Photo credit: Pixabay / CC0 Public Domain
In the first systematic large-scale assessment of the UK’s National Early Warning Risk Score (NEWS) 2 as a scoring system for predicting severe COVID-19 outcomes in patients, researchers at King’s College London found poor to moderate accuracy in identifying patients at risk, to be transferred to intensive care units or to die after 14 days of hospitalization. The accuracy of the predictions in the short term (three days) showed moderate success.
For people hospitalized with severe COVID-19, it is important to quickly identify which patients may be deteriorating and needing to be taken to an intensive care unit (ICU) for organ support or may die. NEWS2 is an early warning value that combines physiological parameters such as respiratory rate, oxygen saturation, blood pressure and temperature. NEWS2 is currently used almost everywhere in UK NHS trusts to identify which patients are at risk of premature deterioration.
The paper was published today in BMC Medicine.
Accuracy of NEWS2
The researchers analyzed data from 1,276 COVID-19 patients enrolled in the NHS Foundation Trust at King’s College Hospital during the first wave from March to April 2020. The team then validated their models using data for over 6,000 patients in eight other hospitals around the world (five in the UK, US). one in Norway and two in China). At all locations in the UK, around a third of patients with COVID-19 were admitted to intensive care or died within 14 days of hospitalization.
The researchers rated how well the patients’ NEWS2 readings measured at hospital admission expected who would have serious COVID-19 results, meaning they will either be admitted to the intensive care unit or die. At all UK locations, the combination of NEWS2 and Age to predict results showed moderate success in the short term (three days) but poor to moderate success in the medium term (14 days).
Dr. Ewan Carr, statistician at the Department of Psychiatry, Psychology and Neuroscience at King’s College London and co-lead author, said: “We conducted the largest study to date evaluating the accuracy of NEWS2 for predicting medium-term COVID outcomes NEWS2 Will Frequent used in UK NHS trusts but little is known about how well it can predict serious COVID outcomes so evaluating its accuracy is important as we strive to improve patient care now and in the future.
By collecting data from nine hospitals worldwide, our results have solid external validation. We found consistency across locations in both the performance of NEWS2 on its own and the complemented model. In short, NEWS2 tended to perform poorly but was improved with the addition of general blood and physiological measures. “
The research team used CogStack, an existing platform developed by NIHR Maudsley BRC and used at King’s College Hospital’s NHS Foundation Trust, which enables rapid extraction and processing of data from the patient’s electronic medical records.
Professor Richard Dobson, Head of Biostatistics and Health Informatics, NIHR Maudsley BRC, said: “The CogStack platform allows us to extract information in real time from the hospital records of the NHS Foundation Trust at King’s College Hospital to investigate really complex issues. This concludes For example, the possibility of extracting information about comorbidities that may only be mentioned in passing in the doctor’s narrative. “
Improving predictive ability
The researchers found that the accuracy of predicting severe outcomes was improved by taking into account routinely drawn blood and patient physiological parameters such as age, oxygen saturation, and neutrophil count, which are an important cell in the fight against the immune system during infections. In models that supplemented NEWS2 with these parameters, the ability to predict difficult outcomes was improved.
Dr. James Teo, consulting neurologist at King’s College Hospital and clinical director of data science and head of the CogStack platform, said, “Our results are the first to validate NEWS2 and show how it can be improved by adding common blood and physiological parameters. Fortunately, this is NHS -The assessment system can be easily adapted and implemented in clinical practice in comparison to other complex risk assessment models.
Along with all of the authors and staff, I would like to thank the patients and members of the King’s Electronic Record Research Interface who made this all possible. ”
Afro-Caribbean patients with severe kidney disease at higher risk of hospitalization due to COVID-19
Provided by the NIHR Maudsley Biomedical Research Center
Quote: NEWS2 was based on a large international study (2021, January 21), accessed on January 21, 2021 from https://medicalxpress.com/news/2021-01-news2-severe-covid-outcome-large explored predicting severe COVID-19 outcomes. html
This document is subject to copyright. Except for fair trade for the purpose of private study or research, no part may be reproduced without written permission. The content is provided for informational purposes only.