Phenotypic Characteristics and Development of a Hospitalization Prediction Risk Score for Outpatients with Diabetes and COVID-19: The DIABCOVID Study

Abstract

Diabetes mellitus (DM) has been identified as a risk factor for severe COVID-19. DM is highly prevalent in the general population. Defining strategies to reduce the health care system burden and the late arrival of some patients thus seems crucial. The study aim was to compare phenotypic characteristics between in and outpatients with diabetes and infected by COVID-19, and to build an easy-to-use hospitalization prediction risk score. This was a retrospective observational study. Patients with DM and laboratory- or CT-confirmed COVID-19, who did (n = 185) and did not (n = 159) require hospitalization between 10 March and 10 April 2020, were compared. Data on diabetes duration, treatments, glycemic control, complications, anthropometrics and peripheral oxygen saturation (SpO2) were collected from medical records. Stepwise multivariate logistic regressions and ROC analyses were performed to build the DIAB score, a score using no more than five easy-to-collect clinical parameters predicting the risk of hospitalization. The DIAB score was then validated in two external cohorts (n = 132 and n = 2036). Hospitalized patients were older (68.0 ± 12.6 vs. 55.2 ± 12.6 years, p 2, 9.7 vs. 3.5%, p = 0.03), hypertension (81.6 vs. 44.3%, p p = 0.009), and lower SpO2 (91.6 vs. 97.3%, p p 27 points predicted hospitalization (sensitivity 77.7%, specificity 89.2%, AUC = 0.895), and death within 28 days. Its performance was validated in the two external cohorts. Outpatients with diabetes were found to be younger, with fewer diabetic complications and less severe obesity than inpatients. DIAB score is an easy-to-use score integrating five variables to help clinicians better manage patients with DM and avert the saturation of emergency care units.

Keywords

• COVID-19
• diabetes
• outpatients
• hospitalization risk score
• DIABSCORE