you understand why and to be informed consumers of the tools that are available. So this is not going to be an exhaustive review of all the screening tools and pros and cons. I also feel very humbled presenting this material, so I expect that there are some on the call with significantly more expertise. And I'm particularly grateful to these talented women who have provided help over the past few years in sorting through and organizing the relevant literature. So I want to give a note of kind of the motivation for this talk and review some key screening concepts that I think are important but often poorly understood. And then I'll give a very brief overview of what we know and don't know about psychotic experiences in adolescents and with some considerations for clinicians. But I'm most interested in hearing your thoughts and questions. So I do hope that you'll stay on to that part. So what motivated me was thinking about adolescents as a distinct period. And I want people on this call to think how many of you have given a psychosis risk self-report screen to an adolescent or asked a community provider to give an adolescent a psychosis risk self-report screen and then try to interpret the response? Did you assume that a published threshold, three items, or a distress score greater than six had the same meaning for a 14-year-old and a 21-year-old? That the threshold was appropriate for the racial, ethnic, or other subpopulation from which the adolescent was or with which the adolescent identifies? Did you wonder whether the adolescent understood the wording of the items as they were intended? Did you provide recommendations to the adolescent, the parent, or the clinician based on their response? Or tell them that they were likely at high risk for psychosis or that their responses were concerning or that they shouldn't worry? I became interested in this topic because I was trying to find answers to these questions and I began to worry about the lack of good data to inform what clinicians and programs are doing fairly often. One of the key concepts to this is that screen thresholds are sample specific. So the answers to some of those questions that I pose are that we can't make those assumptions and that there are differences. I think there is potential for us if we don't know what we're doing to do some harm, but I also want to stress that I don't think that's inevitable. So I'm distressed by people who just kind of see the data and suggest that we should not be screening adolescents at a population basis. I think it's more nuanced and I'm really looking forward to a discussion about this issue. We have to persevere and I think we have a very important goal to reach. So people may be familiar with this diagram of pulling people out of the metaphorical river rather than trying to prevent them from falling in in the first place. And I think adolescence is a period of development particularly important to this concept. We need to be recognizing the signs of early psychosis during adolescence for some pretty important reasons. And some of you who have tried to look at incidence rates or find incidence rates on adolescence is very hard. This is the best graph that I could find illustrating incidents across age. So these are new cases of schizophrenia that appear at different ages in years. And when you look at this graph, you can see that there's a sharp increase in age both with the males on the top line and the females in the bottom line in the new cases that appear during the adolescent period. And one estimate has that incidence increases about four times between ages 10 to 14 and doubles again between ages 14 and 18. So what you notice is that the risk of having a new onset at age 14 is very different than having a new onset at age 22. Now if you look at this slide, this is thinking about the onset of psychosis. So prior to schizophrenia onset, and I would argue since we know that often psychosis is unrecognized for a year and a half, and if we think of prodromal periods as beginning even before that, that the sharp increase in incidence is entirely during the adolescent period, probably from 10 to age 18. So that not only are adolescents different from adults, but young early adolescents are different from older adolescents if we're thinking about both incidence and prevalence rates. The other piece that I think is highly motivating is that there's meta-analytic evidence now that our specialized treatments may be able to reduce the short-term risk of converting to a psychotic disorder. And during a very active developmental period, this has pretty significant long-term implications. I read with interest a recent paper out of the Australian group suggesting that reduced rates of transition are associated with receipt of treatment. So it does seem that the treatments we are offering now in many of our clinics do make a difference. And importantly, we can reliably identify those for whom these treatments are appropriate. So with SIPs and CARMS interviews, we can predict more reliably than for many physical disorders for which we actively screen that someone is appropriate for treatment. And then finally, we have a number of psychosis screening tools that are available to us. So I'm going to be using a set of figures that come from papers by fellow Luisa Foley and colleagues. And I'm going to orient you now to some of the components of these figures so you can follow along. So when we think about general population screening, we are looking to identify people within the full general population sample who have certain characteristics. And the figures that you see here in color are figures that are colored according to their diagnostic outcome three years later. So in dark teal, people who will have had a first episode of psychosis by that three-year point. In light green are people with persistent but non-psychotic mental disorders. And in orange, individuals who will have persistent, what we call attenuated psychotic syndromes that linger through that three-year period. And then on the far left are the gray figures, which are people who are healthy at that three-year follow-up