How to Run a Denials Teardown Your Team Can Repeat
Most independent practices work denials the way they arrive: one claim, one appeal, one phone call at a time. It feels productive. It rarely moves the number. In our recent session on what is actually working in independent practices, Jana Fullbright, Director of Revenue Cycle for the State of Franklin Healthcare Associates, described a different approach, and it’s worth stealing.
The arrival-order approach fails partly because there is no single villain. When we asked 360 independent practice leaders which denial categories hurt them most, the answers clustered between 25% and 39%. Eligibility, prior authorization, documentation, coding, and submission and follow-up all registered, and none dominated, which is part of why denial management resists a single fix. Nearly half of those leaders, 48%, also said their denials are rising. Working the queue in order spreads effort evenly across a problem that is not distributed evenly in dollars, and leaves you no way of knowing whether any of it worked. What follows is the alternative, a repeatable denial management process you can run in an afternoon and run again next month.
Sort on dollars, not dates
Start by refusing to work the list in order. Pull your denials, sort them by dollar value, and look at the top of the pile. You are not trying to fix everything. You are trying to find the one category that is quietly costing you the most.
Joseph Jacot, CEO at Michigan Vascular Center and a clinician for his first 20 years in healthcare, described a version of that sort which needs nothing but a spreadsheet. Dump the data on a regular schedule, build a pivot table, and run it two ways. One pass looks for where the zeros are coming in the door. The other looks for where you have charged out big numbers and received nothing back. The first group tends to be small dollars carried by volume, the kind of thing you can change quickly. The second is the big holes. Both are worth working, but they call for different fixes, and confusing them is how a practice spends a quarter on a category that was never going to move much.
Narrow until there is one thing to look at
In her case, that category was lab denials, and in her words, the practice was drowning in them. So the team narrowed further. They sorted the lab denials by highest dollar and landed on a single test that kept coming back unpaid. Then they asked the only question that matters: why does this keep happening?
The test was vitamin D. The reason they started with lab at all was volume, since the denials were high enough to bog the team down, which made the category worth a dedicated look before anything else. And the narrowing is what made everything after it possible. Lab denials is a category you can stare at for a month. One test is a question with an answer.
Trace it upstream
The answer was upstream, not in the billing office. A small number of providers were ordering the test in situations where it would not be covered, sometimes as part of a blanket order set. The denials were a symptom. The ordering pattern was the cause.
That was about eight providers out of roughly 215 across the practice, which is why the fix could be narrow too. But denial root cause analysis takes patience, because the pattern surfaces late. Jacot described the lag plainly: you do not find a problem until you start seeing a pattern, and a pattern means time has already passed, so his team goes back 60 days, runs the analysis, and plans the fix from there.
His own worst categories make the point about where causes actually live. Prior authorization and documentation that does not support the claim are the two that hurt him most, and neither one is a billing error. The cause is sequence. He has separate teams billing hospital care and office care, and you cannot bill postoperative care when the team ahead of you has not billed the intraoperative care yet. The denial looks like a documentation problem. The fix is telling one team to wait.
Fix it where it starts
That distinction is the whole game. Because the problem was concentrated in a handful of providers, the fix could be too. The team retrained those specific providers on when the test was appropriate, retrained the nurses and lab techs entering the orders, and added edits in the lab system so the orders stopped before they became claims. Then, and this is the part most teams skip, they measured whether it worked and moved on to the next category on the list.
What made that fix stick is that it did not rely on memory. The retraining ran through the coding team and through compliance, so it had owners rather than volunteers, and the lab system itself was changed to stop the orders that should not have been placed. The second part matters more than it sounds. Jacot made the related point about turnover: you can retrain an employee, and you will still have turnover, and the next person arrives knowing none of it. A rule built into the system is the only version of a fix that survives the people who learned it. Any denial management process that depends on everyone remembering the training is a process with a shelf life.
Prove the fix held
Here is the principle underneath it. Working denials without fixing the cause is, as she put it, like a mouse running on a wheel. You can put in enormous effort and end up exactly where you started. What changes the number is accountability: a specific owner, a specific change, and a report that tells you whether the change actually held.
Fullbright’s team is already on its second pass. After vitamin D came lipids, and the plan is the same: retrain, then run the reports in her practice management system that confirm whether the retraining brought the volume down. Her closing advice was the shortest version of the method. You can have confidence in the process, she said, but it is important to have data that confirms it. A denial management process that looks right and one that is working are two different claims, and only one of them can be measured.
You do not need new software to start. You need your denial data sorted by dollars, the discipline to trace the top category to its root cause, and someone whose job is to confirm the fix stuck. Do that four or five times and the shape of your denials starts to change.
If you want to see how other independent practices are approaching this and where your own denials rank against your peers, the full session and the 2026 survey are worth an hour.