Advocate Health's research arm pilots KelaHealth's predictive software to assess surgical risk, improve outcomes

Advocate Aurora Research Institute is leveraging KelaHealth, a predictive analytics company, to improve surgical care outcomes across Advocate Health’s sites. 

The 12-month pilot project relies on KelaHealth’s Surgical Intelligence Platform, which uses machine learning and predictive analytics to evaluate surgical techniques and help reduce variability and improve patient outcomes. It will be deployed to Advocate Health Care and Aurora Health Care sites across the Midwest.

The project will study whether KelaHealth’s platform can successfully deliver insights that reduce risk and improve outcomes. “We are sure that payers will be paying attention,” Debra O’Connor, vice president of quality and clinical effectiveness at the health system, told Fierce Healthcare in an emailed response. “Whether that results in reimbursement remains to be seen, but it is certainly a cutting-edge approach, and we hope they are successful.” 

While the health system has an enterprise data warehouse and claims data to inform retrospective analyses of outcomes, performance and cost, KelaHealth’s approach prioritizes predictive analysis instead.

The institute will feed clinical and robotics data into KelaHealth’s software, which will use AI to predict patient risk based on historical data. KelaHealth’s risk scores, which are continuously updated, integrate with the EHR so doctors can see a patient’s risk for a given surgery and their post-surgery recovery.

The project will use robotics data from the health system’s Intuitive da Vinci Systems, the maker of robotic-assisted technology for surgeons. “Having access in real-time to data that compares performance to peers on the robot, and provides cost analysis, gives surgeons valuable insights that will allow them to refine their surgical technique, deliver better outcomes and promote fiscal responsibility,” O’Connor said.

“In many ways we’re just scratching the surface of how data from all sources—enabled by AI and machine learning—can surface to health care providers actionable insights for each patient’s unique risk factors, thereby helping health care providers plan for the best surgical results,” Brian Miller, EVP and chief digital officer for Intuitive, said in a press release.

Elective surgeries can account for nearly half of gross revenue at hospitals. And complications are not uncommon and can be costly.

“It is a big problem area that I don’t think we as a society talk much about,” KelaHealth’s founder and CEO Bora Chang, M.D., told Fierce Healthcare. “Where is our ‘Surgical Moonshot’?” 

KelaHealth originated at Duke University, spinning out in 2018. The company says its predictions are accurate between 80% and 95% of the time. It also offers final outcomes and cost data on episodes of care, potentially powerful for providers and payers alike. Especially if parties are pursuing value-based care contracts, KelaHealth helps “really put numbers to that.” 

KelaHealth primarily works with health systems like Duke University and Cedar Sinai but also partners with med tech companies like Intuitive and 3M. While it hasn’t worked directly with payers, the company wants to.

“Our vision is to kind of create a much larger ecosystem where all players are connected,” Chang said.