or possibly who had psychotic-like symptoms that have now remitted. So this is the group of individuals that we're primarily thinking about when we're screening. We want to identify people who are at very high risk for having a first episode of psychosis. But I want to make the point that identifying people with persistent non-psychotic mental disorders or persistent attenuated psychotic syndromes is not a wasted effort, that we have treatments that may be able to improve the trajectories and the lives of these individuals as well. So here is a clinical high-risk paradigm. I'm going to be using this to talk a little bit about some screening concepts that you're hearing more about in the news with the COVID pandemic and that, for most of us, we need a little bit of review just to be clear on what they mean. And I want to think about the fact that the sampling is important. So in our clinical high-risk paradigm, we are seeking to identify people who are on the cusp of developing a psychotic disorder, and that's this group of syndromal individuals. And we're pulling those individuals from a very specific population, a help-seeking population. So typically, clinical high-risk centers are not recruiting from a broad population sample, even if they do outreach talks in schools and primary care settings, they are eliciting referrals of people for whom there is an active concern of developing psychosis. And this is different than recruiting from a general population. And the point here is that the density of individuals who are going to meet your target criteria varies according to the sampling strategy. So if you sample in a group of individuals who are actively help-seeking for psychosis-related services, the likelihood of identifying somebody who's very appropriate for those services is much, much higher than if you are sampling in a general population sample. And we call this sample enrichment, and there are good reasons that we sample specifically from help-seeking samples, and that is specifically because our tools are not good enough yet for screening in general population samples. So let's think about the screening concepts using the target of psychosis. So if we're specifically interested in identifying those who will develop psychosis, sensitivity is the ability, the degree to which our screening is going to identify everyone in that target group. And that could be if we're screening the help-seeking group, those who are psychotic at that initial screen or those who will develop psychosis over the three-year follow-up. And I want to orient you to the size of the circles in that bottom three-year follow-up section of this figure, because the size of the circle reflects the proportion of that clinical high-risk sample that will fit into each of those categories. So the predominant outcome is the light green, the persistence of non-psychotic mental disorders. And first episode psychosis is actually the next most prominent outcome for this group. With persistent attenuated psychotic symptoms, a slightly smaller group in that sample. So sensitivity is identifying the dark teal group. Specificity is the degree to which our tool does not identify those outside the target group. So in this case, if psychosis is the group that we're targeting, it's not identifying those who are going to be healthy or remitted, who have non-psychotic mental disorders, or who have persistent attenuated psychotic syndromes at the three-year outcome. And again, the validity in this example is longitudinal diagnostic outcome. When you interview people at three-year follow-up, what can you ascertain about their clinical status? So let's compare that to screening for clinical high-risk. So if the target is to identify individuals at risk or with high-risk syndrome, the validity now is not your long-term outcome of whether they develop a psychotic disorder, but it's determined by whether they meet syndromal criteria on interview. And most of our screening measures have this as the validation, not eventual diagnosis with a psychotic disorder. And so that's really important to keep in mind when we're thinking about these screening tools and when we're using them. The sensitivity with this as the outcome is identifying everybody who has a clinical high-risk syndrome. And typically, we use these screening tools to identify people who we want to invite in for an interview. And that makes a lot of sense. So we want to capture everybody who's at high risk for developing a psychotic disorder. But importantly, we're trying not to bring in for interviews those who are not going to meet these syndromal criteria. And we're not bringing in those who already have a psychotic disorder. So that's an important distinction when you're thinking about tools that are validated against a clinical high-risk syndrome in the interview. So what are the implications? I want to think about a COVID-related example, if you've been listening to the news on antibody screens. The positive predictive value of a screen is the degree to which that screen positive is reliable, that if you have a positive COVID screen, that you have had exposure to COVID. From what I understand, that is fairly good for the antibody tests that are out there, that it is highly likely that you have had exposure to COVID. What's of concern is the negative predictive value. So if you get a negative screen, what percent of people who have those negative screens actually are truly negative? And this is where there has been some concern. And so just to understand that when we're screening, the degree to which that screen captures people who are truly at risk or have a psychotic disorder and doesn't capture people who are not at risk is really critical to thinking about what thresholds we use and how we understand the usefulness of that screen. And these rates depend on the prevalence of what you're screening for in the population. So if you think of the density back in that population sample, if you have low prevalence, you're going to have different PPV and NPV for the same screen than if you have higher prevalence. What's the implication? The same screening