In-Hospital Mortality and Resource Utilization of Patients with COVID-19 by Ethnicity

Abstract

Background

To date, investigations have focused on age-, race- and sex-related differences in COVID-19-associated outcomes. However, differences in COVID-19-associated outcomes related to ethnicity remain underrepresented in the literature.

Methods

In this clinical cohort study, we included adults hospitalized with COVID-19 and confirmed SARS-CoV-2 infection across the Northwestern Medicine health system in the Chicago area between March and June 2020. Individuals were initially identified utilizing NM Electronic Data Warehouse with details of presentation, symptoms, key therapies, and hospital outcomes verified by clinician case review. We assessed outcomes of oxygen use, hospital length of stay (LOS), ICU care and LOS, intubation and duration, and COVID-related death comparing across ethnic groups using multivariable Cox models.     

Results

Of the 1,947 adults hospitalized with COVID-19 included in our analyses, 37.49% identified as Latinx, and 24.16% preferred Spanish as their primary language. Compared with non-Latinx patients, Latinx patients were younger, more likely to be men and less likely to have asthma, COPD, T2DM, and CVD. Yet they required similar days of supplemental oxygen support and required a longer ICU LOS. In unadjusted time-to-event analyses, being Latinx was associated with a lower risk of COVID-related, in-hospital mortality (HR .512 [95% CI .370-709]). After adjustment for age, sex, insurance, comorbidities, and biomarkers, ethnicity was not independently associated with ICU admission, COVID-19-related mortality, or need for mechanical ventilation.

Conclusions

This study demonstrated that ethnicity is not an independent prognostic factor in patients hospitalized with COVID-19 disease after controlling for age, sex, insurance, comorbidities, and biomarkers.

Other

Funding source: Gilead Sciences Inc. [Prot#GS-US-540-5807]