Research, Practice, and Data Informed Investigations of Child and Youth Suicide – Day Two

Transcript

Agenda Item: Welcome and Orientation to Day 2

LISA COLPE: Good morning and welcome to day two. We made it. We had a really great day yesterday, we covered death investigation processes, what the research in child and youth suicide is telling us. We heard about the data systems that collect information about child and youth suicides. And today we have lined up for you some really great panels on special risk populations, as well as some talks from people who are representing states or large counties, and how they pull all the data available to them at their state level into their suicide prevention program. So really looking forward to hearing these talks, and having another great day. So I am going to turn it over to Stacia Friedman-Hill, who is going to introduce and then be the discussant for our first session. Stacia?

Agenda Item: Special Populations

STACIA FRIEDMAN-HILL: Thank you. Good morning, everybody. I am a program officer in the Division of Translational Research at NIMH, and I am also the colead of the NIMH Mental Health Disparities Research Team. This session focuses on youth from populations that have elevated or disproportionate suicide rates or increases in suicide rate. 

And it is important to recognize that their identity is not the risk factor. These populations have the same individual level risk factors as the general population of youth, but they have higher level of exposures to interpersonal, community, and societal factors that elevate risk through aggregated social and life stressors. 

And in particular these youth often experience disparities in screening for suicidality and access to services, as well as many other social, institutional, and structural inequities. 

Dr. John Blosnich and Dr. Michael Staley will discuss LGBTQ youth. Dr. Arial Sheftall will talk about youth from minoritized racial and ethnic populations, and in particular her research on black youth suicide.

Dr. Joette James will present an overview of youth with autism spectrum disorders. We know that systems involved youth also experience high rates of suicidal thoughts and behaviors. Dr. Cynthia Fontanella will focus on youth in the child welfare system, and Dr. Gail Wasserman will talk about youth involved with the juvenile justice system. I am looking forward to listening to these important presentations and moderating our panel discussion.

One reminder for the speakers, you will each introduce the speaker who follows you. Now, it is my pleasure to hand the podium over to Dr. Blosnich for the first presentation. Thank you.

JOHN BLOSNICH: Good morning everybody. Early morning to folks who are also on the west coast. I am going to be covering today kind of the chasm that we have for LGBT research. So you will all pass the final at the end, because the quick answer is we don’t have a lot of information on LGBT suicide specifically. We have a lot of information about ideation and attempt, like reams of it, stacks of it. What we don’t have is the actual phenomena of suicide mortality and suicide death.

So I have disclosures to make. I am funded by NIMH. And I will start quickly by saying the importance of mortality data is that it mobilizes resources in ways that are critically important in terms of awareness of issues, policy, funding, and just intellectual capital around solving a problem that we know exists, because we actually have data to show , account that a problem does exist. 

And there is evidence of this happening for all sorts of populations in the United States for whom we have mortality data. This includes groups from minoritized racial and ethnic identities. This includes veterans as well. 

That we get this information is going to sound a little like a tautology. You need the numbers to know how many people that you have, right? But this is actually a very pernicious problem for LGBT populations because we actually don’t have vital statistics for LGBT people in this country. I mean, let me take a spin on that. We do actually, but they are hidden in the pile of information that we have. 

We cannot find them in the three million or so people who die every year in this country, because we do not collect in a standardized way sexual orientation or gender identity at the time of death. It is not in the death certificate, and it is not part of in most places standardized investigation procedures for violent and investigated deaths.

That is not to say that we don’t know anything about mortality in LGBT population. What I’m going to do this morning is take you quickly through four types of comparative studies that helped us to put together sort of signals to see if we could find information about mortality in LGBT populations. 

I will say a lot of these studies don’t have youth in them. Maybe the last part of them do. But just to give you a sense of the data gymnastics that we have to go through in terms of LGBT health equity researchers of trying to answer these questions of we have a lot of information about social adversity, health disparities for LGBT populations. 

We have almost nothing when it comes to the actual kind of ultimate outcomes for thinking these adversities do not really bode well for thriving, but we don’t have information on the toll it may take on longevity and wellness in people. 

So those four, I have walked through them now really quickly. So we have findings from survey data that have been paired with mortality data. I’m going to go through a couple examples here from the general social survey. MSM stands for men who have sex with men, WSW is women who have sex with women. It’s a behavioral measure of sexual orientation. 

I guess I should have said, prior to all of this, sexual orientation is a related construct, but is distinct from, gender identity. The problem is, it’s not really a problem, but we often say LGBT. And the abbreviation can kind of mash these things together in people’s heads. But sexual orientation is separate from gender identity. A person that identifies as gay does not mean that they are transgender. And someone who identifies as transgender does not mean that they are gay or lesbian or bisexual. 

So keeping an eye in mind there are multiple ways that we sort of have to operationalize when we think about defining sexual orientation and gender identity. So MSM, WSW, that is one measure, one way of thinking about how we would define a person’s sexual orientation, albeit rooted in behavioral measure. 

So you can see here from the general social survey data there was no detection of a greater risk for suicide death for MSM versus non MSM, but a pretty big effect found for women who have sex with women compared to women who do not have sex with women. The women’s health initiative however did not detect a difference for WSW people. 

And then the National Health and Nutrition Examination Survey found that LGB identified people, this is an identity measure, they had twice the risk of all cause mortality compared to their heterosexual peers, but they couldn’t calculate suicide mortality because it was too rare of an event in this dataset.

So we switch from survey data paired with mortality information to registry data. So national registry data, this isn’t something that we have in the US, these are studies from Denmark, Sweden, Nordic countries have really great registries. So you can see here this table summarizes some findings from studies using registry data. 

So there seems to be, at least in these countries, some signal. But there are mixed findings from the actual magnitude of differences between the groups. It is unclear if that may speak to some sort of cohort difference that they’re detecting in the data over time, or if it is sort of a distinct effect.

So this is some work that I did in my previous life with the US Department of Veterans Affairs, looking at medical record data. So we had to find it probably, well not probably, definitely limited ways of defining transgender veterans and medical record data and LGB veterans and medical record data. 

So for transgender veterans, to try to find them we had to rely on ICD coding, so International Classification of Disease coding that is related to gender identity disorder. Which is not a name that we use anymore, the DSM replaced that with gender dysphoria. But ICD still includes GID as a diagnosis code. And VA data rely on ICD coding and not DSM coding.

So we have pulled a sample of folks who had this code, versus pulled a random sample of folks who did not have this code. And so that is how we sort of defined the groups here. And you can see when we looked at causes of death in this group, a very simple metric in public health. 

I kind of joked to my students, I’m like public health folks are just really good at counting at the end of the day. We just have to make sure that we’re counting the right people, that’s the key. The grey line that runs across the diagonal is the top ten causes of death in the general US population. The people with the squares are the transgender veterans, and the people with the circles are who we presume to be cisgender veterans. 

You can see for transgender veterans’ suicide was the fourth leading cause of death for them versus it was the tenth leading cause of death for the general US population and the tenth leading cause of death for the cisgender veteran sample that we had here.

And we detected a similar sort of finding when we looked at LGB veterans. Now this, we did not really have a lot of structured data to go on from this. What I mean by structured data is sort of the ICD codes or any sort of, we didn’t have at the time demographic codes, if you think about people check boxes on forms. 

So this used natural language processing and machine learning to look at clinical progress notes, or basically sort of written text notes and medical records to find signals for veterans who would have what we presume to be an LGB identity. 

So again, thinking about the gymnastics I said earlier about having to go find people in data. We could basically be like the Simone Biles of analysis here because of all the things that we have to do to find people. We can see here that suicide was the fifth leading cause of death for LGB veterans. So we think that we are seeing these signals, but I will get to this in a little bit about the (inaudible) of these methods.

And then last we land at NVDRS. So, NVDRS to their credit introduced two fields for sexual orientation and transgender status. But you can see here this illustration of an earlier study using NVDRS data, there is a massive problem with missingness in these codes. 

Upwards of between 70-80 percent of the data are missing in sexual orientation and transgender data fields here. So I don’t know about the rest of you, but it is very sobering to try to make conclusions about a population when you only have 20 percent of the data about them.

So, to run through what would be a litany of limitations for all of these studies, I have them summarized here, but for the survey registry data we have to wait a long time to get this information. So we talked about the problem with delays in information, along mortality information. 

Imagine having to wait, the NHANES data was based on information collected in the late ‘80s and through the ‘90s, and that paper was published in 2011. So imagine if you were a population waiting to figure out what was the leading cause of death, or where should you focus your resources. You had to wait decades to get that information. So significant amounts of time have to pass for survey registry data, the multiple definitions of SOGI. 

The samples are really small. When you’re talking about suicide, it is a rare event outcome, minoritized sexual orientation and gender identity, it’s a small population too, even though we think it might be disproportionately affected by this outcome. You need a lot of people to get this. The issue of the identity captured at the time might not be the identity at the time of issue. Registry, single people are left out, sorry about it. And of course, missing data, EHR data is plagues NVDRS data.

So, lots and lots of problems. So I am going to talk really briefly about one potential sort of solution that we have been trying, and that is around this post-mortem SOGI project. So this is really the brainchild of Ann Hass and Andrew Lane, and they developed a meeting in 2014 to bring together people from CDC, from the VA, coroner, medical examiner folks, around the table to figure out how can we address this problem that we don’t have mortality data for LGBT populations. 

So the group decided that they thought is it law enforcement, is it funeral directors, and they sort of landed on medical legal death investigation. And thinking about those are the folks who turn up at an investigated death. And at that time they’re asking questions of folks, they’re documenting evidence at the scene, and that this may be a critical window in a critical population to get this sort of information to so that they could be gathering this.

So, unsurprisingly, whenever we surveyed a convenience sample of death investigators, most of them were not asking this question, although most of them said they had cases in which they suspected that the decedent was LGBT, 60 percent said they didn’t ask sexual orientation, 75 percent didn’t ask about gender identity, and none of them reported ever being trained on how to collect specifically SOGI data.

So, our sort of envisioning this, improving the pipeline of data, we spend a lot of time, a lot of iterations on developing this protocol together, alongside advocates and folks who thought deeply about how do we measure and operationalize sexual orientation and gender identity, along with death investigators who are like “this is what our life is like in the field, this is what we go through. 

I’ve shadowed death investigators on a number of scenes, and it is hands down I think probably one of the hardest jobs on earth. So we had to take all that into account when we were developing sort of this guide, this protocol. 

So thinking that we could do this training, they would implement it in the field, that this would make it, that SOGI data would make it, sexual orientation and gender identity data would make it into the reports, and then that would give the abstractors a very clear, the MDI abstractors a very clear signifier that they could code them and help to address this missingness that was occurring in NVDRS.

And so how do we do this training? We have, it is kind of a three part. We do some didactics, quickly sort of going over these constructs, and then doing case discussions, some roleplay exercises. We offer a guide and a checklist for folks. They don’t have to do this. It’s just a way of helping them to organize the evidence that they might gather at the scene. 

But this is where epi folks’ hair might curl a little bit. The way that death investigators get their information, they’re not surveyors. They don’t go in with a clipboard and just run through something with folks. They have to form a rapport with people, they have to quickly, on the spot, prioritize information they have to get, and do a very, sometimes quick, it could be a longer conversation, but it is a conversation. 

And the way that they embed information in this conversation and find the quick darts into how I can get this information is an organic process that doesn’t lend itself to how we think at epi about here, take this 65-item survey with you and just administer to somebody who has just lost somebody, it’s not possible. 

So there is this give and this compromise where epi people want standardization, because we think it makes the most valid instrument possible. I just can’t impress upon you enough that being in that space of working with an investigator when they’re talking with somebody who is having the worst day of their life and trying to get this information, those data collection methods are just flat out not possible, and we have to be able to be flexible about that.

So where we are currently, we’ve run a pilot in LA County and the State of Utah. We are looking at documentation across time. We are learning from the investigators what went well in the field, what doesn’t go well in the field, what would you tell a new investigator who has just joined your force about how to gather this information and how you handle it, and doing some expert review of the documentation. California is embarking on a six-county pilot of this. So we are working hard to get some of these pivots and changes and edits from the current evaluation ready for the next pilot. 

So just some parting thoughts on this. This other issue is that everyone has a sexual orientation, and everyone has a gender identity. This measure, this protocol is not meant solely for LGBT people. 

The other wonderful thing I tell my students about public health is, beyond being really good at counting, we really default to numerators and denominators. So it is really kind of a way to bait people into public health, thinking it is super simple. Then you hit them with things like age standardization  but you tell them it’s like you just count things and you make like, fractions, they’re all about it. 

But when you don’t do this for everybody we have a major problem where you don’t have the right denominator. You cannot assume that everyone who is undetermined for this is heterosexual or cisgender. Make sure that people know. 

And I’ve had investigators tell me very proudly, they’ve documented this for a case, they were like, this person is LGBT. I’m like, that’s great. Can you tell me about some heterosexual and cisgender cases that you had, and it’s sort of like,”oh”. So impressing that this is, everyone has got these characteristics and it needs to be documented for every scene every time.

It can be challenging to get this information. I’m not going to sugarcoat it, I’m not going to be Pollyanna about it. But a lot of information that investigators have to get is difficult and challenging information to get. So it should not be given a heckler’s veto of we should not do this because it’s hard.

Leadership buy-in is critical for this. As the system is being prepared for this, if you have folks onboard to do this they’ve got to have a place to put the data. If they feel like they’re gathering this with nowhere to put it, they’re not going to do it anymore. 

So that leadership buy-in and making sure that the data systems can capture this in a way that puts it into the pipeline is also critically important when you get people onboard for this, otherwise they feel like they’re going on a fool’s errand.

And then my parting thought is that without these data we cannot test and evaluate prevention and intervention at scale for these populations. I could tell you tomorrow, hey, I have an intervention that will prevent LGBT people from dying by suicide, right? And if you buy my argument, and I’ll have all the ideation and attempt literature next to me to sell you my idea, you might be like, oh my gosh, that is great. 

And then you would be like, tell me how you are going to evaluate that, what’s your outcome? I would be like, I don’t really have one. I can’t measure, I can’t tell you how I’ll drop the rate of suicide for LGBT people, because I don’t have a rate. Would you still cut me a check for my project? Anybody? Any generous folks in the room, please? You wouldn’t, and that undercuts so much of what we do in LGBT health without having this ultimate outcome of mortality. 

Suicide is where we are focusing a lot of our efforts, because we think that there is a lot of signal saying that there is a huge need here. But we also know that LGBT populations smoke like houses on fire. We have no idea if lung cancer deaths are disproportionate in this group. 

So, ways that mortality information again are so crucial to these efforts in health equity just underscore why we think that this is one way, one strategy for how to improve this pipeline of data, at least for suicide mortality. And with that, I will stop and I will turn it over to Dr. Staley.

MICHAEL STALEY: Thanks, Dr. Blosnich. It is a pleasure to be here this morning and to talk about our experience in the State of Utah collecting SOGI information. 

Two things came to my mind already in listening to Dr. Blosnich talk that I just want to emphasize. The first is that being a minority of any kind does not cause suicidality. These are external forces, I’m a sociologist, so of course I would say these are external forces, harassment, discrimination, a whole plethora of things. So I want to make that clear, because that has been a confusion sometimes talking to families and other folks.

I’ll just get started here. We wanted to get this information, and a lot of other information. So, why Utah of all places? Utah is a centralized medical examiner state. We are one of 16, and Washington DC. And we investigate all violent, sudden, unexpected, and unattended deaths. 

And in one single investigation system. We’ve got nine pathologists in one office in the suburbs of Salt Lake City. And so you would think this system is really ideal for implementing systemic change. I would point out though that there are some coroner systems that are also - Cook County, LA County and others that can implement this kind of change on a large scale and make a big impact as well.

So that’s the good part. This is the little bit messy part. So within our statewide death investigation system there are nine fulltime death investigators. They work in our office in Salt Lake City, and they cover basically from 6:00 AM until 6:00 PM every day, and then after that we switch over to part-time death investigators. There are 15 of them, and they cover both Salt Lake City and then the surrounding suburban counties, which is an hour north and an hour south. 

And then after that, Utah is a very big state, and there are 89 vendor death investigators. And these vendors for a small amount of money are called in the middle of the night to respond, they never of course know when this is going to happen. 

And some of these vendor death investigators will go to one or two scenes a year, others will go to maybe 30-40 a year. And so they get varying degrees of practice. And they’re remembering all of those things in the ‘every scene,every time’ book that they were trained with, not just sexual orientation and gender identity.  So it gets messy when we start talking about training a death investigation workforce.

So there are three ways, in my mind, that we can collect this information. And we have been talking a lot about collecting this information on the scene of the death. And that’s one way, for the death investigator to ask the family. And I am going to run through each one of these pros and cons. The next is funeral directors or other death certificate certifiers, and then the last is next of kin interviews.

So, a pro of getting this information at the scene of the death is that it is timely. We are getting that information the day of the death. The con is we oftentimes respond hours after death has been pronounced. There is no family, they have already left. Or this is a body that we don’t even know who this person is yet, and so of course we don’t have anybody to ask this information of. 

But one pro here is that the investigator may have an opportunity to speak to the family and getting that one shot at the data. So if we delay this later, the family may not answer our phone calls. Sometimes they get really angry at the medical examiner for removing the body, or for whatever other reason that they may have. So sometimes this is the only shot we have. And I still think that we should try to do it at this level when we can. But oftentimes we just can’t. 

Sometimes the family is just too distraught, and these interviews can get quite lengthy, because we also want to know about their housing status. We want to know about their employment and education, and stressors in their life and their financial situation. And we could keep going on and on and on. 

And this takes a lot of time to get. And sometimes our death investigator, there is one on duty every night, and they’ve got to go to the next scene, and they don’t have time to sit there for three hours and get this kind of information.

Sometimes our death scene investigator doesn’t have the right kind of training to ask a lot of sensitive questions. We have really worked hard on this. I think we are making a lot of progress, but you can imagine in a workforce of 89 vendor investigators and a bunch of part-timers there is a bunch of turnover there. 

And so making sure that they’re all current on that training is really important. We’ve said you should ask about sexual orientation and gender identity, with kind of no follow-up training. The results of that were not great. 

This was something that we really do want to train people on how to use the right language. You don’t have to be perfect, these are evolving, all these labels and terminology it changes, and you don’t have to get it perfect every time, but if we are at least in the ballpark it definitely helps us out.

And oftentimes there is no formal place for death scene investigators to record SOGI information. In our current case management system there is no field for this. We are completely rebuilding it right now, and the new one will have this information, but we rely on the death investigator to put this in their narrative.  And they have kind of a standard template, and it is one of those demographic lines, but it is a copy and paste from Word, so things happen.

The other place that we could get this information is where we get parts A and C of that death certificate, from the funeral home, which is usually from death certificate informant. These are usually family members, somebody really close. And also death certificates are universal, they have to be filled out. So that is a pro.

But we know from other kinds of fields that these aren’t always the best way to gather data. I am pointing specifically to pregnancy status, as well as tobacco as a contributor to death. These get marked unknown all the time, or especially in the area of pregnancy they’re just incorrect. 

If you talk to any of our maternal health mortality folks they’ll tell you just ignore that, or if they’re female we’re going to look through their medical records to make sure that we are including them in our surveillance and fatality review processes.

The other thing is that funeral homes don’t have any formal training in collecting this kind of information either. And if we think about Utah, there are I believe 81 police jurisdictions, and there are probably at least two or three funeral homes for every one of those, and so suddenly the multipliers get really big. And again, that training piece becomes really complicated.

So the third method is next of kin interviews. And by these next of kin interviews, I mean actually calling the family at some point in the future after the death. And we have sort of determined that the best time to do this is the four-to-eight-month mark. This is after we’ve got the bank accounts kind of situated, the shock of it all is starting to wear off just a little bit. Some of that cognitive putting things together is starting to happen.

And this isn’t perfect. We have a Survivor’s Council, we use this group of people to bounce our ideas off of, and they’re like you couldn’t have called (inaudible) full stop. Others are like please, I wish you could have called me two days after this had happened. But on average, on the whole, this four-to-eight-month mark is our sweet spot.

So we try to call families in this four-to-eight-month mark and get all kinds of information. Some of the cons here are the delays. We are not getting this information in real-time, it is a little bit of a delay. 

Another one is that we are relying on one next of kin at one point in time to relay this information. And it is the person who is often listed on the death certificate as the informant, or someone at the scene who is a next of kin as well (inaudible) knowledgeable person. And we learned that in our youth study which I’m going to talk about later today.