tool will have a different PPV and NPV if you are screening someone in a primary care setting or an outpatient mental health clinic or a research center. So the thresholds will have different meaning or you may need to use different thresholds in those settings. So I'm going to come back to this diagram again, hoping this will help pull some things together. In the blue on the right are the percent of each of the samples that will develop a psychotic disorder. If you look at the clinical high-risk samples in the bottom percentage, the 26th percentage, that's the average rate of individuals in our clinical high-risk samples who will develop a psychotic disorder. If we move up that column to the individuals seeking help at early detection centers, the rate is, the prevalence is 15%. And if you go up to the general population, the prevalence is 0.43% for three-year outcome of psychosis. So less than 1%. So again, to just hit home that the prevalence differs remarkably in these different samples. And when we're thinking about screens and you have somebody tell you the threshold is a six, please remember that that threshold is going to vary according to the density. And it's a little counterintuitive in that you need a higher threshold on a screen in the general population samples and a lower threshold in the more enriched samples. And a way to understand this is to think about the fact that there's a greater likelihood that someone may endorse hearing things that no one else hears in a general population sample and be talking about good hearing, or perhaps be talking about hearing something that's unusual but within a normal range, whereas in your more enriched samples, if they endorse hearing something no one else hears, there's a greater likelihood that they're talking about an experience that may be on the psychosis spectrum. So a few other concepts that I think are often neglected that I think are worth thinking about as we're talking about screening in an adolescent sample. So I hope I've made the point that sampling and populations matter and that it makes a difference if it's a help-seeking or non-help-seeking population, what the demographic representation of that sample is relative to the samples that a screen has been tested in, different levels of risk, and age. And I also wanna just highlight that how we screen matters as well. So if you have a self-report tool that you read to someone, you may get different responses than if that person reads it themselves. Or if you have a tool that you can follow up on and ask a question and then ask them to tell you what they mean, you're gonna get much richer data than with a self-report alone. And I don't think it's a surprise to probably anybody on this call that the context also matters. So whether someone has privacy, whether their parent is sitting next to them in the waiting room, whether they're completing this online, or whether there's collateral information, it's all important. And then I've been particularly troubled by the issue of language when we are thinking about adolescents, because the reading level of many of these instruments is actually fairly high. And we often don't think about the capacity for someone to understand the words that are included in the items that we're giving. And then finally, that screen results can change over time. So just like you can test negatively for COVID today and positively tomorrow, your results on a psychosis screen can also change over time. So let me shift gears and talk a little bit about psychosis in adolescents and what we know and don't know. So we know that there's a continuum of experience, just like in adults across the population. We have some estimates for psychotic disorders, but again, these are hard to come by. So the prevalence of schizophrenia is fairly small in this age group, 0.2 being probably the highest estimate. And as you've seen in this graph on the right and before, the incidence to the new cases, the rates of new cases rises dramatically with age. We know there's lower stability of psychotic diagnoses. So someone having a diagnosis in adolescence of schizophrenia has less stability associated with that in that a couple of years later, we may meet them again and have more information and their diagnosis may be bipolar disorder with psychotic features or vice versa. But what's a particular concern to me is that when we look at the duration of untreated psychosis, the delay between identification or the onset of symptoms and the receipt of treatment, these delays are longest in adolescence. So we have a real challenge in that we have a harder time detecting psychosis in the age group, but it's all the more important in this age group. So now I'm gonna shift from disorder to talking a little bit about psychotic and psychotic life experiences. So when we're measuring things on a population level, typically we're using self-reports where we don't necessarily have enough information to know if someone has full conviction in their experience or whether their experience is truly psychotic. So if they endorse seeing things that other people can't see, we don't know if they are actively seeing figures, dark figures outside their window that they fully believe are real, even though they know no one else can see them or whether they're seeing figures and they know that this is not real, which we would consider an attenuated psychotic symptom or whether they're seeing sunspots looking at the sun and that's the endorsement. So in a self-report measure, we don't have this information. So typically we default to thinking about psychotic life experiences, knowing that that's gonna include truly psychotic experiences and experiences that are either mild or not psychotic. So our best estimates are from a meta-analysis by Ian Kelleher that about 7.5% of adolescents and this is 13 to 18 year olds experience psychotic spectrum or psychotic life experiences and this is lower than in younger kids. So the estimate is about 17% in nine to 12 year olds and a little higher than in adults. So you can again see that the prevalence of psychotic life experiences decreases with