Researchers are learning to predict your chances of surviving COVID-19
Not everybody who gets COVID-19 has symptoms, and not all symptomatic patients get equally sick. Hospitalization rates have stabilized in hard-hit areas like northern Italy and New York City, but if the next wave is even bigger and more destructive — one of three scenarios envisioned by University of Minnesota epidemiologist Michael Osterholm and his colleagues — that "would absolutely take the health system down," Osterholm told Stat News. Two studies released last week offer tools that might help hospitals better triage patients.
Researchers in China reported in the journal Nature Machine Learning that an analysis of blood samples taken from 485 coronavirus patients in Wuhan discovered there biomarkers that can predict whether a coronavirus patient will die within 10 days, with more than 90 percent accuracy, Business Insider reports. A computer model the researchers developed looks for high levels of the enzyme lactic dehydrogenase (LDH), linked to lung damage; lymphopenia, or low levels of infection-fighting white blood cells; and a rise in inflammation-signaling high-sensitivity C-reactive proteins (hs-CRP).
Article continues belowThe Week
Escape your echo chamber. Get the facts behind the news, plus analysis from multiple perspectives.
Sign up for The Week's Free Newsletters
From our morning news briefing to a weekly Good News Newsletter, get the best of The Week delivered directly to your inbox.
From our morning news briefing to a weekly Good News Newsletter, get the best of The Week delivered directly to your inbox.
A second paper published last week in the Journal of the American Medical Association found 10 biomarkers they said could predict a COVID-19 patient's risk levels. They turned risk predictors — high LDH levels and low levels of lymphocytes plus history of cancer, age, shortness of breath, and other factors — into a coronavirus risk "calculator."
