AI boosts physician-nurse teamwork at Stanford Hospital
CALIFORNIA, UNITED STATES — Stanford Health Care has implemented an artificial intelligence (AI) model to predict when patients are at risk of deterioration, fostering improved collaboration between physicians and nurses.
The AI system continuously scans vital signs, lab results, and electronic health record (EHR) data every 15 minutes, alerting the clinical team if a patient’s condition is worsening.
Enhancing communication channels
According to a recent study published in JAMA Internal Medicine, this proactive approach has facilitated communication among clinicians beyond just end-of-shift handoffs.
“The big question I want to answer is, ‘How do we use AI to build a more resilient health system in high-stakes situations?'” said senior study author Ron Li, MD, Stanford’s medical informatics director for digital health, in an April 15 news release.
“There are many ways to do that, but one core characteristic for a resilient system is strong communication channels. This model is powered by AI, but the action it triggers, the intervention, is basically a conversation that otherwise may not have happened.”
Reducing clinical deterioration events
After the deployment of the AI tool, Stanford Hospital experienced a 10.4% decline in deterioration events, such as transfers to the intensive care unit, rapid response team incidents, and codes, among 963 high-risk patients.
The study found that by alerting clinicians earlier, the AI system allowed for timely interventions and prevented further complications.
Dr. Li aims to continue improving the technology’s accuracy to reduce alert fatigue among providers. The researchers suggest further expansion and testing of the AI intervention in other care settings based on the positive results.
In a related JAMA study, it was found that AI did not reduce response time but assisted doctors in crafting longer, compassionate drafts.
The AI tool generates an initial draft response based on the patient’s message and medical history. Physicians can then edit the draft for content and tone before sending it to the patient. This “human-in-the-loop” approach ensures accuracy and appropriateness.
