My name is Scott. I have no disclosures for this presentation. I did want to begin by acknowledging my wonderful colleagues at Children's National, without whom this endeavor would not be possible. I am a second year pediatric critical care fellow. I previously completed a combined med-peds program at Cincinnati. And while I have multiple clinical interests, today I'll be spending our time talking about social determinants of health and their impact on clinical outcomes in asthma. Specifically, I'll be discussing a pilot study in which we implemented a new screening protocol for social determinants within our PICU. Our aims were fourfold. One, to identify unmet social needs in patients with asthma. Two, to better understand the social contexts that shape their health outcomes. Three, to recognize health disparities that arise because of these unmet social needs. And four, to ultimately provide resources within the community to help reduce these disparities. Now, before diving into the findings of our pilot study, I did want to take a step back and just discuss the problem. So here's what we know. Social determinants like housing instability and food insecurity have a negative impact on health. It's estimated that up to 50% of health outcomes are actually related to these underlying determinants. And we know that folks who live with adverse social circumstances experienced increased rates of infant mortality, increased rates of chronic disease, and overall shorter life expectancies. Unfortunately, children are not immune to this phenomenon. Children living in poor communities with more limited opportunities for social advancement utilize the health care system at much higher rates than their counterparts. And we see a fairly clear connection between unmet social needs and certain critical outcomes with diseases like asthma, which we know are highly influenced by modifiable environmental factors. However, screening requires a lot of effort and resources, so much so that it proves prohibitive for a lot of institutions and providers. I think there is also a mental toll to contend with. We are obviously, as individuals, not going to cure poverty. We can't independently guarantee quality education for everyone. We are not going to fix the social and structural biases that unfairly stack the deck against so many of our most vulnerable patients. And we don't have the ability, as much as we would like, to guarantee that children can go home to a stable environment, a safe neighborhood. So it's easy to stay confined within our comfort zone and to remain siloed within our areas of expertise. But here's why it's important to try to tear down those barriers. The organizations that we partner with at Children's National are undeniably influencing the lives of patients within our community. So Others May Eat, last year alone, provided stable housing for nearly 1,300 residents, among them almost 300 children. House of Ruth, which provides services for victims of interpersonal violence, was able to help about 64 families per night throughout the region, including about 100 children. Capital Area Food Bank distributed a very impressive 52 million meals in 2020, among them about 300,000 emergency food boxes that went straight to families' homes. And you can peruse the websites of these organizations, read the testimonials, and see how these interventions are having impacts on real human beings. So if we can change the trajectory on an individual level, it stands to reason that we can change things macroscopically if we utilize our resources in a thoughtful and deliberate manner. Put another way, if we extend enough lifelines to enough people, help them navigate the system, we can potentially change meaningful health outcomes in a statistically significant way. And if we can do that to a significant degree, perhaps we can prove a cost-benefit analysis, which is ultimately what we need in order to gain large-scale buy-in. So with that in mind, we began down a much longer path of data collection and analysis in which we wanted to first prove feasibility and utility, recognizing the significant amount of resources that would go into this endeavor. So we took a critical care bent and specifically we're hoping to partially illuminate answers to the following questions. One, are pediatric ICU stays more complicated in patients with unmet social needs? And we'll go into what that means exactly. And do they stay in the ICU or the hospital for a longer period of time as a consequence? And then two, are follow-up rates lower, are readmission rates higher in patients with unmet social needs? And given the established connection between asthma and critical care outcomes, we decided to focus on this patient population in hopes of revealing a disparity and a potential target for future interventions. You can see our inclusion and exclusion criteria listed here. Now, it's important to keep in mind that we actually artificially limited our patient population initially in hopes of not overwhelming our capabilities. And we did so by making a definition for severe asthma that was meant to be something easily querable by the folks who were administering the questionnaire. So these aren't necessarily the things that we would look at for severe asthma scores in general, but we wanted it to be an easy to utilize method for the folks administering the questionnaire. So based off our definition, folks were ultimately screened if they required any advanced medical therapy, which included things like terbutaline, isoproteranol, or isoflurane, if they were on BiPAP for more than 24 hours, or if they had any duration of invasive mechanical ventilation. We used a homegrown screener, which was actually developed by outpatient social workers at our primary asthma clinic called Impact DC. They originally designed this questionnaire to target things that could be immediately intervened upon by the medical