Hi, everyone. My name is Anjali Garg, and I'm a third-year PEDS Critical Care Medicine Fellow at Johns Hopkins University in Baltimore, Maryland. I had to move the mic down just a little bit. I'm shorter than the rest of the people. I'd like to discuss my presentation on the Childhood Opportunity Index Amongst Critically Ill Children, a Multi-Center Retrospective Cohort Study. We have no conflicts, financial incentives, or other disclosures to discuss. I'd like to mention my mentors, Drs. Anthony Sauchet and David Stockwell, who I'd like to thank for all their help and guidance on my research. Our learning objectives include understanding health disparities and how they affect critically ill children, defining the Childhood Opportunity Index. Thank you, Justin, for starting that off for me. Understanding the potential relationship between COI and clinical outcomes, specifically in the pediatric ICU, and exploring areas for further perspective analyses. Social determinants of health are the non-medical factors that affect health outcomes. They're the conditions in which people are born, work, live, and age, and the wider set of forces and systems that shape conditions of daily life. Social determinants of health are an important influence on health inequities and have a huge impact on individuals, including their quality of life and daily activities. The five factors listed on the screen, economic stability, education access and quality, health care access, neighborhood and built environment, and social and community context, are important health disparities that have a basis in social and economic inequity. Specifically within the pediatric ICU, Mitchell and colleagues mentioned that with socioeconomic inequity and racial and ethnic inequity, you have increased PICU mortality, increased PICU readmission, and increased functional disability. Andrest and colleagues mentioned that children from impoverished neighborhoods have increased likelihood of PICU admission and increased severity of illness on presentation. However, existing metrics, such as zip code analyses to score a neighborhood's influence, do not account for enough. The Childhood Opportunity Index is a metric of local and regional neighborhood factors, and it highlights many different social determinants of health that affect a child's health and well-being. The Childhood Opportunity Index was developed by Diversity Data Kids as a United States U.S. population surveillance tool that encompasses specifics of a neighborhood's influence on child health and well-being. This comprehensive assessment of a community takes into account 29 variables that comprise a composite score of that child's health and environment. The variables are divided into three big buckets of categories, social and economic, health and environment, and education. And then these three large subcategories feed back into the overall composite Childhood Opportunity Index score that ranges from very low all the way up to very high. Our objective was to characterize the Childhood Opportunity Index amongst critically ill children and to explore for potential relationships between the Childhood Opportunity Index and clinical outcomes. Our study design was a multi-center retrospective cohort study. It was set in 47 pediatric ICUs within the Pediatric Health Information System Registry. Our study participants were children less than 18 years old that were hospitalized from January 2018 to December 2019. Our primary exposure was the Childhood Opportunity Index, ranging from COI level 1, which is very low, to COI level 5, very high. Our primary outcome measures were clinical outcomes, such as rates of index mortality, mechanical ventilation, and hospital length of stay. Our secondary outcome variables included a United States geographic regional variation analysis. Our statistical analysis included descriptive, comparative, and associative statistics. We had a total sample size of 26,912 patients that were hospitalized in the pediatric ICU from the years of January 2018 to December 2019. We saw that overall in the study cohort that there was more children that were admitted to the PICU from very low and low opportunity neighborhoods, at 25% and 20% respectively, compared to 18% of children who are admitted to the PICU from very high opportunity neighborhoods. When looking at the Childhood Opportunity Index breakdown by race and ethnicity, we note that Hispanic Latino, American Indian, and black children were more likely to be from very low and low opportunity neighborhoods. Notably, 48% of black children admitted to pediatric ICUs were from very low opportunity neighborhoods. When looking at the Childhood Opportunity Index breakdown by insurance, we note that children with Medicaid insurance were more likely to be from very low and low opportunity neighborhoods. When assessing our primary outcome variable, which is clinical outcomes, and looking at the variation by Childhood Opportunity Index, we note that children from very low and low opportunity neighborhoods had higher rates of mortality and higher rates of mechanical ventilation. We did not notice a difference between extracorporeal life support rates amongst COI levels. When looking at differences between length of stay by the Childhood Opportunity Index, we note that there is an inverse relationship between the Childhood Opportunity Index and length of stay. Expected linear regression by age and gender also shows this inverse relationship. We also did significant regional variation analysis, and we note that the east-south-central region had greater than 50% of their admissions from very low and low opportunity neighborhoods. And the same trend was noted in the west-south-central region as well, with almost 50% of their admissions being from very low and low opportunity neighborhoods. This is in comparison to the New England area, where 35% of their admissions were from very high opportunity neighborhoods. This heat map was created by Diversity Data Kids, and it's a more granular representation of the Childhood Opportunity Index, and it's based off the census track data from 2015. We note the same regional variation in this heat map that was seen in our study as well, so showing similar discrepancies in Childhood Opportunity Index regionally. So as we know, critically ill children were more commonly noted to be from lower Childhood Opportunity Index neighborhoods. They were more frequently from historically disadvantaged races and ethnicities, such as Hispanic, Latino, black, and American Indian populations. They more commonly had public insurance, and they clinically experienced worse clinical outcomes with higher rates of mortality and mechanical ventilation. We also noted significant regional variation, with the east-south-central region having the greatest number of PICU admissions from very low and low opportunity neighborhoods, and the New England region having the greatest proportion of admissions from very high opportunity neighborhoods. So in conclusion, the Childhood Opportunity Index is related to clinically relevant outcomes amongst hospitalized children in the pediatric ICU setting. Further analysis into specific sub-features of the Childhood Opportunity Index with those that have the greatest impact on clinical outcomes is necessary. For example, looking into improvement on access to medical resources in rural neighborhoods could potentially reduce the severity of illness that these children present with to the hospital. Our study is not without limitations. Collection of race and ethnicity data is not reported in PHIS, so it's unknown whether that data is self-reported or reported by a third party. PHIS participation and data entry are voluntary within the hospitals that participate in the database, and it's limited to tertiary and quaternary institutions, which could limit the generalizability of our findings, as well as one or two centers could represent a region. Only the overall Childhood Opportunity Index is provided in the registry, and PHIS does not include language preference data, which is a very important health disparity that we were not able to assess. Thank you so much for your time, and I welcome any questions. �

Childhood Opportunity Index

critically ill children

mortality rates

pediatric ICU

access to medical resources