Most practices know their denial rate. Far fewer know whether it's the number that matters. A denial rate is the share of claims a payer rejects on first submission, useful, but it treats a missing modifier and a claim filed past the deadline as the same event. They are not the same event. One is rework; the other is money that is gone.
Separate recoverable from terminal
Split denials into two buckets before you look at any percentage. Recoverable denials can be corrected and resubmitted: a wrong modifier, a diagnosis-to-procedure mismatch, missing documentation. Terminal denials cannot: a claim past the payer's timely-filing window, or a service delivered before an enrollment was active.
A practice with a 12% denial rate that is almost entirely recoverable is in better shape than one at 6% where half the denials are terminal. The headline number hides the difference.
Watch the resolution rate, not just the denial rate
The number that predicts lost revenue is what share of denied claims eventually get paid. If denials are worked consistently, most recoverable ones resolve. If they aren't, they age quietly until they cross a filing deadline and convert into terminal denials without anyone deciding to write them off.
Group by reason code, then by cause
- Cluster denials by CARC code, the same code repeating points at a process, not an accident.
- For each cluster, ask where in the cycle it originated. A bundling denial is a coding issue; an authorisation denial started at the front desk.
- Fix the origin. Resubmitting without fixing the cause means the identical denial arrives next month.
The question to ask your billing team
Not "what's our denial rate?" but "of the claims denied last quarter, how many are still unresolved, and how many of those are now past filing?" If nobody can answer that quickly, the reporting isn't detailed enough to manage from.