United StatesHealthcare and life sciences2023 – ongoingRWJBarnabas Health
AI deterioration alerts for hospital rapid response teams
RWJBarnabas Health researchers reported 18% lower adjusted odds of death among high-risk patients after AI-triggered rapid response
Problem
Hospital patients could deteriorate before clinicians recognized visible warning signs. Earlier assessment by critical care teams could allow earlier intervention.
Approach
RWJBarnabas Health piloted Epic's Deterioration Index at Robert Wood Johnson University Hospital before expanding to other hospitals. The model analyzed electronic health records and alerted rapid response teams when patients reached the highest-risk threshold. Implementation included clinician training and changes to alert delivery.
Results
Takeaway
The reported improvement followed a combined change in prediction, staff training and clinical response. It cannot be attributed to the algorithm alone.
Categories
- Size
- Large enterprise
- Industry
- Healthcare and life sciences
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Vendor solution
- Goal
- Quality and risk
- Evidence
- Company's own claim
Sources
Source summaries
The member account described a deployed deterioration model using electronic medical records. It associated the system with an 18.6% mortality reduction.
Participating researchers reported mortality falling from 23.1% to 18.6% among 23,132 high-risk patients across 11 hospitals. They reported an 18% reduction in adjusted odds of death, a different measure from the earlier member account.
Full details are in the original sources.