Death Risk Score Model of Hospitalized COVID‐19 Patients: A Cohort Study

Introduction: The aim of this study was to establish a predictive model of risk of mortality in hospitalized COVID‐19 patients. Methods: Cohort study in hospitalized patients due to COVID‐19 followed‐up until death or discharge. They underwent clinical evaluation and laboratory tests on hospital admission and were investigated until death or hospital discharge. A death model risk was purposed based on clinical predictors on hospital admission. Results: A total of 215 patients were enrolled, with 39 deaths. The predictors of risk of mortality were age (relative risk (RR): 3,05 [1.65–5.63]), serum creatinine (RR: 7.39 [3.81–14.34]), peripheral blood lymphocyte counts (RR: 5.94 [3.78–9.31]), platelet counts (RR: 3.26 [1.81–5.9]), serum lactate dehydrogenase (LDH; RR: 5.85 [3.75–9.17), and C‐reactive protein (CRP; RR: 3.36 [1.66–6.79]). The death risk considering the punctuation (points) above was age > 64 years (1 point), platelet counts < 197,000 cell/mm 3 (1 point), LDH > 823 UI/L (1 point), lymphocytes < 373,000 cell/mm 3 (1 point), creatinine > 1.2 mg/dL (1 point), and CRP > 7.68 mg/dL (1 point). The death risk considering the punctuation (points) above was 6: 95%, 5: 85%–92%, 4: 50%–88%, 3: 29%–74%, 2: 17%–39%, and 1: 14%–31%. Conclusion: The routine laboratory on hospital admission may be a good predictor of mortality during the hospitalization follow‐up of COVID‐19 patients. A score risk may be adjusted and calculated based on these markers.