Sometimes these next of kin are also really invested in preserving a legacy. They want their person that passed away to be known in a certain way, and so they’re not telling us, I don’t want to use the L word, they’re not lying to us, but preserving a legacy, this is the person that they want to live on in their memory and in ours too. 

But there are a lot of pros here, and that is obviously why we do this. So this is somebody who knew this person really well, and they’re poised to disclose this kind of personal information. Speaking specifically about sexual orientation and gender identity though, a child, a teen may not have disclosed this information to their parent. And so I think that’s kind of where some of the gaps are still in this method. 

In the family’s quest to understand why someone close may have died, they may unearth some SOGI information postmortem though too. We have had that happen frequently, where they’re like but after they died, I looked through their phone and realized they were struggling with same sex identity or being attracted to somebody of the same sex, or their gender identity, things like that.

And another pro here is that we’re asking these personal questions that might be hard to explain on the scene of a death of somebody in immediate shock, and we have the time to sit down and explain why this information is important.

The other pro, and I take a great deal of pride and honor in this part of these phone calls, these interviews, is that at this four month mark we kind of know where this grief ship is heading. And we can point that ship, we can’t fix it of course, nor is that our goal, but we can aim the boat in one or two or three or ten degrees in a slightly better direction. We’ve got the right resources, we’ve got the list of therapists who are trained in this kind of work, and we can connect these families to these resources. And again, that timing is really important.

So in practice, for now, we found that the best way for us to get this information is in these follow-up interviews. And we are only doing this for people who died by suicide or by unintentional drug overdose, or by undetermined drug overdose. We also do this for maternal mortality, that is a relatively small number overall. I would love to be able to expand this into other groups if we get the resources to do that, but of course that is a limitation.

And specifically, because Dr. Blosnich did mention these multiple aspects of sexuality, this three-legged stool of attraction, behavior, and identity, we’re asking about identity. This observable, something might have pulled something not observable, but some identity label that somebody may have attached to themselves. That’s what next of kin are probably most likely to know. So some death scene investigators routinely ask about SOGI information in Utah. I can rattle off their names in my head, I won’t. And they went to Dr. Blosnich’s training, and that worked out well. 

We are really getting this buy-in, making sure that these fields are in our case management system. These are all I think a really key component to where we’re going, and I can only see this improving as time goes on.

And generally, with funeral homes we have tried to approach them about this and other issues. It is sort of like sorry, we are already getting education for you, we cannot be bothered to get anything else. So these are some of the reactions that we’ve got. 

And I want to say that the vast majority of people either at the scene of the death or the phone when we’re calling answer this question in the same way that they answer every other question that we ask: Did they serve in the United States military or other uniformed service, were they a first responder, did they identify as gay, lesbian, or bisexual, did they identify as any gender other than the one that they were assigned at birth. 

And people don’t have any problems answering this question, and they answer it and we move on, and it’s not exceptional. Wait a minute, why are you asking that? We can explain that. Or this one, do you know something I don’t, when I ask this question? No, I ask everybody this question.

People live in political environments, people live in media environments. They ask, I suppose you’re going to, trans is not a verb, but in this sentence it is: I suppose you’re going to trans someone close to me. No. 

Or, this is a strange twist which I really didn’t anticipate. I thought, name, my son, was straight. But after they died I heard LGBTQ people are at a higher risk of suicide, so maybe he was gay, I don’t know. And it’s this sort of narrative that LGBTQ people die by suicide more than their peers, and survivors are constantly asking themselves that why question. 

So of course, I don’t fault this parent for having that response to this question. All of the evidence that she presented would suggest to me that her son was straight, but this was something that she was using to perhaps explain why he may have died by suicide. 

So even when they do have these concerns, they’re easily addressed, and we’re not having these explosive responses of how dare you ask me that question. This just goes about as smoothly as every other question that we ask. I just want to talk about how where we’re at with this process. 

So again, we interview every, or attempt to interview every family after a suicide and after an unintentional drug overdose. We started doing this in October 2018 for suicides, and in Fall 2020 for unintentional overdose. And so we’ve interviewed a little over 1400 next of kin, and we have SOGI information for this just under 1400 decedents. In the year 2021 that is our most complete data year. 

We attempted to reach 630 families following a suicide, and we completed 65 percent of those interviews. We were never able to find 22 percent, and just 11 percent refused. And there were two percent that we were still ‘call me back in a year’ kind of thing that we were still working through.

Drug overdose, you had a little bit higher refusal rate, a little bit lower completion rate, a little bit harder to find in some cases. But really numbers that I’m very pleased with. So. I don’t have the actual SOGI data, that’s not quite ready yet, but hopefully very soon. So thanks.

BRENDAN WEINTRAUB: Next we have Arielle Sheftall, we will pass it on over to you.

ARIELLE SHEFTALL: Thank you. Today I am going to be talking about suicidal behavior in minority youth. Just so everyone is aware, I do receive funding from the NIMH as well as AFSP, and I have no other conflicts to disclose. I am going to go through some epidemiology as I did yesterday for just youth in general, but just also focus primarily on black youth and talking about some prevention opportunities for black youth as well. 

And as stated yesterday I think this is one of the most important slides that I show in all of my presentations. I do believe that every single person that is watching today that is in the room has an opportunity to prevent suicide and suicidal behavior. 

So please, please know that these numbers are available 24 hours a day, seven days a week, and 365 days of the year. There is always someone there to speak with someone, whether it be yourself or a friend or family member that is in need. And again, these are textable as well, so you don’t have to call, you can always just text.

And another reminder, the topic of youth suicide is very, very heavy, and we really need to take that time to take care of ourselves. It is very important to do so. We can only do so much if we’re running on empty. So please if you need to take a break or what have you, please, please know I will not be offended and that is okay, self-care is important.

So, looking at the data, this is suicide by race for five- to 19-year-olds from 2000 to 2021, which is the most recent data that we have right now. As you can see, when you look at the bars here, green is for indigenous Native American and Alaska Native youth. And for quite some time, now we have seen them to have the highest rate of youth suicide. But when you look over time in terms of the increases, we actually see a different story. 

So for white youth we’ve actually seen an increase of 44 percent over this period of time. For Asian and Pacific Islander youth we’ve actually seen a 52 percent increase. For indigenous youth they’ve seen the second largest increase at 62 percent. And then for black youth they’ve actually seen the largest increase, at 100 percent.

When you look at indigenous youth over time, this is data plotted from 2010 to 2020, and what you can see is I’ve broken it down by male and female, so our blue bars are males, and our grey bars are our females, and I have some trendlines present here. 

And what you can see is that for our females actually we don’t have rates present for 2012 and 2013, but we do have rates for all other years. And this is just because the rate was not a reliable rate, the number was quite low for those specific years. But an increase over this time period for both males and females in indigenous youth we have actually seen a 15 percent increase. For females we’ve seen an increase of seven percent, and then for males we have actually seen an increase of 19 percent.

So my brain got working last night when we were talking about multiracial youth, and this morning, at 5:56 AM I was making slides. I know. So I apologize, I will get these to the folks so that you guys will have the most updated slide deck. 

But I wanted to talk about multiracial youth, because that is a group that we do not see that much in the literature. Nationwide, I should say, there are studies that are being done with multiracial youth. But for most of the data that we see it is specific to a specific sample or a specific state. 

So for the nation, when you look at CDC data, Wonder, NVDRS, WISQRS, you don’t actually see a multiracial category until you look at CDC Wonder, and it starts in 2018 actually. So from 2018 to 2021 we’ve actually seen an increase of seven percent in the suicide rate for multiracial youth specifically.

And I went again, a little nontraditional, but I went to the YRBS this morning and downloaded all this data because I wanted to see what they had. And the YRBS actually has been collecting data with high school youth and middle school youth for many years now. 

And when you look at multiracial youth there is a subcategory that is present for the YRBS. And this looks at suicidal thoughts as well as suicide attempts. And what you can see for the middle school youth is not all 50 states actually participate. 

For high school youth the majority of the data is with all 50 states, which is great. But for middle school youth, which is 10 to 13 years of age, we don’t have as much data present for multiracial youth, or just youth in general. But these are the states that actually have data present for 2009. 

And what you can see is that I’ve also plotted it out for 2019. And we actually did increase in terms of the number of states that have middle school multiracial youth data when you look at considering suicide. But again, we are not collecting as much data as what I think we need to do. And PA had the highest rate in terms of seriously considering suicide for multiracial youth, at 33-34 percent of the youth who were surveyed.

And then, when you look at suicide attempts, this data unfortunately is not that great, so I apologize in advance. But we did get better. So in 2019, which is the orange bars, we have more states that are actually reporting the attempted suicide for multiracial youth in middle school specifically. But for 2009 as you can see we only have a couple of states, and unfortunately there are not a lot of states that have overlap. 

New Mexico actually has been pretty consistent with their middle school data, and this does make a significant difference. So for New Mexico there was a significant difference for attempted suicide, self-reported, as well as considering suicide for multiracial youth with 2019 having the higher percentage compared to 2009.

Then when you look at ethnicity, so this is data from 2021, broken down by Hispanic versus non-Hispanic youth. And I also broke it down by male versus female. And we have actually seen an increase in Hispanic females, at 118 percent, from 2000 to 2021. And for our male youth we have also seen an increase in our Hispanic youth at 71 percent. Whereas for our non-Hispanic youth we’ve seen an increase of 55 percent. So again, another group of youth that unfortunately is increasing and we don’t have a lot of data to help us to understand why these increases are occurring.

So black youth suicide is an area that I’ve focused some of my research on. And I just wanted to give you some insight on what is happening there. So in 2021 a suicide was the third leading cause of death for black youth five to 19 years of age, and accounted for 411 deaths in 2021. And again, increased, at 100 percent, from 2000, for this specific age group.

And then when you start looking at the data for males versus females, you actually see a different story for our black males over this time period. So 2000 to 2021, we’ve actually seen an increase of 57 percent for five- to 19-year-olds. 

But for our females we have actually seen an increase of 402 percent, which is just hard to even fathom when you think about youth suicide just in general. But when you start breaking it down, looking at males versus females, looking at race you start to see different stories and different trends occurring.

And then in 2020, when you look at the leading causes of death for black girls 12 to 14 years of age, suicide was actually the leading cause of death that was tied with unintentional injury. So 12- to 14-year-old black girls in 2020, suicide was number one.

This is a paper that Dr. Jeffrey Bridge and colleagues and myself, wrote where we actually examined the age by race disparity that was present in youth specifically from five to 17 years of age. And what we found is that for black youth, five to 12 years of age, they were approximately two times more likely to die by suicide than their white counterparts. And that actually flipped at age 13 when they were less likely to die by suicide. 

And then this led to a number of other papers that we wrote together. And I started to think, you know, let’s stop comparing black youth. Let’s actually just look at black youth only, and consider what is actually going on in terms of the trends by age, and looking at the trends by sex, and also gaining a better understanding of what are those circumstances that happened prior to that child’s death, so we can gain some more insight on how we can actually do better in terms of our intervention programming.

So this paper, which was conducted with myself and colleagues, we looked at black youth suicide only, from 2003 to 2017, and what we found is that by age we saw increases present in every single age group that we examined. For our five- to 11-year-olds they had an annual percent change of 3.9 percent. For our 12- to 14-year-old group they had the lowest annual percent change at 2.9 percent. And then for our 15- to 17-year-old age group they were the largest increase, which was at a 4.9 annual percent change over this period of time.

When you looked at males versus females, we saw again both groups increased over this time period, but for our female group they actually saw an annual percent change that was almost double that of our male group. So their annual percent change was 6.6, whereas for our males it was 2.8.

And then we wanted to look at circumstances that had occurred prior to that child’s death. And we did that by age group as well as by sex. And what we found was that there were a lot of similarities between our five- and 11-year-olds and our 12- to 14-year-olds, the majority of them who had died by suicide did so in their own homes, had a family problem or school problem prior to dying. And then also the injury that caused their death did occur within 24 hours of an argument.

Some differences did actually present themselves. For our 12- to 14-year-old age group they were more likely to be girls than boys. And then for our five- to eleven-year-old age group, if they were diagnosed with a mental health concern they were more likely to be diagnosed with ADHD.

For our 15- and 17-year-old age group they were more likely to have a boyfriend or girlfriend concern, a recent criminal or legal problem that had presented itself prior to the death of that child. They were also more likely to leave a suicide note compared to the other age groups, and if they were diagnosed with a mental health concern, it was either depression, dyssomnia or a substance use concern that they were diagnosed with.

When look at precipitating circumstances for girls versus boys, we found that girls were more likely to have a relationship problem or crisis prior to their death. They were also more likely to die within 24 hours of an argument. 

They were also more likely to have history of suicidal behavior, or a history of treatment, mental health treatment, or even be in current mental health treatment. If they were diagnosed with a mental health concern, it was either depression, dysthymia, or anxiety. And last but not least, they were also more likely to have a physical health problem present compared to their male counterparts.

And then for our boys, they were more likely to have a recent criminal or legal problem compared to the girls, and then also if they were diagnosed with a mental health concern, ADHD was that diagnosis that was present more compared to the girls.

So that is a lot of information. And I do know that it is a lot to process, so I understand that completely. But one thing if anything that I would like you all to take from this presentation is that I believe 110 percent that suicide is preventable. Having those numbers in your phone is just one way. 

But I do believe that there are other ways that we can be thinking about prevention and doing it in different ways for specific youths of color. And these are just some of those ways that I think might be relevant for black youth specifically.

So first and foremost, I think we have to figure out what are those specific culturally relevant risk factors that are present for black youth specifically. Because if we don’t know what those risk factors are, how can we actually implement interventions that are going to address those risk factors? 

So I think that it is really important for us to do some more research and to understand how racism, discrimination, and other culturally relevant risk factors are related to suicidal behavior in black youth specifically. We have seen some research recently that has addressed discrimination, racism, those experiences in black youth and how those are related to suicidal behavior. And I think we are starting to get some insight, but still more research needs to be done so that we can really address these concerns and do a good job at doing so.

I think also some of the theories that do exist in the field do a good job at explaining how suicidal behavior may occur, or how suicide may occur, but we don’t really know if these universal theories are technically universal for minoritized groups. 

So it is really important for us to test these theories in groups that identify as minoritized, to make sure that we’re using the right theories to actually do the research that we’re doing. If they don’t, then we either need to adapt them, or we need to create our own theories that can actually benefit and decrease the rates of suicide that we’re seeing in black youth specifically.

Also something else that I think is really important is to investigate this pathway. Because what we found in another paper that myself, and Dr. Michael Lindsay at NYU conducted, we found that looking at the YRBS data that black youth are more likely to self-report suicide attempts and more likely to report on suicide attempts that actually required medical attention compared to other racial groups. But they were not more likely to actually report more suicidal ideation. 

So in my brain I think you have the thoughts that come first and then you have the actual behavior. But for black youth it may be that there might be a specific pathway that we’re missing, that we don’t really quite understand. Or it may be that we’re not assessing suicidal ideation correctly for black youth specifically. 

And I think that is really going to take into consideration these mixed method approaches, talking to kids, talking to black kids specifically about their experiences, about how they got to a space where they thought that suicide unfortunately was the way out, gaining some insight and being more qualitative specific in the research that we’re doing, and then pairing that with the quantitative data that we have to get a better understanding of the problem of black youth suicide.

And these are just some promising avenues I believe that could be beneficial for preventing suicidal behavior in black youth. I think community really is key. I think it is so important for us to engage community, whether it be gatekeeper training programs, after school programs, boys’ and girls’ clubs, or even training barbers and beauticians or even other places where black youth are present, at urban leagues, coaches. 

Giving them the information to know what are the signs that they should be watching for for suicidal behavior and when they should take the next step what are those next steps to take. So I think it is really important for us to think outside the box. Schools are a great place to do interventions, but kids also are other places besides schools, and I think it is really important for us to consider bringing prevention to the community where these kids reside.

Also, faith-based organizations is another opportunity for us to engage with suicide prevention. I think now we are in a space where more places that are dedicated to faith-based, like synagogues, churches, mosques, they’re interested in talking about mental health more than what they have been before. 

And I think it is really important for us to engage with that leadership and think about how can we take our prevention programming and our knowledge to those spaces so that we are helping youth and also are just the congregations and as a whole. 

So there was one program by Sherry Molock called the Haven Project that I have been very lucky to actually learn more about here at the University of Rochester Medical Center where she, Peter Wyman and Sydney Hankerson, our payer, are actually coming together to consider how this project can be tested in churches, in black churches specifically, and how we can actually take that knowledge and expound upon it so that we can actually think about how we can engage other faith-based organizations to do suicide prevention.

And then also thinking about school-based preventions as well. Kids are in schools. So how do we take some of these school programs that have been doing a good job, and make sure that we are addressing those again culturally relevant risk factors for black youth specifically. 

So Dr. Robinson actually took the Coping With Stress course, and put the curriculum on the table, and had black youth talk to her about the curriculum, what is it missing, what do we need to add, how can we make this relevant. And now she has a grant that is testing that.

Also there is Dr. Goodwill, who I just recently found out is looking at the Signs of Suicide Prevention programming, and how that is relevant for black youth. So again, taking these programs and ensuring that they are actually speaking to the masses versus just a specific group of kids, and making sure that they are relevant for all youth.

So without further ado, I just want to again thank my team and Nationwide Children’s Hospital as well for the work that I have conducted, and I am going to pass it on to Dr. James.

JOETTE JAMES: Good morning everyone. It is a pleasure to be here. It is great to have the opportunity to talk a little bit about something I do every day, I’m a clinician, not a researcher, but I really think this is such an important discussion. Adding into it I’m going to talk about autism spectrum disorder, and that is a population right now that is getting an incredible amount of attention. There is in some ways a zeitgeist happening with respect to ASD. 

And what I am going to talk about is ASD primarily, but also about intellectual disability and thinking about the symptoms that are often associated with autism spectrum disorders and how we can conceptualize them in thinking about suicide and self-harm and non-suicidal self-injury. 

So I am a neuropsychologist by training, and I do evaluations of youth and adults and preschoolers with often now the main question or referral concern that I get being about autism, ADHD, and intellectual disability. And I apologize for the eye bleeding slide here, but this is the DSM. 

There may be a few eye bleeding slides, the DSM text revision. The most recent version of the DSM and its definition of autism. And at its core, autism spectrum disorders are about challenges in socioemotional reciprocity, that back and forth, whether that is in conversation, sharing of emotions, sharing of interests. 

Secondly, individuals with autism spectrum disorders often have difficulties with nonverbal communication, using eye contact, using tone of voice, using facial expressions, and being also able to interpret those in others and show them themselves. And I recently evaluated a three-year-old for autism. Thank goodness she did not have an autism spectrum disorder. 

But one of the examples I gave to the parents was out in the waiting room where we see this incredible even at a young age back and forth and use of nonverbal communication to really send a message. I had just done some testing with this three-year-old child, very adorable, and it came back, she had a break in the waiting room with her mom, I came back out because there were some more tests that I wanted to do with her. She had a truck in her hand, and her mom was giving her some water. 

And I knew she did not want to come back with me again and do more work. She had the truck in her hand, she’s looking over at the toys in the waiting room, she was taking her sweet time drinking that glass of water, looking at her mom, looking back at me, an incredible amount of communication going on there in a three-year-old child. These are behaviors from a very early age that are incredibly complex. 

The thing about when we see individuals that have autism, they may not have all of that coordination happening at a high level or may just be struggling with those aspects of nonverbal communication. 

We think also about this idea of being able to develop and maintain friendships and relationships in different social contexts. Sometimes that can be folks that are active in their socialization but may be missing some of those cues from others, but also individuals who may not show interest at all. This is what makes autism incredibly complicated, we’ll get to that in a moment. 

Again, another eye bleeding slide. So we look at things like repetitive movements, insistence on sameness, difficulties in transitions, and rigid thinking patterns, fixated interests, sometimes those interests are unusual, like lights, or toilets, or street signs, or I have had patients who are really interested, I have had a seven year old who was really interested in the paintings of Edward Hopper, and that was his fixed interest. 

And this is important to think about when we think about in term of context of suicide, because individuals can have these really strong interests or focus, and being distracted from them and not being allowed to engage in those interests can be a source of emotional distress which can lead to some self-harm and self-injury. 

Similarly, with hyper- or hypo-reactivity to stimuli, being attracted to certain sounds, or being attracted to bright lights for example, or having difficulty managing loud sounds or crowds can be again, another source of emotional distress which can lead to self-injurious behaviors.

The key with autism is the importance of these behaviors being present in that developmental period, which I’m going to talk a little bit later about why that might be problematic in a death investigation or suicide when we don’t really have a diagnosis. 

