How To Leverage Data Analytics to Improve Suicide Risk Assessments

February 20, 2024 - Behavioral workforce shortages and access barriers to mental healthcare continue to drive the United States’ mental health crisis, leaving many at risk for adverse outcomes like suicide without the care they need.
Health systems play a pivotal role in closing these gaps and improving outcomes, but preventing suicides remains a challenge. Kimberly Roaten, PhD, ABPP, Professor in the Department of Psychiatry at University of Texas Southwestern Medical Center and Associate Chief Quality and Safety Officer for Behavioral Health - Parkland Health, and Jacqueline Naeem, MD, Senior Medical Director at Parkland Center for Clinical Innovation (PCCI), discuss how health systems can build a risk stratification program to prevent suicide in their populations.

Shania Kennedy 

Hi, everybody. Welcome back to health care strategies. I'm Shania Kennedy, assistant editor of HealthITAnalytics and today I'm here with my colleague, Kelsey Waddill, multimedia manager and managing editor.

Kelsey Waddill 

Hi!

Shania Kennedy 

And actually, before we get into today's episode, I would like to actually provide a brief content warning because we're going to be chatting about suicide and suicide prevention, which is obviously incredibly important. But if those topics could potentially be triggering for you, please take care of yourself and I encourage you to feel free to skip this conversation. And if you or a loved one are experiencing mental health related distress, please call the suicide prevention or crisis hotline in your country. In the US you can call or text the 988 suicide and crisis lifeline for free confidential support 24/7.

With that said, let's go ahead and get into today's topic. As the mental health crisis in the US persists, health care providers can play a key role in supporting patients and preventing suicide. Data from the CDC indicate that suicide rates have been on the rise since the early 2000s. Only dipping briefly from 2018 to 2020, before peaking again in 2021. This increase has led to calls for improved mental health care and robust suicide prevention campaigns.  

One strategy for suicide prevention is identifying and supporting those at risk which health systems have an opportunity to do at the population level across patient groups. We're going to chat today about one of those initiatives which is led by Parkland Health and Hospital System [PHHS] and Parkland Center for Clinical Innovation [PCCI] in Dallas, Texas. To tell us a little bit more about the universal suicide screening program, we have Jacqueline Naeem, MD, PCCI's senior medical director, and Kimberly Roaten, PhD, ABPP, professor in the Department of Psychiatry at the University of Texas Southwestern Medical Center and associate chief quality and safety officer for behavioral health at Parkland Health.

Thank you both for coming onto the show today!

Jacqueline Naeem 

Thank you for having us.

Shania Kennedy 

Okay, so to kick things off, I was wondering if you guys could just give the audience a sort of brief overview of the program and what it does. Dr. Roaten, let's go ahead and start with you.

Kimberly Roaten 

Sure. So in 2015, we decided to start a universal suicide screening at Parkland, as you've heard. This really came from our recognition that many individuals who eventually die by suicide have contact with our healthcare systems in the days, weeks, and months leading up to their death. And we really came to the recognition that it was a missed opportunity to identify risk and do something to prevent suicide attempts and suicide deaths.

So, fortunately, Parkland Health fully supported the effort to screen for risk and all of our patients. So in 2015, we added a standardized validated suicide screening measure for all patient encounters in our health care system for both kids and adults. So every time somebody shows up for clinical care in our system has a doctor's appointment, we ask them a standard set of suicide screening questions, that information is recorded in their chart. And then we provide the appropriate clinical resources and safety precautions to take care of those patients.

Shania Kennedy 

This is definitely a proactive approach. And you mentioned you're screening everyone, regardless of the presenting problems, which I think is unique in this type of program. Of course, other programs already exist, but they're more heavily focused on screening patients who are presenting with mental health problems or other sorts of psychiatric complaints. So I think that's part of what's unique about this program. And with that in mind that you are screening all patients, can you guys walk us through what the risk stratification sort of process looks like for that, because it's certainly large scale, and how do you guys conduct patient outreach?

Kimberly Roaten 

Sure. One of the things we realized, as we were trying to plan this process was that in a system this big, we had to be able to efficiently deploy the appropriate resources. Certainly, we didn't want to violate anybody's rights. We want people to be able to make choices about their own health care, but we also want to provide them with access to the resources they need to address things like depression and suicide risk.

