As per the outcomes of a large population-based study published in The Lancet Digital Health, the use of the QCovid risk prediction model displayed outstanding discrimination for COVID-19 linked deaths in both genders for the specified time periods. QCovid has the potential to be dynamically efficient with the evolution of the COVID pandemic and, hence, exhibits a promising role in guiding national policy.

Vahé Nafilyan and study researchers investigated the QCovid risk prediction algorithm for the estimation of mortality outcomes from SARS-CoV-2 infected adults. A cohort of people (34 897 648 adults) in the age group of 10 to 100 years as per the 2011 census was included.

The time to COVID-19 death (confirmed or suspected) as per the death certificate was considered as the primary outcome. This research study was stratified into two time frames January 24 to April 30 2020, and; May 1 to July 28 2020. QCovid’s performance was based on degrees of discrimination and calibration.

The 90-day risk of COVID-19 death helped estimate the calculated r2 values, Brier scores, and methods of discrimination and calibration over given time frames. On the whole, 26 985 (0·08%) and 13 177 (0·04%) COVID-19 deaths were noted during the first and second time frame, respectively. Good discrimination and calibration were found. The r2, D statistics, and C statistics values were high and similar (table 1) in both the time frames:

The second time period had similar outcomes. The sensitivity for identifying deaths in the first and second time frame was 65·94% and 71·67% for men and women, respectively, in the top 5% of patients with the highest anticipated probabilities of mortality.

Thus, Qcovid represents a novel prediction model that can be utilized for assisting population risk stratification with respect to public health interventions, such as vaccine usage.