The insurer prior authorization pledge of June 2025 so far rests on its own numbers. In April 2026, AHIP and the Blue Cross Blue Shield Association reported that the health plans behind the pledge had cut prior authorization by about 11%, or roughly 6.5 million fewer requests. The reduction in Medicare Advantage was more than 15%. That is the first measured result of the most ambitious voluntary prior authorization commitment the industry has made.
It is also the only one. The pledge contains 6 commitments, and by January 1, 2026, 4 of them were due or already in effect. Only the cut in scope has a published number, and that number comes from the plans themselves, built by applying the codes they removed to the requests they received in 2024. In an American Medical Association survey released in May, only a third of physicians said they expect the pledge to make a meaningful difference.
The pledge is measured here against its own milestones, then against the industry’s 2018 commitments and the surveys that followed them.
Key takeaways
- The scope cut is the only reported result. Plans reported about 11% fewer prior authorizations, or 6.5 million fewer requests. The cut in Medicare Advantage was more than 15%.
- The 11% is modeled, not observed. Plans counted the codes they removed and applied those removals to their 2024 request volumes. The report does not publish the base or say whether any requirements were added.
- Most of what is due has no public measure. Beyond the scope cut, the plans have reported no numbers on the commitments that took effect by January 2026.
- Physicians are skeptical. In the AMA’s survey, 33% of physicians said they believe the pledge will make a meaningful difference, and 24% said denials are consistently reviewed by appropriately qualified clinicians.
- The 2018 pledge set a low bar. After the industry’s 2018 consensus statement, AMA surveys found little change in exemption programs or continuity of care over the following years.
- An independent check is already scheduled. The prior authorization metrics that CMS requires plans to post each March will show 2026 volumes for Medicare Advantage and exchange plans in March 2027.
What insurers promised
On June 23, 2025, AHIP announced that health plans covering about 257 million people had agreed to 6 commitments to simplify prior authorization, developed with HHS and CMS. The announcement listed 48 entries, some representing several companies, across commercial and public coverage. Later reports put the number of participating plans at about 50 to 60, depending on how affiliated companies are counted.
| Commitment | What the plans said | Target date |
|---|---|---|
| Standardize electronic prior authorization | Common, transparent submission standards for electronic requests | Operational by January 1, 2027 |
| Reduce the scope of prior authorization | Plan-specific reductions to medical prior authorization | Demonstrated by January 1, 2026 |
| Continuity of care when patients change plans | The new plan honors existing authorizations for benefit-equivalent in-network services for 90 days | Beginning January 1, 2026 |
| Clearer communication of determinations | Easy-to-understand explanations of decisions and of appeal options | Fully insured and commercial coverage by January 1, 2026 |
| Real-time responses | At least 80% of electronic approvals answered in real time, with the needed clinical documentation | In 2027 |
| Medical review of non-approved requests | Denials based on clinical reasons reviewed by medical professionals | In effect from June 2025 |
From AHIP’s June 2025 announcement. Descriptions paraphrase the commitments.
The commitments are sensible, and most of them track problems this journal has written about. Continuity of care is the subject of our February piece on approvals that last as long as the treatment. Scope is the argument we made in January for reviewing less. The commitments point the right way. What matters now is how anyone outside the plans can tell whether they were kept.
The first progress report
The first report arrived in April 2026, in a fact sheet from AHIP and a joint announcement with the Blue Cross Blue Shield Association. According to the fact sheet, the plans’ changes were projected to produce about 11% fewer prior authorizations in 2026, or 6.5 million fewer requests. The projected reduction in Medicare Advantage was more than 15%. Coverage of the report by AJMC described the cuts as focused on services with established clinical guidelines and predictable use.
The fact sheet also described changes at individual plans. The examples vary in what they count and when the changes were made.
| Plan | Change reported to AHIP |
|---|---|
| Cigna | Removed 345 services from its lists in 2025, an estimated 521,000 fewer medical prior authorizations (about 15%) |
| Kaiser Permanente | Removed more than 3,600 codes and expects nearly 20% fewer prior authorizations in 2026 than in 2024 |
| UnitedHealthcare | Cut volume for common services by nearly 20% in 2023 and another 10% in 2025. Its gold card program covers 514 codes |
| Humana | Removed more than 340 codes across 2024 and 2025, about a third of its 2025 outpatient diagnostic prior authorizations |
| Centene | Began imaging changes in late 2025, expected to eliminate about 300,000 prior authorizations a year |
| EmblemHealth | Removed 1,149 codes between 2021 and 2025 |
From the AHIP fact sheet (April 2026). Figures are as the plans reported them.