So what we ended up creating was, we integrated the screening tools into our electronic health record. So rather than using paper based interventions, we made it fully electronic and then we built in a clinical decision support system on the back end. So essentially, the process for us is a patient arrives for their health care visit. They're checked in or triaged as they would normally be--the things we all experience: get your blood pressure checked, your weight, your height, why are you here today? And then as part of that process, and usually in the context of other sort of sensitive topics we asked about drug and alcohol use. And then we asked about suicide risk.

The nurse typically asks those questions and then those answers are recorded in the electronic health record directly. And the electronic health record then tells the provider sitting in front of the patient what to do next, based on their responses. So it's all automated. So if a patient indicates imminent risk factors for suicide, somebody we need to be concerned about right now, the nurse in that example would be prompted to keep the patient in eye contact, make sure that they're safe, and immediately let the rest of the healthcare team know that we need to immediately respond to safety issues and do a full suicide risk assessment.

Shania Kennedy 

Yeah, and as you mentioned, you're balancing that respecting the patient rights with of course, providing that care and flagging that risk. And, again, I think that's a really great aspect of this program. As somebody who struggled with suicidal ideation as a teenager, for me part of my fear of reaching out to a medical provider about it and getting the help that I needed was this fear of, "are they going to force me to be hospitalized?" Questions of "what are they going to make me do? What choices are they going to take from me?" Which is scary when you're in that situation and, as a teenager--you of course don't want that any time, but especially not in a situation like that. So I think being able to come at it with this compassion to realize that, yes, this is going to be difficult to approach when you do this screening, but doing so is really important. So I think that's really admirable. And I think that's really difficult. So I do want to commend you guys for that.

But to that end, as well, there is a health equity sort of question here, of course, because mental health is so heavily impacted by social determinants, and those factors that aren't just biological. And I know that PCCI and PHHS have initiatives already that are focused on improving health equity in areas outside of this. But for this initiative, in particular, I'm interested at looking at how you guys are taking that SDOH data and incorporating that and considering it, and how does that sort of influence the screening process?

Jacqueline Naeem 

So yeah, so PCCI, that's an area that we're hyper focused on is social determinants of health or non-medical drivers of health or health-related social needs, whatever terminology you want to use. And really, as part of the analysis that we're working with Parkland on is we have access to all that demographic data. And that can help provide some insights to how those screenings performed for different groups. And that might give us some insights in future initiatives and things. And when you talk about kind of equity, I like how you touched upon kind of the mental health aspect, because I think that's an area where lots of times people don't--that's not the first thing that comes to mind when they say the word "equity." So I love that you've included that in your thought process as well. Kim, is there anything else to add about the actual screening part?

Kimberly Roaten 

No, I think all of that is really important to think about. And one of the things that we talk about a lot at Parkland, in general and specifically with this program, is that the interventions we provide are only as good as our patients' ability to use them. So it is not helpful or effective for us to check a list of things that we're supposed to do for suicide risk, without actually checking to see if those things are accessible to a patient. For example, if we recommend outpatient psychotherapy to treat depression, that the patient has no ability to pay, no transportation to get to that appointment--that is not going to be an effective evidence-based intervention for that patient. And so I think the relationship with PCCI has really allowed us to try to explore the overlap and those variables and suicide risk so that we can do a better job of providing the care that people actually need and can use more effectively.

Shania Kennedy 

I guess I should have touched on this a little bit earlier, but for the audience's sake, can you guys talk a little bit about what some of the risk factors that you guys have actually identified are in this use case, just for context?

Kimberly Roaten 

Sure. Not surprisingly, the risk factors that we've identified in the Parkland health system are very similar to the established risk factors for suicide. Generally, what's unique I think about this program and the Parkland Health population is that so many of our patients are from historically marginalized communities, people who have either not had access to appropriate health care or have had bad or traumatic experiences with health care. Recognizing the unique aspects of our patient population in this project has been really helpful for trying to figure out, again, the best way to deploy the things that people really need.

Shania Kennedy 

Yeah, and I think you've emphasized really well, the personalized aspect, because I think, obviously, when you're going through it, suicidal ideation is a very personal thing. And it's hard to explain it and talk about it at times. So knowing that the healthcare provider that you are going to, that you are relying on, is going to take a personalized approach to try to help you, I think, is probably--it would have been reassuring for me, and I'm sure a lot of people would also agree. So I think that's interesting as well. And of course, with any risk stratification program, it's going to have to be tailored to the patient population, just in general. But I think it's especially important for this.

