RWD152 Real-World Evidence on the Use of Ecmo in Hospitalized COVID-19 Patients: Results from a Linked Closed and Open Claims Database in the US

Abstract

To describe the use of extracorporeal membrane oxygenation (ECMO) among hospitalized coronavirus disease 2019 (H-COVID-19) patients in a linked closed claims (CC) and open claims (OC) database. This analysis identified H-COVID-19 patients between April 2020 (Q2 2020) and June 2022 (Q2 2022) in CHRONOS, a linked CC and OC database. The index event was the date of hospitalization, defined as an inpatient claim within 21 days of a COVID-19 diagnosis in the CC. The occurrence of ECMO 30 days after index was identified using CC data alone and then CC and OP data in combination to assess missing data. Study exclusions included patients under the age of 18, a first COVID-19 diagnosis that did not result in hospitalization, and less than 12-months of continuous enrollment in the CC before index. Study criteria were defined by the presence of an ICD-10-CM, ICD-10-PCS, or CPT code on a claim. Results are reported as percentages and 95% confidence intervals. Of 321,687 patients with H-COVID-19, the mean age was 50.1 (SD:12.8) with the highest proportion of hospitalizations occurring in Q3 2021 (19.4%). Overall, 0.50% (0.48%-0.52%) of patients in the CC data received ECMO, increasing to 0.61% (0.58%-0.64%) with the inclusion of OC data. The use of ECMO to treat H-COVID-19 patients decreased between Q2 2021 and Q2 2022, with the highest rates occurring in Q2 of 2020 (0.78%) and Q2 2021 (0.80%). The addition of OC data increased rates to 1.12% and 0.89% in Q2 of 2020 and Q2 2021. Although use of ECMO decreased in the later months of the pandemic, it represents a substantial burden. The current analysis demonstrates that CC data, often sourced from payers, may underestimate the use of ECMO in real-world settings. Opportunities exist to mitigate issues of missing data by linking CC, OC, and other real-world data sources.