Ecart AI tool outperforms other warning scores for patient deterioration

In a cohort study of 362,926 hospital encounters, the Ecart early warning score demonstrated superior performance in identifying patient clinical deterioration, achieving an area under the receiver operating characteristic curve of 0.895. The study compared three artificial intelligence scores and three traditional scores, revealing that Ecart provided better predictive values and lead time for interventions. Notably, the non-AI National Early Warning Score (NEWS) outperformed the Epic Deterioration Index in accuracy.

Journal Article by Edelson DP, Churpek MM (…) Rhodes DJ et 6 al. in JAMA Netw Open

read the whole article in JAMA Netw Open

open it in PubMed