The Denial Rate You Reported Isn’t the One Payers Are Reporting

Last week’s number was 31%. Denial rates, up year over year, in the first six months of a rule built to bring them down. That number came from your own claims data. This week, look at the number payers are putting out instead.

“Transparency” arrived, and it does not explain anything

Under the 2024 rule, impacted payers started publicly posting their own prior-auth metrics this year — approval rates, denial rates, appeal outcomes, decision timeframes. On paper, that is the accountability layer everyone asked for. In practice, KFF already flagged the problem: the reported data is aggregated at a level that tells you almost nothing about what is actually getting approved or denied, by whom, for what service line.

So you now have two denial pictures that do not talk to each other. Yours, granular enough to hurt — broken out by payer, by CPT, by reason code, sitting in your AR aging report. Theirs, rolled up into a topline number clean enough to put in a press release.

A payer can report a flat or improving denial rate at the portfolio level while your specific service line gets worse — and both numbers are technically true.

Grading your own homework, at scale

Call it what it is: the entities being measured designed the measurement. There is no independent audit layer sitting between a payer’s internal denial data and what gets published. If your practice’s experience does not match what a payer reports publicly, you have no formal channel to reconcile that. You just have your own numbers, and a rule that assumed publishing would be enough.

This matters right now for one reason: CMS is treating this transparency requirement as done. It is live, it is checked off, and the agency’s attention has moved to the next build.

Where CMS’s attention actually is

On 1 July, CMS named 29 healthcare organisations — health systems, EHR vendors, physician practices, networks — as early adopters in a new Electronic Prior Authorization Acceleration initiative, working ahead of the 1 January 2027 API deadline. Some of the largest payers in the country are already at that table too.

That is the readiness conversation happening without you, unless you are asking your EHR vendor directly where they stand in it. Not “are we compliant” — “are we in the room.” The organisations on that list are shaping how the workflow gaps get closed before the deadline forces everyone else to react.

What to actually do with this

• Stop treating a payer’s published denial rate as a benchmark for your own. It is a different denominator, reported by a party with no incentive to make it more granular.
• Ask your EHR vendor, by name, whether they are part of CMS’s Acceleration initiative — or what their equivalent readiness plan looks like if they are not.
• Keep logging your own denial reason codes by payer and service line. That dataset now does double duty: your operational tool, and the only honest counter-narrative to what is being published on your behalf.

The 2030 view

Transparency rules work when the party being measured has something to lose from hiding the number. Right now, payers have something to lose from a slow decision — that is what the API forces. They have nothing to lose from an aggregated one.

The practices that come out ahead will not be the ones waiting for CMS to close that gap. They will be the ones who already know their real number cold, so a rounded, portfolio-level press release from a payer does not get to define their year.