Hospital segregation, critical care strain, and inpatient mortality during the COVID-19 pandemic in New York City

Abstract

BackgroundHospital segregation by race, ethnicity, and health insurance coverage is prevalent, with some hospitals providing a disproportionate share of undercompensated care. We assessed whether New York City (NYC) hospitals serving a higher proportion of Medicaid and uninsured patients pre-pandemic experienced greater critical care strain during the first wave of the COVID-19 pandemic, and whether this greater strain was associated with higher rates of in-hospital mortality.MethodsIn a retrospective analysis of all-payer NYC hospital discharge data, we examined changes in admissions and inpatient mortality, stratified by use of intensive care unit (ICU), from the baseline period in early 2020 to the first COVID-19 wave across hospital quartiles (265,329 admissions and 23,032 inpatient deaths), based on the proportion of Medicaid or uninsured admissions from 2017-2019 (quartile 1 lowest to 4 highest). Logistic regressions were used to assess the cross-sectional association between ICU strain, defined as ICU volume in excess of the baseline average, and patient-level mortality.ResultsICU admissions in the first COVID-19 wave were 84%, 97%, 108%, and 123% of the baseline levels by hospital quartile 1-4, respectively. The age-adjusted mortality rates for ICU admissions were 269%, 353%, 375%, and 387%, and those for non-ICU admissions were 355%, 500%, 633%, and 843% of the baseline rates by hospital quartile 1-4, respectively. Compared with the reference group of 100% or less of the baseline weekly average, ICU admissions on a day for which the ICU volume was 101-150%, 151-200%, and > 200% of the baseline weekly average had odds ratio of 1.17 (95% CI=1.10, 1.26), 2.63 (95% CI=2.31, 3.00), and 3.26 (95% CI=2.82, 3.78) for inpatient mortality, and non-ICU admissions on a day for which the ICU volume was 101-150%, 151-200%, and > 200% of the baseline weekly average had odds ratio of 1.28 (95% CI=1.22, 1.34), 2.60 (95% CI=2.40, 2.82), and 3.44 (95% CI=3.11, 3.63) for inpatient mortality.ConclusionsOur findings are consistent with hospital segregation as a potential driver of COVID-related mortality inequities and highlight the need to desegregate health care to address structural racism, advance health equity, and improve pandemic resiliency.