Algorithm Predicts Death, Hospital Admission From COVID-19

In HealthDay News
by Healthday

Final risk algorithm includes age, ethnicity, deprivation, body mass index, range of comorbidities

THURSDAY, Oct. 22, 2020 (HealthDay News) — A population-based risk algorithm performs well for predicting death and hospital admission due to COVID-19 in adults, according to a study published online Oct. 20 in The BMJ.

Ash K. Clift, M.B.B.S., from Radcliffe Observatory Quarter in Oxford, England, and colleagues conducted a population-based cohort study to derive and validate a risk prediction algorithm to estimate hospital admission and mortality outcomes for COVID-19 in adults. The derivation dataset included 6.08 million adults aged 19 to 100 years and the validation dataset included 2.17 million individuals. The derivation and first validation cohort period was Jan. 24 to April 30, 2020, while the second temporal validation cohort period was May 1 to June 30, 2020.

The researchers found that there were 4,384 deaths from COVID-19 in the derivation cohort and 1,722 and 621 in the first and second validation cohort periods, respectively. Age, ethnicity, deprivation, body mass index, and a range of comorbidities were included in the final risk algorithms. Good calibration was seen for the algorithm in the first validation cohort, explaining 73.1 percent of the variation in time to death for deaths from COVID-19 in men. The results were similar for women for both time to death and time to hospital admission and in both time periods.

“This study presents robust risk prediction models that could be used to stratify risk in populations for public health purposes in the event of a ‘second wave’ of the pandemic and support shared management of risk,” the authors write.

Several authors disclosed financial ties to the pharmaceutical and medical technology industries.

Abstract/Full Text

Editorial

Copyright © 2020 HealthDay. All rights reserved.