age. So they're more common in children and young adolescents but more indicative of disorder in older adolescents and adults. So again, the incidence of disorder increases with age and the incidence of psychotic life experiences decreases with age. So you have something that's more prevalent and less predictive in the younger group which is a challenge. We also know that psychotic life experiences have low stability. So they have decreased what we call persistence with age. If you measure, ask the same questions at different time points, the likelihood that someone is gonna endorse that item again at an older age is fairly low. Most of these experiences appear to be transitory but there is some persistence across time and that persistence is generally associated with more psychopathology later in life. So psychotic and psychotic life experiences are meaningful and they're associated with increased rates of psychotic disorder later. They're still our best predictors of transition in clinical high-risk samples and even in the absence of predicting psychotic disorders, they do predict later psychopathology and importantly, suicide attempts. I wanna also talk a little bit about basic symptoms. So there's been work primarily by European researchers trying to understand some of more of the subtle phenomenological changes that people experience prior to the onset of a full psychotic disorder. So basic symptoms refer to a set of disturbances in basic thought process and perception. And there are two subscales for this, the cognizant is examining cognitive symptoms and COPR, the perceptual experiences. And there's a really fascinating paper if you have a chance to look at this that just came out looking at this at a population level and trying to understand what it means. And these symptoms are assessed using the SPIC-E and in a 12 to 18 year old's psychiatric sample. So that included EOP, early onset psychosis, ultra high risk and non-psychotic psychiatric help seekers. You can see that there is a fairly good prevalence of these experiences in this adolescent sample. Interestingly, they found that even though the common belief is that these subtle experiences precede attenuated psychotic symptoms, it seems like the clinical significance, the predictive of these experiences really doesn't come into play until older ages than APS. So it may be that these experiences, it may be something about how we're measuring these experiences in the population or it may be something about neurodevelopment. But I think this is a really fascinating area of research. And related is work looking at alterations in self-experience in adolescents. And I think this is particularly challenging given that adolescents experience a lot of changes in their perceptions of self. Anomalous self-experiences refer to alterations in a person's core sense of self-integrity and experience of the world around them. And in a recent study, well, not the 2005 study using an interview in adolescents with non-help-seeking community adolescents, there were high endorsement of at least a single item of anomalous self-experience. And so again, emphasizing the fact that this is fairly common in this youthful population and is gonna make it harder for us to prospectively identify which changes are meaningful. But again, an area that is actively under investigation. So what are the unknowns about psychosis in adolescents? I think there's a lot that we still really don't understand about incidence and prevalence. We have loose estimates for these. And we certainly don't have good incidence and prevalence rates in specific subpopulations. The predictive validity and specificity is still largely unknown particularly for younger adolescents and what the trajectories and developmental applications of these are is still largely unknown. So I wanna briefly review what we know from self-report screening tools. And there are quite a number of tools out there. So in thinking about what to present, I focused on symptoms that include positive symptoms. Many of these screens also include negative symptoms or nonspecific symptoms but they had to include positive symptoms. They had to include adolescents in the sample and I prioritize those that had adolescent only samples but I looked at instruments in both health seeking and general population samples. So this table, I just am presenting some data on five of these screens. The second column has a description of the screen itself, how many items, what kinds of questions that it asks. And then the third column talks about what the adolescent sampling is with some notes about specific thresholds and PPV rates when those are available. Importantly, it's only the top three that have data specific to adolescents. So the bottom two, which are two examples of screens that are fairly commonly used, the data are on mixed samples with adolescents and adults. The prime screen R is barely into adulthood. So I would say that's probably more likely to reflect adolescents. But we all know some of these screens are tested in samples up to 35. And so again, the relevance of those screening data to a sample that is under 18 or under 20 really is unknown. So for the measures that we have, we do have some preliminary data, both in comparison to a DSM-IV psychosis diagnosis in the case of the PP16, and in relation to the SIPs and CARMs for the PP92 and the K42. But in health seeking samples, which is a lot of the work that I suspect is done by people on this call, I just want to highlight, there's really very little data specific to adolescents. There's a little bit more data looking at general population samples with self-reports and screens. But again, it's only the top three that have been tested in adolescent-only samples with any kind of validation. And all three of those are validated with the KSADS, which is an interview that would specifically probe psychotic experiences. So the self-report items are compared with rates of actual psychotic symptoms. And interestingly, one of the best performing measures is the seven-item APSS. And actually, just three items on that measure perform fairly well, probing