0
0
Stanford, CA — In a major step toward improving physician-patient communication, Stanford Medicine has launched a new artificial intelligence (AI) tool designed to assist physicians in messaging patients about their clinical test results. This innovative technology drafts plain-language interpretations of lab results, which physicians can then review and approve before sending to patients.
This AI tool is part of a larger initiative to help doctors spend less time on administrative tasks and more time focusing on meaningful interactions with patients. It aims to reduce the workload for physicians, particularly when delivering often complex and technical lab results.
David Entwistle, President and CEO of Stanford Health Care, emphasized the transformative potential of AI in healthcare. “Artificial intelligence has tremendous promise to enhance the experience of both patients and clinicians in the healthcare setting—and this tool is one of many ways that we are unlocking that potential,” he said. “At Stanford Medicine, we are proud to be at the forefront of implementing responsible AI in clinical care, with a focus on advancing and empowering better health for all.”
How the AI Tool Works
Developed in-house by Stanford Medicine, the AI tool utilizes Anthropic’s Claude 3.5 Sonnet language model through Amazon Bedrock, Amazon’s platform for deploying and scaling generative AI applications. The system functions similarly to an AI-powered message generator already in use for creating responses to patient inbox inquiries.
The tool generates drafts of messages to explain lab results in easy-to-understand language, aimed at helping patients understand complex medical data without the need for a medical degree. Once the draft is ready, the physician reviews it for accuracy and can make edits before sending it to the patient. This reduces the time spent composing detailed responses while ensuring the message remains clear, personalized, and reassuring.
“The goal is not for the tool to replace the physician’s messaging interaction with the patient but rather provide the physician with a draft that’s either ready or close to ready to send,” said Dr. Michael Pfeffer, Chief Information Officer at Stanford Health Care and the Stanford School of Medicine. “It’s really about continuing to give back to the patient-provider relationship by reducing administrative tasks and fostering better and faster communication.”
Pilot Success and Physician Feedback
The new AI tool has already gone through multiple rounds of testing with positive results. In a pilot program, 10 primary care physicians used the technology for a month, and with their feedback, the team expanded the test to 24 additional physicians for another two months. All drafts generated by the tool are reviewed by physicians, ensuring that no message is sent to a patient without approval.
“The feedback we’ve received so far has been overwhelmingly positive,” said Aditya Bhasin, Vice President of Software Design and Development at Stanford Health Care. “Physicians appreciate the drafts’ concise, accurate nature, and the fact that they are personalized and reassuring when communicating normal results to patients.”
Dr. Christopher Sharp, Chief Medical Information Officer at Stanford Medicine, highlighted the emotional benefit of the tool. “I’ve had patients say to me, ‘Dr. Sharp, you always write a comment on my result, and it makes me feel so much better.’ This tool will make it easier and more efficient to provide those empathetic interpretations, which are so important to our patients.”
Aiming for Broader Use
Currently, the tool is available to primary care physicians at Stanford Health Care, with plans to expand to specialists later this year. The team also plans to monitor the tool’s effectiveness by measuring time saved and evaluating its usage in clinical settings.
“We’re excited to learn from these early implementations and continue to refine the tool,” said Dr. Pfeffer. “This technology is still new, but with continuous feedback from physicians, it will only improve and become more valuable to our healthcare team.”
By introducing AI tools like this one, Stanford Medicine hopes to enhance the quality of care while improving the efficiency of daily tasks, ultimately benefiting both physicians and patients alike.
Provided by Stanford University Medical Center
