AI Model Quadruples Delirium Detection in Hospitals, Improving Patient Outcomes

AI Model Revolutionizes Delirium Detection in Hospitals

Researchers at the Icahn School of Medicine at Mount Sinai have developed an artificial intelligence (AI) model that has significantly improved the detection and treatment of delirium in hospitalized patients. The study, published in JAMA Network Open on May 7, 2025, demonstrates the first real-world benefits of an AI-powered delirium risk assignment model in clinical practice 1

1

Understanding Delirium and Its Impact

Delirium, a sudden and severe state of confusion, affects up to one-third of hospitalized patients and often goes undetected. If left untreated, it can lead to prolonged hospital stays, increased mortality risk, and worsened long-term outcomes 2

2

The AI Model's Approach and Implementation

The research team, led by Dr. Joseph Friedman, took a "vertical integration" approach, working closely with Mount Sinai clinicians and hospital staff from the start. This collaboration ensured that the model was both effective and practical for clinical use 3

3

The AI model analyzes a combination of structured data and clinicians' notes from electronic health records, using machine learning to identify chart data patterns associated with high delirium risk. It also applies natural language processing to identify patterns in staff-written chart notes, capturing subtle mental status changes that might indicate delirium or heightened risk 4

4

Related Stories

AI Outperforms Nurses in Predicting Hospital Admissions, Promising to Reduce ER Overcrowding

AI-Based System Significantly Reduces Hospital Deaths by Identifying High-Risk Patients

Revolutionary AI System InfEHR Transforms Medical Diagnosis Through Advanced Pattern Recognition

AI-enabled tool cuts hospital deaths 18% by detecting patient deterioration earlier

Impressive Results and Improvements

When deployed at Mount Sinai, the AI model dramatically improved delirium detection:

1. A 400% increase in identified cases without increasing screening time
2. Safer prescribing practices, reducing potentially inappropriate medications for older adults
3. Strong, reliable performance in a diverse, real-world hospital setting

The study, involving over 32,000 patients, showed that monthly delirium detection rates improved from 4.4% to 17.2%, allowing for earlier intervention. Patients identified by the model also received lower doses of sedative medications, potentially reducing side effects and improving overall care 1

1

Implications for Healthcare and Future Directions

Dr. Friedman emphasizes that the AI model is not meant to replace doctors but to provide them with a powerful tool to streamline their work. By analyzing vast amounts of patient data, the machine learning approach allows healthcare providers to focus their expertise on more effective diagnosis and treatment 2

2

Dr. David L. Reich, Chief Clinical Officer of the Mount Sinai Health System, sees this research as a significant step towards becoming a learning health system. He highlights the importance of developing, testing, deploying, and fine-tuning AI tools that seamlessly integrate into healthcare workflows 3

3

While the AI model has shown strong results at Mount Sinai, further validation in other hospital systems will be necessary to evaluate its performance in different settings and make any needed adjustments 4

4