Effectiveness of an AI–Enabled Intervention for Detecting Clinical Deterioration

This study, published in JAMA Internal Medicine on March 25, 2024, examines the effectiveness of an AI-enabled intervention aimed at detecting clinical deterioration among hospitalized patients to reduce escalations in care, such as rapid response team activation, ICU transfers, or cardiopulmonary arrests. Conducting a cohort study of 9,938 patients at an academic center using a regression discontinuity design for analysis, the researchers found that exposure to the intervention led to a 10.4 percentage point reduction in the risk of such clinical escalations. This significant finding suggests that integrating AI deterioration models into patient care can decrease the likelihood of emergency interventions, supporting the potential for broader application and further research in different healthcare settings.

Opens on JAMA Internal Medicine