AI Platform eCARTv5 for Predicting Deterioration in Hospitalized Patients Superior to Epic’s Deterioration Index

A recent study compared the performance of several early warning scores for detecting clinical deterioration in hospitalized patients, focusing on AI-based and non-AI-based tools, using data from seven hospitals. This retrospective cohort study centered on the Yale New Haven Health System. The authors documented all consecutive adult medical-surgical ward hospital encounters between March 9, 2019, and November 9, 2023, comparing six early warning scores across 362?926 patient encounters, with eCARTv5, a machine learning mode used to identify clinical deterioration best with an area under the receiver operating characteristics curve (AUROC) of 0.895 and the highest positive predictive values at both the moderate- and high-risk matched thresholds. The National Early Warning Score, a non–artificial intelligence score with an AUROC of 0.831, was the second-best performer at both thresholds, while the Epic Deterioration Index (EDI), a predictive tool integrated within the Epic electronic health record (EHR) system, was one of the worst, with an AUROC of 0.808 and the lowest positive predictive values. The latter Epic system uses patient data to assess and predict the risk of clinical deterioration, aiming to alert healthcare providers early to patients who may need intervention to prevent worsening conditions.

Brief review of eCARTv5