Increasing Rigorous Evaluation of Interventions to Reduce Gender-based Violence Victimization

 

Lauren Reed: Welcome everyone it's good to have you all here as we're waiting for people to jump into the webinar we'll get started in just a minute.

Lauren Reed: and welcome everyone we'll get started in just a moment, as we wait for folks to join.

Lauren Reed: Okay welcome everyone, and thank you so much for joining us today.

Lauren Reed: And my name is lauren read and i'm an assistant professor of social work at Arizona State University in the office of gender based violence and.

Lauren Reed: I am here moderating this webinar today so i'll be introducing our wonderful presentation in just a moment, and thank you for coming to this webinar today, which is.

Lauren Reed: put on by the team at and I J and the violence against women consortium.

Lauren Reed: As an ongoing series to highlight the innovations and exciting work being done in the field of.

Lauren Reed: Violence against women, and because this is February teen dating violence awareness month we are excited to bring a presentation by Dr Joe Cohen.

Lauren Reed: on redefining trauma informed assessments for teen dating violence and introduction to the VIP study, which I will let Dr Cohen explain in just a moment what the VIP study it.

Lauren Reed: And we want to acknowledge that this webinar is a team effort from the records, violence against women consortium the University of Maryland school of social work and the office of continuing professional education and the and I and J, so thank you to this team for supporting this webinar.

Lauren Reed: We also want to provide the disclaimer that the opinions or points of view expressed in this presentation, are those of the presenters and do not necessarily represent the official position or policies of the US Department of Justice.

Lauren Reed: And our objectives, today, so this is what we will be covering.

Lauren Reed: You will be able to define teen dating violence and summarize its prevalence and impact on society this presentation will distinguish between trauma focused and risk factor prevention paradigm.

Lauren Reed: or rfp informed approaches for risk assessment, it will explain from a developmental trauma perspective, what it means for dynamic risk or strength to be trauma informed.

Lauren Reed: And finally, described thresholds for how to define whether a screening measure is clinically useful and statistically fair.

Lauren Reed: so excited to get into all of that, and this webinar is expected to take 75 minutes, including this introduction that we're in right now the presentation will be about 30 to 40 minutes, and then the rest of the time will be for Q amp a.

Lauren Reed: So if you have questions Please submit those there's a Q amp a box.

Lauren Reed: That you can submit questions to, and unfortunately we don't have an opportunity for attendees to jump on and ask questions through audio so make sure to.

Lauren Reed: submit your questions, through that Q amp a box and we'll be excited to have Dr Cohen, to answer those the slide deck from today's session that you're seeing and the recording of this presentation will be made available and about 10 to 15 business days, so this will be recorded.

Lauren Reed: And I would like to present our speaker for today, Joseph Cohen, who is coming to us from the University of Illinois at urbana champaign, so we are honored to have him with us today.

Lauren Reed: and Dr Cohen, is a licensed clinical psychologist and associate professor in the department of psychology there at the University of Illinois urbana champaign.

Lauren Reed: Dr cohen's research seeks to elucidate the pathways to distress and impairment, within the context of interpersonal trauma.

Lauren Reed: And how to translate these findings into trauma informed and culturally sensitive risk assessments for youth, so we are honored to have Dr Cohen here with us today, and we look forward to hearing about his work, thank you.

Joseph R. Cohen: All right, can everyone see my screen okay.

Joseph R. Cohen: Maybe maybe.

Joseph R. Cohen: All right, i'm gonna go ahead and get started well, thank you very much lauren for that introduction and thank you to the team and I J University of Maryland and my Alma mater rockers.

Joseph R. Cohen: for hosting this and including me in the program and i'm excited to be part of the programming here for teen dating violence awareness month, so the title of my talk is redefining trauma informed assessments for teen dating violence and introduction to the VIP study.

Joseph R. Cohen: One second here I think i'm visible try to correct that quickly.

Joseph R. Cohen: Hopefully that's better.

Joseph R. Cohen: we're going to go with it, so I thought we'd start with trauma informed to something gets bantered about a lot and it's on the.

Joseph R. Cohen: verge for some of kind of losing meaning and what we're talking about and what we're not talking about so I thought it would be a good place to start to talk about why trauma informed solutions are important for behavioral health.

Joseph R. Cohen: So the first reason is, is that trauma is one of our most robust predictors of distress and impairment in adolescence.

Joseph R. Cohen: And, as a result of it being such a strong predictor of so many behavioral outcomes it's important for us to consider trauma and potentially traumatic events within our risk models.

Joseph R. Cohen: A second reason is that individuals who are exposed to trauma tend to be more treatment resistant to the prevention protocols, we develop from an evidence based perspective.

Joseph R. Cohen: They have more difficulties engaging with the protocols themselves, they also have reasons that they might feel less comfortable in the settings in which these prevention's.

Joseph R. Cohen: are being established, and so, as a result of that there's reasons for us to focus on trauma informed solutions to make sure we are creating preventive solutions that could be inclusive of those who experience into personal trauma.

Joseph R. Cohen: And the last reason is that we have significant health disparities for a myriad of behavioral health outcomes, this includes teen dating violence as we'll talk about today.

Joseph R. Cohen: And the most data driven and comprehensive explanation for why we have health disparities for these behavioral outcomes.

Joseph R. Cohen: Is because exposure to potentially traumatic events is not evenly distributed in our society, and so, in order for us to start to attenuate these health disparities that we have it's important that we build risk models around these environmental toxins.

Joseph R. Cohen: At times I could feel like we're running in circles doing trauma informed work, but I think there has been some really good.

Joseph R. Cohen: signs of progress in this field and we're going to use Sam says overarching definition of what is trauma informed to discuss where we've had some good progress and where we might be realizing some gaps in our work.

Joseph R. Cohen: So the first is, is that we are seeing a huge increase within our public mental health and other public settings for reinforcing the need for trauma informed work.

Joseph R. Cohen: There was some recent policy work out of drexel university that looked at federal bills from 1973 to 2015 and what they found was in 2015 alone.

Joseph R. Cohen: More than 50% of the bills that discussed trauma informed care were represented, that means we have increasing awareness at the federal level and policies level for trauma informed work.

Joseph R. Cohen: In my own anecdotal experience i'm routinely impressed as a clinical psychologist working with our local juvenile justice system or school systems.

Joseph R. Cohen: Even our local hospitals how intuitive it is too many of the people who are providing services to youth that the behavioral problems they're encountering are a reflection of the environments that they've been exposed to.

Joseph R. Cohen: However, I think conversations with those providers at the local level also elucidate where we're falling behind.

Joseph R. Cohen: So what what they're constantly looking forward as tools or protocols or policies that allow them to be more proactive in the way they act in a trauma informed way and to be provided more specific recommendations on how to respond in a trauma informed way.

Joseph R. Cohen: So the goal of our research is to to try to inform specific practices and how they might be able to become more trauma informed specifically from a developmental trauma perspective.

Joseph R. Cohen: In the practice we're going to be talking about today is what is called a risk assessment for health.

Joseph R. Cohen: Now risk assessments are increasingly being used across a variety of contexts, the juvenile justice system, schools and hospitals.