team. So it is by no means comprehensive, and it also excludes certain domains that are captured through other means in our hospital, things like interpersonal violence, drug use, et cetera. Ultimately, folks who screened positive using this questionnaire met with a family services associate who would provide applicable resources to them based on their specific needs. As we went along, we collected certain screening metrics, performance metrics, to better understand the strengths and weaknesses of our strategy and its impact on overall workflow. We also looked at the social needs of our particular patient population with a focus on what we characterized as high-risk patients, those folks with three or more unmet social needs. In addition to the geomapped resources that we provided everyone with needs, these individuals also received a social work referral for more nuanced and visualized assistance. Most importantly, we looked at outcome metrics to better understand how folks with unmet social needs fared compared to their counterparts. And we looked at specifically BiPAP duration, PICU length of stay, hospital length of stay, follow-up rates, rates of visitation to the ED within six months after presentation, and readmission rates within six months. Lastly, we looked at and collected potential confounders, which included basic demographic information, medical comorbidities like obesity and obstructive sleep apnea, insurance status, as well as other surrogate markers of social complexity, including estimated income and childhood opportunity indices. During our three-month trial period, a total of 98 patients presented to the PICU with asthma. Because our screening was limited to business hours, we missed about 22 of these patients. Of the remaining 76 patients, half of them were ultimately deemed eligible. We did have quite a few wrinkles to iron out initially. So some patients who were eligible were actually missed initially, and some ineligible patients were screened. But collectively, eight of these screened patients were deemed high-risk and received a referral to social work. On average, it took about 42 minutes to screen patients, and that increased in time as the number of social needs increased. Nearly a third of our patients reported a need with housing or utility assistance, followed by food insecurity, employment support, and finally, public benefits assistance. In terms of clinical outcomes, on average, patients with unmet social needs spent about 39 hours on BiPAP. Those without needs spent about 37 hours. The difference of 1.23 hours was not statistically significant. Additionally, patients with unmet social needs spent slightly longer in the ICU, about three full days, versus 2.8 days for those without needs. Again, not statistically significant. And similarly, they spent longer in the hospital, though this was also statistically insignificant. But keep in mind, this was all initial trial data, not really power to detect the difference between the two groups. Now, for the next couple of slides, I am actually including data to date. We've started analyzing what we have, and it's just a little bit more interesting. So in terms of follow-up, about 41% of patients with unmet social needs followed up with a specialty service after their hospital admission, whether it was pulmonology, Impact DC, or allergy and immunology, versus about 51% of patients without social needs. So while this did not meet our threshold for statistical significance, there did seem to be a trend, which I've decided in the last 30 minutes that I'm okay using in this context, since it is very preliminary data. Additionally, and most interestingly, I would say, is that those patients who screened positive were much more likely to present to the ED within six months of discharge. An impressive 74% of these patients presented back to our ED within that timeframe, versus only about 32% of patients with no social needs. However, when looking at readmission rates to the hospital, it was nearly identical between the two groups. So as mentioned, this is really just the start of a much longer endeavor. Within the next 12 months, we plan, or sorry, actually starting next month, we plan to look at the past 12 months of data, really trying to hone in on whether there is a difference between these two groups. But in addition, we are planning on looking at patients who were screened and given applicable resources versus those patients who were not screened, who were missed because of our staffing limitations, to see if there's a difference in downstream effects between those two groups. That's probably the thing that I'm most interested in looking at. Around the same time, we are transitioning over to electronic screening, and we're gonna look at whether or not frequency of reported social needs differs with anonymous electronic screening, realizing that some of these topics are a little bit stigmatizing. And we're also going to look at the effects of electronic screening on workflow. Within the next six months or so, we are going to start looking at patients who've been admitted and screened, who then report to Pulmonology Impact DC, looking at their social circumstances reported on the social screener when they follow up, seeing if there is any difference, if we've made any impact on these patients, essentially, and if that differs between the groups. And then for those folks whose social circumstances did not change, we are going to follow up to get some narrative information about what barriers they experienced in order to better understand what is causing issues with sustainable change. So with that, I would again like to thank my colleagues. There is still much more work to be done. I will gladly accept any questions, comments, or suggestions for future work. Thank you so much.

social determinants of health

clinical outcomes

asthma

screening protocol

unmet social needs