For Community Medical Centers of Fresno, as with many other healthcare facilities, denials are a constant source of pain, and managing them is an ongoing struggle.
THE PROBLEM
In roughly the last 10 years, payers have increased their delay tactics – one method has been to do that with denials, said Eric Eckhart, director of patient financial services at Community Medical Centers of Fresno.
“This increased volume translates to an increased workload for my team,” he explained. “We knew that many of our historical workflows were not efficient enough to handle this new level of denial volume. This situation led us to start seeking out new systems we could implement internally or ones that involved a vendor-based system.
“For the last five years, we have been on a journey to put the team in the best position possible to succeed by streamlining staff reporting structures, improving denial reports, updating EHR workflows and eliminating paper processes,” he continued. “In addition, there has been an increased focus on preventing denials upstream.”
Despite all the positive changes implemented, Eckhart knew it was not the end of the improvement journey. The team needed to keep looking for new tools to add to its toolbox.
PROPOSAL
Eckhart was approached by his clearinghouse vendor in late 2022 to be a beta tester for a new AI-based system that would predict denials before claims submission and would also score incoming denials with a “probability of recovery” score.
Both parts of this new product used Community Medical Centers of Fresno’s claims and payer remittance data to learn how the payers were denying and paying claims.
“Being that the vendor was our clearinghouse, the data was readily available and it didn’t require any additional lift from our end,” Eckhart noted. “The denial prediction tool used the same direct connection into our EHR as the claim edits/rejections from our clearinghouse, so additional programming wasn’t needed. A similar pathway was also used for the ‘probability of recovery’ score.
“By having this tool, we could stop claims on the front end, which gave us one last chance to fix any outstanding items that could trigger a denial,” he continued. “This allowed our billing team a tool that called out areas that needed a second look.”
And if a denial did occur, the follow-up staff had a scoring tool that helped direct their workflow to areas with a high probability of being paid quickly after an initial appeal.
“The overall intent of these tools was not to eliminate all denials, but to provide my staff with another resource to help avoid denials and also to help guide their appeal efforts after the denials are received,” Eckhart said. “The primary reason I chose to move forward and partner with Experian Health was because this system wasn’t a new workflow for staff.
“There is nothing worse than trying to get your staff to disrupt the workflow they know and log into another system to use a resource,” he added. “This system also allowed for the ability to customize what edits and scoring my staff were able to see in the EHR.”
MEETING THE CHALLENGE
Community Medical Centers of Fresno started with the denial prediction piece of the tool in early 2023. The initial rollout of the predictions was slow and very intentional to ensure the leaders had buy-in from the billers that would be working these edits.
“Over several weeks, I reviewed the prediction data by IP/OP, payer, CARC code, etc.,” he recalled. “The tool is very good at predicting future denials, but not all are preventable; therefore, only a select set of predictions are relevant for the team.”
The team decided to implement two CARC code predictions:
197 – Precertification/authorization/notification/pre-treatment absent. (Commercial payers only.)
109 – Claim/service not covered by this payer/contractor. One must send the claim/service to the correct payer/contractor. (Medicaid payers only.)
“The 197 prediction allowed us to ensure an authorization process upstream was being followed and it allowed us to ensure the auth number was making it on the claim – a technical issue at the time was creating this problem,” Eckhart said. “The 109 prediction was a second check for managed Medicaid registration issues occurring upstream and was needed before we were able to implement a coverage automation system on the front end.
“The ‘probability of recovery’ scoring piece was implemented later in the year with our commercial follow-up team,” he continued. “There were some initial challenges in ensuring this score was easily accessible from our current work queues. But after these were resolved, we were able to integrate this into the staff’s daily workflow and it also provided a method for ensuring we are able to get the easy cash in the door as quickly as possible.”
RESULTS
Community Medical Centers of Fresno saw significant results almost immediately. In the initial six months of implementation, it saw a 22% decrease in 197 denials and an 18% decrease in 109 denials.
“Both of these metric improvements have resulted in more than 30 hours per week in additional work that has been eliminated from the follow-up staff workloads,” Eckhart reported. “This one tool has been able to free up staff time and allow for additional appeal work in the future.”
ADVICE FOR OTHERS
The key piece of advice Eckhart offers to other organizations is to ensure they get staff buy-in and start slow.
“AI tools are just that, tools,” he noted. ” We may get to the day where they do everything, but we aren’t there yet. Human intervention and guidance are key to get a successful result. Also be sure the AI is trained on data that is relevant to your organization. If the model is not trained on relevant data, it defeats the whole purpose of AI. You might as well go back to your analyst team with a bunch of spreadsheets.
“Pick the right tool for your situation and needs,” he concluded. “Similar to other technologies, some AI tools make sense for some and not others. I know many of us like to be on the cutting edge of technology, and AI is that buzzword we all feel we need to be a part of. Don’t fall into this trap; find the AI technology that helps and is not just the next vendor offering.”
Follow Bill’s HIT coverage on LinkedIn: Bill Siwicki
Email him: bsiwicki@himss.org
Healthcare IT News is a HIMSS Media publication.
