Document Type

Conference Proceeding

Publication Date


Publication Title

Journal of General Internal Medicine


BACKGROUND: The COVID-19 pandemic has resulted in over 1 million deaths globally. Prognostic tools to identify high risk patients are crucial to guide resource allocation efforts. We aimed at developing a risk assessment tool for patients with COVID-19 based on the risk factors with most significant effect on hospital admission and in-hospital mortality METHODS: We performed a retrospective analysis of patients with positive COVID-19 presenting in between 3/31/2020 – 5/15/2020 at Beaumont Health’s 8 emergency departments (ED). Data was abstracted using automated reports. The electronic health record (HER) embedded risk score previously externally validated was modified based on risk factors, with different points given those that were statistically significant. Two outcome variables were measured, both using a yes/no binary scale: hospital admission and in-hospital mortality. Hospital admission, on the first encounter to the ED, was evaluated for the entire cohort, while mortality was evaluated only for inpatients discharged prior to 5/12/2020. Descriptive statistics, univariate/multivariate analyses by logistic regression were performed and presented in terms of Adjusted Odds Ratios (AOR) with corresponding 95% confidence intervals and P-Values. Any P-Values < 0.05 were considered as statistically significant associations. All analysis was done in SAS 9.4 (SAS Institute Inc. Cary, NC). RESULTS: 2,735 encounters were extracted from EHR. 68.06% were hospital admissions and 9.97% experienced in-hospital mortality. 61.23% were <69 years old. 58.07% had hypertension (HTN), 46.29% had chronic pulmonary disease (CPD), 37.81% had diabetes (DM), and 6.71% had end-stage renal disease (ESRD).Mean length of stay was 8.43 days. In the multivariate model to predict admission, ESRD (AOR 1.97), liver disease (AOR 7.77), CPD (AOR 1.63), DM (AOR 1.70), HTN (AOR 1.97) and nursing home residence (NH) (AOR 1.90) were independently associated with admission. For prediction of in-hospital mortality in the multivariate model, CPD (AOR 2.35), and NH (AOR 1.58) were significantly associated with in-hospital mortality. The modified risk score recognized the statistically significant comorbid conditions and attributed 0 points to non-significant values. The cross-validated CStatistics for the modified risk score model showed good discrimination for both hospital admission (C=0.72 vs 0.70) and in-hospital mortality (C=0.74 vs 0.70) when compared to the automatically generated risk tool for this cohort. CONCLUSIONS: The modified risk score model created using statistically significant risk factors yielded a better scoring system than the scoring system automatically generated in Epic. This risk scoring model may help predict hospital admissions and in-hospital mortality for COVID-19 patients. Further external validation in a different cohort is recommended.




Suppl 1

First Page


Last Page