Because now we are at a point where we are seeing more adults getting diagnosed, but then it becomes really difficult to make a diagnosis definitively when we don’t have information about that developmental period.

Here on this slide I am talking about behaviors that are sort of the translation of the DSM criteria. And the ones I’ve had in yellow are the ones that are related to challenges in social interaction and communication, which can make it difficult to communicate and know an internal state, how someone is feeling in terms of their emotions, their anxiety, and perhaps symptoms of depression that might be there. And really communicating that internal world.

Similarly, when we talk about that second piece of stereotyped behaviors or fixed interests, we have emotional distress as I mentioned, at any change in routine or change in schedule that can then cause or result in a behavioral or emotional outburst, and then produce, lead to self-injurious behaviors.

As with fascination with certain kinds of stimuli, for example a child that may be fascinated by a spinning colorful ball and end up running across the street into traffic, or impulsively trying to get away or dart away from an adverse sensory experience. Many autistic folks are very sensitive to loud sounds, may cover ears, may have trouble in crowds, with fireworks, et cetera. 

And so may be in a mode, especially when language and communication skills are not well developed, to using behavioral means of escaping and getting away from those things that might bother or disturb them that can lead to injury.  So either a hyper-sensitivity or a hyposensitivity. I have had patients who have had hyposensitivity to pain and touching hot surfaces, that can lead to injury in that way.

I think what in talking about autism and suicide and self-injury, sometimes I feel like I’m making the landscape more complicated and raising more questions, but I think that it is really a nuance here when we’re talking about developmental disabilities and conditions.

This statistic is actually old now, because if anyone has been listening more recently, the 2020 data tells us that, from the autism and developmental disabilities monitoring network, tells us now that it is more recently one in 36 children identify, this would be children at the age of eight, identified with ASD.

In the United States this would be the year 2020, so that’s even more than the data that you’re seeing on the slide, it’s more prevalent now, and we are just seeing increases. We’re not exactly sure why this is happening, but the most recent would say one in 36. 

It is much more common for white children to be diagnosed and to be diagnosed earlier than black children or Hispanic children. And 33 percent of individuals with autism are also diagnosed with an intellectual disability. Intellectual disability is a separate diagnosis that speaks to primarily an IQ under about two percent, an IQ of 70 to 75 or below, the second percentile. 

And then is accompanied by deficits or problems in everyday functioning. In communication, in functional academics, in leisure skills, in self-care. And again, that can vary in severity. But a sizeable proportion of individuals with autism also have that comorbidity. 

And I think one of the things that we do know is that comorbidity really rules the day in autism, and I’ll talk about that in a moment. Again, there are discrepancies in when autism is identified. And so this is problematic when we’re thinking about, again, whether or not someone has a diagnosis at the time of death. 

You might think that now we were in 2023 and that has really resolved, it really hasn’t, there are huge discrepancies amongst different communities in terms of when the diagnosis happens. And again, because autism itself is so variable and is a spectrum, that contributes to the challenges. With boys still being much more likely to be diagnosed than girls.

There are a number of comorbid physical and medical conditions that are associated with autism at an increased risk. Those would include primarily fragile X and tuberous sclerosis, and the other risk, and particularly when it comes to accidental deaths, would be an increased risk of seizure disorders. And I used to do some work with a brain bank where we looked at just individuals who had died who often had comorbid autism and a significant seizure disorder. So it is a much increased risk in this population. 

And these are some of the others, just looking at quick facts about ASD in terms of increased risk with older parents, and with the comorbidities that are in the larger phenotypes in families, typically anxiety, and depression as well, and ADHD.

So really understanding ASD is important, because as a neurodevelopmental disorder it is important to be able to assess the behaviors in the developmental period in childhood, and that might not happen as I said for an adult. 

I have actually in a forensic context evaluated individuals who are 50 years old, 60 years old, 70 years old, and it gets more and more challenging to make a definitive diagnosis, because you don’t have individuals who can speak to behaviors in that developmental period if they’ve passed away. 

And so that can be really challenging in terms of making that diagnosis. And then in the context of suicide that individual as an adult may not actually have a diagnosis of ASD. It is changing, but it is still very tricky to make that diagnosis, especially if there is no-one able to speak to that time period.

I think that in my career I would say that one of the biggest challenges with autism is that the wide variety of presentations. I have seen folks who are nonverbal with significant intellectual disability. I’ve also seen folks with IQs of the 99th percentile who have incredible vocabularies but extreme difficulty functioning in real life situations, and their actual adaptive everyday functioning is nowhere near what you would expect given their intelligence, and that bears out in employment and just other areas of life. 

This wide variety of symptom presentation makes it challenging because making a definitive diagnosis, it is not really looking at the criteria and seeing how it may apply in a particular individual, especially as the symptoms can change over time, especially in individuals who have higher intellectual ability, and they can mask and they can learn certain behaviors that they know are socially expected, and can present with challenges in being able to make again that definitive diagnosis.

Comorbidities are incredibly important. As our last speaker had mentioned, ADHD is often an impediment risk factor for suicide that is certainly in autism. There are approximately a third of individuals with autism also have comorbid ADHD, ADHD being a risk factor in part because of the impulsivity associated with ADHD. But as you can see here, there is incredible overlap with learning disabilities. About 20 percent of individuals with autism also have co-occurring anxiety disorders, and 11 percent have co-occurring depression.

So when we look at autism and suicide, I think the theme here that I have been hearing in these special populations, really it is a relatively new area of research, and as a result we are just at the beginning of understanding these relationships, particularly with respect to youth. 

And some of the issues that we have with the studies are really variabilities in IQs in the samples, small sample sizes, not accounting for things like race and ethnicity. This data is often not included in studies, or how the autism spectrum disorder diagnosis was actually made. 

There are just a wide variety of ways in which the diagnosis can be made in individuals making that diagnosis, whether a psychologist or a pediatrician, and different measures used to make that diagnosis. And studies often are not reporting that data. Combining suicidal ideation with suicide attempts often happens. 

And so we still have issues with the characterization of these samples. But we definitely know there is a higher risk of suicidal ideation and behavior in individuals with autism, and you can see one in four under 25 years of age reporting suicidal ideation out of this O’Halloran--

(Signal frozen)

Oh, my internet connection is unstable again. There are definite differences between self and parent report. We see this all the time clinically. Why that is the case, we don’t know yet. This is really preliminary. We see this with teacher and parent report. Often teachers will report more or less than parents and vice versa. Why this is happening in this population we don’t know yet, but there may be some discrepancies there that we still need to account for. 

Interestingly, factors associated with increased risk for suicide in this population include adverse childhood experiences, lower socioeconomic status, being a person of color, and interestingly bullying. That is one of the questions that I ask often from a clinical standpoint, especially when I am looking at young adults and adults, and I’m looking backwards to part of an overall assessment or evaluation of whether someone has ASD. 

Bullying is very, very common in the histories of individuals with ASD. They often don’t know why they were bullied, or they have trouble reporting on the specifics of that, but it is a very common thing that we see. And of course comorbid mood disorder.

There has been a little bit of work on protective factors for suicide, including social skills training programs developed to help individuals with autism develop better social cognition and understanding.

So in terms of thinking about self-injury versus self-harm, non-suicidal health injurious behavior is very common in ASD, with higher odds than those without. And these include things like hair pulling, hand biting, head banging. And I’ve seen patients do this to the extent to which they are actually detaching retinas. 

And this is very different from what we might see in someone who is engaging in cutting behavior as part of an emotional mood regulation situation. It is really about escaping situations where there is sensory overload or sensory seeking as a way for someone with autism to modulate their experience of sensory stimulation. And ASD is associated with increased morbidity and mortality. 

And I think some of the things that are important when we think about suicidality is individuals with autism often have trouble communicating their emotional life and talking about their internal experiences, and that’s part of having this condition. So recognizing their emotional experience, being able to communicate it effectively to others.

And so questions that might be on a suicide assessment instrument may just not work well for someone with autism. For example, have you had any intention of acting on these thoughts of killing yourself as opposed to you having the thoughts but definitely would not act on them.  In someone with autism that is just going to go right over their heads, the complexity, the language, and the abstraction, as well as the focus on the internal life.

The last thing I am going to talk about is just diagnostic overshadowing. And I think, again, and this is a new area of research, and I think that there are obviously the comorbidities with anxiety and depression that would increase suicide in an individual with autism or suicidal ideation at the very least, but I think sometimes the focus on autism is such that it overshadows everything, we may be missing some of those critical conditions that might help us understand better what the risk might be.

And I am now just going to turn over the podium to new Dr. Fontanella.

CYNTHIA FONTANELLA: Hello, today I am going to be talking to you about suicide among youth in the child welfare system. One of the things that I want to talk about is describe the rate of suicide among youth in the child welfare system, and risk and protective factors, share some data from an NIMH study that we’re conducting. And to describe some of the challenges of studying suicide among youth in the child welfare system.

Why should we focus on youth in the child welfare system? Well, youth in the child welfare system represent an important boundaried population at high risk for suicide. Youth in the child welfare system are a vulnerable group at high risk for suicide compared to the general population because they have higher rates of mental health disorders, addictions, risk and exposure, risk-exposure history such as poverty, child maltreatment, parental psychopathology, substance abuse, and exposure to community violence. 

Now I’m going to talk to you about what are some of the rates of suicide among youth in the child welfare system. In terms of suicide deaths, one study found that youth involved in child welfare or juvenile justice were three to five times more likely to die by suicide than in the general population. A largescale study in Sweden found that youth are more than four times, 4.3 times more likely to die by suicide than youth in the general population, youth in foster care.

In terms of suicide attempts, adolescents who have been in foster care were nearly four times more likely to have attempted suicide than other youth, and experiencing trauma or child abuse increased the risk of attempted suicide by two to five-fold. In terms of suicide ideation, adolescents who had been in foster care were nearly two and a half times more likely to seriously consider suicide than other youth.

Now I just want to briefly go over what the definition of risk and protective factors. A risk factor is an attribute or characteristic that makes it more likely for an individual to consider to attempt or die by suicide. And this can be a combination of individual relationship, community or societal factors.

Protective factors are characteristics that make it less likely for an individual to consider, attempt, or die by suicide. These same characteristics can also reduce a risk factor’s impact. And I like a socioecological approach to look at risk and protective factors. 

You can see at the individual level some risk factors are mental illness, substance abuse, previous suicide attempts, impulsivity and aggression at the relationship level, high conflict or violent relationships, family history of suicide at the community level. Poor access to healthcare resources and supportive relationships at the societal level. Availability of lethal means for suicide. Unsafe media portrayals of suicide.

And what are some risk factors for youth in a child welfare system? A lot of them are very similar to the general population. Some big risk factors are mental illness, substance abuse. An extremely strong risk factor for suicide is prior suicide attempt. Self-injury. Abuse. History of abuse and neglect. Trauma. Family conflict and dysfunction.

Protective factors are psychological or emotional wellbeing, family connectedness, safe school or school connectedness, exposure to a caring adult, connected to supportive parent and coping skills.

In terms of the gaps in the research, there are few population-based studies on suicide among youth in the child welfare. To date there are only three studies that have examined suicide deaths among youth in the child welfare studies. 

And all of these studies have been outside of the United States, mostly in Nordic countries. While there is quite a bit of research on risk factors for suicidal behavior, there are minimal risk factors on risk factors for suicide among youth in the child welfare system. 

So now I want to talk a little bit about our study. We integrated data from multiple data sets in Ohio. Ohio Medicaid, child welfare, juvenile justice, Ohio Violent Reporting Death System, education data, and death certificate data. And the data on administrative data was from 2010 to 2020 for over 2.2 million youth ages five to 25.

So now I want to talk a little bit about some of the findings that we found. We looked at mortality rates for kids in the Child Welfare System compared to the general population and Medicaid. We looked at all-cause-mortality, suicide and overdose and other causes of mortality. 

As you can see the risk of dying by suicide is in Ohio 2.5 times higher for youth in the child welfare system than the general population. And the risk of dying by suicide is nearly three times higher for youth in the child welfare system compared to the overall Medicaid population. 

The risk of overdose death is also quite high. For overdose deaths the risk is 4.8 times higher for youth in the child welfare system compared to the general population. Among females, the risk of dying in child welfare system is 3.78 times higher than the general population. And among youth 5 to 14 the risk of suicide is 3.4 times higher than the general population.

We also looked at demographic and placement factors associated with suicide for youth in the child welfare system. And removal from the home, because of parental problems with the child, the odds were 3.7 times more likely, youth were 3.7 times more likely to die by suicide. 

Living in a detention center the odds of dying by suicide was 5.2 times more likely to die by suicide. Living in congregant care, the odds were 3.4 times higher than other types of placement. As you increase the age you’re more likely to die by suicide.

Non-Hispanic black youth are 72 percent less likely to die by suicide compared to white youth. And males are more likely, 2.2 times more likely to die by suicide compared to females.

We were also interested in looking at health service use because health providers are critical to suicide prevention strategies. And we looked at health service use for youth in the child welfare system at one month and six months. 

As you can see, 90 percent of youth who died by suicide had a physical health or mental health visit within six months of suicide. So this provides an opportunity to intervene, to reduce suicide. And at six months 48 percent had a visit within one month of death. 50 percent had a mental health visit within six months before death. And a little over a quarter had a mental health visit within a month of death.

We were also interested in looking at what are the clinical characteristics of suicide decedents compared to match controls. And as you can see, most if you were in an out of home placement that increased the odds of suicide by 2.2 times. Having a mental health diagnosis was strongly associated with suicide death. The highest odds were for schizophrenia, bipolar disorder, depression, and other diagnoses. 

As you increase the number of mental health diagnoses that increase the number, the odds of suicide, having a substance abuse diagnosis, youths were four times more likely to die by suicide compared to controls. And in terms of medical diagnoses, having a seizure disorder that increased the odds by 3.5 times more likely to die by suicide compared to controls. Also, previous self-harm, 16 times more likely to die by suicide compared to controls.

In terms of what is the availability of data, what is the data source and what are the strengths and limitations of some of these data sources, AFCARS has some data on mortality. Unfortunately they don’t break it down by suicide. And hopefully in the future they’ll do that.

The NSCAW database includes some variables on suicidal behavior, suicidal ideation, and suicidal attempts. And the NVDRS has data on all suicides, but it tends to be under-reported for kids in the child welfare system. National Medicaid data, with mortality data. And they have variables on kids in foster care. And of course state child welfare systems can be linked with child mortality data.

What are some barriers to research? Lack of data on suicide surveillance for youth in the child welfare system. And child welfare systems have disparate systems, which makes it complicated to collect data. Youths in the child welfare systems are also involved in multiple siloed systems, and it is very difficult to integrate and to link data across these systems. And suicide among youth in child welfare is often underreported by coroners.

In terms of prevention, where do we go? I am a strong advocate for universal screening in the child welfare system. The universal Zero Suicide Approach developed for health and behavioral health care systems is one approached. 

Also, trauma informed approaches such as Trauma-Focused Cognitive Behavioral Therapy may offer an integrated treatment for youth in child welfare and co-occurring trauma and suicidal behavior. And that’s all I have. I am going to pass it on to Dr. Wasserman.

GAIL WASSERMAN: Okay, well, I’m going to give something a little bit upbeat. Good, we need this. Okay, so these are my acknowledgments. Thank you all. So, clearly there are barriers for those of you who want to do work in the juvenile justice system. Some of those have to do with the operation of the system itself, we will talk a little bit about that, that’s not so upbeat. And it is sort of parallel controls related to confidentiality. That’s also not so upbeat. Clearly don’t wade in the water without a guide. 

And the take home is that you should be aware there is no single justice system. There is no single authority to deal with. Kids get arrested. They go to an intake service, either in probation or in the court system. 

These are separate entities. Separate agencies, separate authorities, separate administrations, separate record systems. Either they go to court, or they don’t. Either they go to detention, or they don’t. Either they go to probation, or they don’t. And either they go on to secure care or they don’t. 

At each juncture there is a full, hard stop in recordkeeping. Each state, each county has a totally different system of operation. This is an extraordinarily, there are comparabilities, parallels, but it is really very difficult to get some handle on what it is that is going on.

Most kids who enter the justice system are passed on to other sectors, with little follow-up or follow-back of information that would be available either to a practitioner or to a researcher. So as I say, there is a full, hard stop. Once the kid has moved on, say into a program, it is very difficult to go back and figure out information about where he has come from and accessed those records, because the justice records for adolescents are sealed generally. 

If we do want to know where they go on the other hand, first of all about half of those who enter really don’t go any further than, they don’t even get to that intake visit. They sort of get diverted by law enforcement before they even enter probation or court intake. 

But once they do, as you can sort of see by the themes here, probation is going to be a very common theme one way or another in terms of services that are provided for them.

There are great flowcharts that depict the falloff of cases. If you want to see the proportions of kids that go from one system to another that you can see if you look at those references that are published every four or five years by the Office of Juvenile Justice that you can track based on court statistics. But among the cases that go forward to intake services, so mind you half of them get off of the charts before they get there. About 45 percent are followed by their local community probation authorities. 

So cases that are monitored informally, in lieu of appearing in court, have to agree with certain conditions with probation authorities, and then they may be followed only irregularly for a few months, and then the case is closed, the case is sealed. It may be very difficult for the agency to access the information even if they wanted to. 

Cases that go on to the court, that are adjudicated and found delinquent, may get more formal probation supervision as part of their sentencing. They may be followed more systematically and from a longer period of time, usually for a year or two, at which point then the same full, hard stop occurs. 

Case is wiped, sealed, closed, it is really very difficult to get in at that point and find out what has happened prior and before. If they are instead diverted into some kind of community program or some other sanctioned or mental health services or some community program, again the records really don’t go with a kid necessarily. And the length and type of supervision from probation authorities varies by jurisdiction, but it can be quite casual and quite informal in most cases. 

So depending on the consequences for the young person, he or she may move on to supervision by another agency or no supervision at all. So 33 percent of the kids that go into intake before court will be dismissed. Eight percent only will go on to be placed in secure facilities. 

So most kids do not go into residential care. One percent are sent up to adult criminal court if they’ve been engaged in a very serious offense. 25 percent will receive some other sanction. Each transition means a new organization, little or no information feedback back and forth. 

So when they transition from court to some other agency, the court very rarely gets ongoing reports. This just ends and you move on to another setting, and a strong reluctance to break confidentiality because of stigma associated with juvenile justice status. So that’s a good thing for the kid who is supposed to be able to have a fresh start, but a very difficult challenge for researchers or for clinicians who want to get a full picture of what is going on.

So, despite the barriers, looking at suicide risks and prevention in this population is however critical. Almost everything we’ve talked about thus far is elevated as a risk in this population. Neighborhood risks are elevated. Family risks are elevated. Frankly listening to people talking about having conversations with families after a suicide, I honestly can’t imagine those conversations with families of youth in the justice system, many of whom are so alienated from public authorities that I can’t imagine them wanting to have an open and fair honest conversation with any of you. That is a real challenge.

There are peer networks, obviously school problems, definitely mental health challenges in terms of loss, trauma, and the like. High levels of behavioral health problems, elevations. 

This is work that we did looking at self-reported mental health problems of almost 10,000 kids in 18 states at three levels of justice penetration. Risks for mental health and suicide attempt history increased as sort of severity, we call it penetration of the justice system as you move from being in community probation to detention to secure care, as you penetrate the justice system, as risk increases, reports of all sorts of mental health problems and suicide history increase significantly. 

These kids also have increased means and opportunity, access to firearms certainly is increased, although it is is interesting to hear that suffocation is now, which is sort of an equal opportunity suicide means, you don’t need to be a juvenile justice participant to have access to that means certainly. We can do supervision at home. Kids who are not in school, who are roaming around loose and have a lot of free time on their hands.

So, as a result of work like that, which described the level of mental health problems in these kids, the justice system moved in the 1990s and early 2000s to consistent screening initiatives in most of its settings. And therein lies a very interesting conundrum.

So here is data from our JJ-TRIALS work, which really is based on 8000 probationers in six states. So what you can see, we have 8000 kids. These are all kids in community supervision. 

Almost all of them, 72 percent of them get screened for mental health services. About half of them have some sort of mental health need. But only 22 percent of those kids are getting referral to mental health services, even though it is their agency practice to screen them. Leading me in great frustration, every time I have a training for those probation agencies, to say the purpose of screening is not to screen.

So I think this is typical, and maybe you all need to keep this in mind as you plan your work with public agencies. And I think this is typical of the interface between public health and public agency policy, that there can be the adoption of a practice without really grasping the underlying purpose. The purpose of screening is to identify a kid who has a need, and if he has a need to address that need. So what on earth agencies are doing, screening, full stop, and doing nothing about it. 