Several of these examples include changes made before the pledge. UnitedHealthcare’s figures begin in 2023, and EmblemHealth’s code removals run from 2021. That does not make them less real for the physicians affected. It does mean the examples describe a longer trend, not only the response to June 2025.
The fact sheet’s examples also show how differently plans approach the work. Some removed codes outright. Others, such as Highmark, which described a program for 25,000 clinicians it designates as trusted providers, exempt selected clinicians instead. Both approaches reduce requests, and each would show up differently in a plan’s data.
What the 11% does and does not show
The fact sheet explains its method in a short note. Plans counted the distinct procedure codes they removed from prior authorization lists as of January 1, 2026. They then applied those removals to the requests they received in 2024 to estimate how many requests would no longer be needed. Requests handled by delegated vendors were included. The figures cover Medicare Advantage and fully insured commercial coverage, including exchange plans.
That method is reasonable for a first report, and it has 4 limits that readers should keep in mind.
It is a projection. The 11% is an estimate of requests that would not occur in 2026, based on 2024 volumes. It is not a count of requests that did not occur. Enrollment and utilization change from year to year, so the actual effect could be larger or smaller.
The base is not published. The report gives a percentage and a total but not the number of requests behind them. KFF counted 52.8 million prior authorization determinations in Medicare Advantage alone in 2024. Without the plans’ own base, a reader cannot tell what share of the market the 6.5 million represents, or how the Medicare Advantage figure relates to the total.
It counts removals, not net change. The method counts codes taken off prior authorization lists. The fact sheet does not say whether plans added requirements for other services during the same period, which would offset some of the reduction.
It is self-reported. The plans supplied the figures, and the report does not describe any independent review. None of this suggests the figures are wrong. It means they are the plans’ estimate of their own progress, which is where any progress report starts and where a credible one should not end.
A pledge that grades itself can be accurate. It cannot be convincing.
There is also a question of which requests were removed. A code removed from the list because nearly every request for it was approved saves real work for physicians and reviewers, and it changes little about what care is delivered. That is the kind of change this journal has argued for. But it also means the 11% says more about administrative effort than about access to care. The services where denials are frequent, and where patients feel delays most, are by design the ones that remain.
The commitments with no public measure
By January 1, 2026, 4 commitments were due or already in effect. Besides the scope cut, none has a reported number.
Continuity of care. Since January 1, a plan that enrolls a patient in the middle of treatment is supposed to honor the patient’s existing authorizations for benefit-equivalent in-network services for 90 days. No report has said how many transitions the plans have handled or how many authorizations they honored. AJMC’s coverage of the April report described the policy but gave no numbers for it. The commitment is also narrower than it may sound. It covers in-network services that are equivalent under the new benefit, and it lasts for a transition period, not the length of treatment.
Clearer explanations. Plans committed to easy-to-understand explanations of determinations, with information on appeals, for fully insured and commercial coverage by January 1, 2026. That is hard to measure from outside without seeing the letters. It is easy to measure from inside, for example as the share of denials that name the specific criterion not met. Federal rules already require a specific denial reason for Medicare Advantage and other impacted payers starting in 2026, as we described in January.
Medical review of denials. The pledge said all non-approved requests based on clinical reasons “will continue to be reviewed by medical professionals,” and that this commitment was in effect from the start. The only outside measure is the AMA’s survey, in which 24% of physicians said medical necessity denials are consistently reviewed by appropriately qualified clinicians. The 2 standards are not the same. The pledge did not define medical professionals, and the AMA’s question asked about appropriately qualified clinicians, a narrower standard.
The other 2 commitments, standardized electronic prior authorization and real-time responses, are due in 2027.
What physicians report
The AMA released its 2025 physician survey in May 2026. It surveyed 1,000 practicing physicians. The release does not say when the survey was fielded, which matters because the first milestones took effect on January 1.
