Research Scientist Harvard Pilgrim Health Care Institute Harvard Pilgrim Health Care Institute Cambridge, United States
Background: Systemic factors predispose United States (US) racial minority groups to a greater risk of severe COVID-19 outcomes.
Objectives: To determine the association between race and COVID-19 outcomes adjusted for baseline demographic, clinical, and socioeconomic characteristics.
Methods: We conducted a cohort study using data from the rapid COVID-19 Sentinel Distributed Database. We identified a cohort of individuals with COVID-19 defined as those with a COVID-19 specific ICD-10 diagnostic code or a positive PCR test between April 1, 2020, and March 31, 2021, and at least 6 months of enrollment in a health plan prior to cohort entry. This cohort was followed for 30 days for the occurrence of hospitalization with COVID-19. We identified another cohort of individuals hospitalized with COVID-19 and followed them, separately, for the occurrence of critical COVID, and all-cause inpatient mortality in 30 days. Critical COVID was defined as hospitalization plus admission to the ICU, on mechanical ventilator, ECMO, or with acute renal failure and receiving renal replacement therapy. Race was the primary independent variable, and it was self-reported in the database. Ethnicity was not well captured in the database. The association between race and COVID-19 outcomes was determined in Firth logistic regression models, where White race was the reference race category, after adjusting for age, sex, socioeconomic factors (income, property value, and unemployment in the neighborhood), and comorbidities.
Results: During the study period, we identified 770,924 individuals with COVID-19, 45.5% of whom were White, 42.1% of unknown race, 9.5% Black, 2% Asian, 0.5% Native Hawaiian or Other Pacific Islander (NHOPI), and 0.3% American Indian or Alaska Native (AIAN). All racial minority groups had increased odds of hospitalization compared to Whites (AIAN, adjusted Odds Ratio 1.45 [95% confidence interval 1.11-1.90]; Asian 1.58 [1.38-1.80]; Black 1.58 [1.40-1.78]; NHOPI 1.32 [1.22-1.42]). There were 128,167 individuals hospitalized with COVID-19, 56.2% of whom were White, 23.4% of unknown race, 17.6% Black, 1.5% Asian, 0.9% NHOPI, and 0.4% AIAN. Over half of all hospitalized individuals progressed to critical COVID (58%) and 14.2% died within 30 days. Progression to critical COVID was significantly higher in NHOPI compared to Whites (1.17 [1.03-1.32]). AIAN (1.52 [1.20-1.91]), Asians (1.30 [1.13-1.49]), and NHOPI (1.24 [1.06-1.45]) had significantly higher odds of 30-day inpatient mortality compared to Whites.
Conclusions: Our study based on a large and geographically diverse US database demonstrated significant racial differences in COVID-19 hospitalizations and inpatient mortality in the first year of the pandemic.