So if only 22 percent of kids are getting a referral, not a great surprise that only 15 percent of kids with needs are actually initiating treatment in their community service behavioral health agencies.

So I had an epiphany in the 1990s. Peter Jenson came from NIMH to work at Columbia in the Division of Child Psychiatry where I worked, and he said stop counting and start figuring out how to fix this. Very good advice to a psychologist who is working doing mostly epidemiology. 

So we began trying to draft a way to A, probation offices to fix the linkage problem. They knew how to screen, and they were starting to screen, and they didn’t want us to do anything about it. They kept saying we’re not clinicians, we don’t know how to do this. Mental health doesn’t want to take our kids. What do we do about this?

So we came up with Project Connect, which was very clutzy, in the 1990s it was paper and pencil, and it was very slow. It worked. It was a suicide screening program under the Garret Lee Smith blueprint. It worked. A lot of kids still didn’t make it to care, and because it was so clutzy it wasn’t sustained.

So fast-forward, years go by, and we came up with an app in more recent years, which is a web-based clinical decision support system which is installed on tablets, seamlessly combining screening of a history of suicidal behavior and related behavioral health problems that classifies clinical need and allows county specific real-time referrals for behavioral health services based on need, and assists program officers who are non-clinicians, and identifying risk, referring to care, and achieving that cross-system linkage. 

So what does it look like? So a kid completes a voiced, because we know these kids have reading issues, it’s voiced, they hear it, very short, 5.6 minutes web-based screen. There is a risk classification, which we have three layers, because not all suicide risk is the same. There are pants on fire kids, and there are kids who you can wait a little bit longer. The results go up to the cloud, they come down, they say this is class one, class two, or class three. 

Each county has specific risk pathways defined in advance by probation and behavioral health agencies. What are you going to do about these kids, what are you going to do about these kids. That appears on the handheld so the probation officer doesn’t have really any discretion or decision, it tells him what to do, get Billy to this agency, call this person, call Joe, take him to the green door. It determines those pathways. It gives more information, sends along a referral form, helps connect with a provider, sends that referral form along with a kid to the correct provider. And here are the results. 

So first we have screening. So yes, compared to baseline kids and under implementation, screening is about the same. Identification is about the same, but the adjusted odds ratio, adjusting for gender, race, and county, we have about an elevenfold increase in referral of kids who have unmet behavioral health need, and a 16, almost 17 percent increase in treatment initiation. 

So if you sort of know what you have to do and take each step of the way out of it, and try to address and micro deal with the problems as they exist, you can perhaps eliminate some of the shortfall.

BRENDAN WEINTRAUB: So I am going to have all the folks who just presented come on their cameras, and I am going to pass it over to Stacia.

STACIA FRIEDMAN-HILL: Thanks Brendan. So first of all, I wanted to thank all of the presenters for a really wonderful session, just filled with thoughtful, informative presentations. This session included a really diverse set of populations. I saw several common themes that were running through the presentations, including themes such as challenges in obtaining the relevant data, under-reporting, and disparities in access to mental health services. 

And I want to give you all a chance to engage in conversation with each other, but I also wanted to ask a question myself to kick off the discussion. And I’ll apologize, it is going to take me several sentences to ask this question, so I hope that you will bear with me. 

In many cases the description of certain populations of youth that are at higher or increasing risk for suicide is based on demographic information that is a categorical checkbox on a medical examiner’s report or death reporting system form. 

And that may tell us or identify who is more likely to have a higher risk or increasing risk levels, but it doesn’t necessarily tell us why. And we know that there are likely to be many contributing factors which co-occur. 

And some of the talks today and yesterday have mentioned YRBSS data or life stressors categories and child death reports or educational data. I wanted to ask you if you could talk some more about what is your approach to using epidemiological data to incorporate multilevel, multifactor perspectives, and additionally what information or tools would make it easier for you to use this work. So this is open to any of the panelists from today’s session.

JOHN BLOSNICH: At least whenever we think about LGBT youth, I think Dr. Staley had a very good point about that is not the risk factor. I would describe, it’s like my students. It is sort of how astronomers know that there is a planet around a star, they can’t see the actual planet. You look for signs like the star’s light dims, and the light dims at a certain interval you can measure, and that there’s a pull on that you can measure. You get these clues, but you can’t actually see the thing. Yet. 

I think it’s when we get down to things like if we think about experiences of intergenerational poverty, systemic racism, historical racism, heterosexism, things that we don’t do well at measuring those experiences are the things that are kind of behind these identities that we collect from individuals. 

So I think that is probably one of the most important caveats in what we have, is making sure that when we discuss these findings to folks is that we as researchers, to the extent of practitioners that print out reports, that we couch these in ways that folks know how to interpret the data in kind of its context. 

And knowing that as much as we can’t even gather LGBT information we are certainly not good at having valid instruments at measuring these other sorts of largescale, largely hidden for a lot of say some nefarious reasons, these discriminatory historical and present structures that are aimed at vulnerable and marginalized populations, and they go really covert very quickly. 

And so as much as we can go after the things and the characteristics we can see, a lot of the things we can’t see. But I think it’s on us to be able to have that conversation and to be able to put that.

CYNTHIA FONTANELLA: Well I think that we need to look at these risk factors as you were saying, Stacia, at multiple levels. Not just the individual level and the context as John was saying. And the contextual level factors, community level factors. 

And some of the tools that we can use at the community level are the area resource file, the census data, and area deprivation index, and certainly where people live, that geographical location is really important too. And so we need to look at both individual, family and community level factors of risk.

GAIL WASSERMAN: It strikes me that there is almost a parallel about inquiring about some of the information that you’re asking about, the medical examiner is asking about, which is very sensitive. And somebody earlier had mentioned that they feel uncomfortable asking certain questions. 

And I just say that I’ve heard that from probation officers who don’t want to talk about suicide with the individual, with the adolescent. And I wonder if the modality of having, another sense we heard, a prerecorded set of questions with a very sensitive, engaged interviewer in that format, might take some of that discomfort out of some of these questions, where both the interviewer and the respondent doesn’t want to have this conversation. It is so much easier to talk to an anonymous computer. 

We’ve done work for a long time with the DISC in its voice version, people are much more comfortable talking to a computer, even if they know a human is going to review the information, than they are face to face with a person. And that person is very often, if it is a law enforcement person or like our probation officers, they are very uncomfortable raising certain topics. 

So I wonder if that is a modality that people might consider using, and it might get you more uniform information about sensitive topics. Not really an answer to your question, but a suggestion.

STACIA FRIEDMAN-HILL: Thank you. I think Dr. Luby has her hand raised.

JOAN LUBY: Thank you. Going back to what John just said and building on it. The issue of, to me, measures of discrimination, marginalization, sort of experience of social hostility. I mean if we sort of think about the question of why are these rates of suicide going up so dramatically, of course that’s the overhanging issue. 

And you think about the sessions this morning, and you look at these marginalized groups who have these extremely high rates of suicide, the one thing that they’ve had in common is they’re marginalized, they’re invalidated, they experience a lot of social hostility. That signal is increasing over the last four to six years where in our society there are these organized efforts to marginalize, reject, and take away the rights of these people. I do think that that is a really important thing for us to measure.

STACIA FRIEDMAN-HILL: Thank you for that. Ms. McGivern?

LAURI MCGIVERN: I just want to comment again on Dr. Wasserman’s information gathering. And I think death investigators in general, I’m not going to speak for everyone for sure, but they do interview a lot of families that are in distress at the time of the incident. So I think that they do have experience with interviewing skills. 

And although the questions can be very difficult, there is a systematic way that we ask questions. And I thank both John and Michael, it is great that you have worked with medical examiners and actually been out on scenes to understand the dynamics and how stressful these scenes can be. 

So I don’t know if it’s a matter of, and you may not be speaking specifically about death investigators being uncomfortable asking these questions. I don’t know that it’s a matter of being uncomfortable, it’s a matter of whether the time is right, and you have to take that opportunity when you can, but the time may not be right in all of these cases to be asking these questions beyond what we need for cause and manner of death. 

So I just wanted to kind of put that back out there to remind folks that these scenes are really emotionally charged and extremely challenging. So it is not necessarily that the interviewers are uncomfortable with the questions.

JOAN LUBY: But Lauri, it is also unsafe. It can be an unsafe environment too.

LAURI MCGIVERN: It could be an unsafe environment as well. There is a lot of deferred anger, especially early on, and that anger can be pointed at the investigators. So as you’re asking these questions, we’re often asked why are you asking this, it doesn’t make sense to the people, the family members that are sitting there when we’re asking questions that seem a little bit outside in terms of the death, the incident that occurred. 

Why are we asking them about housing, even gender identity, which I do think is extremely important. I also do like that there have been some follow-up with these families. And it sounds like those follow-ups have been successful, and that that may be definitely another avenue to consider.

MICHAEL STALEY: I think that certainly the death scene is one opportunity, but even if a hundred percent of death investigators and a hundred percent of families showed up and we had the opportunity to do the interviews on the scenes of these deaths, I would still keep our program calling families three months down the road. 

We might even ask some of these questions again, because I think that more repetition is good. But also that they might have learned something in those four to eight months that they didn’t know beforehand, but also that postvention piece. I really think that we view these and we sell these as an epidemiological gathering session, but it’s also postvention. It is an intervention in the future, preventing future suicides as well in those scenes. It’s just such a mixed bag as well. 

So I think the consistency, even though it’s a phone call it’s not in person, it does afford us just another opportunity, a consistency opportunity to get the data. 

STACIA FRIEDMAN-HILL: Dr. Boyd, I want to make sure that we acknowledge that you had a question as well, or comment.

RHONDA BOYD: It was more of a comment. I just want to reiterate what John said, and I think it is the wording about sociodemographic factors, instead of calling them risks, they’re not necessarily risks, we don’t want to think of race, ethnicity, gender as risk factors. They can be characteristics. And so we take that into account. 

One thing I realize, many of the things that happen to many marginalized groups, there is a socioeconomic poverty impact, and we haven’t talked about that at all. So it is not clear, but many times in addition to the discrimination that many times that groups also are impoverished or economically disadvantaged based on sort of the systematic structures in our country. And so we need to take all of this into account when we think about sort of the contextual factors that impact suicide risk.

STACIA FRIEDMAN-HILL: Thank you very much for that. I do want to say that there has been some criticism of the field that there is too much emphasis on individual level factors and not enough acknowledgment of racism. So instead of focusing on race we should be focusing on racism and discrimination, both in terms of the legacy of structural racism, but also perceived discrimination, and looking at time trends, following for example national events of racial violence or policies that stigmatized LGBTQ youth. So that seems to be an area that I think research has only recently started to really illuminate, and we need much more in that area. Dr. Warner, you have had your hand up, and down, and up again, so I want to make sure we get to you.

MARGARET WARNER: I keep raising and lowering it depending on where the conversation has taken us. But since you brought that up, it does lead into what I wanted to say back in the beginning when I first had my hand raised, which is I 100 percent agree with what you’re saying in terms of structural racism or whatever we’re calling a risk factor. The goal here is to talk about how to collect more accurate information on particular, when somebody dies and where you would find, what phase we might be able to identify those risk factors, or characteristics of the individuals. 

And I understand that those are just proxy measures, which I think Dr. Blosnich made a good analogy for why it is so important. But I think the other reason why it is important to think about this is because we know that from our epi point of view, differential misclassification.

So in other words if we’re not able to identify individuals in a study, we’re going to end up with bias for those particular populations. And when you have a bias, this differential misclassification bias, typically that will result in a bias toward the null, so you won’t be able to see higher (inaudible). 

So I think that for these populations there is potentially identified high risk, but we just don’t know how high risk it is, and that is sort of, even though there are proxy measures for something else we still can’t see the sort of the magnitude of the problem with those special populations. 

So that’s why the focus is on the individual level factors for this group. I think that’s sort of a fundamental discussion here, and I think there are a lot of people in the audience, not in the audience but in the wider webinar who may be wondering why we’re focusing on this when clearly these are proxy measures for something else, but that’s the reason because we want to understand the risk for those populations. That’s sort of my soapbox bit.

JOHN BLOSNICH: In my earlier comments I was not suggesting that death investigators or folks at the scene ought to be documenting these other sorts of constructs I was talking about or these other exposures. I think that is incumbent on the research community. 

But to Dr. Warner’s point, the ability to accurately get these numbers, that you can aggregate them over time, and researchers sort of do their work in terms of area level exposures, community level exposures, and bringing in these other, the way that we operationalize and define these other constructs, to kind of marry them with the mortality data, to find that information and to look at those associations, I think that’s sort of where, I was going, not saying we’re going to send investigators out to get these other sorts of, here is a 55-item scale on these experiences, it just would not work.

JAN GORNIAK: So my question is, what I have been thinking about is so what, maybe that is the next steps, but we are gathering this information, is it just for numbers? So when I think about having investigators go out and collecting this information, I can see, and I am envisioning that is too much. This is our charge, this is what we’re going to do. 

But if I collect this information, and what? What does it really mean? Is it just a number? Is this going to be a statistic? What are we going to do with that information? So in that sense, when we’re looking at the death investigation systems, whether it’s a coroner system or a medical examiner system, we really have to get the buy-in from the leadership. Or it’s not going to trickle down. 

I know in my experience when I was in leadership, yes, here and there I’m going to do it. I’m not sure, and this is my own opinion, so please don’t write my boss, that if I go back with all this information that my cup is overflowing and I’m so excited, but if I go back, so what? Do you know what I mean? I’m just stuck. So that was my one point. 

And then my next was that we hear about the diagnoses of these children that live with depression, anxiety, ADHD. Are you looking at yes, they have the diagnosis, are we drilling down also to their treatment and their follow-up? Are they continuing, they have a diagnosis, are they taking medication or are they compliant with deployments and things like that. So are we still losing them. We’re checking to have this diagnosis, but is there that follow-up on what they’re doing with that diagnosis?

ARIELLE SHEFTALL: I think I could speak to some of that. I think the so-what piece, for me anyway, is that if we have this information and we can actually gear interventions that will actually address some of these risk factors. So that is pretty much what I think the so what is about. 

NVDRS for instance is one of the databases that I’ve used, and I can tell you the quality between one state versus another in terms of the narrative data is drastic. Sometimes you can get a narrative that has one sentence, that’s it. 

And sometimes you can get a narrative that is like two pages long, and that helps us researchers understand okay, what was going on in that child’s life, what are those risk factors that we can actually gain insight on, and then how can we actually intervene on those risk factors moving forward with interventions that can actually make a difference in decreasing suicide and suicidal behavior. So I think that is the big so-what on my end. 

And then in terms of the compliance with medication, with treatment, you can actually gain that information from NVDRS, like if they are in current treatment, if they have had past treatment, if they are on medication, but that is if it is actually coded as such. So there are again those limitations that are present within the data. That doesn’t allow us to answer those questions unfortunately. 

So if we have a child that does have ADHD, do we know that they were taking medication? Maybe. It really depends on that data that’s being collected for that specific database. So again I think it is unfortunately the data that we have can’t answer these really important questions, and we are hoping that these conversations that we are having today can help to provide some insight on the so-what and where we would like to go from here.

JOHN BLOSNICH: Is anyone here a policy maker? In FEMA, couching it in terms of the public health function that coroner examiner medical offices operate in, how much we do not support that financially, the best way to keep a population alive is to figure out how people are dying. But death is largely unpleasant. We don’t like to talk about it. it is tragic. 

But finding some way to communicate to the folks that control the purse strings that can get more funding to folks, because I think it is an absolute critical point. I am going to take this all back and say hey ya’ll, you are going to ask like 15 other domains when you have like five minutes on a scene before you have to get to your next one. Or you’re in an unstable scene where you have to (inaudible) there’s a lot of things that can happen there. 

But as researchers, as practitioners, how can we get the policy makers to this table as well, to fully understand the impacts of what is going on, what we lack, not just in research data, but what happens in practice, what do you lack, what tools you lack, what training, what personnel needs you have, and how can that be leveraged for more resources for this, to make sure that policy makers know not a long but a very steeled bullet point, I think all been told that is the best way to communicate with policy makers, is in instill the importance of this, the absolute importance of this, and to remind people of the critical role that ME and coroner offices play in thinking about when you see a risk that’s emerging in a community of how people are dying, that they are often the first people to let the community know we’ve got to mobilize around this, we’re seeing a pattern here, we’ve got to get out in front of that. And that starts with that death scene.

STACIA FRIEDMAN-HILL: I want to give Dr. Pinneri a chance to comment.

KATHRYN PINNERI: I am actually going to respond to what you were saying John, in that our funding for most offices, unless you’re a state office, like Michael’s, comes from county funds. So we are a county funded agency. So when we are tasked with obtaining data for federal partners or federal things on county funds, it makes it hard. We don’t have funding, we don’t have the money or the resources to add on more and more and more things. We are all struggling to do the work that we are supposed to do and that we are paid to do for our jurisdictions. 

And obviously, yes, there are grants out there. But we are not grant writers. We do not have grant managers in our offices. I manage one grant, and every time it pops up that I have a report due my heart just drops a little bit. I’m like, oh, crap, again? I just did one three months ago. And so that’s not the answer. 

For our offices, we don’t have the ability to do that. We don’t have the time, frankly the grants themselves have gotten shorter to write, but it is an onerous task for an overworked medical examiner’s office.

The other thing I just want to say is that a lot of our offices somewhat lose site of our public health role because we don’t have time unfortunately, we’re tasked again with the county duties of cause and manner of death for our constituents and testifying in court, and all the other things that go along with it. A reminder, isn’t a bad idea, but we just don’t have the resources to do a lot of this stuff that we want to do. It’s not that we don’t want to do it, we want to do it. 

And then the third thing is that the money that does come to states from the Federal Government doesn’t trickle down to our offices. We don’t see it. It gets stopped somewhere else. 

Now I know there are some offices that are very successful in getting it, Cook County is very successful in getting the funds to go to them, but in Texas nothing, none of the offices are getting any of the federal money that are designated towards these type of common health functions to assist with this. 

So that is kind of a struggle that we deal with all these requests, and you all are not the only ones asking for it, everyone wants our data because it is that important. But we need health and resources as john was saying, we need resources, we need somebody to come in and help us out to provide this important data.

STACIA FRIEDMAN-HILL: We have time for one or two additional questions. Dr. Warner, do you want to ask another question or make a comment?

MARGARET WARNER: I was going to make a comment similar to follow on from what Dr. Pinneri and Dr. Blosnich said, but now I am a little bit tongue tied because they made such great presentations about some of the issues. At CDC we are trying to increase our profile and help strengthen the ties with medical legal death investigation, but it is a very small effort at this point. 

We have various other activities that we have already been doing in this area for many years like the NVDRS activities and sudden and unintentional deaths, we’re strengthening those activities. 

One of the things that we are working on a lot, which I did not get a chance to mention, was data modernization. And I know we have six minutes before lunch, so I don’t want to talk too long about that, but that might be something that could help some of the specifics on at least when an investigator is reporting finding information they can report that. That is at a very nascent stage of (inaudible) possible again, resources being 100 percent here.

KATHRYN PINNERI: I think on a grand scale the fact that people who die can help people who live gets lost. And they think I don’t want to find a bunch of dead people, I’ve got all these living people to take care of. And then they don’t realize that that comes through those who have died who are helping those who are living. 

And I think that that message just gets stuck and lost. And I think someone could be more important maybe than paying someone to collect this data, but ultimately it becomes full circle, we are all part of this circle of life, whatever you want to call it, but the information we give helps those who live, and I don’t think that message gets portrayed.

STACIA FRIEDMAN-HILL: I really wanted to close out this session on a positive note. And so I think that is a lead-in to something I want to say, which is we have been focusing on the death statistics. But I think Dr. Blosnich, made a really important point when you started your presentation, about this data can show us trends in order to inform what we do with the data in terms of policies or going forward with research questions. 

And several of the presentations during the session included discussions of prevention. We are talking a lot about the death statistics, but really what we want to take from that is focusing on populations that are at increased risk and understanding why, and more importantly understanding that trajectory from ideation attempts and preventing the trajectory progressing to the point where we’re talking about suicide deaths. So I think that we can highlight where the need is the greatest, and then that can inform our research and prevention and changing the statistics in the long run.

So I wanted to thank everybody for this session. And I’m looking forward to this afternoon. And I’m going to hand it back to Brendan I think to take us into lunch.

(Luncheon Break)

Afternoon Session

BRENDAN WEINTRAUB: Hi everybody. We are going to do a quick little Mentimeter poll. So let me go ahead and share the screen here. So the question that we have today is, what other child or youth special populations need further consideration. So you’re just going to go to www.MentiMeter.com  and use the code 31733. 

