Laura Wilt, chief digital officer at California-based Sutter Health, recently spoke with Healthcare Innovation about the health system’s efforts to use generative AI to integrate patient-facing summaries into the electronic medical record.
The nonprofit Sutter, which has 23 acute-care hospitals, is partnering with Pittsburgh-based generative AI startup Abridge on this effort.
Abridge, which has launched deployments of its technology at Yale New Haven Health System, UCI Health, Emory Healthcare, the University of Kansas Health System, UPMC and other, said studies have shown that 40 to 80 percent of the medical information patients hear during office visits is forgotten immediately, and that clinical documentation does not always cover the details from medical encounters that patients wish to revisit and better understand.
Wilt, who took the position at Sutter last year after 10 years at Louisiana-based Ochsner Health where she was most recently system vice president and chief information officer, said the motivation to work with Abridge on automating documentation is driven both by provider experience and patient experience.
“We think there's a big opportunity to use technology to help bridge some of that gap. Our providers might have seen a lot of patients during the day and some of them are probably doing their notes in real time but others are waiting until the end of the day,” she said. “And that work can feel very administrative, especially if you're doing it at the end of the day. Abridge really helps with that. You don't have to pile that work up at the end. You can really do it as you go.”
She said this is “one of the first things that we're going to do in terms of this type of technology across the entire Sutter Health footprint and have both primary care and specialists be involved.”
Wilt spoke about the innovation with Abridge on patient-facing summaries. “At the end of your visit, Sutter has been releasing the clinic notes into our portal for a number of years, just like most people have. As a patient you get that and you can read it but often it does not translate to what regular people know about healthcare,” she said. “One of the technologies Abridge has is to really translate that into a patient-friendly summary of what happened. Generative AI and large language models are great at translating, so this is really translating to more patient- friendly terms.”
“Clinicians often don’t have time to write detailed patient Instructions and summaries that reflect the details from their conversation that would most benefit the patient,” said Shiv Rao, M.D., CEO and founder of Abridge, in a statement. “This is a prime use case for generative AI, and it’s a privilege to team up with those from Sutter Health on this new patient-centered initiative.”
Wilt said she thinks there's a huge opportunity to study how much more engaged patients might be in their care if they understand the documentation and notes better.
She described Sutter’s approach to other use cases for AI. The health system has hired a chief data and analytics officer, Kiran Mysore. It also has a group that includes physicians and others that reviews use cases such as the Abridge implementation. “Their focus has been more on traditional machine learning and how we're applying that with a lot of focus on research, and making sure that we're learning as we go,” Wilt said. “So we're excited to expand that and grow as different types of AI become available to us.”
Wilt said she expects this effort to have an impact on provider satisfaction, joy of work, as well as patient satisfaction, “with providers getting to be a little bit more present in that conversation. I think we're very focused on collecting some of those metrics before and after so we can really understand the impact.”
