The research indicates a collaborative future for AI and nursing expertise in enhancing care delivery, with plans for future live clinical testing to validate the findings.

Published in the July 9 issue of Mayo Clinic Proceedings: Digital Health, the research aims to provide practical solutions to enhance emergency care delivery.

Lead author Jonathan Nover, who is also the Vice President of Nursing and Emergency Services at Mount Sinai, emphasized the need for a forecasting tool in healthcare, similar to reservation systems used by airlines and hotels, to improve operational efficiency.

A recent multi-hospital study by the Mount Sinai Health System reveals that artificial intelligence (AI) can significantly enhance the prediction of hospital admissions in emergency departments (ED), facilitating earlier interventions and better resource allocation.

Interestingly, the AI's performance as a standalone predictor of hospital admissions was robust, showing that it did not significantly improve when combined with nurses' assessments.

The authors stress that the AI tool is intended to support, not replace, clinicians, thereby enabling them to concentrate on delivering compassionate care rather than logistical challenges.

Future studies will focus on implementing the AI model into real-time workflows to evaluate its impact on operational metrics like boarding times and patient throughput.

The study received support from various grants and resources from the Icahn School of Medicine at Mount Sinai and the National Institutes of Health.

This research highlights the ongoing challenges faced by emergency departments in the U.S., particularly issues related to overcrowding and delays in patient admissions due to limited inpatient bed availability.

Co-senior author Eyal Klang noted that the AI model is designed to support clinical decision-making by transforming complex data into actionable insights, allowing clinicians to focus more on patient care.

The study aims to improve patient care, enhance the overall patient experience, reduce overcrowding, and optimize resource allocation in hospitals.

Involving over 500 ED nurses, the study analyzed data from more than 1 million past patient visits, comparing AI-driven predictions against traditional nurse triage assessments.