This morning we heard excellent presentations on LGBT youth, suicide and minority children and youth, those with neurodevelopmental disorders, those in child services and welfare systems, and juvenile justice. So we have heard a lot of great information on special youth populations, but just trying to see what else we could consider. 

We have trans youth, folks that are disabled, children living in poverty, rural youth, chronic health. Children and veterans. It looks like the two that are popping up right now are rural youth and trans youth. Definitely something that we should consider a bit further. We’re going to leave this open for a few more minutes. Feel free to continue adding as you are thinking about it. But for now I am going to pass it over to Diane Pilkey just to bring us into our next session. Thank you. And here we go.

Agenda Item: State-Based Approaches to Child/Youth Suicide Investigations and Suicide Prevention Programming

DIANE PILKEY: Hi everybody. My name is Diane Pilkey. I work at HRSA, the Maternal and Child Health Bureau, and I’m the Project Officer for the National Center for Fatality Review and Prevention. And I’m very excited to just briefly introduce this next session. We have had some great information on suicide data and risk factors, special populations, and now we’re going to have some states come up and provide some examples of their approaches to doing child and youth suicide investigations and suicide prevention programming.

So the speakers coming up will be Dr. Michael Staley, Utah Office of Medical Examiners. Dr. Holly Wilcox, Johns Hopkins University. Dr. Angie Hayes, the Harris County Fatality Review Team in Texas. Ms. Kelly Dougherty, the Colorado Department of Public Health and Environment. And I’m going to turn it over to Dr. Staley.

MICHAEL STALEY: Hello everybody. Welcome back from lunch. I am excited to talk about a project that we have been working on for some time at the Utah Office of the Medical Examiner. And I first want to acknowledge my longtime mentor, Dr. Doug Grey, who was at the University of Utah for many years and recently retired. 

And this should really be titled the Utah Youth Suicide Research Study 2.0 because of the work that he did in the early 2000s and late ‘90s. And incidentally we’re finding a lot of the same things, which is robust, but also kind of like, shoot, 20 years later, right?

So, in 2017 the Utah Department of Health, now the Department of Health and Human Services, saw that we had this hugely rapidly increasing rate of suicide among 10- to 17-year-olds. And they called our friends at CDC, who said we need help with this. 

And they got an Epi-Aid to come to Utah. And they looked at various different data sources, and we had about 141 percent increase in teen suicide between 2000 and 2015. Suicide across the nation of course in the same period of time is increasing, but our trends just were exponentially increasing even faster than those in our neighboring states and across the country.

So using NVDRS, OME death scene investigation narratives, and morbidity data sources, these authors were able to describe youth who died by suicide. But they also acknowledge that there were significant and critical gaps in knowledge that remain. 

And I have to credit this study with really getting me a job. Because the legislature at this point in time said well we are now supposed to make decisions on data from this EPI AID that ended in 2015, our data is two years old, and we want something more real-time to do this. 

So they created my position, a psychological autopsy examiner position, and they said you’re going to do psychological autopsies on 70 percent of people in Utah who died by suicide, which if you did the math I would have to have a phone on each ear and never leave the building 365 days a year to meet that metric.

So I thought critically about this and I said, hey, I want to be able to call everybody who loses somebody to suicide in Utah. I talked a little about that this morning. But this is part two, because we really wanted that in-depth information that is validated as a can-be of collateral information about significant groups. 

And I’m wondering okay, where do we go first, do we go to veterans, do we go to LGBTQ people, American Indian people. And then standing behind then Governor Herbert, he said and Dr. Staley will be studying youth. Okay, so we’re going to study youth.

So there are some big opportunities here, and funding wasn’t really one of them. I described this as a bubblegum and dental floss operation. We have a great relationship with the University of Utah School of Medicine Department of Psychiatry, and I had third- and fourth-year medical students and first- and second-year residents at my disposal to help me do these interviews. We would be at the OME talking to families sometimes until 1:00-2:00 in the morning Tuesday evenings, and we did this, we actually just finished, we had a few families that wanted to wait a while, and we just did our last interview for this study last week. So we really tried to be as inclusive as possible.

So we had significant public support on this issue. Suicide among youth, if you google that in Utah, even though youth represent about six percent of our overall deaths every year, they represent about 40-50 percent of all of the articles written about suicide in Utah. So there is a lot of buy-in here, and people are really wanting to know this information and engage in this kind of prevention work.

And significantly we also have legislative champions who understood the need to not just act immediately, that’s the charge of public health, but also improve the state’s ability to measure and target future prevention and intervention work. 

And I would say that postvention is something that was still kind of new to legislators, and we have to sell them on that piece, but I really think we’ve made a lot of headway in that domain in the last couple of years, by just talking about doing this work.

So they created my position, and they said you’re going to be at the Office of the Medical Examiner, because we want you to have the proximity to the deaths, and we also want you to have the authority of the medical examiner to get the information that you need. And I already talked about that goal, which thank goodness that’s not in statute. 

And so we have this two-part program about data aggregation and psych autopsy, it’s what I’m talking about now, and then the program I talked about earlier. So Utah youth suicide research project is a two-phase project. And the first phase is to de-silo and aggregate the data that we already have. And if you think about that, that is a lot of data. 

So who are we going to do this on? We focused on all ten- to 17-year-olds, plus 18-year-olds who graduated high school within three months of death who died by suicide between 2016 and 2019. And we excluded non-resident youth. Utah has two trauma level one centers. I grew up in Montana. If you break your leg in Montana, you get flown to Utah. 

And so a lot of people with significant injuries are brought to Utah, and unfortunately many of them succumb to their injuries, and then that death is investigated by the Utah Office of the Medical Examiner. And we excluded those deaths. So there in total were 201 teens that met this case definition in our study period. 

Going back to that phase one de-silo and aggregating project. So there are several places to get this data. And this is even with state statute that says the medical examiner may request any record in writing and receive it under penalty of a class D misdemeanor, et cetera, it is not that easy still. 

So we went to the Utah State Board of Education, USBE, and we got their data. We got everything but disciplinary and I believe attendance data, and we are still working through how to analyze that. It turns out that is really complex data to work through. 

We are also working with DCFS and Child Protective Services, Juvenile Justice, the Utah Criminal Justice Information Database, UCJIS, and then there was also looking at healthcare data that is already coded. 

There is obviously some limitations to that, the Health Facilities Database and the All-Payers’ Claims Database, both of which are managed by the Utah Department of Health and Human Services. So we’ve got a good, this is an interdepartmental transfer. So hopefully we can aggregate all of these things. 

And surprisingly this is taking longer than actually contacting 201 families and then friends for all these people and conducting these interviews. I didn’t anticipate how heavy of a lift this would be in terms of the data sharing parts of this. I am also not an attorney so I just sort of break it and beg for forgiveness and try and try again.

So I am going to talk mostly today about he second phase, which is the psychological autopsy phase. So we attempted to interview at least one parent or guardian and as many as three friends for every youth who died by suicide. 

And we used a snowball approach. So we would ask mom or dad who was your son or daughter’s best friend, or two or three of their friends. And sometimes we would get a list, and sometimes we might get one, and it is actually a finding where they might say they didn’t have friends. Heartbreaking to hear, but we did encounter that frequently. 

And then we asked those friends also, who was Johnny hanging out with near the time that he died? Those were oftentimes different people than the list of people that mom and dad gave to us, so we would reach out to them, and through the sort of process you start to get different parts of the story. But also to validate the story. Because we’re asking the same question to all of the people that we interview. That’s how you validate collateral information post-mortem.

So of this 201 teens that fit our case definition, 34 cases had at least three interviews. 78 cases we were able to complete one or two interviews. 30 next of kin refused to participate. And 59 did not respond to our invitations to participate.

And every time I do a news story on this study I’ll get a call from a parent that says I never got your letter, but I lost my son or my daughter in 2018, can I participate? I don’t want to say there is this passive rejection of the study with these 59 people, but I actually believe there are some people in this group who moved after they lost their child, who changed their phone number, and we just have never been able to find them. We do have access to LexisNexis as of last week though. I wish I had that much sooner.

So I am going to talk about some preliminary results from the study. And I am going to focus today specifically on the information that we learned just from parents, because that is the data that we have had the most time to spend with.

And I am breaking these down into some major domains. The first of those domains is community, family, and connectedness. Some of this by the way really agrees with what we’ve found in NVDRS. Others build upon it, some of it adds to data completeness. And others is unique on a lot of its own. So roughly one in four youth who died by suicide were in a romantic relationship that was on the verge of ending or had just ended. 

Most parents indicated their teen was wiling to talk to them and/or their friends when they needed to talk, but nearly half of all youth who died by suicide were reported to feel that they did not belong in their community. And to be honest I am not exactly too sure just what that means. What does that translate to, especially when they’re saying yes, they had great friends, they had a great friend group, but they felt like they didn’t belong in their community. 

Half of youth who died by suicide were bullied. We obviously know that that was a big contributor, and over half of their friends witnessed their friends being bullied. And I think that is a really unique addition to the bullying story, is that witnessing your friends being bullied for the way they dress or their sexual orientation or something else is basically as good as being bullied itself in many ways, because it is sort of a social control that says I can’t be fully myself in this way. Obviously the direct impacts of bullying can be much worse than that, but it does have this repressive effect.

I think this next bullet point, and I added it just because I get asked about this all the time, about online bullying interactions and cyber bullying. According to parents one in four of these teens was bullied online. And I think we are probably going to be able to add to this from the information on peers, because they were in that online environment more with their peers than their parents were. But keep in mind a lot of parents did have the opportunity to go through their kids’ cellphone after they died.

Physical, emotional, and sexual abuse. We knew that these would tell a part of the story going into this, and we expanded all of these questions beyond ACEs to the early traumatic inventory ETI shortform, we adapted that for collateral data use. These indicators came through way stronger than I even could have anticipated. 

And these, by the way, not sexual orientation, these are the hardest conversations, about the abuse that these kids suffered, and their parents’ powerlessness to stop it, or they didn’t know that it happened until after the death occurred, or whatever the case might be. The tragedy has so many layers.

Again, more on community, family, and connectedness. Just under half of youth who died by suicide were reported to be impulsive decision makers. We know that youth are impulsive, but here is some actual empirical evidence of that. And they could give us examples of impulsive decisions their kids were making in the class.

Another thing that we have talked about is suicide contagion and diffusion in this meeting, and again here is some more empirical proof. One in five of these teens who died by suicide experienced the loss of a friend, and one in ten lost a best friend. A third lost a family member to suicide at some point in their lifetime, and two out of five knew someone at their school, church, or neighborhood who died by suicide.

I am really interested in this idea of mobility. There are of course these developmentally mobile youth who are going from elementary school to middle school to high school. That is normal, we would expect that, so we took those out of our analysis and looked at other kinds of school changes. Over half of these youth had one school change at least where they were going from a public school to a charter school and then maybe back to a public school. 

Which tells me, and we can verify this in some of our education data, things weren’t going well at the public school, so then they went to a charter school, but then they lost the services that the public school had, and then they had to go back to a different public school. And I don’t know about you, but if I had to change friend groups three times in the course of four months, I would also have some connectedness issues, to say the least.

Nearly a third of the youth who died by suicide had a record of frequent truancy or extensive school absences. I went to graduate school in New Hampshire, and some of my pals were truancy officers on the side. I don’t know how you get that gig, but you know like you missed three days of school, you get a knock on the door, your kid missed school for three days. In Utah that doesn’t happen. 

And I heard from many of these parents and their friends that their son or daughter would go to school, they would check in at first period, and then they were out, and they were in the parking lot or they were off doing something else. So this comes again from parents telling us the story, not from the actual school data. But we know some of these youth at least were really disconnected from some of the supports that we would expect.

Nearly a third of the youth who died by suicide had disciplinary actions taken against them at school. And most of those were really proximal to the time of death. I have given a similar presentation to educators, and my advice to them was make sure you discipline the action and not the youth. You’re not a bad person, but you did a bad thing, and that’s why you have to be suspended for three days. There are some good adjustments that we can make in this field.

In healthcare, I work a lot with healthcare partners, nine in ten youth who died by suicide saw a healthcare provider in the year prior to their death. And a third of them saw a healthcare provider within two weeks. What a tremendous opportunity to engage youth. 

Over half who died by suicide were seeing a mental health provider at the time of their death. Over a third of the youth who died by suicide attempted suicide once or more before the time they died, of course we knew that’s a strong predictor. But here is the thing that we added is that a third never received any behavioral health treatment following an attempt.

Firearms are a huge part of our suicide prevention strategy, and in the cases we studied, two thirds of firearms, in suicide deaths involving a firearm, which is about half of these deaths, two thirds of those deaths the firearm was stored unlocked. 

And in suicide deaths that did not involve a firearm, which is roughly the other half, two thirds reported that the child did not have access to a firearm, either one was not present, or it was locked and the child had no access. Yes, putting time and distance between kids and firearms is a meaningful strategy.

So this is the end here. What is next, where do we want to go? Well, first we’ve got to finish this study. The second is there is this little pandemic that happened between the end of where these teens died by suicide and now, and we know that has really changed the landscape for how this looks. So I am asking my partners in the legislature for a generation 2.0 of this study. I would hope to make it a little more bubblegum, a little more dental floss this time around, but we will see what happens.

I also want to say that we are making this data available to our suicide child fatality review committees. I am deeply involved in those in Utah. And how does this really in death information which we’re getting quite a bit after the fact inform those more timely decisions. 

We are also making sure the NVDRS and where applicable SUDORS gets this information. And we’re creating a model like HOPE that other jurisdictions and states, especially those that have access to or work in partnerships with medical schools can take this model and duplicate it for their own use. And I am going to introduce Dr. Holly Wilcox next.

HOLLY WILCOX: Hello everyone. Thank you for inviting me NIMH and HRSA folks, I appreciate it. I wear a few different hats, and today I’m here to speak in my hat as co-chair of the Suicide Commission in Maryland, and also a member of the schoolboard, but I’m also a professor at Hopkins and have spent my whole career trying to think about issues of youth suicide prevention.

So, Maryland at a glance. Maryland has been called American in miniature because of its diversity. We have beaches on the eastern shore that has a feel like Delaware, and then we have sort of mountains in the west of the state. 

So it is very diverse in terms of its topography, geography, demographics. The population is about six million. We have a centralized Office of the Chief Medical Examiner. We have had sort of a revolving door through that office of some of the key central staff, including the Chief Medical Examiner. 

And we have one of the lowest suicide rates in the United States, but we have a very high proportion of deaths that are classified as undetermined. And it has been that way for a while, and it is just a manner of taking a more conservative approach to this issue.

So I wanted to show you some data right out of the gate. The bar that is on the top, the orange-ish bar is the total number of suicides in Maryland. One major advantage to being able to work with the Office of the Chief Medical Examiner is you can get data in almost real-time. We’re not doing the two- or three-year lag like with some of the data that we have been talking about in this meeting. 

So here you can see on the top that is what our trend looks like in Maryland. And on the bottom, the blue line is youth. But when you dig a little bit deeper into these data, what we have been seeing over time is whereas youth suicide was about three percent of the total suicides in Maryland, now it is a little over six percent. And so we are seeing an increase unfortunately that looks like it could align with the pandemic.

The other thing to note, although our suicide rate is one of the lowest, we do notice major issues in terms of our youth risk behavior surveillance data. So if we look at those data in terms of suicide ideation and attempt, here you can see some data looking over time, these are high school students reporting serious thoughts of suicide over the past 12 months, and you can see what the data look like, and trending upward, in the wrong direction, but not for planning in high school.

And this is our middle school data. This is lifetime ideation also. We are seeing really steep increases in these data. The same thing for suicide attempt, I don’t have enough time to show you everything, but I have them in the back of my presentation if we want to get to this later.

So what are we doing about this issue? We have decided to take somewhat of a cohesive approach to this, thinking about working with the Office of the Chief Medical examiner, our Suicide Prevention Office and other folks in our health department, and thinking what we can do to make progress.

One thing that we have done is, with the support of the NIMH, we have been able to build the Maryland Suicide data Warehouse. We’ve been working on a shoestring budget to try to do some psychological autopsies. This is not funded by any source, we just cobbled some funding together and medical trainees and residents have been helping do a lot of the psychological autopsy interviews. 

We recently have legislation to establish a suicide fatality review, which is starting very soon. And then we have been thinking about interventions. And of course for us firearms are top of mind, because they account for in Maryland almost half of all suicides are by firearms. So we have a lot of concentrated efforts in that area. And of course, we are thinking about how to reach young people at risk. And one great place to reach them is at school.

So here is our Maryland Suicide Data Warehouse in a nutshell. This figure just shows you what it includes. But it is exactly like Utah, it is thinking about the data silos and how do we de-silo the silos, and how do we aggregate those data together. And so Hadi Kharrazi is an informaticist, and he is a data person, and he knows how to do this data linkage and integration, harmonization, and predictive analytics.

And so you can see the grant numbers that are in the bottom of this slide, and that is what we have been doing, we have been trying to do this. And the way we do it is we help with our health information exchange, which for us in the mid-Atlantic it’s CRISP, Chesapeake Regional Information System for our Patients. 

And what CRISP does is it’s a HIPAA protected agency, it takes all these data, it puts them together for each person, there is a line for each person, and all the data are loaded in across these various sources. So here we have Medicaid and Medicare data, we have VA healthcare system data that is integrated in, which was a lot of work to get in place. 

Hospitals, healthcare data is in, so electronic healthcare records. Medical claims data. Commercial claims, and public insurance. Census derived geodata are part of this. And of course central to this is a partnership with the Office of the Chief Medical Examiner, because with them we have the manner of death, we have the narratives, which are essential to us, and we have toxicology results and so forth.

And so in terms of these data, a lot of it is coded data, some of it is unstructured data, so there are the discharge summaries where a lot of the meat is in terms of risk factors and protective factors. Social work notes, discharge summaries, and so forth. So we have all that, and this is being used for predictive analytics. 

But the other idea is that these data are going to inform the suicide fatality review process, and the idea is to get these data together and to do a warm handover of the data to the health department, so the health department and the suicide prevention office will manage these data and will be hopefully refreshing it, and it will be used for action, which is the purpose of what we’re hooping this is going to achieve for us. So this gives you, I know I don’t have a ton of time, but you can get an idea when we put these data together what our numbers look like in terms of the suicide deaths, and by race/ethnicity, and by the manner of death. 

One finding that we’ve been seeing, this is a recurring theme in this meeting, is looking at these disparities. In Maryland one thing we’re seeing, when we look at the adults on this left side of the figure, we kind of see one pattern of data. But when we look at the young people what we’re seeing is the percentage of the suicides for white people is declining, for white young people, and for black young people we see a steep increase. 

And so this is in front of mind for all of us in thinking about how we can do a better job of identifying people more upstream and so forth. So that is part of what we have been thinking about. And one strategy to try to have a plan in place is the psychological autopsy interviews that we’re doing. We are not as advanced as they are in Utah. 

And these have started, and really we are using these as a method for understanding suicide, kind of a short-term risk factor. Because from the data we have in hand with the warehouse we have an idea of the long-term factors, but the proximal factors are what’s missing for us, what happened in the minutes, days, hours before the death, and we want to be able to talk with the families and learn from them and understand what was going on.

And so we have Dr. Lanny Berman who is part of our team, and he used to be the Executive Officer of the American Association for Suicidology, and did a lot of the psychological autopsy training. And so he has trained our whole team in Maryland, and he is available to train other teams I’m sure out there in other states. 

And so we work with the Office of the Chief Medical Examiners to have them tell us what their procedures are for engaging families. And this was a long IRB process and a long process of sorting out what the transitions of folks in the Office of the Chief Medical Examiner what our process would be for engaging families.

So the OCME sends out letters to families who suffered a suicide loss. And then, if they agreed to be contacted, we at Hopkins can go ahead and contact the families and ask them if they would like to participate in interviews. And so that is how we’ve identified our families. 

Right now we are focusing on young people who have died from firearm suicide, but we have a parallel study that is of people who had an opioid overdose with unclear intent. And that’s mostly adults.

And so, some of the findings, this is very new for us. Of the first eleven cases we’re seeing these social interactions, the same as Utah, the social interaction themes, interpersonal conflicts, social isolation, and also some conversations that young people were having about suicide with their siblings and parents.

Strange video game, alarming video game behaviors, where kids are in these dark rooms and playing video games on the day of their death or the week of their deaths, playing a lot of violent video games and actually jumping in the line of fire in the game and killing themselves, which came up two or three times interestingly enough in these interviews.

