DERIVATION AND VALIDATION OF A CLINICAL SCORE TO PREDICT DEATH AMONG NON-PALLIATIVE COVID-19 PATIENTS PRESENTING TO EMERGENCY DEPARTMENTS: THE CCEDRRN COVID MORTALITY SCORE

Hohl, C. M.; Rosychuk, R. J.; Archambault, P. M.; O'Sullivan, F.; Leeies, M.; Mercier, E.; Clark, G.; Innes, G. D.; Brooks, S. C.; Hayward, J.; Ho, V.; Jelic, T.; Welsford, M.; Sivilotti, M. L.; Morrison, L. J.; Perry, J. J.; Network, Canadian COVID-19 Emergency Department Rapid Response

Background: Predicting mortality from coronavirus disease 2019 (COVID-19) using information available when patients present to the Emergency Department (ED) can inform goals-of-care decisions and assist with ethical allocation of critical care resources. Methods: We conducted an observational study to develop and validate a clinical score to predict ED and in-hospital mortality among consecutive non-palliative COVID-19 patients. We recruited from 44 hospitals participating in the Canadian COVID-19 ED Rapid Response Network (CCEDRRN) between March 1, 2020 and January 31, 2021. We randomly assigned hospitals to derivation or validation, and pre-specified clinical variables as candidate predictors. We used logistic regression to develop the score in a derivation cohort, and examined its performance in predicting ED and in-hospital mortality in a validation cohort. Results: Of 8,761 eligible patients, 618 (7{middle dot}01%) died. The score included age, sex, type of residence, arrival mode, chest pain, severe liver disease, respiratory rate, and level of respiratory support. The area under the curve was 0{middle dot}92 (95% confidence intervals [CI] 0{middle dot}91--0{middle dot}93) in derivation and 0{middle dot}92 (95%CI 0{middle dot}89--0{middle dot}93) in validation. The score had excellent calibration. Above a score of 15, the observed mortality was 81{middle dot}0% (81/100) with a specificity of 98{middle dot}8% (95%CI 99{middle dot}5--99{middle dot}9%). Interpretation: The CCEDRRN COVID Mortality Score is a simple score that accurately predicts mortality with variables that are available on patient arrival without the need for diagnostic tests.