Joseph R. Cohen: They all share one common goal risk assessments are intended to give some kind of insight into how the youth is currently functioning.

Joseph R. Cohen: In addition to give some kind of prediction for what their risk might look like in the future within the context of the juvenile justice system, these risk assessments are mandatory.

Joseph R. Cohen: The front line purpose of these risk assessment is intended to guide preventive interventions for youth, how can we kind of tailor.

Joseph R. Cohen: Rehabilitation services to meet the needs that they have it can also be used to inform legal decision making and so risk assessments play an outsized important role within the context of the juvenile justice system.

Joseph R. Cohen: For the past 30 years risk assessments have been predominantly informed by what some called the risk factor prevention paradigm.

Joseph R. Cohen: This approach, seeks to focus on the dynamic individual differences that are targeted by preventive interventions so made popular by farrington and others with the goals of these ours to see.

Joseph R. Cohen: What can we bend in shape that is related to the outcomes, we hope to ameliorate and then use those on the front line of screening and assessment.

Joseph R. Cohen: And for many years, this approach went unquestioned as it was seen as a big improvement compared to previous generations of risk assessments, which focused on things like static demographics parent functioning things like this.

Joseph R. Cohen: However, more recently, the risk factor prevention paradigm has been increasingly question by investigators.

Joseph R. Cohen: One has been the question of what individual differences do we choose to include in our assessment and whether these individual differences actually equally predict things like violent behavior across different adolescent subpopulations.

Joseph R. Cohen: The second thing and something that was very motivating to me and getting into this field was that there's no clear.

Joseph R. Cohen: delineation of whether the risk factor prevention paradigm is in fact trauma informed there's no assessment of the structural risks or the environmental toxins that we know from a research perspective are critical and protecting predicting behavioral health outcomes.

Joseph R. Cohen: As a result, we've seen a recent wave of recommendations that suggest including in the risk assessment, an explicit measurement of adverse childhood experiences, some people might experience.

Joseph R. Cohen: So this could include the original aces screening measure that has become very popular which assesses five forms of.

Joseph R. Cohen: treatment and five other forms of family adversity related to impaired caregiving and poverty related stressors and there's many modifications of the aces screening.

Joseph R. Cohen: And while i'm sympathetic to this movement, I, like the fact that it causes some of the structural risks that me as a developmental psycho pathologist know is so potent.

Joseph R. Cohen: As a clinical psychologist I have some translational concerns and ethical considerations related to this.

Joseph R. Cohen: So, from a translational perspective, one thing that was really nice about the risk factor prevention paradigm.

Joseph R. Cohen: was in many ways it was ahead of its time we hear a lot now about precision medicine or personalized medicine.

Joseph R. Cohen: The risk factor prevention paradigm when exemplified by models, like the risk need responsibility model are intended to tailor the interventions to the actual areas of risk the kids need the most help with.

Joseph R. Cohen: within an ace is screaming or trauma focused meaning, we lose that piece, because we know that trauma can impact kids and so many diverse ways.

Joseph R. Cohen: More concerning, for me, though, are the ethical considerations of asking kids in these these questions places like the juvenile justice setting places like the school places like the hospitals.

Joseph R. Cohen: Do we want these settings to understand these adversities.

Joseph R. Cohen: That these families might have impacted do we want them to orient services or legal decision making, based on the static events in the past, I think there's significant concerns.

Joseph R. Cohen: That we could even talk about at the end about moving in that direction.

Joseph R. Cohen: So this tension both the strengths, the pros and the cons of both of these approaches was very much the motivation for the VIP study.

Joseph R. Cohen: How do we develop a trauma informed risk assessment protocol that retains the benefits of the rfp in which we're focused on dynamic risks, but can better elucidate or better highlight the structural and the structural risks and the environmental toxins these kids face.

Joseph R. Cohen: And so we sought out to focus on individual differences that stem specifically from potentially traumatic events.

Joseph R. Cohen: We saw to focus on industry individual differences that could not only inform clinical decision making, from a quantitative perspective, but also focused on the indices that are targeted specifically by trauma informed interventions.

Joseph R. Cohen: And finally, we thought it was important from the get go to be focusing on individual differences that are invariant across significant demographics so trying not to replicate some of the biases that are involved in the current rfp risk assessments that are routinely used.

Joseph R. Cohen: In order to do this we focus on what's called a latent vulnerability approach, so the latent the concept of a latent vulnerability.

Joseph R. Cohen: has been around for many decades, but the actual terminology was coined by human mccrory was a psychiatrist in England and what he suggests in.

Joseph R. Cohen: In his own research and also in Meta analyses and little literature reviews that he's conducted.

Joseph R. Cohen: Is that there's measurable alterations of neurobiological individual differences due to exposure to adverse environments or potentially traumatic environments.

Joseph R. Cohen: He suggests the three different deficits that might emerge are hyper vigilance to threat, this is defined by biased responses to neutral stimuli or mildly negative stimuli.

Joseph R. Cohen: deficits in reward processing, this is how we approach or how we experience positive environments or positive stimuli.

Joseph R. Cohen: And finally, a camp of discrete risks associated with social emotional well being that can stem anywhere from emotion regulation to psychological well being.

Joseph R. Cohen: what's exciting about my priorities model is that has also been extended into behavioral paradigms so we know it is not just biological risk factors.

Joseph R. Cohen: That can be organized this way, but we can also start understanding individual risk of the behavioral level associated with these three camps of risks.

Joseph R. Cohen: The other thing that's really nice about mccrory is model.

Joseph R. Cohen: Is that it highlights risks that are taught the key theoretical mechanisms of risk within our frontline trauma informed care models so hyper vigilance to threat.

Joseph R. Cohen: can be intervened with things like exposure based therapy or narrative therapies reward processing can be promoted and improved through behavioral activation and contingency management protocols.

Joseph R. Cohen: and emotion regulation can be targeted to a wide swath of cognitive behavioral paradigms that see to promote self and social understanding, so it hits a lot of our kind of translational aims in developing a dynamic trauma informed assessment model.

Joseph R. Cohen: And so big question for the VIP study which i'll talk about in this in a second is can we measure these different domains of risk using self report data in in vulnerable adolescent populations.

Joseph R. Cohen: And the first extension of the VIP project was specifically focused on teen dating violence.

Joseph R. Cohen: As an outcome so so a lot of my training was actually not in the violence literature i'm a developmental.

Joseph R. Cohen: trauma researcher developmental psychopathology and we look at various outcomes that might stand from potentially traumatic events, so the study i'll talk about today was in very in many ways, one of my initial steps into teen dating violence.

Joseph R. Cohen: But it was a compelling area for me to look into for several reasons, one, we know that this is a robust predictor of poor health and social functional impairment, not only an adolescence, not only the distressing individual experiences.

Joseph R. Cohen: In adolescence if they're involved in a violent romantic relationship, but how many of those outcomes persist into emerging adulthood and adulthood after that.

Joseph R. Cohen: teen dating violence can be defined by a wide range of harmful partner directed behaviors today we're just going to be talking about to.

