Late Breaking Abstract - Relevance of first wave derived COVID-19 prediction scores, in the UK second wave, for mortality and safe early discharge

Abstract

<b>Introduction:</b> COVID-19 prognostication scores derived from the first wave requires prospective validation due to the evolving second wave with prevalent B.1.1.7 variant altering demographic and outcome. COVID-19 virtual hospital (VH) models preventing hospital admission requires a safe triage tool. <b>Objective:</b> Prospectively validate relevance of&nbsp;COVID-19 first wave derived prognostication scores, SOARS and 4C Mortality Score, to determine mortality and safe early discharge in the evolving second wave. <b>Methods:</b> Prospective observational cohort study of SOARS and 4C Mortality Score in 1383 single site (PREDICT) and 20595 multi-site (ISARIC) patient cohorts, in UK second wave. <b>Results:</b> 1383 (median age 67y, mortality 24.7%) and 20595 (mortality 19.4%) patient cohorts had AUC for mortality of 0.8 and 0.74 (SOARS) and 0.83 and 0.91 (4C Mortality Score) in the PREDICT and ISARIC cohorts respectively. 19.3% (231/1195 PREDICT cohort) and 16.7% (2550/14992 ISARIC cohort) with SOARS of 0-1 were VH candidates. Applying SOARS in the VH pathway resulted in low re-admission, 11.8% (27/229), and low mortality, 0.9% (2/229). Use is still suboptimal, as 8.1% in PREDICT and 9.5% in ISARIC cohorts were admitted despite SOARS of 0-1. <b>Conclusion:</b> SOARS and 4C Mortality Score remains valid despite SARS-CoV-2 variants altering mortality. Both scores should be implemented for admission avoidance through a VH platform.