Respiratory Data from 175K Covid-19 Patients is Base of Severity Prediction

According to a study published in JAMA Network Open, the most extensive data Repository of COVID-19 patients in the United States was utilized to construct a model to predict clinical severity based on first-day admission data. The study used data from almost two million medical records held in the National COVID Cohort Collaborative’s Data Enclave, or N3C. The study was the first to employ the N3C database, which was created expressly for COVID-19 research.

The National Center for Advancing Translational Sciences, a center for this type of research at the National Institutes of Health, created the N3C. The N3C release set contains 1,926,526 patients from 34 sites across the United States as of December 2020. According to the NCATS website, the electronic health record Repository included data from 6.3 million patients as of July 2021.

Researchers said, “This cohort is racially and ethnically diverse and geographically distributed. We evaluated COVID-19 severity and associated clinical and demographic factors over time and used machine learning to develop a clinically useful model that accurately predicts severity using data from the first day of hospital admission.”18.6 percent of the nearly 175,000 adults who tested positive for the coronavirus in the research were admitted to the hospital. Among them, 6,565, or approximately a fifth, had a “severe clinical course,” including invasive ventilatory support, extracorporeal membrane oxygenation, hospice discharge, or death.

From March and April 2020 through September and October 2020, the researchers discovered a drop in inpatient mortality. Treatment patterns shifted as the epidemic progressed, with antibacterial and immunomodulatory drugs being used in different ways. Researchers could create accurate machine learning models to predict clinical severity based on data accessible on the first calendar day of admission using the N3C data. Patient age and generally available vital signs, and laboratory values are the most powerful predictors.