Joseph R. Cohen: Which are just to have many subtypes in which it can manifest will be talking about physical forms of teen dating violence.

Joseph R. Cohen: acts such as shoving slapping kicking choking or burning and sexual acts related to teen dating violence which can range anywhere from unwanted sexual touching to forced penetration.

Joseph R. Cohen: Maybe the most surprising thing to me as somebody who was coming to a bit new is how prevalent teen dating violence is.

Joseph R. Cohen: So as a depression researcher, we are often oriented to depression as a public mental health crisis because of how many adolescents it impacts.

Joseph R. Cohen: That we see teams examples of teen dating violence or almost justice prevalent or in some examples even more prevalent than rates of depression, based on epidemiological research.

Joseph R. Cohen: It was also compelling to me that we see significant significant demographic disparities, both with regard to gender, as well as regard to race and ethnicity.

Joseph R. Cohen: In response to these alarming trends for the past 20 years there's been a lot of attention on what our risk factors for who goes on to develop.

Joseph R. Cohen: teen dating violence perpetration The goal of this research is if we can better identify who's more likely to develop these behaviors we could intervene with those individuals earlier on.

Joseph R. Cohen: virgie from the centers of disease control has a very useful.

Joseph R. Cohen: Review of the literature and, within that review they identified 53 different risk factors that might be related to individuals who go on and develop physical and sexual TV perpetration.

Joseph R. Cohen: Back in 2018 we sought to distill that list further we saw to we sought to examine which of these predictors might be incrementally valid.

Joseph R. Cohen: and which of these predictors might be clinically useful which we'll talk about a little bit later within the VIP study.

Joseph R. Cohen: And what we found was that domestic violence exposure maltreatment exposure TD victimization in addition to to relationships specific variables conflict resolutions and acceptance of teen dating violence were the most salient predictors.

Joseph R. Cohen: And so from this study, it was clear to me that a trauma focused assessment for teen dating violence prevention might have to be trauma focused and the fact that it is.

Joseph R. Cohen: It is one of our strongest predictors in fact that trauma can be used to explain the significant health disparities, we see and could a latent vulnerability approach be helpful, with this outcome.

Joseph R. Cohen: And so, this is our theoretical model for the VIP study is looking at whether the risk that is conferred by family violence examples of male treatment or domestic violence exposure, as well as TV victimization can be explained by these areas of latent vulnerability.

Joseph R. Cohen: And so, for the VIP study I collaborated with the University of Texas medical branch, and so we recruited adolescence from a school sample.

Joseph R. Cohen: outside of Houston that's where the majority of our kiddos come from 507 and we also recruited a high risk sample from here in urbana champaign at the at the diversion program we recruited 77 adolescence.

Joseph R. Cohen: We were successful in recruiting a diverse sample approximately half female about even for adolescents who identified as white or black representative latinx population as well as close to 20% of adolescents who self identified as a gender minority.

Joseph R. Cohen: We were able to retain of that 584 approximately 70% so 458 adolescence, and I very purposely put the dates up there, as you can see our follow ups.

Joseph R. Cohen: happened over the covert pandemic both the onset really have a social response to the pandemic, as well as the continued effects of it during the school and i'll talk about that for a little bit.

Joseph R. Cohen: here's our measures for the study, one thing I want to point out is, if you look under hyper vigilance to threat there's two different measures that are used.

Joseph R. Cohen: hyper vigilance to threaten youth can kind of manifest in two ways, it can manifest in kind of the classic way we think of it in that avoidant way.

Joseph R. Cohen: In which kids might be easily fearful or avoidant of experiencing the stimuli that's neutral or mildly negative in nature.

Joseph R. Cohen: This we measured with the post traumatic recognitions inventory, we also know that it could result in kids having an overactive approach system.

Joseph R. Cohen: That they are that they are impulsive and that they are, they are, they are drawn to conflict when provoked by neutral stimuli, and so we use two different measures to capture those two aspects of hyper vigilance to threat.

Joseph R. Cohen: OK, so the aims of the study.

Joseph R. Cohen: aims of the study where what potentially traumatic events and or layton vulnerabilities predict physical and sexual subtypes of teen dating violence which significant predictors may be clinically useful in which significant predictors are statistically fair.

Joseph R. Cohen: To give some context for this study at baseline about 22% of those in relationships reported being involved with.

Joseph R. Cohen: perpetrating teen dating violence physical subtype in the past 12 months about 12% reported perpetrating sexual subtypes of teen dating violence in the past 12 months.

Joseph R. Cohen: This is remarkably similar to some of the past estimates me and my team.

Joseph R. Cohen: have collected study, led by range shorey at the university Wisconsin milwaukee in which we found about 22% and 11% reported perpetrating physical and sexual TV respectively.

Joseph R. Cohen: One thing i'll point out is that you'll see our 12 month rates at the end of the study significantly decreased about 13% for.

Joseph R. Cohen: teen dating violence, physical and 8% for teen teen dating violence, sexual now people who do longitudinal research are accustomed to this trend in research, in which our main outcomes often decrease, the more we assess them.

Joseph R. Cohen: what's interesting about these statistics is that we don't see a similar trend for things like psychological teen dating violence, and so we see the decreases here in teen dating violence being related to social isolation and the covert pandemic.

Joseph R. Cohen: Okay, I am going to briefly go through the data analytic plan, because it can get a little bit in the weeds but certainly it could be something that we revisit in terms of questions.

Joseph R. Cohen: So our main aim was tested within structural equation modeling environment, a first step of this study, as we talked about was just can we measure these latent vulnerabilities in a way that's consistent.

Joseph R. Cohen: With the theory and we can we found good temporal and demographic and variants across our three late vulnerabilities.

Joseph R. Cohen: One thing that we found was that we did not find a significant slope in the latent vulnerabilities which prevented us in our preliminary data collection, we test our full mediation model.

Joseph R. Cohen: However, more importantly, we did find significant intercept for the latent vulnerabilities at the intercept level.

Joseph R. Cohen: Which means we could compare kids along with the temporal and demographic and very and some of the confirmatory factor analysis he could compare kind of the means that are derived from the scores is meaningful and different amongst adolescence.

Joseph R. Cohen: We also found good model fit for what some call simplex model.

Joseph R. Cohen: Which is an example, the latency trade model, which suggests that are latent vulnerabilities when measured over time share enough of variants for it to be considered a predisposition.

Joseph R. Cohen: They also have significant state differences.

Joseph R. Cohen: Over the course of the year, which suggests that there might be individual growth patterns that exist at a nonlinear way in which we could measure in the future it's also consistent with the risk factor prevention paradigm.

Joseph R. Cohen: that these are predispositions that can be dynamic in nature, potentially traumatic events and tdd perpetration across the analyses were treated us to fix the effects as we don't conceptualize those as leading variables.

Joseph R. Cohen: Okay, so for our preliminary findings in terms of what actually predicts teen dating violence in the VIP study we found teen dating violence of victimization was the most incremental predictor for both subtypes of teen dating violence at baseline.

Joseph R. Cohen: While we found hyper vigilance to threat, specifically the approach system predicting physical forms of teen dating violence and deficits in reward learning predicting both physical and sexual subtypes of teen dating violence.