auditory hallucinations, visual hallucinations, and suspiciousness. This presents a nice opportunity for screening, but obviously, we need a lot more data across a wider population and different subpopulations to better understand if people are answering three questions, what the implications are for further response. And finally, I think it's important to remember that there are general mental health screens that include items that tap into psychosis risk. The BASC-2 outperformed other psychosis-specific screens in a small help-seeking sample of both adolescents and young adults. And higher scores on the self-report associated with the CBCL are associated with higher risk for later non-affective psychosis. So here are two of the leading reviews on screening tools that, just for your reference, I think the range of sensitivity and specificity and positive and negative predictive value says a lot about how variable these tools operate in different samples. There's some recommendation for the prime screen in the Addington article and the PCP, but again, this really is sample-specific, and so I want to offer some caution in terms of automatically taking thresholds for recommendations for every population. And then a reminder that there are a lot of understudied issues when we think about screening tools, the role of neighborhood and trauma exposure and endorsing suspiciousness, for example, different interpretation of items according to racial, ethnic context or religious backgrounds, little understanding of how substance use may impact rates of endorsement or even the rates of experiences that are endorsed, and different biases that we need to better understand when we're screening different populations and particularly younger populations. If you're thinking about screening tools, obviously logistical considerations are very important, the time to complete, the reading level of the instrument, non-availability of non-English translations. So screening tools, there are a lot of unknowns. The validity of positive responses is one of the particularly concerning unknowns, and I think most of the samples for which the tools have been developed are fairly small, and the diagnostic follow-up of those samples, if there is any, is short. The samples are typically enriched so that the data from these screening tools cannot be readily applied to more general population samples. And the translation of item endorsement to actually symptom or syndrome or disorder isn't easy to make. So how do you interpret response patterns? I think there are some genuine questions that I hope you will consider. But before you get hopeless about screening for psychosis, I wanna go over this really very interesting document diagram in a recent paper by Fusifoli and colleagues. So these are two prominent world programs, OASIS in London and Headspace in Australia, who have spent years establishing very strong community relationships and using very aggressive techniques to identify young people at very high risk for major psychotic disorders. The big populations of teal people at the bottom of the arrows are those individuals who develop a first episode of psychosis and are undetected prior to the onset of that disorder. And the small teal groups to each side are those subgroups that are identified prior to the onset of a first episode of psychosis. This is pretty sobering to think that we are capturing with our clinical high-risk programs really a very small portion of those individuals who are gonna develop first episode of psychosis. And it's particularly sobering when you consider that there's a good estimate that about two thirds of individuals who develop a first episode of psychosis will have a prodromal period. So they will have early warning signs that presumably we could detect. It's sad, I know some of you are familiar with this ad that was used in the peer program to educate the community, but unfortunately it will never be this easy. But I'm gonna add one more added complication to the diagram on the right side now to illustrate something that I find hopeful. So the dots, the colored dots represent different risk factors that we know are associated with later emergence of psychosis. And if you can just think about the number of colored dots as the density of risk factors in that sample. So you have the highest density of risk factors in the bottom group of clinical high risk who develop a psychotic disorder and the lowest density in the general population. But if we can start merging self-report and experience data with other known risk factors, I think that we're going to get much closer to our goal of being able to identify risk. So I want people to be hopeful, but I want you to be cautious. So if you're going to use screening tool with adolescents, consider the population. Are you screening someone in a primary care setting or in a online survey for your clinical high risk research? Think about the item endorsement as the starting point, not the end point. So when I give a community provider a screen, a self-report screen to give to their client, I don't give them a threshold. I ask them to ask that person what they are experiencing when they endorse an item to have them tell the clinician more about that experience and how they interpret it and how it makes a difference in their lives, how it has changed its dynamics. Don't overvalue thresholds. So again, I'm not saying don't screen, but I am suggesting that we need to be cautious about the published thresholds that go with screening tools that have been tested in often very small samples. But do take serious if someone is indicating distress or impairment due to an item, but this does appear to be much more predictive of later concern. And do take time and encourage clinicians who are using these screens to talk to the adolescents, the parents, about their experiences to find more about it, more about the texture of that experience and the impact of that experience. And just increasing people's ability to talk about these experiences, I think, can go a long way. And then finally, I would really encourage us not to use categories like