Some other things. Of course, firearm is a major theme, and there were some questions that were directed at the families. The families for the most part did not consider the gun in their home to be a risk. And sometimes mothers didn’t even know their child had a firearm, it was a gift from the father that was given to the child. 

And so this issue of psychoeducation and lethal means counseling was something that posed to the family, like is this something that would be receptive to or is it something that you received. The general consensus is that we’re not getting information about lethal means access and firearm safety. And they seemed open to it. And also, if they had known that information they might have taken some action to secure their firearms.

We do have some exciting news in Maryland. We have a new governor, Wes Moore, who is a Democrat. And this picture I love, because in this picture, this is when he signed the bill to end child poverty in Maryland. So the Fair Wage Act of 2023 and the Family Prosperity Act was signed, and I just like this picture very much.

We also know from the CDC’s technical package that poverty and economic factors are so central. So I also wonder if there will be an impact of this legislation on outcomes for young people.

We have been working really hard. I feel like I never learned in graduate school how to intervene with policy makers, how to write testimony, or any of this. But I am spending more and more and more time doing that, is providing oral testimony or written testimony for specific legislation. And so our whole team, including Paul Nestadt and others, really we have been focused on this as our mission. 

So the Extreme Risk Protection Order, I provided testimony for that, and it passed back in 2018. So we’re one of the close to I guess maybe half the states now, maybe fewer than that have ERPO red flag legislation. The trouble is how to implement it. 

In Maryland, family members can file an ERPO. Family members, law enforcement, and health professionals. So that’s different from a lot of other states, health professionals can’t file. And so most health professionals, clinicians, can’t take an entire day off to go sit in front of a judge in court to get this ERPO through, to get the court order to implement the ERPO. 

And so we have been trying to think creatively, mostly Shannon Frattaroli and Paul Nestadt, my colleagues, have been thinking about how to implement this so that it can be effective and it can be used as a solid tool in Maryland.

We also have legislation that is in the education field in Maryland that is very exciting, the Blueprint for Maryland’s Future, which is a huge allocation of funding in Maryland is going to improve the educational system. 

We also had a closure, there was a long gun loophole that firearms for the most part, handguns were very well regulated in Maryland, and long guns were not at all. So we were able to, Paul Nestadt and others were able to close that long run loophole, including the age at which people could purchase a long gun, and many other factors related to that. 

So we have the fatality review, we have the people that have been appointed to that, we are going to start relatively soon. We were able to get a suicide prevention coordinator position in Baltimore City, right now it is unfunded, as our fatality review was unfunded, and then we had to rally the troops to get the funding for these things.

And we had suicide loss survivors, including some representatives and delegates in our state who are suicide loss survivors, who really came forward and were like we’ve got to get this money. And the money for the suicide fatality review is for the data preparation and the coordination of it. So you need a person who is going to be able to coordinate those meetings and keep the ball rolling and everything.

And then, very exciting to me, is Jaelynn’s law that just went through this legislative session, which is taking the child access prevention, but most of child access prevention thinks about unintentional injuries, they don’t think about suicide, so the age for that legislation stops at 12 or 13, and this filled the gap up to age 17. It’s now 17 and under, there are penalties for parents when their child uses a firearm, when they can access a gun. So it is firearm safety-oriented legislation.

And of course, we like to see this slide, this pattern of okay the legislation is going all the way forward to the governor. And this one, I saw this happen for the suicide fatality review, a bunch of us were jumping up and clapping to see this happen. 

And in Maryland of course we have pretty strict firearm legislation by Gifford’s Law Center, we get a grade of an A in terms of our scorecard. So we have pretty stiff firearm penalties in laws. But the other thing to notice is the access is just out of control. When you look at our data, this is just Maryland data, and you look at the Nix, which is the FBI database on firearm sales, you can see what it looked like before the pandemic with the kind of lines along the bottom of this slide. And then you can see the more recent data in the red, that’s this year. And then the yellow bar is 2022.

So you can see that the firearm sales are way up. And this only counts one. So somebody can go in and buy 10 guns, and it is only counted as one. So this is a very big issue for all of us in the field is to be thinking about the access of firearms out there right now.

One thing that we have done is we have a Safe Gun Storage Map, Cass Crifasi and folks from our Center for Gun Policy Solutions, put together this map with students who were in our School of Public Health, trainees, got together as a project, and they created this safe storage map which is highly practical. 

So anyone in Maryland, if you’re a grandparent and you have your grandkids coming in, your son says I want all the firearms out of your household, my kids are staying with you, they need to be out. Or if you’re at risk for suicide or have a loved one at risk for suicide there are sites that agree to be safe storage sites. 

And you can go online and you can click on these little buttons, and you can see where these sites are located, and you can go ahead over and store your firearms rather than having them taken away by ERPO, this could be the first option. For somebody through their own volition to go ahead and take their firearms. I don’t have time to go over all this right now, but we can get to this maybe another time, or if folks are interested you can reach out to us. 

But we’ve been trying to think through this idea of how to implement the red flag laws, and this gun violence navigator with the staff who are hired, in this case our Mayor’s Office in Baltimore provided funding for this trial, two fulltime social workers who will be located in our Johns Hopkins Healthcare System, and they will be the people who will be interfacing with families who are treated in our healthcare system who have firearms. 

And oftentimes in a safety plan we can find out somebody has a firearm in their home, and they can be the person who goes in and has the conversation, and they can help with all of the next steps if somebody is at risk for suicide and has access to firearms in their home. 

And we have this great Maryland Consortium on Coordinated Community Supports, which is part of the Blueprint for Maryland’s Future, and we are thinking through various ways to tap into this legislation for suicide prevention and youth. And I just wanted to thank several people on this slide. And I left off Chris Kitchen because I couldn’t find a picture of him. Anyway, thank you to our extended team, and thank you all for your attention.

Staff: Next up we will have Angie Hayes come present.

ANGIE HAYES: Alright. Hi y’all. I am Angie Hayes, I am both a researcher and the Chair of the Harris County Child Fatality Review Team. I have no conflicts to disclose, but if you happen to be a dragon with a horde of cash I have research projects I would love funding for.

And as you know, our entire workshop has been around the topic of suicide. But however in this presentation I will be discussing one particular death in a little more detail. And that can be a little bit more emotionally harder than talking about suicide from a 10,000-foot view. 

So if you need to step out, turn off the volume, take a break, please feel free to do that, I’ll give you a heads-up when we get to that slide. And as you have heard before, their hotline number is 988, and you can also text HOME to 741741.

So, everything is bigger in Texas, and we love to brag about it, as you all know. Harris County is on the southeast side of Texas, shown in red here. It is the third largest county in the US, and the county seat, Houston, is the fourth largest city in the US.

To give you a better idea of the geography of Texas, that little blue dot that just bounced in is El Paso, on the west edge of Texas. El Paso is closer to San Diego, California, than it is to Houston. It is enormous. And Houston is the home of the Texas Medical Center. This picture shows you part of the Texas Medical Center. It is larger than a small city might be. In the far background you can see the skyline of downtown Houston.

Inside the Texas Medical Center, we have Texas Children’s Hospital, which is the largest children’s hospital in the US, arguably the world. And then Texas Children’s Hospital and Baylor College of Medicine host the Division of Public Health Pediatrics, which is the only one of its kind.

So nearly one out of every 50 children in the US lives in Harris County. That’s one and a quarter million children. One out of every six Texas children live in Harris County. So not only are there a lot of children, we have a large proportion of children. The US has about 22 percent children. New York City has about 20 percent children. In Harris County we are over 26 percent.

Unfortunately, that means we see over 250 pediatric non-natural fatalities in a year. Now, this term non-natural, refers to a death that might be a homicide, or an accident, or a suicide. These are not things like cancer or pneumonia or illnesses that may come about. One out of 11 of those deaths on average is a suicide. And in the last six years there have been 25 to 35 deaths a year, adolescent suicide deaths a year, and it is still trending upwards right now.

So these deaths affect different races and ethnicities differently, most notably for Harris County is that white children make up 22 percent of the child population and 42 percent of the suicide deaths that we see. To note, Hispanic is an ethnicity and not a race, but these are the data classifications we get from the ME’s office, so it is the data that we have to analyze.

So as we all know, and as we have been talking about, yesterday and today, every state and local jurisdiction has different processes and methods for ruling the manner of death. In Texas it is by county, and there are 254 counties in Texas. Of those counties, 16 of them have a medical examiner or office of forensic investigation of some sort. 

And then even within those offices, where we have more than one pathologist, we sometimes find discrepancies in rulings between them, because humans will human, and that’s the way things go. But Harris County is staffed with eight forensic pathologists, which is approximately half of what the ME’s office needs to function at full capacity for a population of our size. So that means some of our rulings take a little bit longer than we would like, especially difficult rulings like suicide or infant sleep-related deaths like SIDS.

The other 238 counties in Texas rely on their justices of the peace. A justice of the peace is elected, they are not medically related in any way, they are a legal entity. And they are not required to have inquest training in any way. Training is available through the JP training courses, but it is not required. 

So that sounds kind of a lot, 238 counties with JPs who rule deaths. But there is more. There are 817 JPs in those 238 counties who also sometimes rely on the county judge to help them out when the JP can’t make it there. 

And you may have heard this process publicized when Supreme Court justice Antonine Scalia died in 2016, the JP couldn’t make it to certify his death, and they called the county judge, and lots of fun phone calls about fundraisers got publicized and things. But a lot of people got a peek into the Texas process there.

Now, within Harris County there are some things that I would like to talk about. For research we would absolutely love to have toxicology screens on all of our suicide deaths, because we would love to know more about what substances are influencing and affecting these children before their death. Alcohol, opioids, benzodiazepines, that sort of thing. But capacity and funding, as we have heard over and over, is a restricting factor that keeps us from gathering that information.

Next there are a couple of rules that I just find interesting I wanted to bring up. One is that in Harris County if an adolescent dies playing Russian roulette it’s ruled a suicide, the logic being that they’re old enough to understand the consequences of their action, and this is what happens when you play that game. 

However, a motor vehicle collision with excess speed, for instance a teenager driving 100 miles an hour down the freeway without an explicit suicide note does not have the same ruling, it’s considered an accident. All the counties are different. This is just something Harris County does that I think is interesting.

So, a Harris County Child Fatality Review Team is made up of about 40 organizations, including experts in public health, medical examiners, child protective services, mental health providers, law enforcement. We even have community representatives and people who work directly with disadvantaged populations, young mothers and that sort of thing, who come in and work with us. We are in the process of reaching out to our public defender’s office looking for a broader perspective and greater representation for our kids.

Now, I told you, everything is bigger in Texas. Unfortunately, that includes some of the barriers that we face. So Texas does not prioritize healthcare access. Double down, they really don’t prioritize mental healthcare access. Mental Healthcare America ranked Texas number 51. It doesn’t get any lower, there are 51 to be ranked in mental healthcare access.

One of the barriers there is the sprawling nature of our state and of our county. It can take more than two hours on a decent, sun shiny day to get from one side of the county to the other, and there is an absolute dearth of public transit. There is just no way to get from here to there without having your own car or paying an obscene amount for an uber. So it makes it really hard to access even when there are resources becoming available. 

Next is the implementation of prevention strategies and response protocols, whether they be evidence based or innovative. We ran into an issue with our population size. There are so many people, this is a great way to make a big impact. 

But Harris County houses 25 public school districts, and 388 private schools. Houston Independent School District has 187,000 students in 39 middle schools and 37 high schools. And to get into any one of those schools you have to talk to the district and get permission, and after the district gives you permission you go to the school and ask the principal if you can work in the school. 

And then you have to find changemakers and people who will buy into your project within the school to work with you. So that in and of itself creates a barrier to entry, even if you could get in, or when you can occasionally get in, in Houston ISD the children speak over 100 different languages, and 80 percent of them are economically disadvantaged.

So, as is true across the country, what we have been talking about here, we run into a lot, is silos. We do not know the work other people are doing in other industries, in other areas, where we could collaborate and create wonderful symbiotic relationships and improve things. We also don’t know when there are duplications of work. Someone else is already doing almost the same thing I’m trying to implement here, it makes it really hard because there is so much going on and it is so widely spread.

Another opportunity for improvement in Harris County would be expanding our trauma informed forensic interviews. We’ve heard about how valuable those are, and how important those are, and the more funding we can get for that the better it would be. 

But then also we would love to see an improvement in our ability to offer healthcare resources and access to the bereaved families when we go and talk to them and work with them. Those are resources that are not currently terribly available.

And then last for this list for now, I am going to mention natural disasters. And natural disasters are not unique to Harris County. But it is an issue that we have faced, and it is something that we continue to face.

So I am going to tell you a little about hurricanes. For those who did not grow up in hurricane land, anywhere along the coast where hurricanes may come at you, I’m going to give you a little personal anecdote here. 

At about six or seven years old I had what is called a go bag packed in my room, I mean at six it was my favorite teddy bear and my favorite blankie. But as I got older the go bag was updated with hurricane warnings and that sort of thing, and my parents did what they could to emotionally prepare me for there might be some really bad weather, and we might leave, and you’ll never see your house again. 

And for a seven-year-old to attempt to be emotionally prepared for that kind of thing, it affects the way that you develop and grow and think about the world, the security that you feel or don’t feel.

So now that I have given you a little bit of a picture what it is like to be a kid growing up with these hurricanes, I am sure you all remember, everyone in Houston remembers hurricane Harvey. So Harvey came at us August 24th through the 27th of 2017. There were entire swathes of the city flooded. And it wouldn’t be Houston without seeing cattle herded through a neighborhood in flooded streets. There were 17,000 people rescued from their homes. 

George R Brown Convention Center, one of the largest convention centers in the nation, was turned into a shelter, and was at capacity. Every shelter was at capacity. And then some of the shelters flooded. There were 42,000 people in shelters, 336,000 people without electricity. This led to 185,000 damaged homes, 17,000 with major damage, and over 9,000 completely destroyed.

So this is where I’ll be talking about a suicide death in a little more depth, more specific. And if you need to step away, take a break, turn off your volume, now is a great time to do that. Remember your hotline number is 988, or you can text HOME to 741741. 

So after Harvey, tragedy followed tragedy. There was a young mother who had been in an abusive relationship. Just before Harvey she was able to get out with her son. Then Harvey hit. They lost their home, and the mother had no choice but to send the son back to live with his father. 

I have presented this many times, and I have yet to get all the way through it without choking up. I am sorry. He felt very trapped. And he died by suicide. And he left us some of these details in a note that he left behind. So in all of the ways that we think about preventing suicide, housing resources during a natural disaster and after a natural disaster can help families in unimaginable ways.

Now, Harris County struggles to support their population in many ways. But there are some ways we do okay, and some places we are improving. To start, our Houston Area Suicide Prevention Coalition recently has been revitalized by a grant from the Episcopal Health Foundation that has allowed us to start bringing together different organizations across the city and helping them to build symbiotic relationships.

We have also implemented new training programs for employees at Texas Children’s Hospital and expanded the number of beds available in the ER for patients coming in struggling with mental health.

Something that has already made a big difference in Harris County over the last five years is gaining funding for our child fatality review team housed inside of a research medical hospital, which allows us to blend research implementation and review all together.

This led to a recent grant where I helped develop a natural language processing algorithm that has helped us to analyze the stories behind the youth suicides that we have. Some of the interesting pieces that have popped up are the very high prevalence of suicide following an argument with a parent who took away their cell phone. 

And then, the less surprising, the high prevalence of prior CPS involvement before the death of an adolescent. And then also LGBT youth. And we don’t know all of them, but even the ones that we do know we see a high prevalence.

And then lastly for now, in a shocking turn of events there is legislation from Texas that is worth modeling and has done some good things. Our state legislation that permits county child fatality teams has specific wording permitting organizations to share data and information about cases and meetings without needing an MOU or DUA. So it breaks down some of those walls, and it makes going together and getting through these reviews a little bit easier.

So, this is my contact information. This is an updated email address from the bio that I have on the website, if you want to snap that. If you have further questions, ideas for your report, collaboration, or if you are the aforementioned dragon with a horde of money, please contact me here. And now I will pass it over to Kelly Dougherty to talk to you about Colorado.

KELLY DOUGHERTY: Hello everyone. Thank you for inviting me to talk about child and youth suicide prevention in the context of Colorado’s death review system. My name is Kelly Dougherty, I use she/her pronouns, and I’m the injury prevention coordinator for the Colorado Child Fatality Prevention System at the Colorado Department of Public Health and Environment.

And I am going to start today by giving you all an overview of the CFPS and how we conduct child death reviews in our state, as well as a few statewide statistics related to child and youth suicide.

So, in Colorado we have been conducting public health child death reviews to some extent since the 1980s. However, our reviews were able to become much more thorough and comprehensive in 2013, with the passage of the Colorado Child Fatality Prevention Act. And this legislation gave our system dedicated yearly funding to support local death reviews that are typically facilitated through the local public health agency. Our system is statewide, multi-disciplinary and multi-agency effort to prevent child deaths. 

And we have funding, although of course we view it as limited funding, but just really thankful because I know in many states they’re doing death review work without funding, so we’re fortunate in that respect in Colorado. We have 43 local review teams that represent all 64 counties. And these teams identify risk and protective factors and opportunities for prevention, which inform prevention at both the local level and statewide level.

Our system specifically reviews deaths of children and youths under the age of 18 who die by injury, violence, and undetermined causes, examples of which are listed here. These deaths typically represent about 35 to 40 percent of child deaths in our state each year. So the deaths we don’t review are typically due to things like congenital malformations, perinatal conditions, disease related deaths. 

And our system consists of our local teams, governor appointed state review team, and then the small support team at our public health department, which that’s where I work. 

And I am going to share some statewide statistics on child and youth suicide. But before I do that, I always want to acknowledge that these statistics represent real children who needlessly lost their lives, and their families and friends are going to forever be grieving their deaths. So the statistics from the science writer Paul Brodeur, that statistics are human beings with the tears wiped away, I try to keep that in mind when thinking at the statistical level to kind of counter that inherent dehumanization.

And then, in addition, it is really important to always keep equity front and center in our child fatality prevention work. We observed disparities for almost every cause of death that we could review by race and ethnicity, place of residence, urban versus rural, sexual orientation and gender identity among other disparities. 

So when we’re leading this work across the state, we always try to remind our partners about the social, economic, and environmental injustices, and that we can stop this structural violence by looking past the individual level to the system and policy level.

So now we’re going to look at those deaths that were reviewed by our child fatality prevention system. So this figure shows the leading causes of death occurring among those under age 18 in Colorado that were reviewed by our system for the most recent five-year period that we have data, so 2017 to 2021. 

And you can see the leading cause of death for children and youth was suicide, and then motor vehicle and transportation deaths, followed by sudden un Chris Kitchen(ph.) Chris Kitchen(ph.) expected infant deaths, child maltreatment, then firearm, homicide, unintentional poisoning, and drowning. 

The categories aren’t mutually exclusive, so the totals on this slide would sum beyond the total number of deaths reviewed. We pull out firearm deaths in particular to look at these. So this firearm category would include suicide deaths as well as homicides, child maltreatment, and accidental injury. 

And so as we saw on the last slide, suicide is a leading cause of death among all children and youth ages five to 17 in Colorado. In total 313 children and youth died by suicide in the state from 2017 to 2021, and among those 310 were Colorado residents. So we’re looking at rates based on the Colorado residents. 

So you can see the rate in Colorado has fluctuated, but the overall crude rate was 6.4 per 100,000, which was twice as high as the national rate of 3.2, which is shown by this green dotted line. So in Colorado we consistently have a high suicide rate among young people.

Now, when we look back at the means used in child and youth suicide deaths, we can see that the majority in Colorado were via asphyxia or hanging, followed by firearm, and then drug overdose or poisoning. 

But when we look at these by sex we see differences, this was national, but this is what we see in Colorado, is that the blue bar is representing males, and why asphyxia is still the leading cause of death for young males in Colorado, that is followed very closely by firearm, at 46 percent, and then very small percentage of poisoning. 

And then when we look at females, asphyxia was most common at almost 63 percent, then followed by about 20 percent of firearm, and then drug overdose or poisoning. So you can see that males in Colorado are more than twice as likely to use a firearm than females. So by analyzing this type of data we can better target our prevention strategies to specific demographics and specific risks.

We also collect data on circumstance information, including details of youth history of suicide-related behavior and personal crises. So the most common finding associated with deaths by suicide was that the young person has communicated suicidal thoughts or intentions. 

In addition, over a quarter of young people who had ever experienced the death of a peer, friend, or family member, and among that quarter of young people who died by suicide who has experienced that loss, 82.1 percent of those deaths were known to be suicide deaths. So I know some other folks have talked about that experience of losing someone close to suicide, and that kind of information that we collect really enforces the importance of suicide postvention.