Joseph R. Cohen: In a simultaneous model, we found that the specificity concerning hyper vigilance to threat and reward held for physical and sexual subtypes respectively, and we also found that TV victimization remains significant predicting both at baseline.

Joseph R. Cohen: In terms of our prospective models, we found that both TD victimization as well as emotional maltreatment emerges significant predictors.

Joseph R. Cohen: We also found for our perspective models, most of the latest vulnerability models from baseline replicated we saw hypervigilance to threat for approach.

Joseph R. Cohen: predicting physical subtypes of teen dating violence, one year later, we also found the same for reward learning predicting sexual subtext of teen dating violence.

Joseph R. Cohen: Initial finding concerning psychological well being was no longer significant once we considered the significant adversities to in our simultaneous model we saw the same thing with hyper vigilance to threaten reward predicting physical and sexual subtypes of teen dating violence respectively.

Joseph R. Cohen: So summary from the predictive model is that trauma focused algorithms TD victimization emerged as a robust predictor at baseline we see different factors predicting future TV perpetration.

Joseph R. Cohen: Which is a common thing we see in risk assessment indicators for the hearing now we're not always the same for the indicators in the future.

Joseph R. Cohen: From the latest vulnerability model perspective that we do see consistency.

Joseph R. Cohen: We see hypervigilance to threat specifically approach predicting physical forms at baseline and 12 months later and anticipatory reward or reward learning predicting sexual subtypes of teen dating violence across the study.

Joseph R. Cohen: A major emphasis of the VIP study is to begin using a translational analytic plan.

Joseph R. Cohen: So structural equation modeling creates a fantastic opportunity to test our theories related to teen dating violence.

Joseph R. Cohen: They are difficult to translate into clinical practice, and so a big goal of this study is to use a structural equation modeling framework to test our underlying.

Joseph R. Cohen: Findings but then use an evidence based medicine perspective to see whether these things might actually be clinically useful when we treat our outcome around.

Joseph R. Cohen: treat our outcomes categorical to facilitate decision points and decision making.

Joseph R. Cohen: And so just for today's talk we're going to be focusing on to indices from the evidence based medicine perspective.

Joseph R. Cohen: These are receiver operating characteristics, where we have we generate an area under the curve to determine how good are we at classifying.

Joseph R. Cohen: Who might go on to be a perpetrator versus who is not a perpetrator and we also use diagnostic likelihood ratios, which provides insight into if somebody is going to perpetrate how many times more likely, are they to perpetrate.

Joseph R. Cohen: These are orthogonal constructs.

Joseph R. Cohen: Okay, so from a clinical utility perspective for our baseline we see some patterns emerging, we see that TD victimization is really a robust classifier for, who is currently perpetrating violence.

Joseph R. Cohen: So we see that both for physical and sexual subtypes of teen dating violence as a reminder point nine o is viewed as an excellent classifier within our oC analyses above 4.0 for diagnostic likelihood ratios suggest clinical utility.

Joseph R. Cohen: But we also find our latest vulnerability approach also exceeding those benchmarks, we see Point six eight for the combination of reward and hypervigilance to threat for physical subtypes of teen dating violence and a DLR that exceeds 4.0.

Joseph R. Cohen: In terms of our prospective predictors.

Joseph R. Cohen: We really only see a latent vulnerability approach reaching the benchmarks that are using with an evidence based medicine perspective.

Joseph R. Cohen: in which the predictor might be useful, so we see an area under the curve of Point six, nine and a DLR above two we see both emotional maltreatment and reward for sexual subtypes of teen dating violence failing to reach those benchmarks.

Joseph R. Cohen: And so what are some of the findings from terms of clinical utility we do find that a combination of reward and hypervigilance to threat to approach.

Joseph R. Cohen: Maybe a clinical useful predictor of TD V physical at baseline and that hyper vigilance to prep approach on its own, maybe a uniquely useful predictor of prospective physical subtypes of teen dating violence.

Joseph R. Cohen: The last thing I want us to introduce us to is this concept of statistical fairness.

Joseph R. Cohen: This is something that us in our lab have recently taken quite an interest to like many on this call, I was maybe oriented to the concept.

Joseph R. Cohen: That, if you if you include an interaction between fill in the blank demographic and your predictor and it was not significant You conclude that your findings are equivalent across subpopulations.

Joseph R. Cohen: But there's been a lot of good work recently on the University of Pennsylvania specifically within the context of risk assessment.

Joseph R. Cohen: That suggests that the odds ratios that regression equations generate can obscure all kinds of examples of statistical unfairness and bias within the risk assessments.

Joseph R. Cohen: And so we have done some work recently as suggest using the aces or using a trauma focused approach might violate several principles of statistical fairness.

Joseph R. Cohen: And so, for the today's presentation we're going to find.

Joseph R. Cohen: Statistical fairness in two ways the first is, does the algorithm classify risk differently, an adolescent subpopulations we're going to compare area under the curve, in order to make that determination.

Joseph R. Cohen: Secondly, is are the odds of perpetrating differentially calibrated across adolescent subpopulations.

Joseph R. Cohen: To put simply This suggests that if you have an indicator that suggests if boys are three times as likely to perpetrate if they have X risk factor our girls also three times similarly like we to perpetrate where's the risk associated with them much greater.

Joseph R. Cohen: In terms of gender, we find we find striking when we find striking differences in how our latest vulnerability approach works, but not how the trauma trauma focus model.

Joseph R. Cohen: works and so here for TD victimization we see very similar areas under the curve both hovering around Point seven zero.

Joseph R. Cohen: However, when we look at our latest vulnerability stratified by gender, we see the latest vulnerability approach is much more successful in classifying risk for males compared to females.

Joseph R. Cohen: Lots of talk amongst our research team and our investigative team, whether this is a strength, whether this is a weakness, whether this is neither a strength or weakness.

Joseph R. Cohen: And so that could also be a discussion point for us in terms of how should we consider the development of risk assessment for teen dating violence differently.

Joseph R. Cohen: Across the dimension of gender.

Joseph R. Cohen: In terms of race, the findings are very inconclusive, we see for area under the curve, we see both a trauma focused and late invulnerability approach better classifies risk in those adolescents who identified as white.

Joseph R. Cohen: And that's indicated by the higher area under the curve that are bolted there.

Joseph R. Cohen: But then we see higher diagnostic likelihood ratios for those who do not identify as white.

Joseph R. Cohen: And again, you know we're still we're still kind of diving into kind of how can you have this reverse pattern of area under the curve and diagnostic like we've had ratios like I said the orthogonal but it suggests that there might be.

Joseph R. Cohen: Areas that we could kind of explore their to understand how the algorithm might be performing differently as a function of race and ethnicity.

Joseph R. Cohen: So, in terms of summarizing the statistical fairness, we can conclude that late and perform late vulnerabilities perform better for classifying perpetration in males and that trends concerning race and ethnicity or difficult to detail at this time.

Joseph R. Cohen: Part of the reason I wanted to include in this talk, though, was just to get us all thinking about what should the goals for our risk assessment perspective be from a statistical fairness.

