Utilizes a lightweight algorithm and machine learning collected for 8 variables during the first 48 hours of hospitalization to predict the risk of 6-month mortality. Researchers at the University of Minnesota have developed a lightweight algorithm using machine learning for predicting the risk of 6-month mortality at the time of hospital admission. Using just 8 different variables collected during the first 48 hours of hospitalization, this algorithm predicted death within 6-months with an AUC of 0.92. The discriminative ability of this algorithm has been shown to be significantly better than historical estimates of clinician performance. This algorithm can be a critical tool in supporting clinical decision-making at admission and in evaluating suitable options such as transfer to tertiary referral center, serious illness care-conversations in high-risk patients, patient/family counseling, and palliative care utilization.