The results describe a burden that has not yet moved much. Physicians reported an average of 40 prior authorizations a week, taking about 13 hours of physician and staff time. Of the physicians surveyed, 40% employ staff who work only on prior authorization. Most said prior authorization delays care and contributes to burnout, and 26% said it had led to a serious adverse event for a patient in their care.
On the pledge itself, the AMA reported that 33% of physicians believe it will make a meaningful difference for patients and doctors. In the AMA’s summary of the results, AMA President Bobby Mukkamala said physician trust in voluntary pledges “is deeply eroded after years of unfulfilled promises,” and that insurers must rebuild it with action that can be measured.
Physicians also rated insurers by burden. UnitedHealthcare drew the highest share of “high” or “extremely high” burden ratings, at 75%, followed by Humana at 65%. Survey ratings reflect physicians’ experience and are not measured performance.
The survey also includes 2 measures that track how slowly the underlying conditions have changed. Only 5% of physicians said they contract with plans that offer exemption programs such as gold cards, and 88% said prior authorization interferes with continuity of care. The same questions, asked after the last industry pledge, produced similar answers.
The 2018 pledge is the baseline
The June 2025 pledge was not the industry’s first. In January 2018, insurer and physician groups, including AHIP and the AMA, signed a consensus statement on improving prior authorization. It contained 5 commitments, and several of them read like the 2025 list. They included exempting clinicians with strong approval records and protecting continuity of care. The statement also called for faster automation of prior authorization.
The AMA tracked the result. Its December 2018 survey, published in March 2019, found “little to no progress.” A December 2021 survey, reported by Medical Economics in May 2022, found much the same.
| Measure in AMA surveys | December 2018 | December 2021 | 2025 survey |
|---|---|---|---|
| Physicians contracting with plans that offer exemptions or apply prior authorization selectively | 8% | 9% | 5% |
| Physicians who say prior authorization interferes with continuity of care | 85% | 88% | 88% |
| Physicians whose EHR offers electronic prior authorization for prescription drugs | 21% | 26% | 24% |
From the AMA’s published summaries of each survey. Question wording varies across years.
The comparison is imperfect, and the 2025 survey may predate the January milestones. But it sets the standard the new pledge has to beat. On the measures the AMA has tracked since 2018, the experience physicians report has barely moved.
The 2 pledges differ in ways that may matter. The 2018 statement, as the AMA summarized it, described goals without target dates. The 2025 commitments have target dates and a public progress report, and federal officials were involved from the start. Several of them also overlap with federal rules that plans must meet anyway. Those features give the 2025 pledge a better chance than its predecessor. They do not substitute for evidence.
An independent check is already scheduled
The plans do not have to rely on their own estimates for long. The CMS interoperability and prior authorization rule requires impacted payers, from Medicare Advantage organizations to exchange issuers, to post prior authorization metrics every year by March 31. The first postings, for 2025, appeared in March 2026. Most of the plans we reviewed published counts of requests as well as rates.
The second postings, due by March 31, 2027, will cover 2026. For a plan that removed codes in 2025, 2026 requests should be lower than 2025 requests, after accounting for changes in enrollment. That comparison will be imperfect. It will also be public and comparable across plans, which the April report is not.
A reader could run the check in a few steps once the 2026 postings appear.
- Find the plan’s 2025 and 2026 postings for the same contract or issuer.
- Compare total standard and expedited requests between the 2 years.
- Adjust for enrollment, using CMS enrollment files for the same contracts, to express requests per enrollee.
- Compare the change with the plan’s claim in the AHIP report, where the plan reported one.
The check has a gap. The CMS metrics do not cover fully insured employer coverage, which is a large part of the pledge. For that market, the only public evidence will come from state reporting laws, such as Minnesota’s requirement to publish denials by service, or from whatever the plans choose to report.
The 2027 commitments line up with federal deadlines too. Impacted payers must have a Prior Authorization API in place by January 1, 2027, under the same CMS rule. The pledge’s commitment to standardized electronic prior authorization extends that expectation to commercial coverage, and its commitment to answer 80% of electronic approvals in real time sets a performance target the rule does not. Plan executives told MedCity News in December 2025 that the technology can be expensive for smaller practices, which is where the 2027 goals will be hardest to reach.