Joseph R. Cohen: Okay, there are some limitations, I just want to mention quickly.

Joseph R. Cohen: So one is that the you know I don't know if you heard but we've been dealing with this pandemic and the study overlay overlapped with this pandemic and so you know.

Joseph R. Cohen: Not only was there decreased the TV perpetration but we're doing some work now to understand how kind of qualitatively.

Joseph R. Cohen: teen dating violence might have changed during the pandemic and look differently and so from a risk perspective or risk assessment perspective, you might be predicting a different qualitatively different outcome during a pandemic, as opposed to when a pandemic is not going on.

Joseph R. Cohen: You know and i'd also like to highlight that our outcome in this preliminary VIP study was relying on a static cadre score and the cadre itself has limitations and anytime you're relying on a score at a single time point that also has limitations.

Joseph R. Cohen: Here are some of the projects, concluded that the.

Joseph R. Cohen: January 1 of this year, so we still are knee deep in ongoing analyses or curious about the Cross cutting nature of our latest vulnerability approach for other outcomes that might be of interest, not just the juvenile justice system, but also schools and hospitals.

Joseph R. Cohen: One goal of our one of our overarching goals is to align risk assessments across Community settings for at risk adolescence, we think this will allow for more cross cutting preventive interventions as well.

Joseph R. Cohen: we're also one one aspect from the study that produce no findings was that we had a lot of.

Joseph R. Cohen: We had some angels to protect and promote factors that we did not find to be significant.

Joseph R. Cohen: And so we have some other secondary measures that we collected, to see if that could buffer some of the risk associated with layton vulnerabilities or directly predict TV perpetration outcomes and so we're continuing to examine those things in an exploratory way.

Joseph R. Cohen: We were fortunate January 1 to receive continuing funding to follow this group of 584 adolescents for the next five years for annual assessments, we are super excited and grateful that we get to continue to do this work.

Joseph R. Cohen: The main aim for this is to test whether a latent vulnerability model can predict outcomes across the adolescence adulthood transition.

Joseph R. Cohen: And so, this will also allow us to conceptualize risk video graphically we have good reason to think that we will find individual growth.

Joseph R. Cohen: terms, and so we are curious what that means and predicting risk.

Joseph R. Cohen: we're also looking forward to developing more holistic algorithms for TV perpetration.

Joseph R. Cohen: Like I said, because our hypotheses concerning protective and promoted factors were now significant we've certainly expanded our consideration of what could be both culturally relevant and developmentally appropriate for protective and promoted factors for this outcome.

Joseph R. Cohen: And we are also focused on having a more nuanced take turpin concerning identity so using dimensional measures of gender identity of racial ethnic identity.

Joseph R. Cohen: As well as using an analytic approach, called the might approach.

Joseph R. Cohen: which allows us to nest our models within intersecting identities, so we can better understand if there are health disparities how might these health disparities be manifesting across the dimensions of these categorical variables.

Joseph R. Cohen: are so not across these categories, but instead of relying on categorical variables.

Joseph R. Cohen: And then finally we're seeking to examine how pandemic related experiences might have impacted perspective tdd perpetration the pandemic did allow us to collect some data on experiences adolescents were having.

Joseph R. Cohen: loneliness anxiety related to the pandemic stress pandemic related stressors and we are curious to see what are the long term effects of those experiences.

Joseph R. Cohen: The VIP study was also really important in getting us involved, working with our local juvenile justice system and really learning about the risk assessment and risk communication.

Joseph R. Cohen: At an applied level, and it was a fascinating experience for me in a very motivating experience for me.

Joseph R. Cohen: And so we are now doing some work around we realized, we made some assumptions in the VIP study.

Joseph R. Cohen: We assumed that families would prefer to be asked about latent vulnerabilities as opposed to explicit adversities.

Joseph R. Cohen: we'd like to test that hypothesis we'd like to talk to them about what are the domains that they would feel comfortable answering about within a risk assessment process.

Joseph R. Cohen: And we also want to know better, about what are some of their community nerd narratives concerning risk and resiliency that might be important to include in our models from a risk assessment perspective.

Joseph R. Cohen: We think this is twofold one we don't think our risk models are comprehensive we think there's several variables that we're missing from a scientific perspective.

Joseph R. Cohen: That can help promote resiliency in this Community, we also think that the risk assessment process is not an engaging process for families many times.

Joseph R. Cohen: it's sometimes conducted by very engaging people but could be a very stigmatizing process, it could be a very hands off process.

Joseph R. Cohen: And we think that if we are assessing and measuring domains that are relevant to the families, we might see a better hand off to preventive services if they feel included as part of that process and so that's what the rhyme study is seeking to do.

Joseph R. Cohen: So I think yes same towards my end of my time here so so some conclusions about the VIP study and the overall kind of framing for it.

Joseph R. Cohen: Is that we do think risk assessments for TD V perpetration need to be trauma informed and I gave one behaviourally specific way, while how we think it should be framed as trauma informed.

Joseph R. Cohen: We don't think it's the only approach I would love to hear other approach, but I think all of us using greater behavioral specificity about what we mean when we say trauma informed is critical for this field we think it's important to increase accuracy and also to address health disparities.

Joseph R. Cohen: We found preliminary support exists for a feasible latent vulnerability approach being valid for TD V perpetration assessment in particularly found that hypervigilance to threat approach and reward.

Joseph R. Cohen: may be particularly useful in future research we're looking to further determine what, what are the clinical utility and statistical fairness of these indices, as well as integrating more positive mental health indicators and Community narratives into the risk assessment process.

Joseph R. Cohen: So I just want to give a few acknowledgments the research that I spoke about today, the the data that I presented was funded by a w EB dubois early career award from the National Institute of justice.

Joseph R. Cohen: This was a really impactful funding mechanism for me Tina crossland who helped organize this talk.

Joseph R. Cohen: was really instrumental and kind of had me as an early career investigator and getting that study off the ground.

Joseph R. Cohen: and being able to talk about it with, and now we are going to continue to follow those kids through continued funding through the National Institute of justice, the rhyme study, which I mentioned quickly is funded by the national institutes of health, specifically, and I see hd.

Joseph R. Cohen: And finally, we had an awesome team working on this.

Joseph R. Cohen: Jeff temple and Ryan shorey were collaborators and co-investigators on this project and my graduate students J one choice enough, the core Morgan starts have been an awesome team.

Joseph R. Cohen: To work with around this and have been instrumental in us not only collecting this data, but continuing to push ourselves to make sure we're ensuring that we're in fact translational in our research and many, many ways over the years who've been really helpful and executing the study.

Joseph R. Cohen: And so that is it for my talk, one of the reasons I was so excited to talk to everyone, because they allotted so much time.

Joseph R. Cohen: For us to talk and discuss and so now that this talk is out of the way I can have some fun and hope we've learned from all of you about some of the cool stuff you're doing in the field of teen dating violence, and of course answer any questions you might have.

Lauren Reed: Thank you so much sector going for sharing this work, this is really interesting as as someone who's been.

Lauren Reed: Doing teen dating violence work for a long time and.

