Use of AI in radiology revenue cycle management
Artificial intelligence (AI) use in radiology has been seeing a lot growth on the clinical side, but AI is also making inroads to improve efficiency on the practice management side of medical imaging. This was a topic of discussion in sessions at the Radiology Business Management Association (RBMA) 2024 meeting.
Radiology Business spoke with Dave Walker, senior director of revenue cycle, Radiology Associates of North Texas (RANT), who spoke at the meeting on the use of AI for revenue cycle management. His practice began using AI about five years ago with coding and it expanded to other areas on the business management side. He said AI is becoming more prevalent at his practice and across radiology. Most importantly, they have had success using AI to capture more revenue and improve efficiency..
"Some of our big successes have been in our self-pay collections arena, doing a lot of modeling around propensity to pay, what's the best fee schedule to charge for patients. We have seen an 8% increase in our self-pay revenue, which for a practice that's primarily hospital-based, it's a big leap and has really done a lot. It's really been a big win for the practice," Walker explained.
He said the AI enabled them to work harder on the front end before a collection agency gets involved. Walker said this increased patient satisfaction, while at the same time increasing the practice's satisfaction because they are collecting more. AI is also being leveraged to take a better look at data analytics and insurance claims processing.
"We also use a lot of data analytics machine learning to look at our charge capture ratios. Five years ago our charge capture ratio was at 98.5%, but we now consistently stay at 99.7% because of our reports database. We're able to really quickly isolate what hasn't been billed, figure out why, and get it out the door. That alone has meant millions of dollars per year for us," Walker explained.
AI bots help reduce need for as many humans in claims processing
Walker said another innovation in radiology revenue cycle management is the use of AI bots to automatically process claims to help speed and streamline workflows.
"If you go back 20 years, in a typical large revenue shop there would literally be hundreds of people following up on claims processing. What we are seeing more now are systems becoming smart enough through machine learning to say, okay, this happened with this claim, what's my next best action? And then doing it automatically, versus having to pull somebody into the process to do it," he said.
In automated claims review, the AI can take a range of actions. For example, Walker said they get a claim return rate of about 8-9%. He will automatically look and go ask if they had a claim paid for the patient during the last 60 days. He also asks how did they pay and what insurance was it under.
"Then we will just do an auto update where all of a sudden you don't have a follow up person involved, it's just automated. And really it is a bot that takes action. AI identifies what action needs to be taken, the bot takes the action and takes a lot of the human element out of it to move the revenue cycle much faster, and you have less holes in the process," Walker explained.
Machine learning helps to manage data analytics
Radiology revenue cycle management has struggled for years to have access to data at their fingertips and reporting software to hone into where the problems are with billing claims. But, Walker said this has rapidly changed due to AI enabled software.
"Because of machine learning and AI we're seeing a big change in how we think about and look at data. I focused my whole career on a claim-by-claim basis, just following up on the claim, getting it paid, knowing that 80% was going to pay itself. And then you go after the 20%. What machine learning starts asking is why didn't the 20% get paid? What can we do to resolve it? And it also isolates it so we can see it faster. We're able to use big data to solve problems, so instead of moving the needle a half a percent, we can move the needle all of a sudden several percentage points when it comes to collections. We now have the right tools and we can actually see what's going on," Walker told Radiology Business.
Next steps to improve RCM and the need for innovation
What the use of AI has taught Walker is that there are more efficient ways of doing things if a practice leverages new technology. With declining reimbursements, practices are looking to increase revenue while reducing costs, and AI technology can enable both.
"I think what we really have to do is focus on innovation. We tend to be stuck in the way we've always done things in the past. We need to forge ahead and be state-of-the-art, to keep pushing the boundaries, and to demand our vendors push the boundaries with us. We need to really figure out how we can best use data, how we can do things better and make things faster and cheaper. Our role is to effectively collect things and we have to do it in a more cost effective manner," Walker explained.
Marty Stempniak has covered healthcare since 2012, with his byline appearing in the American Hospital Association's member magazine, Modern Healthcare and McKnight's. Prior to that, he wrote about village government and local business for his hometown newspaper in Oak Park, Illinois. He won a Peter Lisagor and Gold EXCEL awards in 2017 for his coverage of the opioid epidemic.