Admission clinical parameters in predicting in-hospital mortality in COVID-19

Background: COVID-19 is a highly heterogenic disease. We aimed to investigate which clinical admission parameters predict disease-related mortality in COVID-19 patients.

Method: Admission clinical parameters in patients admitted between 10/03/20-09/04/20, with confirmed COVID-19 infection were recorded. Data was extracted from clinical records and 30-day mortality calculated.

Results: A total of 286 patients were identified: 63.3% male, 86.0% Caucasian, mean (SD) age: 68.1 (16.1) years, 60.5% were overweight, 38.5% had hypertension, 15.7% asthma, 11.2% chronic obstructive pulmonary disease, 19.2% diabetes and 13.6% ischaemic heart disease. The 30-day mortality rate was 24.5% and 16.8% required intensive care admission. Patients who died within 30 days post-admission had higher oxygen requirement (p<0.001), higher respiratory rate (p=0.008), higher CRP (p=0.023) and lower lymphocyte counts (p=0.049) on admission compared with those alive. Patients who died within 30 days were also older (p<0.001) and were more likely to have diabetes and COPD (p=0.047 and 0.030 respectively). Age (Odds ratio (OR): 1.05, p<0.001), oxygen requirement (OR: 1.03, p<0.001) and respiratory rate (OR:1.06, p=0.003) on admission were significantly associated with 30-day mortality in univariate logistic regression. An age ≥ 70 years (OR:3.4, 95%CI [1.9-6.1], p<0.001) and oxygen requirement on admission of ≥ 35% (OR:3.3, 95%CI [1.7-6.1], p<0.001) were independent predictors for 30-day mortality; 51.9% of patients who met both criteria died at 30 days.

Conclusion: COVID-19 poses high inpatient mortality, using clinical admission parameters to predict outcome may aid risk stratification and resource allocation.