Background: Two oral antiviral therapies have recently received FDA emergency authorization for COVID-19 patients at high risk of disease progression. Limited real-world data are available on the medical conditions associated with hospitalization.
Objectives: To assess the rate of hospitalization and explore the baseline medical conditions associated with the risk of COVID-19 progression.
Methods: Patients presenting with a diagnosis of COVID-19 in the outpatient setting (index) were identified using the Optum® de-identified Clinformatics® Data Mart database between 04/01/2020 and 05/31/2021. All diagnoses prior to index were extracted from a 12-month baseline period with continuous enrollment and no evidence of pregnancy or prior COVID-19 infection. Patients were followed until inpatient hospitalization or 30 days following index. Those who died or disenrolled prior to day 30 were censored. Any baseline comorbidity with at least a 5% sample prevalence was included as a covariate in a regularized (LASSO) cox regression model predicting progression to hospitalization, controlling for demographics and year-quarter of diagnosis.
Results: Of 487,060 patients identified, 16,392 were hospitalized within 30 days (3.39 hospitalizations per 100 patient months) and 3.05% were censored. 76.51% of hospitalizations occurred ≤10 days of index, with pneumonia or other diseases of the respiratory system present in 20.14%. Diseases of the circulatory system were the most common (18.84%) diagnosis for hospitalizations >20 days after index. A regularized cox regression identified 29 highly correlated baseline diagnoses including known risk factors such as hypertension, diabetes, kidney disease, COPD, cardiovascular disease, liver disease and obesity. Sleep apnea (HR: 1.20, 95% CI: 1.14-1.26), fluid/electrolyte disorders (HR: 1.15, 95% CI: 1.08-1.21) and history of mental health or substance abuse (HR: 1.10, 95% CI: 1.05-1.15) were also highly predictive.
Conclusions: Using real-world data and machine learning techniques, this study confirmed previously known risk factors, while identifying additional diagnoses, that increase the risk of COVID-19 hospitalization. If confirmed, these findings could be added to preventative strategies for reducing the risk of severe illness.
