Abridge, a company that sells AI software for medical documentation, on Thursday announced the largest deployment of its technology to date. The University of Kansas Health System will adopt the software for more than 1,500 of its physicians across 140 locations.
The Pittsburgh-based startup was founded in 2018 by UPMC cardiologist Shiv Rao and two Carnegie Mellon University researchers, Sandeep Konam and Florian Metze. They created the company to alleviate physician burnout by decreasing “pajama time,” which refers to the hours clinicians spend entering clinical notes after work.
With the Rise of AI, What IP Disputes in Healthcare Are Likely to Emerge?
Munck Wilson Mandala Partner Greg Howison shared his perspective on some of the legal ramifications around AI, IP, connected devices and the data they generate, in response to emailed questions.
U.S. physicians spend an average of nearly 2 hours each day completing documentation outside work hours, according to a recent study in JAMA Internal Medicine. At the University of Kansas Health System, this average is slightly higher — the health system estimates that its providers spend 130 minutes per day completing documentation outside of work hours.
Documenting the visits that patients have with their providers is important, but that task shouldn’t be stealing upwards of 10 hours per week from physicians.
Healthcare is driven by conversations, but neither physicians nor patients can remember these conversations very well, Rao said in a recent interview. His company offers software designed to help both patients and physicians with this problem.
Abridge’s first solution, rolled out in 2020, is a patient app that listens to visits and bookmarks parts of the conversation where a physician gives instructions or next steps. The tool lets patients listen to a recording of the conversation and gives them a transcript of it. It can also provide patients with more information on their physician’s recommendations, such as what a certain procedure entails or the side effects of a suggested medication.
A Personalized Approach to Medication Nonadherence
At the Abarca Forward conference earlier this year, George Van Antwerp, managing director at Deloitte, discussed how social determinants of health and a personalized member experience can improve medication adherence and health outcomes.
The company’s physician-facing software, launched last year, listens to visits and creates a near-instant summary for physicians that follows their prototypical note structure, along with a full transcript of the conversation.
Let’s say a physician at the University of Kansas Health System has a 30-minute visit with a patient. About 30 seconds after the conversation ends, the physician will be able to access an accurate, easy-to-read summary of the visit through Abridge’s app. This summary weeds out all the irrelevant “small talk” parts of the conversation and includes categories for the patient’s allergies, medications, social history, symptoms, treatment plan and next steps, Rao explained.
This tool’s AI was trained on a proprietary dataset derived from more than 1.5 million medical visits, he added.
Dr. Gregory Ator, a surgeon and chief medical information officer at the University of Kansas Health System, said in an interview that Abridge’s technology “brings some of the joy back” to practicing medicine by eliminating the drudgery of the note creation process.
“We didn’t go to medical school and spent half of our life in school to be distracted with these clerical tasks,” he declared. “That’s really what we’re doing — being a typist. That can be good and everybody needs to know how to type but that is not the heart and soul of what we’re trying to accomplish.”
Dr. Ator didn’t name any of the other software vendors that his health system was looking at when trying to address its clinicians’ “pajama time” problem. But companies like Nuance (acquired by Microsoft), DeepScribe and Suki also offer technology designed to automate clinical documentation by leveraging AI and humans.
Abridge’s technology stood out as the most trustworthy because it was designed with a physician at the helm, Dr Ator said.
Rao, the aforementioned physician, thinks his company differentiates itself from its competitors through a couple main factors.
Most of his startup’s competitors create transcripts and then use human scribes to put together useful notes, Rao claimed. Abridge is different because its technology produces a note summary at the same time as the visit transcript. The company’s software is also the only one to integrate with telemedicine and call centers, whereas Abridge’s competitors only work with in-person conversations, he declared.
He also argued that Abridge publishes more peer-reviewed papers on note-related physician burnout and how its technology is working to solve it than any other competing company. Rao said this was important because all healthcare AI companies should make an effort to be more transparent.
To measure the success of its partnership with Abridge, the University of Kansas Health System will be tracking how its clinicians’ “pajama time” decreases over time. The health system will also solicit feedback from its providers about how easily the software integrates into their workflow and how much editing they have to do to Abridge’s transcripts, Dr. Ator said.
Photo: megaflopp, Getty Images