In addition to circumstance information we also collect information on life stressors. So quickly I wanted to highlight some interesting differences in personal crises via age group. So starting at the bottom of the figure, when looking at intimate partner prevalence such as an argument or a breakup, this was the contributing factor in 19.5 percent of suicide deaths among youth 15 to 17 compared to little under eight percent of youth ages 10 to 14. 

Other interpersonal crises such as being bullied and arguments with friends were selected more for youth ages 10 to 14 compared to the older age group. However, school failure was more of a stressor seen with the older age group.

And then finally looking back at those family-related stressors, we see that a parent’s divorce and family discord were pretty comparable for both age groups, but an argument with parents or caregivers was selected more frequently for the 10- to 14-year-old age group.

Our system also collects data about if any young person, regardless of cause of death, had a history of child maltreatment, meaning child abuse or neglect prior to their death. And a history of maltreatment for our purposes means that there as a referral or a substantiation from child protective services, or a documentation in the autopsy, or law enforcement or medical records.

So, notably 34.2 percent of children and youth who died by suicide in Colorado had previously experienced child maltreatment as a victim. And among those we see neglect, then physical abuse, followed by emotional abuse, and 8.6 percent experiencing sexual abuse. 

One of the ways that we improved our data collection around this area is that for this five-year period we didn’t have any missing or unknown child maltreatment history for these suicide deaths. And this was a collective effort to improve our data quality with our human services partners.

And I am speaking on data quality. I am now going to review some of the efforts we have undertaken, specifically with coroners and medical examiners’ offices to improve our data quality on child and youth suicide, with the ultimate goal to prevent these deaths.

So what is missing in terms of our data? A very important and often unknown piece of information is related to gender identity and sexual orientation of a young person. We know from other data sources, such as our Healthy Kids Colorado Survey, that LGBT plus youth report higher rates of suicidal ideation and attempts than their cisgender and heterosexual peers. However, we don’t have good data on whether or not these youth died by suicide at higher rates, because this information is difficult to collect after a death. 

We also have a lot of room for improvement when it comes to collecting more complete data on mental health history, the storage and ownership status of guns used in suicide deaths, and life stressors from the experiences of racism and discrimination to the impacts of social media. So while we try to collect this information we definitely have a lot of room for improvement.

And one of the initiatives I wanted to highlight today that we’re involved in is the creation of a suicide death investigation forum. The forum was first created in the 2016-’17 fiscal year in partnership with the Office of Suicide Prevention and the Colorado National Violent Death Reporting System. So we piloted this form, and then revised it last year in 2022. 

A main purpose of the form is to collect information on gender identity, sexual orientation, life stressors, behavioral health, and child maltreatment history. And to support the use of the forum, our system and the Office of Suicide Prevention have been offering coroner mini grants, which I will focus on for a couple of slides.

We are currently in the second year of our mini-grant program. We offer $5000 to urban counties and $2500 to rural counties. It is part of the Office of Suicide Prevention’s postvention efforts. And the money that goes to the coroners and medical examiners can be used for using the investigation form, participation in prevention coalitions, as well as providing resources to suicide loss survivors. 

In total, 19 of our 64 counties are either currently participating or participated last year, which is great, but definitely more room to engage more communities if we’re able to continue providing these mini grants. And so far the participating counties have submitted 137 suicide investigation forms for deaths that occurred last year, and those include eight forms submitted for children and youth under the age of 18, and another seven for young adults ages 19 to 24.

So, some of the challenges. This Suicide Investigation Form is incredibly useful for prevention, but some of the challenges that coroners that have used it identified are that sometimes family and friends aren’t ready to answer the questions right away. Definitely true. 

And I think this highlights the need to follow up with families a few weeks or months after death, yet this points to the problem highlighted in the second quote, that there is a huge delay in completing the forms secondary to staffing. 

Our coroners and ME’s offices are notoriously understaffed and under resourced, so even if a coroner wants to collect this information and offer more follow-up support to the families, they may be constrained to focus solely on determining the cause and manner just because of capacity.

And then another challenge is talking about personal potential thorny issues with families. So this coroner said that they found the form to be received negatively by the family, and the investigator was feeling ashamed of asking some of the questions around gender and sexuality. We certainly don’t want investigators to feel ashamed or to cause additional trauma to the family. 

I think partially this points to the fact that having these conversations with a traumatized, grieving family is incredibly difficult, even for experienced investigators, and especially when they are asking questions that they are not used to having asked when it is a new process.

So kind of picking up on that challenge of capacity, we had 10 coroners who completed evaluations so far following their participation, and as you can see here six of the ten spent at least an hour filling out the form, with one of those spending over two hours. 

And then we might assume that over time as investigators get more familiar with the forum and how to talk to families about the questions the time might decrease. However this might be a learning curve that coroners are wary of taking on or don’t feel like they have the capacity.

Additionally we ask them how helpful the form was to their agency in collecting data, which was basically split, with five of them rating it as a four or five, and five of them saying two or three. And we know that this information is helpful for our child fatality prevention and suicide prevention work, but we also need the information to be helpful to the actual death scene investigators that are collecting it. So that’s definitely a goal, is to make this tool useful, not burdensome, for the people filling it out.

But I also want to say in spite of the challenges our death scene investigators are finding the form beneficial. One coroner wrote about how the form has made their staff more aware of the factors involved in suicide and how to investigate these factors on scene and in follow-up interviews.

Another coroner was especially interested in how the data could be used and what it might tell us about the role of ADHD in impulsive suicides. So this is kind of an example of as we collect more and more data the kind of questions that we hope to be able to explore further. 

And one coroner said, a great guideline, as an agency we ask most of the questions either on scene or on follow-up investigation. So kind of going back to that challenge of asking these questions immediately following a death, having the capacity to follow up with families and building the experience to understand on a case-by-case basis which questions are appropriate to ask on scene and which should be asked later, and how that is so important. 

And then in addition to using the investigation form, our coroners and MEs have also done additional things like contacting local hospitals and public health agencies to find out how they can be involved with prevention workgroups or coalitions. Presenting suicide trends and statistics to a local health coalition. And then meeting with other partners like public health schools, hospitals, and mental health providers. 

And then regarding postvention materials, the grant has supported the distribution of care packages and postvention packets and the delivery of resources during the family interview. One agency started offering a survivor report program, and our Denver Medical Examiner’s Office directed some of the funds to support their family advocacy program. In addition, one agency intends to use some of the funds to support their staff in dealing with the vicarious trauma that comes along with investigating these deaths.

Some additional efforts that our Child Fatality Prevention System is supporting are the incorporation of questions about life stressors into the family outreach done by the family advocates of the Denver Medical Examiner’s Office. 

Also, last year the El Paso County Coroner started asking families about experiences of racism and discrimination for children who died from any cause, and from suicide and overdose deaths for all ages. 

We also tracked the use of a suicide investigation form and whether or not it was shared with the review team through a couple additional questions to our case reporting system. And we hold informal, quarterly local coordinator discussions, and support our coordinators with training. And then finally our review teams present this data to their communities, bring recommendations to other community groups, and identify additional opportunities for improvement at the local level as well as the statewide level. So that concludes my presentation, thank you.

BRENDAN WEINTRAUB: Thank you so much. If folks who were part of this session could just turn on your cameras, we will move over to the discussion portion.

JOHN ACKERMAN: Good afternoon, everyone. I appreciate being part of this wonderful presentation group. I appreciate the opportunity to highlight the work of four tremendous suicide prevention innovators. I’m just going to take a few minutes to highlight some of the key themes that came across in these talks on state-based approaches to US youth suicide investigations and opportunities to enhance prevention. 

So each speaker touched on ways to overcome particularly sticky problems that they were facing either across the communities or states, and ways to capitalize from opportunities and partnerships. So I would like for all of us to think about how these different models could be applied to work in our own communities, in our own states.

And then I would like for Holly, Kelly, Angie, Michael, and I, to spend the rest of our time hopefully addressing what Dr. Gorniak very succinctly posed to all of us in attendance: So what? This is really the essential question in youth suicide prevention right now. I would also add two other words, so how, would be another way to do this. 

But we are regularly confronted with worsening rates of suicide among young people and multiple presenters talked about worsening or really alarming inequities among youth of colors, sexual and gender minoritized youth, rural youth, youth with disabilities and chronic health conditions, youth exposed to maltreatment, youth that are bullied, youth with easy access to lethal means, and difficult access to healthcare. Youth who struggle with social connection and a lack of supportive adults in mental healthcare.

So a challenge in the work that we do is that there are lots of different risk factors, and unfortunately there are also lots of different pathways to the tragic outcome of suicide. 

Developmental psychologists sometimes call this equifinality, there are lots of ways unfortunately to reach the same outcome that we are trying to prevent. So what this means is we are going to need a lot of really good data, and that is a lot of what our presenters talked about today. 

We need high quality data and lots of it. So it is really inspiring to hear from such wonderful colleagues about the way that they are overcoming real barriers, real limitations in resources, including crunch timeframes, staffing limitations, political and legislative disruptions, siloed systems and data, and widely varying timeframes among investigation processes, and different methods and ways to share data. So a few themes seem to unify these successful state efforts, which I’ll highlight, and then I’ll cut my part really short and so we can get to the so what piece.

So the things that I kind of noticed was, first, across each of these state based and community-based approaches, there was an intentional elimination of silos, and the creation of a supportive, centralized structure that empowered partners to share their perspectives, address local challenges, and give voice to those impacted by suicide.

The second theme was that there was a focus on collaboration and the creation of lasting agreements between organizations to ensure that data could be obtained with fewer barriers, and then shared out to agencies and stakeholders responsible for making prevention decisions. 

I think this is really key. We have got to get this information back out into the hands of those who are using it, and convince them or support them, and involve them in the data processes. And in several cases there was advocacy that led to legislation that codified agreements and processes so that they remain in place even if there were political or administrative changes. So we need this continuity and sustainability of the work that we’re doing, otherwise we are going to continue to need to start over from scratch each time.

And then my last theme was that each state-based effort focused on leveraging data and outcomes from improved death investigation processes to do things like guide upstream prevention practices, increase awareness, reduce stigma, address inequities, and enhance connection and support among young people, and then very importantly create safer environments so that when young people do face a crisis they have the means and the supports from those around them to address this. 

So each presenter focused on different ways their states confronted these gaps and limitations to create some very innovative approaches. So I then spent time highlighting what each person did, but you just heard them, and we are going to get to some good discussion here.

So I just wanted to start off and allow the presenters maybe if they have any general reactions before I have a few specific questions, and maybe frame this in terms of ways that you overcame barriers or sort of the most important innovations that your state underwent. If there’s no specific, general comments, I can get to a few specific questions.

ANGIE HAYES: What Dr. Pinneri pointed out to me via message before she had to take off a few minutes ago was that there are several counties and states across the US who use some of the same classification with playing Russian roulette and vehicle collisions. But that is not entirely unique to Harris County. But I thought that was interesting, more context.

JOHN ACKERMAN: So I want to get a sense of what each of you felt was the most important step you took to de-silo either the data or the processes, what did it take for you to get initial momentum. Maybe I’ll just start in the order of the speakers so that Michael can start, and then we’ll go to Holly, Angie and Kelly.

MICHAEL STALEY: Tenacity. I think that sometimes we get up against a structural barrier and back away from it, and it took three years of legislation and fine tuning. And we are not even all the way there yet, we’ve got some rulemaking still to do in Utah to allow for the level of data sharing that we’re trying to pull off. 

But you know, you take these questions before legislators, I have not been met with a whole lot, I won’t say no resistance, but not a lot of resistance, when you’re saying the goal here is to prevent youth from dying by suicide. And I think when you present an option for them to engage, you’re going to find reasons to engage rather than the other way.

HOLLY WILCOX: For me, I think it was kind of my role as the suicide commission co-chair. So I was kind of a known commodity and kind of trusted, I think by a lot of the agencies that are part of that. They’re people native to it from the VA health system, and a lot of other departments across the state. 

And so I think that was establishing partnerships and trust I think are really key, and patience. Because everything takes so much longer than you would expect. Like the data use agreements, and the IRB, two steps because it has to go through Hopkins and through the state of Maryland and there are so many delays and just not getting frustrated and trying to really take a deep breath before you nudge somebody. So I think that is all part of it. 

The other thing for us that has been helpful is trying to leverage senators and other policy makers that we know have lived experience, having a family member who was lost to suicide, and trying to leverage their support. 

We also had one senator who was on our suicide commission was fantastic, a republican, got everything that we were going for. She was the person who was behind the Suicide Fatality Review Board. And so having her support in getting through the legislative hurdles was really key.

The other thing is leveraging the American Foundation for Suicide Prevention, and some other family advocacy groups. Because people with lived experience oftentimes want to help but they don’t know how to help. But if you go to them and say we’re getting lots of barriers and obstacles right now with this legislation, and we don’t have funding to implement this bill or this law. Is there anything you can do? 

And they can really help. I mean, we had one dad who lost his son to suicide who contacts us every once in a while, and he’s like, what can I do, what can I do? We’re like we don’t have any money for the suicide fatality review, can you help? And he is like, yes, I can help. And so there are other examples too, but I will let others weigh in.

JOHN ACKERMAN: I will just mention in terms of you talked about the timeframes, which can be really extended and don’t always map on to the funding cycles that folks get. So I think trying to be patient, but also trying to do a little bit of advocacy to help sort of local, state, and federal folks understand that this will take time. So if there is a last-minute funding opportunity that needs to be used within a certain timeframe that can be challenging in this type of sphere. So I appreciate the persistence and the perseverance with advocacy. Angie, did you want to add?

ANGIE HAYES: I would just echo the tenacity, persistence, that is how we have gotten as far as we have, is just dig my heels in and keep trucking and keep trying, and if I can’t get in this door, I go knock on another one. If I can’t get in there, I go find somebody who knows somebody who is inside there. 

It has been several years, I’ve been at this one for five years in a way. We’ve worked together and build it up into a whole lot more than it was when I first came in, and that has just been little bit by little bit, it is more than I thought it would be when I first came in, which is fantastic. It didn’t happen overnight, we are still growing and making changes and setting that long-term view of maybe it didn’t happen today, this time, this funding cycle, this year, but maybe it will next time.

KELLY DOUGHERTY: One thing I think that has really helped us in Colorado is getting that statute passed in 2013, because that gave us dedicated funding every year to support the reviews at the local level. So prior to that we were doing more limited reviews, and they were all at the state level. So of course, we don’t have this much information, we don’t know what each community in the way that people from that community do. 

So I think it has really improved the quality of our reviews, to have them done primarily at the local level and improve the recommendations, and making the recommendations more relevant and more kind of sometimes actionable by those people locally. And also like the data sharing, that is in our statute, so we are able to have access to data, and we can point to that statute to other agencies. 

And then I think just one piece that we definitely see as missing is more of that family voice and family experience. So I was really glad I learned last week actually from our family advocate and review team coordinator in Denver that there are a couple other coordinators offices in the state who are hiring family advocates, family support people for their offices, or who are expressing interest in trying to get funding to do that. 

So I think Denver ME’s office is really out front in the country for having bereavement supports to families. So I am excited to see that. And we also have great Office of Suicide Prevention staff that help and lead a lot of this work in the state.

JOHN ACKERMAN: You really do have a very wonderful group in Colorado. It is very lived experience centric, which I appreciate. It looks like Michael has a response as well.

MICHAEL STALEY: I agree, Denver is amazing, and we definitely want to take their vast model and implement it in Utah, I got to see that firsthand last week when we visited there. I just wanted to make a general observation in the last three minutes, and that is I think in all three of our data we show the recent breakups for example, I’m just cherry-picking here as something that is a high-risk time for a young person. 

But people have been breaking up for a really long time. So what is changing about that that makes suicide a better option to them, of course their own, reflexively, a better option to them, than something else than people sad for a period of time, what has changed. 

And I have thought a lot about this question. Suicidologist asks why suicide, why suicide now, why by this method, and how could it have been prevented. And the why now part, we haven’t done all of our analyses of our data, but I am concerned we are going to get to a place where we’re like shoot, we still didn’t get the why now part. So I don’t know if anybody has any kinds of, my co-presenters or anybody else, have any kind of reflection about that, or what happened.

HOLLY WILCOX: It is the sheer volume of firearms that are out there right now, the easy access that could be one factor that is making things different now than maybe  breaking up in the past. If you know you have guns in your house and you know how to get them, and you’re in crisis, and you don’t think things are going to improve. 

And also, skills. I mean, not that skills are any different now than they had been in the past. But I think anything we can do in schools to enhance skills of kids to know how to deal with highly stressful situations and know how to deescalate themselves and deescalate a situation, I think that is another area of promise.

JOHN ACKERMAN: It feels like there is reduced connection, reduced support. And lethal means broadly, kids also how to access other ways of dying in a very nuanced way. It looks like I have lots to say on this, but I saw that Dr. Warner has a statement, and Dr. Gorniak. And I know we’re running on the break, so I’ll let Dr. Warner go ahead.

MARGARET WARNER: I was not looking at the time, but I do have a quick comment, as really exciting stuff people are doing out there in states. I am wondering, sort of harkening back on the question I asked during the first presentation about bringing together some of the researchers that are from NIMH and other places that have scales or different diagnostic protocols or screening tools that they’re using in a laboratory setting, in a very controlled environment, or somewhat controlled I guess, thinking about how those tools and screening diagnostics could be brought down to the postmortem setting and blended with some of the forms and various forms and interviews and different techniques people are using in the field and how we can try to blend those together. 

I think that is some of the ideas that were brought together, to bring together this group. But I’ll talk a little bit more about that next. I thought I would while I have the chance to ask the presenters if they have any thoughts on that, although we are out of time, so.

MICHAEL STALEY: Yes, I have already asked Dr. Luby, because I know our instrument really kinds of breaks down for under-14-year-olds, and their experience is different than their older peers. So we need some better tools for that and to build those connections here.

JOHN ACKERMAN: And it looks like we’ve got a quick minute for Dr. Gorniak to ask a question.

JAN GORNIAK: So my comment is the why now and the skills. I think we have to look at the different generations, the Gen Z specific things for example, maybe some of you remember when we would take pictures when we were younger, we had to have them on a camera, turn the film into the drug store, and then wait for them to get back. Now you take a picture and you’re like, oh, what did I look like? I can take my driver’s license picture four times to make sure it looks okay. 

So I think even as adults we have lost how to be patient. And I think we have to look at those type of skills that need to be taught again, because as the world is evolving and changing, technology is wonderful, remember when TV used to go off? Now we get news and everything bombarded all the time. 

So I think we have to look at those type of things also on why now. And exactly, some people get divorced, they have a party, some people get divorced, they put a gun to the head. But what’s the difference? And where are those coping skills that we are losing? That was just my point.

JOHN ACKERMAN: I would like to thank all the presenters who did just a phenomenal job, and I look forward to discussions for the rest of the day. Thank you.

(Break)

BRENDAN WEINTRAUB: Okay. So, as is tradition, coming back from a break, we are going to do a Mentimeter poll. So I just went ahead and shared the screen here. Please go ahead and go to www.mentimeter.com and use the code 31733. Our question is a variation of what Dr. Karen Mack mentioned yesterday, which is what additional training, research or data resources do you need to address child or youth suicide. 

So all folks from the audience are welcome to participate in this as well. And it will be good to mention that we are planning to take a look at these Mentimeter polls a little bit more in depth after this meeting, and use it to kind of inform next steps. 

So far we have seen some policy and advocacy training, a data clearinghouse, data sharing, it seems like there is a lot of data there. Better measures of experiences of discrimination. More research about the issue of social determinants of health. 

Lots of data here, that’s great. Like I did previously, we are going to leave this Mentimeter up, so please continue to add to it. I will pass it over to Dr. Colpe.

Agenda Item: Summary and Future Steps – (Federal Suicide Prevention Partners)

LISA COLPE: Here we are at the end. This is our last session. What we have done is provide some time for our federal partners and panelists to share with you sort of what they’re thinking of for next steps or for resources that are available. And just to be able to give their thoughts about where their work is going. First, I would like to ask Margy Warner to come on up and tell us a little bit about what you’re thinking.

MARGARET WARNER: I am very excited to be the first person to give my thoughts, because sometimes people remember those the most. Be the first and the last, I don’t know about you guys in the middle. I did put together some slides, they were very quick. So I have a lot more thoughts. 