Kelsey Waddill 

If I could just interject a quick question is from the previous question, but I was curious about: we talk a lot about how the workforce is so strained, especially in these areas. Time is such a valuable thing for doctors. And I was curious, especially sitting down and trying to work through what is available to the patient and what they could access seems like a very intensive process and I was curious, who sits down and does that with the patient? Is it the doctor? And if so, like, how do you manage that with the strain that they're already feeling, so much on their plate?

Kimberly Roaten 

Yeah, that's a great question. I think, time pressure in health care is such a prominent issue right now. And particularly in the high acuity environments, like the emergency department, that's a really challenging place to do this sort of work effectively without alienating your entire workforce.

So a couple of things, I think it really helped us with that, as part of our clinical decision support, we created a risk stratification tiered approach. So patients who say "no" to all of our screening questions, they don't get any further intervention. So that patient is probably getting maybe two or three additional minutes of questions from nursing staff, and they don't require a higher level of care from anybody else.

The next two categories are moderate and high risk. So I'll start with high risk patients, because that's a little bit more straightforward, too. So those are the patients who are endorsing items, like "I'm having thoughts of killing myself right now." So those patients are very intuitively people that we need to take care of and keep safe in this moment. So those are the patients that also require higher cost resources, like one-to-one observation, potentially inpatient, psychiatric hospitalization.

But all of our data is telling us that's extraordinarily rare, the vast majority of our patients do not require inpatient psychiatric care, we just need to get them connected to good outpatient care. So for those patients, for the patients in the moderate and high risk categories, those patients are typically offered the opportunity or required to participate in a full risk assessment, suicide risk assessment. And for those patients, that full risk assessment can be completed by a well-trained social worker and experienced psychologist or psychiatrist. So there are a lot of different people who have the kind of training to provide that sort of assessment and then provide the connection to the resources that the patient needs.

So I think other systems who are thinking about implementing these sorts of systems can think creatively about who can provide these interventions, assessment, and resources because it doesn't have to be a physician. It doesn't even have to be a psychologist. It can be somebody else with good mental health training.

I think maybe the other piece that I'll add to that is, one thing that can be really helpful with that is good community connections. So if the point person in front of the patient in the moment when the screening is conducted can connect the person with somebody in the community who then can do the higher level of care systems can spread the resource utilization and the time commitment in a way that's a lot more tolerable for a busy system.

Kelsey Waddill 

Yeah, and I bet if you have somebody who's dedicated, that it's already in their head what the resources are, instead of someone who has to then go on and search and figure it out, that speeds up the connection process. So that's great.

Shania Kennedy 

I was curious about that, Kelsey, as well, so it was glad that you brought that up because, yeah, doing something like this obviously requires a lot of resources and training and personnel. You can't do this kind of thing without obviously the people involved with all health care, but especially something like this, having that point of contact that's really there and really knowledgeable is so, so crucial.

But you guys have been doing this for almost a decade at this point. The program started in 2015. And I think the last that I looked you were up to, what, over 4 million screenings, so that's quite a bit, that is quite a bit of analyzing patient data and hearing patients stories, and obviously, you're flagging people who might have gone undetected and potentially without help and that's a huge deal already. But you've also, I presume, gained a sort of more nuanced understanding of at least the suicide risk for your population and you touched on this a little bit already. And obviously, you've learned how to adjust those care pathways and improve them over time.

With all that in mind, all of this knowledge that you've gained, can you talk a little bit or share a little bit about what you guys have learned since you've implemented the program, because I'd be really interested to hear that?

Kimberly Roaten 

So I feel like we've learned a lot. I, personally, have learned a lot from this program.

Some sort of key lessons that I just didn't think about when I was first getting this started--most of them for me, honestly, center around sustainability. Because some part of my brain, nine years ago thought, I've gotten the program started, "that's the hard part. We're good to go now." And [I] didn't really have the leadership or program development experience to realize that there would need to be continuous efforts to make sure--for example, in all of our hospital leadership turns over, they don't know what this program is, and they don't understand the value. And somebody has to be prepared to keep the momentum going, and to keep the enthusiasm for the program going. So I think that was a really important lesson for me and one that I try to share with other places, because getting it started as a heavy lift. But continuing to keep the program going is something that also requires pretty continuous effort as well.

One thing that's really important for sustainability is sharing the patient stories. So for systems who are considering implementing something like this, or something even smaller scale, going back to the frontline staff who are asking these difficult questions and giving them feedback when you have identified a patient who really needed help, that you might have missed without the questions, is really meaningful to staff, particularly nurses. So that's another piece of advice I would give to anybody trying to start something like this, whether it's related to suicide risk or not: those patient stories and good catches really make a difference for sustainability.

