Impact of age, sex, race, and regionality on major clinical outcomes of COVID-19 in hospitalized patients in the United States.

Background: The COVID-19 pandemic has affected all people across the globe. Regional and community differences in timing and severity of surges throughout the pandemic can provide insight into risk factors for worse outcomes in those hospitalized with COVID-19.
Methods: The study cohort was derived from the Cerner Real World Data (CRWD) COVID-19 Database made up of hospitalized patients with proven infection from December 1, 2019 through November 30, 2020. Baseline demographic information, comorbidities, and hospital characteristics were obtained. We performed multivariate analysis to determine if age, race, comorbidity and regionality were predictors for mortality, ARDS, mechanical ventilation or sepsis hospitalized patients with COVID-19.
Results: Of 100,902 hospitalized COVID-19 patients included in the analysis (median age 52 years, IQR 36-67; 50.7% female), COVID-19 case fatality rate was 8.5% with majority of deaths in those ≥ 65 years (70.8%). In multivariate analysis, age ≥ 65 years, male gender and higher Charlson Comorbidity Index (CCI) were independent risk factors for mortality and ARDS. Those identifying as non-Black or non-White race have a marginally higher risk for mortality (OR 1.101, CI 1.032-1.174) and greater risk of ARDS (OR 1.44, CI 1.334-1.554) when compared to those who identify as White. The risk of mortality or ARDS was similar for Blacks as Whites. Multivariate analysis found higher mortality risk in the Northeast (OR 1.299, CI 1.22-1.29) and West (OR 1.26, CI 1.18-1.34). Larger hospitals also had an increased risk of mortality, greatest in hospitals with 500-999 beds (OR 1.67, CI 1.43-1.95).
Conclusion: Advanced age, male sex and a higher CCI predicted worse outcomes in hospitalized COVID-19 patients. In multivariate analysis, worse outcomes were identified in small minority populations, however there was no difference in study outcomes between those who identify as Black or White.
(© 2022. The Author(s).)