News · Published May 22, 2026 · Updated Sep 2026

AMA's New Prior Auth Numbers: What Changed Since 2025

The AMA released its latest prior authorization survey, and the story is not relief. The time burden held at 13 hours a week, denials climbed, and physicians say the June 2025 insurer pledge has not delivered. Here is what the numbers say, and what is actually in your control.

The AMA's most recent prior authorization survey, released May 2026 and fielded December 2025 among 1,000 physicians, shows the burden essentially unchanged: about 13 hours of physician and staff time per week, roughly 40 requests per physician per week, and 94% citing burnout. The real change is direction: denials are rising, 32% say requests are often or always denied, and physicians report the June 2025 insurer reform pledge has not improved things. Policy is slow; your own workflow is not.

Every year the AMA surveys physicians on prior authorization, and every year the numbers land like a status report on a problem no one has fixed. The latest survey, released in May 2026, is no exception. The burden did not ease. It shifted, and where it shifted is toward more denials, even as insurers and regulators promised relief. Here is what the data shows, and the one part of it you can actually change.

The headline: no relief

The single most important takeaway from the AMA's most recent prior authorization survey is what did not happen: the burden did not go down. Fielded in December 2025 among 1,000 practicing physicians and released in May 2026, the survey shows the core burden holding roughly steady at levels that were already unsustainable, even though the period it covers included a high-profile insurer reform pledge and the run-up to new federal rules. That matters because the expectation, from patients, physicians, and policymakers alike, was that 2025 and 2026 would finally bend the curve, with more than 60 insurers pledging voluntary reforms and the CMS-0057-F rule beginning to take effect. Instead, physicians report that the day-to-day reality in their practices has not meaningfully improved. The AMA's own framing of the release was blunt on this point, headlining that the reform pledge has fallen short in physicians' experience. So the news is not a new catastrophe; it is the absence of the relief that was promised, which for practices already spending a workday a week on authorizations is its own kind of bad news. The context on the rules that were supposed to help is in the CMS-0057-F guide.

The numbers

Here are the core figures from the survey, paraphrased from the AMA's reporting.

The time figure, roughly 13 hours a week per physician, is the one that anchors everything, because it translates directly into staff cost and lost clinical capacity, the breakdown in the 13-hours guide. Note also that two in five practices report staff working exclusively on prior authorization, which tells you the burden is heavy enough that many practices have had to dedicate whole roles to it. All figures here are the AMA's, reported in its May 2026 release.

What changed: denials up

If the time burden is the flat line, denials are the line pointing the wrong way, and that is the real change in this survey. About 32% of physicians now report that requests are often or always denied, up from roughly 27% in the prior survey, and around 74% say denials have increased over the past five years, a striking level of agreement that the trend is worsening rather than improving. Layered on top is a newer worry: a majority of physicians express concern that insurers' growing use of automated and augmented-intelligence tools could push denial rates higher still, with some reporting patterns consistent with systematic, batch-style denials. The significance for a practice is direct, because rising denials are not just a policy grievance; they are money and time. Every denial is potential rework at real cost, and the industry evidence is that most denials are ultimately overturned when appealed, yet many are never appealed, so a rising denial rate against a practice with no rework process means more earned revenue quietly abandoned, the economics in the denial statistics, the rework cost guide, and the appeal success rate guide. The denials trend is where this survey should change how a practice thinks, because it is the part getting worse.

The pledge that fell short

The backdrop to these numbers is the June 2025 insurer pledge, when more than 60 health insurers publicly committed to voluntarily reform their prior authorization programs, with a set of specific promises on staggered timelines. It was meant to be a turning point. The AMA's May 2026 survey is, in effect, an early report card on that pledge, and physicians graded it poorly: they report little practical improvement in the burden, the volume, or the denials that the pledge was supposed to address. The AMA has long noted that similar voluntary commitments in the past produced limited change, and this survey suggests the 2025 pledge is, so far, following the same pattern. None of this means reform is impossible or that the regulatory changes will not eventually help; the CMS-0057-F rule carries real, enforceable requirements on decision timeframes and, from 2027, on electronic processing, unlike a voluntary pledge, detailed in the CMS-0057-F guide. But the survey is a clear signal not to wait on voluntary promises to fix your practice's burden, because as of this data, they have not. Policy may yet move; in the meantime, the burden is still yours to carry, or to reduce yourself.

Where the burden lands hardest

The survey also breaks down where the burden concentrates, which is useful for a practice deciding where to focus. Physicians reported high administrative burden across every major insurer, but not evenly: the AMA's data put UnitedHealthcare at the top for the highest burden categories, at around three-quarters of physicians, followed by Humana, Anthem/Elevance, Aetna, Cigna, and Blue Cross Blue Shield, each still reported as high-burden by a majority. The takeaway is not to single out one payer, since all of them rank high, but that prior authorization burden is a near-universal feature of dealing with major commercial plans, not a quirk of one contract. For a practice, that reframes the problem: you cannot escape it by shifting payer mix, because the burden follows the major plans across the board, which means the only durable relief is operational rather than contractual. The survey's specialty and patient-population breakdowns tell a similar story, with the burden varying in degree but present across the board. Knowing which of your payers drive the most authorization volume and the most denials, from your own data, lets you target your workflow where it will recover the most time, the payer-specific view that pairs with the specialty tactics in the by-specialty guide. The burden is broad, but your response can still be precise.

What is actually in your control

Which is the point worth ending on, because a national survey can leave a practice feeling like a spectator to forces it cannot influence. You cannot, by yourself, change insurer behavior or accelerate federal rules. But the survey measures an aggregate of individual practices, and your practice's share of that burden is far more in your control than the headline numbers suggest. The 13 hours a week is an average of practices handling prior authorization well and practices handling it chaotically, and the difference between them is operational: whether authorizations are tracked in a real system so none expires, whether denials are worked and appealed rather than abandoned, and whether the work is organized rather than ad hoc. A practice that runs prior auth as a tracked pipeline, with clear ownership and a denial-and-appeal process, spends its share of those 13 hours far more efficiently and loses far less to expired auths and unworked denials than one that does not, which is exactly what the system in the Zero-Slip guide and the specialty-specific tactics in the by-specialty guide are built to do. So read the survey for what it is: confirmation that the external burden is not easing, and therefore that the internal fix, the one you control, is where the leverage is. Reform may come. Your workflow can improve now, and that is the part worth acting on. For the full picture of prior auth as an operational problem, start with the prior authorization guide.

Frequently asked questions