NEWS2 was evaluated for predicting serious COVID-19 outcomes in a large international study

Researchers at King’s College London identify patients in the first systematic large-scale assessment of the UK National Early Warning Risk Score (NEWS) 2 as a scoring system for predicting severe COVID-19 outcomes in patients We have found that the accuracy for is low to medium. There is a risk of being transferred to the intensive care unit (ICU) or dying after 14 days of hospitalization. The accuracy of the short-term (3 days) forecast showed moderate success.

For those hospitalized with severe COVID-19, it is important to quickly identify which patients are getting worse and need to be transferred to. intensive care unit (ICU) For organ support or may die. NEWS2 is an early warning score that combines physiological parameters such as respiratory rate and oxygen saturation. blood pressure And temperature. NEWS2 is now almost universally used by the UK NHS Trust to identify which patients are at risk of premature exacerbations.

The treatise is today BMC medicine..

NEWS2 accuracy

Researchers analyzed data from 1,276 COVID-19 patients admitted to the Kings College Hospital NHS Foundation Trust during the first wave of March-April 2020. The team then validated the model using data from more than 6,000 patients in eight other hospitals around the world (five in the UK and five in the UK). 1 in Norway, 2 in China). At all UK facilities, approximately one-third of COVID-19 patients were transferred to the ICU or died within 14 days of admission.

Researchers evaluated how well the patient’s NEWS2 score measured on admission can predict patients with severe COVID-19 outcomes, which means they will be transferred to the ICU or die. .. Predicting results by combining NEWS2 with age on all UK sites showed moderate success in the short term (3 days), but moderate to moderate success in the medium term (14 days) was.

Dr. Ewan Carr, a statistician researcher and co-author of the King’s College London Institute of Psychiatry and Neuroscience, said: Although .NEWS2 is widely used by NHS trusts in the United Kingdom, little is known about how well it can predict the consequences of severe COVID, so when considering improving current and future patient care , It is important to evaluate its accuracy.

By collecting data from nine hospitals around the world, our results have strong external validation. We found that both the NEWS2-only performance and the supplemental model were consistent across sites. In short, NEWS2 tended to perform moderately, but was improved by adding common blood and physiologic measurements. “

The research team used CogStack, an existing platform developed by NIHR Maudsley BRC and used by King’s College Hospital NHS Foundation Trust. This allows you to quickly extract and process patient data. Electronic health record..

Professor Richard Dobson, Head of Biostatistics and Health Informatics at NIHR Maudsley BRC, said: hospital Record in real time at King’s College Hospital NHS Foundation Trust to actually investigate these complex questions. This includes, for example, the ability to extract information related to comorbidities that may only be mentioned when passing a doctor’s explanation. “

Improved predictive ability

Researchers have serious outcomes by considering blood and physiological parameters that are regularly collected from patients, such as age, oxygen saturation, and the number of neutrophils, which are immune system cells that fight important infections. We have found that the accuracy of predicting is improved. Models with these parameters added to NEWS2 have improved ability to predict serious outcomes.

Dr. James Theo, a consultant neurologist at King’s College Hospital, a clinical director of data science, and a leader in the CogStack platform, said: Physiological parameters.. Thankfully, this NHS scoring system is easy to adapt and implement in real-world clinical practice compared to other complex risk scoring models.

I would like to thank all the authors and collaborators for the guidance and contributions of the Patients and King’s Electronic Recording Research Interface members who made this all possible. ”

Provided by NIHR Moseley Biomedical Research Center