Lauren Reed: Including in juvenile justice settings with risk assessment, this is really interesting sort of innovative work and I really appreciate this trauma informed approach, so thank you so much.

Lauren Reed: We have a few questions, so I will be moderating some questions for Dr Colin about this work and things that he discussed so please.

Lauren Reed: If you're thinking about some things you'd like to discuss pop it in the Q amp a box and I will be happy to to moderate that process, so we have a few questions coming in, so the first question.

Lauren Reed: is why isn't emotional abuse also addressed i'm assuming they're talking about as an outcome for this for this study.

Joseph R. Cohen: yeah you could be talking about two things, maybe psychological versions of teen dating violence they're talking about which actually is data we we collected and you know there's always.

Joseph R. Cohen: Things that you kind of trim and cut.

Joseph R. Cohen: For presentation and the original aims of the study were physical and sexual subtypes but we certainly think psychological forms of teen dating violence or an important outcome and we plan on applying this model.

Joseph R. Cohen: To psychological forms of teen dating violence, you also might be talking about emotional abuse, being one of the truck post potentially traumatic events that you experience.

Joseph R. Cohen: That predict teen dating violence and then actually was included the emotional maltreatment measure we use was comprised of both emotional abuse and emotional neglect.

Joseph R. Cohen: The original intent was to assess those separately with the measure we use, which was the seeker actually a one factor solution best represented that risk so when we use that emotional mail treatment.

Joseph R. Cohen: box that's reflecting on both emotional abuse and emotional neglect the child might have experienced previously.

Lauren Reed: Great Thank you.

Lauren Reed: And another question, in your opinion, what are some potential implications of this research on assessment tools used in diversion programs.

Joseph R. Cohen: yeah you know I would I would love to hear thoughts bye bye who asked that by harder or anybody else who has questions or thoughts on this.

Joseph R. Cohen: We think it can be used in two different ways.

Joseph R. Cohen: and

Joseph R. Cohen: yeah i'll limit myself to two different ways, one way, one way that we think it could be informed, within the context of diversion programs is.

Joseph R. Cohen: The clinic that I that I run here in my role as a clinical psychologist as a professor is we work with the diversion Program.

Joseph R. Cohen: And we're looking to get more targeted referrals and so me on the front end of receiving referrals.

Joseph R. Cohen: If I know the deficits might be hyper vigilance to threat or reward or social emotional deficits, this can actually inform the assessment battery were using.

Joseph R. Cohen: which saves time for the families, which is a huge kind of translational barrier, we have is that they do an entire risk assessment.

Joseph R. Cohen: At the diversion program then hand it off to a behavioral health team that then can get a very lengthy assessment in and of themselves so we're trying to cut down that assessment time so assessing the weight and vulnerabilities.

Joseph R. Cohen: From the Community settings themselves, we can be a big service to the family.

Joseph R. Cohen: The other way that we think it might the second way, it could be useful, is that we think these are just robust predictors of the outcomes, the diversion program is trying to attenuate.

Joseph R. Cohen: And then the third way, is that this gives us something to routinely measure or monitor from either the diversion program perspective or the behavioral health setting.

Joseph R. Cohen: To understand whether our preventive interventions are having their intended effects.

Joseph R. Cohen: So things like hyper vigilance a threat, you know things like reward in particular behavioral activation and contingency management activities.

Joseph R. Cohen: These are things that can easily be implemented in Community settings and so we can then start seeing from those Community settings if the improvement in the kids that we're seeing are in line with our theoretical model.

Lauren Reed: Thank you so much, another.

Lauren Reed: Question coming in what factors, if any, did the pandemic impact this study, so, for example, and this is a similar question that I had you were very successful in retaining the sample.

Lauren Reed: And I was reading a little bit about your work, and I know you put a lot of effort into your attention, and it was very impressive, especially during the pandemic, so the question is how did the pandemic impact the study and also how did it impact your attention, and do you have tips.

Joseph R. Cohen: My number one tip is recruit excellent ari's and graduate students now it had a it had a big impact on our study, a lot of our initial plans.

Joseph R. Cohen: For retention, a lot of the ways across both the juvenile justice setting and the schools were were dependent on those systems still being up and running.

Joseph R. Cohen: And so, when the pandemic happened from a methodological perspective, we had to be.

Joseph R. Cohen: very quick on our feet and i've never spoken to my IRB more I hope you know all nice people hope to never speak to them that much again but lots of kinds of changes in how we were going to retain these adolescence.

Joseph R. Cohen: And, and it became just a lot more work, I mean it became you know developing short videos and texting those short videos to see if they were interested in completing it, because nobody was checking their school email.

Joseph R. Cohen: We found down in Texas, and so that was a really useful tip that actually Jeff temple, who was a Co investigator came up with that was quite successful.

Joseph R. Cohen: And then, in terms of the actual study you know, we do think forms of teen dating violence decreased, as a result of it and then like I was saying, you know one thing we're still sorting through the data is.

Joseph R. Cohen: You know, we believe that kind of the nature of teen dating that definitely didn't go away during the Panda.

Joseph R. Cohen: It still happen and there's still you know we're still looking at you know over 10% of adolescents are reporting being perpetrating teen dating violence.

Joseph R. Cohen: But, but what did this look like and how might have been different highlight of some of the.

Joseph R. Cohen: person centered work in terms of the constellations of teen dating violence across the different subtypes how many of some of those constellations change during the pandemic, this might be a way for us to kind of.

Joseph R. Cohen: Understand qualitatively How did the context of teen dating violence change during the pandemic.

Joseph R. Cohen: From a quantitative perspective and then.

Joseph R. Cohen: If we have those profiles of teen dating violence change that has significant implications for our risk models right we're now predicting kind of different behaviors and in different contexts and were originally intended.

Joseph R. Cohen: But we think that's worthwhile stuff you know we know, violence against women increases greatly, and as it has all kinds of increased negative impacts during national disasters.

Joseph R. Cohen: We know that a lot of the systems that keep women safe.

Joseph R. Cohen: Are compromised in some way during national disasters So even if we're building risk models to understand the nature of teen dating violence during the pandemic we don't see that as a static thing we think that is an important public health aim in and of itself.

Lauren Reed: Thank you so much, I really appreciate that and as a as a follow up, we have a question about how was the data collected during the pandemic.

Lauren Reed: Compared to before so i'm a follow up question on the pandemic impacts yeah.

Joseph R. Cohen: Fortunately, this was a self report study which which helped a lot before the pandemic.

Joseph R. Cohen: The schools, the kids will come from the schools in Texas completed all measures via paper and pencil.

Joseph R. Cohen: At the juvenile justice system, there was a mix they had the choice of doing it online or doing it at the diversion Center itself, about two thirds did it at the diversion Center itself so paper and pencil.

Joseph R. Cohen: As of wave two except for a few kids who got in just before the pandemic at the diversion program all surveys were conducted on quality next wave two and wave three.

Joseph R. Cohen: Now from this study, because it was all groups doing it together we've done previous research to examine how does kind of doing calls over the phone, as opposed to doing assessments within the lab change the data.