We also learned some sort of interesting concrete lessons that in retrospect now are funny. One of the stories that I like to tell most often is, when we first went live, the whole hospital is prepared for--we didn't know what because nobody had done this before. So we didn't know what to expect in terms of patients and numbers and all sorts of things.

And one of the most supportive groups when we started was our obstetrics and gynecology colleagues. And they are very thoughtful about how they care for their patients. They want to protect moms and babies and families. And so they were really supportive of the program. But they were the most frustrated right off the bat. And we couldn't figure out what was going on. They kept saying, "our nurses hate this. Our patients hate this."

Kelsey Waddill 

Oh no.

Kimberly Roaten 

So, my goodness, what are we doing? And we finally realized that we had inadvertently dropped the questions into the part of the intake process where, for the patients and the nurses, it essentially looked like: "tell me your name, tell me your date of birth, tell me when you had your last menstrual cycle, and do you have thoughts of killing yourself today?"

Shania Kennedy 

Oh.

Kimberly Roaten 

Basically, we just inadvertently sequenced the questions just like that. It again, it's funny now. At the time, we were like, "oh, my goodness, we need to fix this!" But learning those in-the-moment lessons about what is feasible, what makes this acceptable to patients acceptable to our staff asking questions. And I just feel like those lessons have been really interesting to learn and then share with others along the way.

Shania Kennedy 

That's so funny.

Kelsey Waddill 

[laughs] Yeah, timing is so important.

Shania Kennedy 

I can only imagine what it would be like to be like filling out the form and then be like, "oh, okay, that escalated quite quickly." Yeah, I can see how patients might not feel great about that one. That would be a little bit weird. But it is really important to consider though, yeah, like, when should this be incorporated so that it works for our patients? And that's going to be different for different health systems, I'm sure.

Kimberly Roaten 

And I think the part that Jackie will talk more about too, is that the other thing that I didn't really plan for at the beginning but I, fortunately, stumbled into a really good working relationship with is the data piece of this. So if you want to continue to sustain a program like this and to advocate for its viability, you have to have good access to data and good support from people who can help you analyze the data, especially when we're talking in the magnitude of what we're currently doing, which is looking at 7 million patient encounters to compare pre- and post-[programmed periods]. So Jackie can talk more about what that piece of it is look like.

Jacqueline Naeem 

Yeah. And I'd say first of all, I think the amount of data is one of the things that makes us most excited about this because I think you'd be hard pressed to find anyone anywhere that has the the wealth and the depth of data that we have for these patients. And so what we're doing at PCCI is we're able to look through the electronic health record, we're able to understand how many people scored positively on the assessment, how many were no risk, dive into some of the demographic things I talked about before, look at their utilization. So were they people that frequently come to the emergency room or where do they go? So understanding kind of the patient's story and also their screenings time as well has been really interesting.

Dr. Roaten alluded to the fact that we also are really fortunate now because--the hard question to answer is the outcome. Who are the people that went on to die by suicide? And without knowing that information, it's very hard to understand "who are those people?" Right? So, we have to know the end to understand what happened before. And so that's where we've been able to also pull in the state mortality data and be able to match that data across the Parkland patient population as well so we can understand that person's journey is those people that went on to die by suicide. What was happening to them beforehand? And what can we learn from that? And how can we share those learnings across [the system]? We're in the depths of that analysis right now. And it's really exciting. It's also we talk in our meetings about--there are so many questions we could ask, it's hard to stay focused in except because there are so many learnings with with kind of this wealth of data. So we don't have any answers for you today because we're in the midst of that analysis, but I think we will have some really interesting learnings from this program.

Kelsey Waddill 

Yeah. Could you share some of the questions that you're that? What are the ones that you decided "okay, these are the ones we're going to focus on?" How did you decide that? Because that seems like, you have such a wealth of data, where do you even start?

Jacqueline Naeem 

It's evolved as we've gone. I think what we're really curious to understand is, particularly for those patients, what were they presenting, like, in the emergency room? We want to understand, when they have had the assessment, what was the outcome of that? We are going to hopefully pull in some additional data to look at did they also touch hospitals outside of Parkland? We know what happens to them in Parkland, but [are] there also learnings we can pull in from that information as well. So I think really understanding as much as we can, any kind of trends or patterns, and also looking at rates of suicide so can we see any changes when we look over the years. So that's why we're doing the pre-programmed period and the post- periods. That's how we've got 7 million plus encounters that we're looking at. I'm feel super lucky that we get to be part of the work and the journey.

