(787) Assessing Healthcare Disparities in Pediatric ECMO

Introduction: Healthcare disparities are associated with differences in outcomes in critically ill children, with several studies demonstrating these differences in respiratory illnesses, sepsis, oncology and cardiac surgery based on social determinants of health (SoDH). There is a paucity of literature evaluating these healthcare disparities in pediatric extracorporeal membrane oxygenation (ECMO). This project aimed to compare differences in clinical outcomes in children treated with ECMO based on social determinants of health.

Methods: We conducted a retrospective cohort study of patients who were placed on ECMO between January 2014 to December 2021 at a quaternary care pediatric hospital in the USA. Patients were excluded if there was no demographic information. Race, ethnicity, insurance type, child opportunity index (COI) and social vulnerability index were selected as factors of SoDH. The primary outcome of interest was mortality, with secondary outcomes of interest being duration on ECMO, ICU and hospital length of stay and ECMO related complications. A Kruskal Wallis and chi squared analysis was used to assess for independent associations between social determinants of health and and outcomes on ECMO.

Results: From January 2014-December 2021, there were 311 patients cannulated on ECMO at our institution. The majority of patients were male (58%), white (74%), and native English speakers (96%). The majority of patients had government insurance (57%) and had low or very low childhood opportunity index scores (42%). A majority of patients were placed on VA ECMO (80%). We found no significant differences in ECMO mortality (p=0.09), ECMO duration (p= 0.09), ICU length of stay (p=0.22) or total length of stay (p=0.32) based on COI. There is ongoing analysis from two additional ECMO centers, and the combined data will be used to create the multivariable regression model to assess for health care disparities, and then compare all three institutions for regional effects.

Conclusions: We did not find statistically significant differences in mortality and secondary outcome across different COI. We anticipate that there will be regional effect with different outcomes based on location. Further research is needed to better understand these differences in ECMO outcomes as well as regional effects.