COVID-19 infection causes only mild symptoms in the majority of the people affected but proves fatal to some. The ability to predict which patients are most likely to succumb to COVID-19 could help to focus treatment on those individuals and potentially save lives.

A team led by David Altschul, M.D., published a new model for predicting patient mortality from COVID-19 online on October 7 in Scientific Reports. The new model used information from patients at Montefiore Medical Center and included six risk factors and biomarkers associated with disease severity: age, oxygen saturation, mean arterial pressure, blood urea nitrogen, blood coagulation, and high levels of several inflammatory proteins. The ability to predict a patient’s risk of dying from COVID-19 could be particularly helpful in guiding care during the early stages of hospitalization.

Dr. Altschul is chief of the division of neurovascular surgery and associate professor in The Leo M. Davidoff Department of Neurological Surgery and associate professor of radiology at Einstein. He is also surgical director of the Montefiore Comprehensive Center for Stroke Care.