Photo: Mariyariya/Getty Images
Autonomous AI coding improved Mercyhealth’s revenue cycle performance and eased the heavy lifting that went with a greater volume of claims, says Kelly Pierson, director of coding and clinical documentation integrity.
The health system, headquartered in Rockford, Illinois, went looking for an AI partner as patient volumes increased, she said. Physician and ER visits were growing.
The regional health system of seven hospitals in Illinois and Wisconsin serves more than 2.4 million patients annually.
“We could not find enough experienced coders,” Pierson said. “We were in competition with our own organizations. There was a lot of heavy lifting.”
Coding sits in the middle of the revenue cycle, serving as the complex financial bridge for medical claims and documentation. It is key in reducing denials and ensuring hospitals get the revenue they’re due.
More than two years ago, the health system decided it was time for technology to take on the heavy lifting. The first thought was to go with computer-assisted coding, Pierson said. Then, they looked at autonomous coding vendors, interviewing several before going with Arintra.
“They gave us the most customizable logic,” Pierson said. “They also will shift their logic based on payer rules.”
Since automating high-volume claims processing across 10 different specialties, Mercyhealth has seen a 5.1% revenue increase and a 50% reduction in accounts receivable days.
Coders are now working on higher complexity claims.
Pierson said she was nervous at first about trusting AI.
There was a fear, she said, “that AI would run rampant and apply erroneous charge codes.” However, Pierson also knew that Mercyhealth was leaving money on the table.
“We’ve been live with them for two-and-a-half years,” Pierson said. “We were their first health system and Epic client.”
To ease fears, Pierson and team had coders review a percent of the automated coding. Then, the coders were shifted into doing proactive audits.
The fear of AI turned into coding confidence, she said.
“It really changed a lot of us,” she said. “We go into payer audits feeling confident.”
The first sign of improvement was a 5% uplift in revenue month-over-month, she said. Performance shifted from an average 16 days to 7 days for coding work queues.
“Our pre-AR days came down,” she said.
Coding staffing performance also improved.
“Now they’re engaged, looking at ways for additional revenue opportunities,” Pierson said. “They’re looking to improve documentation and shifting in ways to grow instead of being stuck in the day-to-day.”
There’s no longer coder to coder variability.
The health system is currently looking at vendors for other AI solutions.
Email the writer: [email protected]
Newsletter Signup
Healthcare Finance Newsletters give healthcare decision-makers insight into the latest financial and payer developments in the business of healthcare.
Subscribe today!
Add Healthcare Finance News to your network.
© current_year Healthcare Finance is a publication of HIMSS Media