Predictive Performance of the National Early Warning Score 2 for Stratification of Critically Ill COVID-19 Patients

Abstract

Aim: To validate the ability of National Early Waring Score 2 (NEWS2) for predicting the severity of Coronavirus disease-2019 (COVID-19). In addition, we also intend to examine the impact of pre-existing comorbidities to produce an advanced COVID-19 disease.Materials and Methods: A multicenter prospective cohort was performed on 108 patients having moderate-intensity COVID-19 infection during October 2020 and November 2021. NEWS2 parameters were recorded on admission to generate an output score, which then classified in accordance with the NEWS2 reference scale into low, medium, and high-risk categories. Each patient was followed till discharge or death for the clinical progression of COVID-19. The measures of validity and area under the curve (AUC) for NEWS2 threshold scores were calculated to predict the clinical deterioration of COVID-19.Results: Overall, 29.6% patients developed an advanced disease, out of which 21.8% patients died during treatment. NEWS2 score of 6 or more showed the highest sensitivity (78.1%), specificity (94.8%), and the AUC (0.838) for predicting an adverse outcome. Among comorbidities, the majority showed an increased risk of clinical deterioration.Conclusion: NEWS2 score of 6 or more at baseline showed good predictive ability to stratify patients with poor outcomes who may later require escalated care. However, we recommend more research to confirm our findings.