Stanford Health Care is stepping up its partnership with startup Atropos Health to integrate real-world evidence into physicians' workflow and clinical notes.
The pilot project builds on the two organizations' multiyear relationship and will integrate personalized evidence from Atropos Health with ambient artificial intelligence and Stanford Health's electronic health record system.
Atropos Health, which is already installed inside Stanford Health Care’s firewall, will generate evidence that will help clinicians finalize their encounter notes, including those generated by the hospital’s ambient AI provider, DAX from Microsoft, without having to leave their existing workflow, the organizations said.
“Over our multi-year collaboration with Atropos Health, we’ve seen significant benefit when it comes to advancing research and providing evidence-based medicine at the bedside,” said Michael Pfeffer, senior vice president and chief information and digital officer for Stanford Medicine, in a statement.
“We are seeing additional opportunities to leverage AI-enabled technology to ultimately benefit patients through personalized medicine with this program.”
At Stanford, Brigham Hyde, CEO and co-founder at Atropos Health, Nigam Shah, Ph.D., chief data scientist at Stanford Health Care, and Saurabh Gombar, M.D., Ph.D., developed the "Green Button" technology by tapping the huge volumes of data lying dormant in the EMRs of millions of patients to build clinical decision-making.
Atropos Health, founded in 2020 as a spinout of the Green Button technology, developed a consultation service for doctors powered by publication-grade real-world evidence to guide clinical decisions. In 2023, Atropos Health launched a new operating system, Geneva OS, and a chatbot interface to help generate observational studies rapidly and at scale. That technology, ChatRWD, reduces the time to produce high-quality publication-grade real-world evidence from months to minutes through a chat-based AI co-pilot, according to the company.
Healthcare and life science organizations work with Atropos Health to close evidence gaps from "bench to bedside," improve individual patient outcomes with data-driven care and expedite research, according to the company.
"The Green Button project was designed to answer 'bedside' questions on demand, and it depended on the physician to trigger the question and order this report," Shah, who is a co-founder of Atropos Health, said in an interview with Fierce Healthcare. "Over the years, Atropos has been doing these on-demand studies and now has a massive library of questions already asked and answered."
Advancements with ambient AI for clinical documentation open up opportunities to seamlessly integrate real-world evidence in way that reduces friction on provider workflows.
"Instead of having to wait until the physician thinks of asking the question, now we can get a little bit more proactive because we already have a library of evidence. If there is an existing scenario that matches, we can surface that upfront. So along with the [clinical] note, you get evidence from data, you also get a literature summary," Shah said. "For us at Stanford, the goal always has been to use the best evidence at the proverbial bedside. Ambient AI transcription makes it feasible to reduce the friction in surfacing this evidence."
The pilot will evaluate provider satisfaction, including time savings and use of RWE for treatment decisions. Stanford’s clinicians already rely on the Green Button for RWE delivered in less than 48 hours and Atropos Health's ChatRWD to deliver full observational studies in minutes, according to the organizations.
"Integrating personalized evidence from Atropos Health with ambient AI and the EHR demonstrates Stanford’s commitment to adopting solutions that improve clinician satisfaction and drive better outcomes," Hyde said in a statement. "This takes us one step closer to realizing the dream of personalized care and the Learning Health system while also maximizing convenience and workflow for physicians."
Stanford’s previously announced initiative, Patients Like Mine, will also be advanced through this additional Atropos Health integration, according to the health system. Patients Like Mine is a suite of services designed to equip community physicians with the data and resources needed to provide high-quality, evidence-based treatment to their patients.
The aim of the pilot is to reduce friction in the provider workflow and improve evidenced-based care at Stanford to drive better clinical outcomes, executives said.
"Our goal is to use AI safely, ethically and cost effectively," Shah said. "We don't want to be building AI gizmos just for the flash of it. We want to tie technology deployment to tangible value, either to the physician or providers at large, physicians, nurses and patients."
As healthcare AI evolves, there will be an increasing focus on the value of the output, Shah noted.
"The past two years were more about the technology, and now everybody's under pressure. OK, we got the tech. What is it good for? So I think that's going to be the focus next two years," he said.