screen positive, screen negative, except in research settings or when a screen is needed to be able to prioritize very precious resources. So if you cannot see everybody and you use a screen, that makes a lot of sense, but just be cautious about categories and categorizing people accordingly. So I'm going to stop there, and I really look forward to some questions and comments and weighing in with the screening experts who are on the call. Thank you, Kristen. What a wonderful, thorough, and expansive presentation. And I echo the comments Dr. Taylor just made, being so thankful for you raising these issues and really recommending our caution, really, around this process of screening. And while we also feel like it's very important to be able to try to get clearer about who are the young people who might be at greater risk, there are so many variables and factors that you pointed out for us so beautifully. So thank you very, very much. Some of the initial questions that have come up are really also around, even with the different age groups that we're talking about, even across the adolescent spectrum, one of the questions that came up from Denise very early on is, so you're really also suggesting that if we're looking at potentially different adolescent populations, we're maybe looking at utilizing not only different screens potentially, but part of what you're also talking about is different processes depending on the situation that the young people are in. So I'm wondering if you can just clarify some of that for us again. Yeah. So hopefully I have that understanding. I think we don't have enough data on the young adolescents to be confident about some of the assumptions that we make with our screens that were tested on adolescent and young adult samples. So I just think that as we're interviewing more and more young adolescents, which I think we should be doing, and as we're screening and asking them about these experiences, I think that the data on the screening tools becomes a little less relevant and we need to be a little bit more clinically savvy in asking those questions and listening to what adolescents are saying and a little bit more humble about what we know about what those mean. Yeah, really, really true. And I know that people are also developing broader screening tools and looking at this question of incorporating specific questions related to clinical high risk and early psychosis within the context of broader, more general screens. What kinds of thoughts do you have about looking at these questions in relation to a broader screening process? Yeah, I love that question. I have a couple thoughts on that. So I think that because of the amount of stigma that's still out there for psychosis specifically, I think embedding risk for psychosis in broader screens is really the first step that we need to take because we need to be including these kinds of experiences in the questions that we're asking about mental health changes that adolescents experience so that people get more comfortable talking about them and so that adolescents understand that this happens to other kids. The downside of that is that when you have a broad screen, you can only ask a few questions. And I, as someone who's interviewed a lot of adolescents, as I know many people on the call have, I worry about when we ask just a few questions that we're losing sensitivity, that we're missing a lot of experiences that are initial experiences of psychotic disorders that may not be captured, let's say in auditory hallucinations or visual hallucinations. So ultimately, I hope that we can move towards some broad general screens that ask about psychosis and eventually some much more nuanced psychosis-specific screens that capture a variety of experiences and have good norms. The other side of your question, I think, is how much should we be screening for psychotic symptoms versus other risk factors? And I think that slide with the different risk factors probably best illustrates my thoughts on that. But again, I welcome other people's comments that there's a lot to be said about functional decline, about social withdrawal, about nonspecific risk indicators, sleep disruption, that when we add them and when we get better risk calculators that can incorporate some of those factors with the more positive psychotic-like experiences, I do think that's going to work much better for early detection. The problem is we ask about some of those experiences already and we're not asking about psychosis. I kind of focused on the psychosis piece, but it's not that that needs to be our primary target. Thank you. Thank you for clarifying that. One of the questions I'm trying to go back to, there was a question that was asked about around the risk factors that are included, I think, in sort of your later slide that had maybe the dots on it with the different kinds of potential risks. Can you clarify what any of the risk factors are that were represented potentially by those dots, do you think? That is a great question. I don't have the paper right in front of me at the moment, but I think that there's a mix of demographic risk factors. You have age, you have sex assigned at birth, you have population, whether someone's from an urban setting or rural setting, so socio-demographics, you have exposure to certain kinds of adverse events. There's really a wide range of kind of risk factors that I remember. Some of their work is really looking at risk factors that can be identified within a medical record chart, for instance, so things that may be documented, and moving to a capacity to maybe trigger a psychosis risk screen, just looking at what kinds of risk factors cluster together. Okay. One of the themes that keeps seeming to pop up in the comments and questions is sort of the tension around, for the people that are working in maybe larger clinical programs, how do you appropriately sort of maybe utilize a screen for what is already maybe a general clinical population versus then how do you look at that differently maybe when you're... One question was around like when you're a child psychiatrist working in a clinical program