And we were all talking amongst ourselves, and I was really happy to hear that Michael and Joan Luby had already connected on this idea of bringing together some of their works in terms of what has gone on with the of state of the art from NIMH researchers as well as people working in the field, so that really does bring together this idea of science to service, service to science approach that we were really hoping to generate with this workshop. 

So really hopeful, I’m sure there are other things going on that I don’t know about, with you guys all talking amongst yourselves, and anybody really who has some thoughts on this, I am sure Lisa mentioned there are ways to get in touch with us, which are all going to be all over the place.

So again, the idea was to bring the state of the art of the research practice and data informed investigation of child and youth suicides, we heard a lot of things over the past two days, including the fact that the indications from the data, that both suicide ideation and suicide is increasing among the young is really quite clear, and that there are certain special populations, we discussed minority youth, LGBT youth. And then I’ve listed out the ones that we’ve talked about here, but there are certainly other groups. 

But the investigation and how we get that information is really critical. And there are ways to link data, but really a lot of the information, we can only do so much with linked data, and a lot of the information is gathered during an investigation, whether that’s the initial immediately after a death, or during the course of I think four to eight months was laid out as a period of time. 

And maybe some of the things we could think about would be what are those periods of time within, is a good state, a good time to ask questions. So these are the kinds of things that our groups can talk about, both from the practical experiences, as well as from what we know about people’s experiences after a suicide.

The other thing that came out a lot, in the beginning at least, was this idea of postvention. In other words, what kind of intervention can we do after a suicide. So I started all this and I forgot what my next slide was, I moved around my slides. But let me tell you a little bit, since this slide came up, and then I’ll go back to some of the things we can do afterwards, because some of what we were asked to do in these summing up moments was to talk about resources that are available. 

And there has been a lot of discussion about the medical examiner and coroner community during the course of this two-day workshop. And I wanted to mention that we do have the new Collaborating Office for Medical Examiners and Coroners at CDC. It is very new, we have been working with medical examiners and coroners at CDC through a variety of different programs over the years, but new resources to bring together information for medical examiners and coroners in one area. 

So I encourage people to look at our resources that we have available. We also have a public-facing website with more information about the medical examiner and corner community, for those of you who want to learn more about that. So that is one of the resources from the Federal Government. 

I also wanted to mention there is an interagency workgroup on medical legal death investigation, which many people in the federal space that are participating in this workshop are involved with in one way or another. So we try to facilitate collaboration and peer learning amongst the federal agencies on work that the medical legal death investigation community or the medical examiner and coroners and how it interacts with our federal programs and what programs we have. 

So we are trying to coordinate a little bit more within the federal government, and some of the things that we have done, either as agencies individually or as agencies together, have worked on reports about needs assessments and recommendations. 

So I think that some of the issues that have come up, and some of the barriers that people have talked about, a related to resources and the needs within the offices of medical examiners and coroners. So I wanted to bring people’s awareness to these resources.

So we have some recommendations we are working towards. Some of the recommendations are around data systems, and some of the recommendations are around accreditation and certification of offices. And those are really important ways that we can improve the situation and bring needs, address some of the needs in those offices. Particularly around accreditation and certification there are moves to, well, in those reports you can find more information about that. 

So I’m just going to jump a little bit again. Because we have been talking a lot about data, I wanted to talk about the data modernization activities we have with medical examiners and coroners. And I put up our website around data modernization activities. It is a new activity. We are establishing common data elements for medical legal death investigation offices. 

There currently are not established elements that each office needs to collect, we’re developing data standards around those data elements, and we have real world demonstrations of data modernization within offices. So it is a doable task, it does take time and effort, and when we see a need, we can work towards those.

So in terms of the takeaways, again, I was charged with giving a little bit about the resources that are available. So those are some of the resources that I sort of feel address some of the things that were touched on during this workshop.

But I wanted to talk specifically about some of the takeaways we could have from his meeting, and I really wanted to highlight the idea that understanding the state of the art in understanding of child and youth suicide can really help to flood the ideas and help. We think we are identifying a problem and a trend, and we want to get out ahead of that in terms of our ability to monitor what’s going on in child and youth suicide.

We also, in order to do that, we need good data, and we also need, so we need good data both to identify people at risk, risk and protective factors, as well as a number of times it has been meant in evaluation. 

So if we are going to truly have prevention programs that address these needs, and we would target those data driven prevention programs, so we need to know who is at risk, then we need to develop prevention programs to prevent those deaths, and we need a way to evaluate those. And we can’t evaluate unless we have good data. So just sort of bringing it all around, mortality data is useful for all those phases. 

So in terms of the idea of postvention really came up, I’d never heard that, I’d been doing mortality research for many years, so I love this idea. And the idea that that can be done within sort of during the course of an investigation, whether that is within the medical examiner and coroner’s office immediately upon finding out about the death, or four to eight months I think was mentioned. 

But what’s the science around that four to eight month, that’s really a practice-based understanding. And we can try to sort of hone in on what’s the sort of best case. Or the best case in a bad situation. But I keep throwing out adapting these screening and diagnostic tools to the postmortem setting. I really feel like that’s something that we can do. 

And also collating the individual tools that are out there, and how they’re being documented in medical examiner and coroner systems, and coming back to that idea that data modernization, we really need to think about what are the common elements that need to be collected, and not necessarily on every single death, but on the deaths where we know that they are relevant, and then how were they recorded.

And then, as a final takeaway from this, I think just talking amongst the panelists from the medical examiner and coroner community, we think it is important to bring some of these concepts forward to a wider community. I think there still is concept potentially of an age cutoff. 

I heard it before, and some of the panelists are nodding their heads, we do think that there are some offices that are still abiding by some kind of artificial cutoff, age cutoff, so how do we bring forward this message that both the indications from both the antemortem, so the research from NIMH scientists is showing that there is suicide ideation in younger children, much younger than I think was previously thought, and whether that’s a new thing or an old thing, it is definitely there, and so the idea that there may be suicides among the very young does need to be brought forward, and investigations need to consider that. 

Because ultimately, I’m not sure whether you followed, just wrapping back to my opening comments, what is written on that death certificate in that manner of death checkbox, does influence a lot of the data collection activities. So it is really important that that manner of checkbox reflect the manner of death, and not be restricted by people’s understanding of childhood mental health.

BRENDAN WEINTRAUB: OK, next we will welcome Dr. Jane Pearson.

JANE PEARSON: I am going to present some slides on behalf of Eric Murphy. So, I just wanted to let folks know about the translational research that NIMH has. And Joan Luby has been funded, I think maybe actually all three of the presenters we had, the extramural folks, may have had some funding from this part of NIMH. It is really trying to understand what is contributing to youth suicide risk, and also the protective factors. Stacia Friedman Hill is also part of NIMH.

So we just wanted to raise some of the topics that NIMH has been looking at. And the biomarkers I will get to in a moment again in terms of also understanding some of those risk factors. We brought up a lot of family risk. But understanding a little bit more about what is actually happening in people is really important in understanding those mechanisms. The social interactions, and all the things we’ve been talking about, disruptions, triggering events, links of social interactions and those disruptions across social media, bullying, and then also intersectionality. 

So in addition to race, ethnicity, multiple race, there is also sexual and gender minority issues. And then if you could even of course talk about intersectionality with different levels of SES with these kids. So we are trying to get a better handle on that in terms of how these pieces all fit together in terms of creating risk.

And then a lot of focus on trying to think through research methods. Statistical, using devices, a number of folks mentioned the need for mixed methods or community-based interviews trying to get a sense of what is most appropriate for certain cultures and groups and communities to better understand those risk and protective factors. 

And then the classification piece in terms of for live kids, ideation, attempt, self-harming events, and especially how this plays out in preteens, just to echo what Margy was saying about our challenges in trying to assess very young kids, pre-teens, and how some of these constructs may shift over time. 

So the next slide just refers you all to the suicide research team website, which has a lot of the announcements relevant to suicide prevention on it, and I wanted to share Eric Murphy’s email if you wanted to reach out to him and find out some more about his program. Thanks.

So related to that biological question, I see Abigail is in the room, that is great. And I will let Abigail introduce herself. Really happy to have her at this workshop and suggest another direction that some of the discussions could lead to in terms of improving our data around this problem.

ABIGAIL SOYOMBO: Thank you so much to the organizers for inviting me to present in a kind of different direction in terms of research into suicidality or mental generally or diseases affecting the brain. 

And also understanding, presenting on the NIH NeuroBioBank and the efforts that NIH has put into place to better understand human brain function as well as brain disorders, of depression or mental health disorders. So one of the coleads of the NIH NeuroBioBank resource, I have my other project leads on there. 

So introducing the NIH NeuroBioBank, it’s a network of six brain banks that are funded by a couple of NIH institutes. I have listed the brain banks there. Basically they funded through contracts by the NIH to procure brains from postmortem human brains, and process, characterize the donors through medical records or interviews, and then store the brain appropriately or using a general harmonized process, and then distribute to a central website, which I’ve listed there. 

The brain banks work with a number of partners. The Brain Donor Project is one of the partners we use that helps with outreach. And also Brain Donor Project, sometimes they will get called either by families after death of the donor, and the brain donor project will then refer to the bank that is closest to where the death occurred. So the network works with as well as the medical examiners’ offices, organ procurement organizations, in terms of getting referrals. 

So the overall mission of the NeuroBioBank is to facilitate brain research as well as increase awareness of brain donation. So when people are already aware, after a loved one passes, then they are able to support research into brain health by donating the loved one’s brains. We always tell people brain donation is not really part of the organ donation consenting process, so that is additional. People really need to sign an additional consent form for brain donation.

So on the NIH NeuroBioBank website we have a form for contacting us. And the contact form will go directly to the brain donor project, who will handle the referral, or to the appropriate bank, depending on the location of the family, of the disease donor.

And I just wanted to indicate that the NeuroBioBank, the inventory is very diverse. We have all kinds of disorders, including from donors that died from suicide, as well as individuals that have no disease at all, and they are used as controls. And basically, tissue request, we provide tissue to all over the US. 

The goal, the reason the NIH wrote this NeuroBioBank network as opposed to funding different investigators through grant was to make sure that access is broad to qualified researchers. That way research is also broadened as part of providing equity in terms of access to very valuable tissue for research to support research.

So, a request comes into us, NIH approves, and then the request goes to the brain banks for distribution. And I just wanted to acknowledge, this is not the research group, but just to let you know that we get a lot of requests for the tissues, and a lot of very important publications are made, and a lot of supports of biomedical as well as translational research into disease mechanisms, and trying to understand the brain function in health and in disease.

So I will be happy to take any questions, or in terms of maybe partnership with ME’s office or referral of donors. Thank you. 

KARIN MACK: So, thanks everybody for staying until the end. Thanks for sharing the challenges that you are facing in making suicide determinations amongst youth. As part of CDC’s mission, we are called upon to present reliable and accurate data. So we really appreciate all the efforts that you go through to help make that happen, and to support those efforts.

These are some of the things that we are supporting. So CDC supports a comprehensive suicide prevention program in 17 states that will be expanded a bit in the fall. If you don’t know about that program, or you’re not connected to them, I would encourage you to reach out. They do suicide prevention programming in their states, in their areas.

Another resource we have that we just updated last year is the Suicide Prevention Resource for Action. There was a hard copy that was circulating, and it was actually the first time I had seen the hard copy. And there is a QR code that will take you right to the resource for action. 

And this is a compilation of policy, programs, and practices, evidence-based solutions to help to do suicide prevention work in communities. So it has strategies and approaches. One of the components in there is about postvention. So there is information there on programs that can be implemented in communities for that.

And I wanted to offer thanks to Dr. Staley for mentioning the EPI-AID, and to Dr. Blosnich for calling out being on the frontline, and maybe the first person to connect the dots, because I feel like you setup my slides very nicely, so thank you for that, calling those things out.

So one of the things that CDC does is provide technical assistance to communities and states to assist with suicide clusters. Again, that QR code will take you to the suicide prevention page. Right now there is some high-level information about identifying and responding to suicide clusters. The second part on that slide talks about CDC technical assistance. So that’s for any of the outbreak investigations, CDC goes into states and communities to do investigations with our IAS officers.

And one of the things that we have right now is a 1988 document that we reluctantly share, it is a little dated, for the prevention and containment of suicide clusters. We are in the process of updating that. And if you have time, I know you already spent a lot of time here, if you had time next Thursday afternoon, May 4th, we are presenting that initial content to the Injury Center’s Board of Scientific Counselors. 

So that’s a public meeting, it’s virtual, you register through the Federal Notice, and it would be great to get any feedback that you might have about responding to suicide clusters in communities, what resources are needed, how do you identify something, and then how do you respond to that.

DIANE PILKEY: I just want to thank NIMH for sponsoring these meetings, and I think the presentation and discussion have really highlighted the need for additional resources for death scene investigation as well as the importance of those connections between the medical examiner and coroner community, the death scene investigator, surveillance systems, and fatality review efforts such as child death review, in order to ensure that we have that quality data for research and to inform prevention of suicide.

And a couple things relevant to the National Center for Fatality Review and Prevention that Abby presented on and others have talked about. First I wanted to mention that the Child Death Review Case Reporting System does capture data on all of the special populations we talked about today, including LGBTQ youth, children and youth with special healthcare needs, youth in the welfare system, as well as those juvenile deaths that occurred in juvenile justice systems. And I think the challenges of capturing data, especially on LGBTQ are similar to what other folks have mentioned in terms of there is a lot of missing data in particular.

A couple things I wanted to mention. The Center is planning to convene a mass training for death scene investigators in 2024 focusing on pediatric death.  And their plan right now is to provide some funding for a certain number of death scene investigators from each state to attend. And the focus will be on a train the trainer model, so these individuals can come back to their states and be available to provide trainings within their states.

Again, the focus is going to be on pediatric deaths, with additional emphasis placed on certain cause or manners of death, including suicide. And while the national center will fund and convene this meeting, it is going to rely heavily on national partners, subject matter experts, and child death review teams to provide the content expertise. The tentative target time period for this training would be late spring or early summer of 2024.

And just as a reminder, the National Center is sponsoring a special supplement in the Journal of Pediatrics that is going to focus on child death review and fetal infant mortality review of CDR femur data to inform practice, policy, and research. If you are interested in getting access to a researcher database or submitting an article, information is available on the Center’s website, which is NCFRP.Org, National Center for Fatality Review and Prevention. And that is all. Thank you so much, this has been a great meeting.

JOHN ACKERMAN: I thought it might be helpful, and I can share the screen, about six months ago we were able to - Dr. Lisa Horowitz and I, worked with the American Psychological Association to put out an open access book on Youth Suicide Prevention and Policy. It covered a lot of topics in a very brief way, and covered a lot of the risk factors that we talked about. 

And the authors, several of which have presented here really were empowered to provide some action steps and really talk about what we could do in terms of investments and in terms of figuring out how we could really leverage data to make some changes in the area of youth suicide, with a focus on how do we scale some of the solutions that we have started to think about what are some core ways to implement suicide care and suicide prevention models, thinking about universal options, thinking about integrated care options. 

So it is a little more upstream than some of the death review data that we have talked about, but I think it is important to know that this information is being used in hopefully highly practical ways, and hopefully we can continue to empower people to engage in culturally responsive and really population specific work. 

And then we need to get this information out into the hands of individuals, so I think it is going to be really important to focus on how we scale things. So in Ohio for example we supported over 250 schools in implementing both screening and suicide prevention processes through the support of some hospital and partner resources, and I think we do know enough how to start to scale some of this work, especially with lethal means safety efforts and care models that can work. 

So I think a lot of what we can do is start to move forward in thinking about how we take the information that we do have, and scale it, and make sure folks have access to this work. Again, that was just one resource that we have to look at, and I appreciate the support of all the partners who contributed to this. 

And I will say in terms of plugging open access, the particular book that they had previously has volumes on things like child abuse and foster care and other topics. It gets about 2000 to 3000 downloads, and this one already has 57,000 downloads because people are able to access it. So we can have all the best information in the world, but if we don’t give people the ability to access it in real-time then we’re not going to get the same impact that we could otherwise. Thanks for that call-out Lisa, I appreciate the opportunity.

LISA COLPE: Absolutely. Now to any other panelists that have something that they might want to highlight availability for. Sorry to put you on the spot. But hearing none, we may go into our final words from our fearless leader Dr. Jane Pearson, who is going to give us just a little bit of a wrap before we adjourn. Jane, take it away.

JANE PEARSON: There are a number of different institutes in addition to NIMH who fortunately also invest in suicide research: The National Institute of Minority Health and Disparities would be one that we would like to engage more in, and they often sign on to some of our announcements. Especially some of the ones that I was mentioning that Eric Murphy has for healthy disparity youth.

This one came out in February. And if you scroll down, this notice of interest in terms of mortality for adolescents and young adults, includes some really important topics here that had been brought up at this meeting. 

So I just wanted to make sure you were aware of this, you could search for it, the NOT HD23 001. So this includes looking at accuracy of death coding, improving the research on mortality, and looking at some of the groups that are having increased rates. 

A notice of special interest means there is not specific money set aside for this, but it really helps focus an area in terms of what an institute is interested in. So I just wanted to make sure we got that included in the meeting. So thanks, Brendan. 

And I will just go through a few themes that I heard today. First of all, before we get to that, thanks to everyone who has worked so hard to put this meeting together. We have been talking about this for maybe since 2019, the need to pull these different groups together, and I am so glad it has happened, and really appreciate all the effort that has gone into it. 

And it is allowing us to really I think, look at these data trends in really important ways. We knew the rates were increasing, but now to understand some of the subtleties in this, and also some of the missing data that I think we are becoming aware of. And it just makes me wonder if we should be more forthcoming with the missing data, to emphasize the need for more resources, the need for more infrastructure, and how resources are limiting us in understanding this.

I have been really impressed with what the researchers have been able to do with some of this mortality data, and then look at smaller samples to start to figure out what the processes are for risk and protective factors. 

And then also looking at this longitudinally. So we are hearing about precipitating events, all these triggering events, the social disruptions. We are hearing about kids at risk, who we already know are at risk for so many different reasons, and all the needs we have in terms of mental health services. And then some of the more distal factors, and opportunities where we could start much earlier and do better. And then the measurement approaches to all of this. 

So I strongly endorse what Margy was saying, are there other opportunities where researchers have sorted out some measures that could be informing some of the interviews for the death investigations, and then also thinking about just the realities of the timing of all these questions. 

We might need multiple approaches, we might need to let families know this data might be collected eventually, and then have somebody else follow up again, and think about this in terms of a more staggered way to collect more important data, and finding more about those timeframes that Margy was mentioning would be really important.

We know there is so much already happening, thank goodness, in terms of collaboration across these different data collection systems, and we saw some really great examples where states are really working hard to link datasets together so we have more than just the individual decadent information, but we can start building both their longitudinal story as well as sort of their contexts where they’ve been seen, understanding the social determinants of all that.

So, none of this is new. All I am trying to do is summarize that we now have a better picture of what I think we all could be working on, and now have some contacts, maybe new friends that we could reach out to to better inform all of the different components that we have to have here to do some successful intervention to really change these trends. 

So I just want to say thanks again. If anybody else wants to add any more comments here, I’d be happy to hear about that. I have very serious FOMO, missing out on being here in person and talking to folks. So if there was anything else that popped up that people haven’t had a chance to share, want to share, I’ll leave a few more minutes open before having Lisa finally close us out.

JAN GORNIAK: So, one thing that I was just thinking about, and I think Abbie mentioned it or it was on one of her slides about EDI. So one thing that I think is important is when we are having our child death review team, child fatality review team, the review team, whatever you call it, that there is representation on those teams, just so that people can be culturally aware of different things. 

So it may be different myself going to somebody’s home asking questions compared to Michael, how receptive people are. So I think we just remember to have a diverse team also when we’re looking at mental health.

JANE PEARSON: That is great. And we need diversity in our research teams, absolutely. Excellent point, thank you. Anybody else? I want to thank the audience for sending in some real interesting questions. There were a couple themes showing up in terms of sexually transmitted diseases as being real risk factors for youth who probably aren’t quite certain what to do with it, what to make of this in terms of very challenging conditions, lack of information. 

So we’ll be looking over all those Q&As, so we really appreciate the audience participating in the way they could, and through the Mentimeters. We are going to take a look at those. So thank you to the audience for joining us virtually. Thanks. And Lisa, I will turn it back over to you.

LISA COLPE: So, the purpose of me being up here is to call adjourn, unless there is anything anybody else wants to say. I really appreciate your tenacity staying in this meeting and being able to stick with all of the various and complex problems that we’ve been discussing today. And I really am pleased that we’ve been able to work together to put these sessions together and thank you again for all of the time and attention you’ve given to it. So with that, let’s call it a meeting.