Sex-Specific Association between Underlying Diseases and the Severity and Mortality Due to COVID-19 Infection: A Retrospective Observational Cohort Analysis of Clinical Epidemiological Information Collected by the Korea Disease Control and Prevention Agency

This study is a retrospective observational cohort analysis aiming to explore the relationship between underlying disease and the severity and mortality rate of coronavirus disease (COVID-19) by sex. As sample subjects, 5077 confirmed COVID-19 patients were selected. The dependent variable was each patient's clinical severity, dichotomized into two groups: clinical non-severity group and clinical severity group (including death group). Eleven underlying diseases were considered variables of interest, and each was dichotomized. Binary multivariate logistic regression model analyses were performed. Our results showed that the proportion of male patients (7.1%) in the clinical severity group was significantly higher than that of female patients (4.5%) and that the risk of being in the clinical severity group was higher in patients with specific underlying diseases. The underlying diseases varied: in males, rheumatism and autoimmune (adjusted odds ratio (aOR) = 6.69, 95% confidence interval (CI) = 1.60-27.98), dementia (aOR = 4.09, 95% CI = 2.14-7.82), cancer (aOR = 2.69, 95% CI = 1.27-5.69), and diabetes mellitus (aOR = 1.81, 95% CI = 1.18-2.77); in females, chronic kidney disease (aOR = 5.09, 95% CI = 1.87-13.86), dementia (aOR = 3.08, 95% CI = 1.18-5.23), diabetes mellitus (aOR = 1.87, 95% CI = 1.15-3.02), and hypertension (aOR = 1.73, 95% CI = 1.08-2.78). This study identified certain underlying diseases related to the high risk of being in clinically severe conditions and found that they differ between sexes. Prevention and treatment measure should be developed to reduce severity or mortality in confirmed COVID-19, based on underlying diseases and sex. However, further in-depth research is required to explore whether the findings and suggestions of this study can be generalized to other countries.