where people have already been referred to you for what may already be at least some more in-depth assessment around a particular psychiatric issue, and this issue of how you balance the nuances of sort of the screening of the larger group in a general clinical setting versus looking at the possibility of more intensive levels of symptoms or potential psychopathology potentially. Yeah. I feel like I don't know that I have an answer to that question. I'd be interested in other people on the call. Yeah. I'm chiming in because I am really eager for us to get better data to answer those questions. I think that we're all straddling kind of what the data are and the individual that's in your office presenting with certain symptoms, and you're trying to make your best estimate of what that means. Yeah. One of the questions that also has just come up is when you've been involved in these kinds of situations or really if a provider is looking at using screening tools, do you have recommendations for how they then might have that conversation with that young person and then maybe also with their family about the results of that screening? I'm trying to think about it in terms of your point about not having positive or negative screens per se, but how then would you maybe talk with a young person and then their family about screening results within these contexts? Yeah. That's a great question. In my experience, we try to stay as close to that person's or family's own words and own experience as possible. You really want to start with their level of concern or the changes that they're noticing and validate those. There is usually a measure of normalizing the fact that these kinds of experiences occur on a continuum and that they can be transient or they can be longer lasting, that they can remit completely or they can get worse. There's a lot that we don't know about predicting that, but the treatments that we have are generally helpful for people learning how to manage those experiences, to be able to embrace activities and relationships that are meaningful to them in spite of the experiences if the experience continues, and to really reassure them that they have a partner in trying to manage that. Now, again, having that discussion if you're not the person who's going to do the treatment is different than if you're the person who can provide that treatment, but providing a sense of hope that is not naive to the fact that these experiences make it worse before they get better or that they may not remit entirely is tremendously powerful. I think it's important to not add to that person's meaning-making too prematurely, where we're adding our own labels and our own scientific concepts to something before we have reason to believe that those apply to the person, but that you really stay with the person's direct experience. If they're confused about what's real or not, we want to talk with them about how to become more grounded in reality to be able to question the thought versus the experience without adding too much layer about the science of that right away. Now, that said, there are plenty of families that I've worked with for whom the science and resources on the web are extremely helpful and reassuring, so I don't think there's a one-size-fits-all with this, but I think starting more cautiously and staying nearer to the person's experience is usually good until you get a sense of what their response is. Thank you. I know you mentioned one other area that sort of needs to be further studied and clarified around some of the religious and spiritual kinds of issues and also in many of the cross-cultural issues as it relates to utilization of screening and cultural interpretations of illness as well. I know that as I've worked with different cultural groups and looked at some of these psychosis screens, I've often been given feedback around how some of the questions really are maybe really more perceived more openly and interpreted really differently across different cultural groups. I really wonder at times how one assesses questions and interpretations, whether it's of questions related to voices or questions related to being able to predict things that might happen in the future or other kinds of spiritual or religious or other kinds of experiences that people might have or that might fit within their role within a certain spiritual group or culture, and I wonder just if you might comment on some of the complexities around addressing some of those kinds of issues as we think about building screening into our processes. This is not my area of expertise, but I can certainly speak to my experience that asking about how other people in that cultural group would answer that same question or whether they have similar experiences is a typical starting point, so either asking the young person themselves. Are there other people in your family or in your peer group or your church group that have the same experience? Is there a way that your experience is different from theirs? Also, asking people within, let's say, the religious group whether that young person's experience is typical of other people in that group or whether there are ways in which it is outside of that more common experience, so we're I think always using some clinical judgment to get a sense of whether someone's experience is taking off from something that is very centered in a group experience or shared religious belief and becoming more idiosyncratic, more intense, more problematic, and that's a tricky judgment call, but I think checking it out with other people within a given community is often essential. Well, thank you, and thank you. I know our time is pretty much up here. Thank you for a wonderful, wonderful, and illuminating presentation. I know, Judith, you want to take us through the last couple of slides before we stop, but again, thank you, Kristen, very much for really a wonderful, wonderful presentation for us today.

screening for psychosis

adolescents

Dr. Kristen Woodbury

informed consumerism

screening tools

psychosis prevalence

predictive validity

psychotic experiences

mental health screenings