Health systems entering the world of predictive analytics should start small before leaning so heavily on big data, said Katrina Belt, chief financial officer of Baptist Health in Alabama.
“For us, we needed to ease in,” Belt said. They started with looking at claims, she said. After that, they started analyzing clinical data to help them reduce catheter-associated infections.
“We use analytics to tell us where to begin,” Belt said. “It’s gives us an opportunity for us to address risk.”
Learn on-demand, earn credit, find products and solutions. Get Started >>
Belt and Linda Burt, chief financial officer of the Nebraska Methodist Health System, shared their perspectives during a Healthcare Finance webinar on predictive analytics Monday.
Burt said a program started at Nebraska Methodist Health System in 2010 helps them to be data-centric in per unit cost before going down a path of cost reductions.
“It allows us to be more confident that we’re looking at things properly,” she said.
Belt said Baptist Health started by taking a look at the self-pay population. They learned emergency room visits were decreasing but the number of patients coming back with acute care conditions was increasing in intensive care.
From demographic information, they learned the self-pay population was split equally among men and women with their ages falling mostly between 18 and 26 years old.
“They were going to continue to be our patients,” Belt said. “They went away with doctors’ orders they threw away and prescriptions they couldn’t afford to fill.”
Baptist Health addressed the issue starting with incentives to reduce bad debt and putting a program in place in which the patients agreed to be compliant with their care if the provider helped them pay the cost of prescriptions.
“We began to understand who our patients were,” Belt said.
Burt said at Nebraska Methodist Health System, data opened their eyes to the discrepancy in post-acute care in their three-hospital system. One was a heavy discharger to skilled facilities and the other two were not, she said.
Through application to one of the Medicare bundled payment models, they also learned the amount of skilled care at the one facility was much higher than other systems in the region.
They also found that most of their hospital readmissions were coming out of the skilled facilities. Numbers showed the greatest number returned on a Sunday or Tuesday.
If a patient was discharged on a Friday, and not seen by a home health worker that weekend, it was more likely the patient would be back in the hospital that Sunday, Belt said.
Now they make sure the patient is seen by home health within a day. They may need to start sending a hospitalist to the nursing home, she said.
“That’s been our focus,” Burt said, “getting skilled care under control more so than any other component.”
As inpatient care under a fee-for-service model nets greater revenue than outpatient care, hospitals continue to straddle the reform environment.
“Analytics will make sure we have the right skill sets in the transition of care,” Burt said.