UC San Diego Health is at the forefront of digital health innovation as one of the first health systems in the country to pilot generative artificial intelligence (GenAI) for drafting replies to patient messages within the Epic Systems electronic health record.
Image Credit: innni/Shutterstock.com
A recent study from the University of California San Diego School of Medicine found that although AI-generated replies did not shorten the response time of physicians, they did reduce their cognitive load by supplying an empathetic draft. This enables physicians to edit rather than create responses from scratch.
This study is the first randomized prospective assessment of physician messaging created by AI, and it was published in the April 15th, 2024, online edition of the Journal of the American Medical Association’s Network Open.
We are very interested in using AI to help solve health system challenges, including the increase in patient messages that are contributing to physician burnout. The evidence that the messages are longer suggests that that they are higher quality, and the data is clear that physicians appreciated the help, which lowered cognitive burden.
Christopher Longhurst, MD, Study Senior Author and Executive Director, Joan and Irwin Jacobs Center for Health Innovation, University of California, San Diego
This quality improvement study assesses patient-physician correspondence and indicates that integrating generative AI into digital healthcare interactions can enhance patient care by improving the quality, efficiency, and engagement of communications.
Additionally, by reducing physician workload, generative AI aims to alleviate burnout, allowing doctors to concentrate on more complex aspects of patient care.
This study shows that generative AI can be a collaborative tool. Our physicians receive about 200 messages a week. AI could help break ‘writer’s block’ by providing physicians an empathy-infused draft upon which to craft thoughtful responses to patients.
Ming Tai-Seale, Ph.D., MPH, Study Lead Author and Professor, Family Medicine, University of California San Diego
The COVID-19 pandemic led to an unprecedented surge in digital communications between patients and doctors, a trend that continues to be in high demand. Portals like MyUCSDChart, utilized by UC San Diego Health, enable direct emailing to doctors. This convenience has created increased pressure for prompt responses from providers, a demand many now find challenging to meet efficiently.
Using generative AI to generate patient responses to non-emergency queries has been tried in a pilot study with electronic health record vendor Epic Systems, which began in April 2023 at UC San Diego Health to provide virtual physician assistance to help fulfill the increased demand for patient communications.
For transparency, the replies are clearly marked as AI-generated before being reviewed and edited by the signing physician.
Physicians, often pressed for time and able to provide only brief factual responses, have found that generative AI helps them compose longer, more compassionate replies that are well-received by patients.
AI doesn’t get tired, so even at the end of a long day, it still has the capacity to help draft an empathetic message while synthesizing the request and relevant data into the response. So, while we were surprised by the study’s findings that AI messaging didn’t save doctors time, we see that it may help prevent burnout by providing a detailed draft as a starting point.
Marlene Millen, MD, Study Co-Author and Chief Medical Information Officer, University of California San Diego
The study's findings indicate a potential paradigm shift in healthcare communication through the use of AI. However, further analysis is necessary to evaluate how patients value the increased empathy and length of replies.
UC San Diego Health, in collaboration with the Jacobs Center for Health Innovation, has been intensively testing GenAI models since May 2023. These transformational initiatives will contribute to the safe, effective, and unique application of GenAI in health care.
Sally L. Baxter, Florin Vaida, Amanda Walker, Amy M. Sitapati, Chad Osborne, Joseph Diaz, Nimit Desai, Sophie Webb, Gregory Polston, Teresa Helsten, Erin Gross, Jessica Thackaberry, Ammar Mandvi, Dustin Lillie, Steve Li, Geneen Gin, Suraj Achar, and Heather Hofflick all of UC San Diego; and Christopher Sharp of Stanford are the other study authors.
This research was partially funded by the Joan and Irwin Jacobs Center for Health Innovation at UC San Diego Health, as well as an Agency for Healthcare Research and Quality grant (P30 HS029770-01).
Journal Reference:
Tai-Seale, M., et. al. (2024) AI-Generated Draft Replies Integrated into Health Records and Physicians’ Electronic Communication. Journal of the American Medical Association’s Network Open. doi:10.1001/jamanetworkopen.2024.6565