Online Tool Identifies Patients at Highest Risk of Deterioration From COVID-19

January 13, 2021
Killian Meara

Researchers suggest that It could also potentially be used in other countries for risk-stratification.

Investigators from the UK Coronavirus Clinical Characterization Consortium have developed an online tool which employs risk-stratification to accurately predict the likelihood of deterioration in hospitalized adults with the coronavirus disease 2019 (COVID-19). The findings were reported in the journal The Lancet Respiratory Medicine.

The tool was recently made available to doctors from the National Health Service (NHS) at no cost, and aims to support the clinician’s decision-making process and ultimately helping to improve the outcomes of patients.

The work builds on the consortiums ‘4C Mortality Score’, which they developed to predict the percentage of death risk from COVID-19 subsequent of a hospital admission. The tool, called the ‘4C Deterioration Score’, was created using data gathered from 74,944 adults who were diagnosed with COVID-19 and were admitted to 260 hospitals across the UK. It uses 11 measurements that are routinely collected from patients to calculate a percentage risk of deterioration.

The 11 measurements used include age, gender, physical measurements such as oxygen levels and standard laboratory tests.

"Accurate risk-stratification at the point of admission to hospital will give doctors greater confidence about clinical decisions and planning ahead for the needs of individual patients,” Mahdad Noursadeghi, co-senior author on the study said. “The addition of the new 4C Deterioration Score alongside the 4C Mortality Score will provide clinicians with an evidence-based measure to identify those who will need increased hospital support during their admission, even if they have a low risk of death."

Investigators used a multivariable logistic regression model that used various measures to predict an outcome and tested the 11 measures against the patient cohort to establish to what degree and how the measures impacted the likelihood of deterioration. They also assessed how the toll performed in other regions of the NHS and found that the results were similar, suggesting it has the potential to work across the NHS.

"The scale and wide geographical coverage of the ISARIC4C study across the country was critical to the development of this prediction tool,” Rishi Gupta, first author on the study said. “Our analysis provides very encouraging evidence that the 4C Deterioration tool is likely to be useful for clinicians across England, Scotland and Wales to support clinical decision-making."