Joseph R. Cohen: We haven't found huge significant differences in that, so I don't think the quality of data collection is compromised at all as a result of pivoting to online data collection.

Lauren Reed: it's great Thank you and just a question I was wondering about thinking about and I don't have the answer to this question, but a lot of my work focuses on.

Lauren Reed: Cyber dating abuse and so abuse that's happening through digital media media which, I imagine, has been greatly impacted by the pandemic and so i'm.

Lauren Reed: i'm familiar with the cadre and i'm trying to remember if it specifies.

Lauren Reed: The context, like in person or digital for the abuse so it's hard to know what the what the teens were thinking when they were responding to those items, but I wonder if asking about digital forms of abuse might elucidate what they've been up to during the pandemic that might be different.

Joseph R. Cohen: yeah big regret in the original VIP study we do not disentangle between cyber dating violence and other kinds of dating violence, we do have that for the the VIP continuation study.

Joseph R. Cohen: And we do ask for some retrospective assessment around that during the pandemic, but we didn't explicitly collect that.

Joseph R. Cohen: Information your original VIP study you know you always have a laundry list of things you kick yourself for as you're analyzing the data and that one's like right up there at the top.

Lauren Reed: Sure sure, of course.

Lauren Reed: I would be happy to help.

Lauren Reed: You be great that's really interesting and i'm also really interested in the how the.

Lauren Reed: I think you called them the latent vulnerabilities function different for girls and boys and what you were thinking about and.

Lauren Reed: And perhaps if that had to do with how you were measuring dating violence or the inclusion of of emotional abuse or not, or any other thoughts that your team was having on on those gender differences.

Joseph R. Cohen: yeah one of.

Joseph R. Cohen: that's a big part of our measurement mile you know from a developmental trauma perspective.

Joseph R. Cohen: A lot of times we would expect the latent vulnerabilities to be similar across boys and girls, but then how those vulnerabilities are expressed tend to vary based on gender and societal context right, and so a.

Joseph R. Cohen: Big kind of question and goal this study is are these latent vulnerabilities showing in variance.

Joseph R. Cohen: Across boys and girls and what we found is not only do they showing variance which suggests that the latent vulnerability is manifesting similarly across adolescent subpopulations we don't see any mean differences.

Joseph R. Cohen: In those latent vulnerabilities, which means that we can kind of understand increases and decreases similarly.

Joseph R. Cohen: I don't think the same is probably true for teen dating violence, which is the outcome we're looking at and that's why, when we think about how do we want our algorithms to work, do we want.

Joseph R. Cohen: Gender specific algorithms there's there's big disadvantages to that, but if the vulnerabilities are being expressed differently by gender and if the perpetration of violence by gender is different in many ways in which it is.

Joseph R. Cohen: Is it problematic is it necessary to have algorithms that are tailored to this in order to be accurate, you know that's something that if you know, again, you know the the.

Joseph R. Cohen: We have a lot of teen teen dating violence experts on here, that would be something I would love to hear perspectives on because that's been something we've been bantering about internally.

Lauren Reed: yeah that's really interesting Thank you.

Lauren Reed: We also have some questions coming in, people are really interested in sort of the implications and implementation of.

Lauren Reed: The tools that you are the changes to tools that you're suggesting so you've talked a little bit about.

Lauren Reed: places where this could be used, but, in your opinion, which settings are the best to start with, to start implementing these tools given you know potential training that might be needed to to use them and other factors.

Joseph R. Cohen: Well, I forgot my mute button there um this would be great if we have criminal you know people who are kind of embedded within.

Joseph R. Cohen: The juvenile justice system to hear their perspective and things like that we see the most growth for this within the diversion programs but, but some of that is because.

Joseph R. Cohen: At least in in champaign county we're fortunate to have fantastic Community partners have both the diversion program and at the probation Program.

Joseph R. Cohen: So part of the reasons we're orienting this towards the diversion program is is twofold one the case managers within our county of the diversion Program.

Joseph R. Cohen: are able to work a little bit more closely with the adolescents that are referred to them, we think one of the big benefits from the latest vulnerability model, for example, is that we might be able to.

Joseph R. Cohen: Implement one session very brief interventions with the adolescence at the actual diversion program itself and understand how they're responding from this latent vulnerability perspective.

Joseph R. Cohen: And they have we have a lot of buy in from the diversion program to do this, and so we see this right now is being something that is done, alongside the risk assessment, the latest vulnerability screening is actually quite short.

Joseph R. Cohen: The hyper vigilance to threat approach items are five items the reward learning items are seven items, so this is a total of 12 items we're talking about so it's quite feasible for that to integrate that into it.

Joseph R. Cohen: The probation department, would like to integrate it into what they're doing it becomes a little bit more challenging.

Joseph R. Cohen: For them to do, because the risk assessment protocol and what they use for the risk assessment protocol.

Joseph R. Cohen: Is a actually told to them, it is not up to their decision and things like that, at the State level, and so we we rethink across both of those levels in the juvenile justice system, which are quite different settings.

Joseph R. Cohen: there's a lot of potential for the latest vulnerability approach, but I think that there's a lot of ways, we could demonstrated success at the diversion program level in order to go through the proper channels to see if it can be done at the probation level.

Lauren Reed: really interesting, thank you, we have another question about teen dating violence, so we have someone asking what resources can teens utilize in order to educate them to look for red flags and counseling for victims themselves.

Joseph R. Cohen: yeah I want to put her on the spot, there could there could be other people who chime in on this.

Joseph R. Cohen: beyond me, I think the I think the folks that and I J, this is a big push by them is to try to kind of create public information.

Joseph R. Cohen: That can be disseminated and even lauren if you have thoughts on this, too, I don't mean to put you on the spot, as the moderator, so there is a lot of.

Joseph R. Cohen: Material online that has been published for public consumption, both for providers and for adolescents themselves.

Joseph R. Cohen: As part of kind of teen dating violence awareness month so they've they've kind of tailored the material to be developmentally appropriate and give them really.

Joseph R. Cohen: I mean from a behavioral perspective, I was very sympathetic to kind of the tips they gave for how to talk about teen dating violence with adolescence, and so, those are.

Joseph R. Cohen: Widespread materials across the Internet, you can certainly follow up with me if you ever have trouble tracking them down they're not materials i've curated, but there are materials i've consumed and i've been quite impressed by it, but I think.

Joseph R. Cohen: You know I think one thing if, if I were to add on to that question is that teen dating violence and a lot of times in our public health and risk communication is siloed.

Joseph R. Cohen: If we even look at kind of the frontline preventive interventions for teen dating violence fourth are safe dates things like that it's it's in a lot of ways, separate from a lot of other.

Joseph R. Cohen: kind of behavioral outcomes that we focus on, and as a result, I think that leads to a lack of awareness about both its prevalence and its impact we don't see teen dating violence, including a lot of.

Joseph R. Cohen: trauma screening or aces screening or things like that, and so part of the goal of the latest vulnerability approaches also that it will bring teen dating violence into the fold.

Joseph R. Cohen: For some of our developmental trauma models, as the important kind of behavioral outcome, it is for adolescent phobia.

Lauren Reed: Great Thank you so much, and i'm also posting.

