AI tool linked to lower death rates at 11 NJ hospitals - Central New Jersey News

AI tool linked to lower death rates at 11 NJ hospitals

An AI tool is helping New Jersey hospitals spot patients at risk of sudden medical decline before it becomes life-threatening, and a new study suggests it may be saving lives.

Researchers with RWJBarnabas Health and Rutgers Robert Wood Johnson Medical School found that the use of an artificial intelligence-powered early warning system was associated with significantly lower death rates among high-risk hospitalized patients.

The study, published in NEJM AI, evaluated the Epic Deterioration Index, or EDI, across 11 RWJBarnabas Health hospitals. Researchers found an 18% reduction in the risk-adjusted odds of in-hospital death among high-risk patients after the system was implemented.

The analysis included 23,132 high-risk patients. Mortality rates fell from 23.1% to 18.6% following the rollout of the system.

“Our goal was to identify patients earlier, before they reached a point where intervention becomes much more difficult,” said Dr. Thomas Nahass, vice president of health informatics at RWJBarnabas Health and the study's lead author.

“The deterioration index gives us an earlier point in time. If we can get a critical care eye on the patient sooner, we can change the course of their outcome.”

The Epic Deterioration Index continuously analyzes patient information from electronic health records, including vital signs, laboratory results and nursing assessments, to identify people at risk of rapid clinical deterioration.

The system recalculates risk scores every 15 minutes. When a patient reaches the highest-risk category, a hospital's rapid response team is automatically notified.

RWJBarnabas Health and Rutgers spent several years developing a systemwide approach to using the technology. The tool was first piloted at Robert Wood Johnson University Hospital before being expanded throughout the hospital network.

The broader rollout included automated alerts, clinician training and ongoing performance monitoring.

Researchers found that rapid response team activations for high-risk patients increased from 25.3% to 37.5% of hospital stays after implementation. Despite more rapid response calls, there was no significant increase in transfers to intensive care units.

“This study demonstrates how AI-enabled tools, when paired with experienced clinical teams, can help us identify patients at risk sooner and deliver the right care at the right time,” said Dr. Andy Anderson, chief medical and quality officer at RWJBarnabas Health and a study co-author.

“These findings highlight the potential for innovation to improve quality, safety and outcomes for the patients we serve.”

The study included academic medical centers, community teaching hospitals and community hospitals throughout the RWJBarnabas Health network.

Researchers emphasized that the reduction in mortality was tied not only to the AI tool itself, but also to staff education, electronic health record alerts and coordinated rapid response efforts.

“This is what an integrated academic health system is for,” said Dr. Stephen P. O'Mahony, chief medical information officer at RWJBarnabas Health and a senior author of the study.

“We combined Rutgers methodological rigor with the operational reach of 11 RWJBarnabas hospitals. The mortality benefit was not produced by an algorithm but by the partnership around the algorithm.”

Because the Epic Deterioration Index is already integrated into Epic, one of the most widely used electronic health record systems in the United States, researchers said the findings could have implications for hospitals nationwide.

The team is now studying patients whose risk scores rise rapidly, with the goal of identifying deterioration even sooner and improving outcomes further.

Email: bwadlow@MyCentralJersey.com

This story was created by reporter Brad Wadlow, bwadlow@usatodayco.com, with the assistance of Artificial Intelligence (AI). Journalists were involved in every step of the information gathering, review, editing and publishing process. Learn more.