S19 Relevance of prediction scores derived from the SARS-CoV-2 first wave, in the UK COVID-19 second wave, for early discharge, severity and mortality: a PREDICT COVID UK prospective observational cohort study

Abstract

<h3>Introduction and Objective</h3> COVID-19 prognostication scores are all based on COVID-19 first wave, requiring prospective validation in the evolving pandemic due to SARS-CoV-2 variants (prevalent B.1.1.7 replacing parent D614) and healthcare responses altering patient demographic and mortality. Accelerated COVID-19 virtual hospital (VH) telemedicine model implementation avoids hospital admission, appropriately allocating hospital resources to pandemic needs in tandem with resumption of regular healthcare services, requires a safe triage tool. Widely used COVID-19 first wave derived prognostication scores, SOARS and 4C Mortality Score, with uncertain performance in the evolving pandemic, raises the need for prospective validation. We prospectively validate SOARS and 4C Mortality Score in the evolving UK COVID-19 second wave determining relevance for mortality and safe early discharge. <h3>Methods</h3> Protocol-based, prospective observational cohort study of SOARS and 4C Mortality Score in 1,383 PREDICT (single site) and 20,595 multi-site ISARIC (International Severe Acute Respiratory and Emerging Infections Consortium) patient cohorts during the UK COVID-19 second wave, between October 2020 and January 2021 <h3>Results</h3> Data from 1,383 patients (median age 67y, IQR 52–82; mortality 24.7%) in the PREDICT and 20,595 patients from the ISARIC (mortality 19.4%) cohorts showed SOARS had AUC of 0.8 and 0.74, while 4C Mortality Score had an AUC of 0.83 and 0.91 for hospital mortality, in the PREDICT and ISARIC cohorts respectively, therefore effective in evaluating both safe discharge and in-hospital mortality. 19.3% (231/1195, PREDICT cohort) and 16.7% (2550/14992, ISARIC cohort) with a SOARS of 0–1 were potential candidates for home discharge to a virtual hospital (VH) model. SOARS score implementation resulted in low re-admission rates, 11.8% (27/229), and low mortality, 0.9% (2/229), in the VH pathway. Use is still suboptimal to prevent admission, as 8.1% in the PREDICT cohort and 9.5% in the ISARIC cohort were admitted despite SOARS score of 0–1. <h3>Conclusion</h3> SOARS and 4C Mortality Score remains valid and relevant to their purpose, transforming complex clinical presentations into tangible numbers, aiding objective decision making, despite evolving viral subtype and treatment advances altering patient demographic and mortality. More importantly both scores are easily implemented within urgent care pathways for safe admission avoidance especially to a VH model.