Lauren Reed: Some resources in the chat here so for folks that are interested in seeking out some specific resources from the Office for victims of crime, so this is like a compilation of different resources that you might check out hotlines information.

Lauren Reed: And so I would encourage folks to check that out as well and.

Lauren Reed: As Dr Cohen mentioned there's a lot being pumped out right now, especially on social media that gives you a lot of tips and tricks for identifying red flags, raising awareness.

Lauren Reed: Getting teams access to confidential resources for a teen dating abuse so twitter's a great place to start, for example, and these resources that we posted also give some good tips as well.

Okay.

Lauren Reed: I right i'm looking through other questions I see here Okay, so another question that we have is in the following steps in the study.

Lauren Reed: Looking at different an intersection intersecting identities, are there any thoughts about assessing gender identity, more broadly, that would potentially need to broaden the forms of dating violence as well.

Joseph R. Cohen: yeah that's an excellent question so.

Joseph R. Cohen: teen dating violence among gender minority adolescence is is a big concern, there is, there is a lot of good evidence to suggest that this is.

Joseph R. Cohen: Both a population that has that unique risk to experience teen dating violence, as well as a population that might not benefit.

Joseph R. Cohen: As much from universal protocols related to teen dating violence prevention so actually a consultant on the project, who might be on this call.

Joseph R. Cohen: Ryan shorey, this is a huge passion for his and so within the VIP continued study we've included some measures.

Joseph R. Cohen: that allow us to assess gender and sexual identity across different dimensions in order to understand better.

Joseph R. Cohen: Where our model might be a strength and where there might be some accommodations with that model based on how that one facet of identity may intersect with other facets of identity.

Lauren Reed: Thank you so much, and.

Lauren Reed: That was something I was wondering as well, when you were talking about the follow ups that you plan to do, and you talked about applying an intersection ality lens to this work, and looking at multiple identities at once and analytic methods for doing so.

Lauren Reed: I don't know if you know the details of that at this exact moment but i'm super interested in hearing.

Lauren Reed: A little bit more about that or what you have planned yeah.

Joseph R. Cohen: I could, I can talk, I can talk broadly about and then definitely go into more details about it it's the approach is called a might approach multi dimensional assessment of.

Joseph R. Cohen: heterogeneity and data analysis it's easier just have the acronym but what the the premise of a Meta approach is is it within a multi level framework.

Joseph R. Cohen: And the idea is rather than using a fixed effects approach in which you're going to be limited in all kinds of ways.

Joseph R. Cohen: For example, kind of testing higher ordered interactions and 345 ways which people who want to conduct culturally competent.

Joseph R. Cohen: Research are often hampered by with the model approach allows you to do is create a variable that represents each identity that's represented within your data set.

Joseph R. Cohen: And then you actually nest those models at that level of data, in which, within each level each stratum represents a specific example.

Joseph R. Cohen: And then, if that random effect from that multi level model is significant, you know you can infer.

Joseph R. Cohen: That your risk model varies across identities.

Joseph R. Cohen: And then, once you have that influence within a mix level model there's all kinds of cool things you could do you can create residuals you can create predicted probabilities and you could start comparing estimates from specific stratum in that model.

Joseph R. Cohen: compared to other identities, to understand risk and resiliency and.

Joseph R. Cohen: it's got some faults it's got some cons, but we think it's a really good step forward in building culturally competent models in line with underlying intersection ality theory, which is something that's really difficult to do within a quantitative framework.

Lauren Reed: Thank you so much.

Lauren Reed: it's really fascinating so another thing that I was hoping we'd have time to discuss and it looks like we do is a lot of folks on this call might be familiar with the aces.

Lauren Reed: And might be something that folks are using and so I would love to hear more of a discussion about sort of the strengths and weaknesses of the aces.

Lauren Reed: You know that led you to do this additional research, but especially in the context of teen dating violence So what are the sort of strengths and weaknesses of bases that you wanted to address.

Joseph R. Cohen: yeah and this could certainly.

Joseph R. Cohen: Did floor and you might not notice I don't know if the audience has a chance to talk during this but, but if they do this could definitely be one other people can.

Joseph R. Cohen: Either unmute and talk i'd love to hear it or put it responses in the question and answer box.

Joseph R. Cohen: So the AC study for those of you are not familiar.

Joseph R. Cohen: was done from a public health perspective, like I said during the talk it includes these five male treatment items and these five items.

Joseph R. Cohen: that are sometimes I hate even using these words that are sometimes called the family chaos items i'm going to use some more words I hate using so These include like impaired caregivers poverty related stressors parental criminality.

Joseph R. Cohen: What are some limitation of the aces we think there are numerous one it was initially developed in an adult population.

Joseph R. Cohen: And then just kind of translated down to children adolescence, and so it might not be capturing the adversities that are actually most salient for children, adolescents right, it was created from just adults kind of retrospectively.

Joseph R. Cohen: reflecting on what were some important events in their life.

Joseph R. Cohen: We think that there are many adversities.

Joseph R. Cohen: That the aces doesn't.

Joseph R. Cohen: there's aces that are included within the inventory, such as like parental divorce and loss that.

Joseph R. Cohen: Actually might not be an adversity for millions of kids and then there's other adversities like teen dating violence that just aren't included in it.

Joseph R. Cohen: It also doesn't provide any assessment, besides the subtype of trauma.

Joseph R. Cohen: So it is you know their strength to the aces and one of the biggest strengths of it is a broader incredible amount of awareness to this field so so the pro certainly outweigh the cons in my mind.

Joseph R. Cohen: But one problem is that our audience just a yes, no questions about some types of trauma when there's a lot of aspects of trauma that are much more predictive for something like teen dating violence.

Joseph R. Cohen: Compared to kind of how many traumas to be experienced the age of onset the creepiness city.

Joseph R. Cohen: who perpetrated the violence, all of these things might be either as important, or, in some cases, more important for predicting behavioral health and adolescence.

Joseph R. Cohen: And then, again, finally, like the talk, even if we ironed out those issues, I think we really have to question translational we.

Joseph R. Cohen: Do we want these questions being asked, do we want these answers being entered as part of the Court record, do we want these answers being entered into electronic mental health records.

Joseph R. Cohen: I think there's a lot of stop and pause around that that might not be happening as we kind of push forward in that direction.

Lauren Reed: Thank you so much really appreciate it.

Lauren Reed: And if anyone wanted to jump in and add anything to the chat about this aces discussion or any other discussions we've been having please feel free don't believe people can jump in on audio, but they do have the chat available and the Q amp a box, if anyone wanted to add.

Lauren Reed: Great Thank you so much Tina and the chat.

Lauren Reed: Thank you all for coming, this has been wonderful Thank you so much, any other less lot.

Lauren Reed: Well, I just want to thank you so much, Dr Cohen, for sharing this presentation in this really innovative work, this is exciting, you have a great team, and this data is really rich and exciting, so I look forward to learning more about the follow ups.

Lauren Reed: Thank you so much.

Joseph R. Cohen: Thank you very much lauren have a good day everybody.

Joseph R. Cohen: Thank you.