Kelsey Waddill 

Yeah, that's a hard tension between--you're looking at a lot of data that's represents real tragedy.

Jacqueline Naeem 

Yep.

Kelsey Waddill 

But in the light of all of the good that you can do to prevent future tragedies, that's really huge.

Jacqueline Naeem 

Yeah.

Kelsey Waddill 

So.... And also seems like a great opportunity for hospitals to coordinate or collaborate a little bit because you said that you're looking at other hospitals data as well. Is that just public data or...?

Jacqueline Naeem 

Hopefully. There's an organization DFW Hospital Council, and they collect data across all hospital systems across North Texas. So we're hoping that we can hopefully see some of that.

Shania Kennedy 

Yeah.

Jacqueline Naeem 

That's a whole process, though.

Kelsey Waddill 

I'm sure.

Shania Kennedy 

Yeah, I'm really interested--of course, on my site we cover analytics and AI, all that kind of thing, pretty much exclusively. But I'm really interested to see where you guys go with this. And what you end up focusing on and looking at just because yeah, like you said, there's so many aspects to this, that you can niche down and analyze. But picking the most important and the most actionable is the number one thing, is what I'm getting from what you're saying. And obviously, yeah, that makes sense. Of course, you're helping people.

We talked a little bit about what other health systems can do if they're looking to implement a program like this, or they're exploring how they can stratify risk in their own population. Do you guys have any advice for other health systems who are going down this path or looking into it that would be valuable to them?

Kimberly Roaten 

I think the biggest one to me is the community connections that I talked about a little bit earlier, just thinking creatively about what your resources really are. So many of us work in systems that don't have abundant mental health resources. But that doesn't mean that we can't get patients connected with what they need. We just may need to think more creatively, not only about our connections in the community that can be helpful--like crisis lines, lifelines, peer support, things like that--but also who's already operating in our system who could actually take part in this program?

So for us, that was our social workers. They really weren't doing a lot of our suicide risk assessments, but they can. It's part of their training, it's part of their competency, and we needed to ask them to do that. So I think thinking creatively about how to utilize the resources that already exist, rather than jumping to the assumption that it requires you hiring a lot more people or buying a lot more things to be effective.

And then the only other piece of advice I would give is, don't reinvent the wheel. Find a system that's similar to yours that's doing things that you hope to be able to do and ask lots of questions. You do not have to do this on your own. And you do not have to start from scratch.

Shania Kennedy 

People are always willing to share, trust me on that one. People who are doing this kind of work, like you guys, are usually very happy to talk about it. I would encourage anybody listening, if your health system is interested in something like this, connect [on] LinkedIn, whatever. You want to email me, go for it, because I can probably give you a little bit of information about where you might be able to look just because I have some connections. But yeah, leveraging the resources you have is really key, not just in this, but for anything like this, where you're doing analytics and risk stratification where, you know, sure, you could invest in all of these technologies and all that, but sometimes you really don't need to do that as much as you think you do. And I think that's a really important point.

Kelsey Waddill 

Yeah. Dr. Naeem, do you have any closing thoughts about what these organizations who may now be reaching out to us [laughs] are going to--what they could do from your perspective?

Jacqueline Naeem 

They need a Dr. Roaten to be somebody that is passionate about the work, who drives and drives the work. I think you do need to have a real belief and desire in doing this work and recognizing kind of the impact that it does have. That'd be my recommendation. But don't steal her. But somebody like her.

Kelsey Waddill 

Dr. Roaten, you might be getting your inbox flooded with lots of people asking for your help [laughs]. But the more information-sharing we can all do, the better off I think, hopefully, this will be so....

Shania Kennedy 

Alright, that's all we have for today's episode. Thank you again, both of you for coming on to chat with us about this.

And to all of our listeners, of course, thank you for tuning in. Without you guys, there wouldn't be a podcast, so I always appreciate it. And I always appreciate hearing from you guys, when you email me after the episodes. That's so fun. And if you do want to reach out to me, you can do so by emailing skennedy@xtelligentmedia.com. If you want to share your thoughts or if you want to share any other stories you think that we should consider for coverage. And again, that's s-k-e-n-n-e-d-y@xtelligentmedia.com. And if you like our conversation, let us know by following Healthcare Strategies on your favorite podcast platform and leave us a review.