What credible measurement would look like
The plans can make the next report far more convincing without new rules. Each commitment can be measured from records the plans already keep.
- Publish the base. Report the number of requests in the baseline year by line of business, so that each percentage can be checked against public data.
- Report the net change. Count prior authorization requirements added as well as removed, and report actual request volumes once 2026 data is complete.
- Measure continuity. Report how many new enrollees arrived with active authorizations, and how many of those authorizations the plan honored.
- Measure explanations. Report the share of denials that name the specific criterion or documentation that was missing.
- Measure review. Report the share of clinical denials made by a physician in the same or a similar specialty as the ordering clinician.
- Measure real time. Report the share of electronic approvals returned while the clinician is still in the workflow, with the definition of real time stated.
- Invite verification. Have an independent party, or state regulators, review a sample of the underlying data.
None of these measures requires a new system. Each is a count of decisions a plan already makes and records.
Where Redlake stands
We build tools health plans use in utilization review, and we think the pledge’s commitments are the right ones. Our own rules point the same way. A clinician makes every denial. A machine can say yes, but only a person can say no. Real-time approval of complete requests that meet the criteria is exactly what software should do, and it is how a plan reaches the 80% target without shortcuts on denials.
Our third rule is the one that matters most for the pledge. Every decision can be reconstructed. When each decision records the service and the criteria, along with the reviewer and the time stamps, every measure in the list above is a query. The reports that would satisfy physicians are the same reports that would tell a plan where its own process is slow or unnecessary.
We also think the first report deserves a fair reading. Plans removed thousands of codes from their lists, and many physicians will notice fewer requests in 2026. The industry’s problem is not that it reported too little. It is that the only numbers are its own, at a moment when the people it most needs to convince have learned to wait for proof.
What to watch in the rest of 2026
- The next progress report. Whether AHIP and the Blue Cross Blue Shield Association publish an update before the 2027 milestones, and whether it includes the base and the commitments that have no numbers yet.
- Continuity in practice. Whether physicians and patients report that new plans honored existing authorizations after January 1.
- Readiness for 2027. Whether the standardized electronic framework is ready for providers who lack the technology.
- State action. Whether legislatures write pledge commitments, such as 90-day continuity, into law, which would make them enforceable for state-regulated plans.
- March 2027. The second round of CMS metrics, the first public data covering a full year after the scope cuts.
About this analysis
Basis. The pledge and its progress are described from AHIP’s June 2025 announcement and April 2026 fact sheet, along with press coverage of the April report. Physician survey results come from the AMA’s May 2026 release and summary, and from AMA survey findings published in 2019 and 2022. CMS’s public reporting requirements are described from earlier Rounds analyses and the CMS fact sheet for the interoperability and prior authorization rule.
Data notes. The 11% and 6.5 million figures are projections reported by the plans, based on 2024 volumes. Plan examples are as the plans reported them to AHIP. AMA survey questions changed in wording across years, so the comparison table shows trends rather than identical measures. The KFF figure covers all Medicare Advantage insurers, including some that did not sign the pledge. The verification steps are this analysis’s suggestion, not a method CMS or AHIP has published.
Sources. AHIP, press release (June 2025) and prior authorization fact sheet (April 2026). AJMC (April 2026). MedCity News (December 2025). AMA, survey release and summary (May 2026). AMA (March 2019). Medical Economics (May 2022). KFF (January 2026). CMS, interoperability and prior authorization final rule fact sheet (January 2024).
Frequently asked questions
What did health insurers pledge on prior authorization in 2025?
In June 2025, health plans covering about 257 million people made 6 commitments. They included reducing the scope of prior authorization by January 2026 and honoring existing authorizations for 90 days when patients change plans. Standardized electronic prior authorization is due by January 2027.
Did the pledge reduce prior authorization?
The plans reported in April 2026 that their changes would produce about 11% fewer prior authorizations in 2026, or 6.5 million fewer requests. The figure is a self-reported projection based on 2024 volumes.
How will the pledge be verified?
There is no independent audit yet. The prior authorization metrics that Medicare Advantage and exchange plans must post under CMS rules, due each March, will show 2026 volumes in March 2027.
What do physicians say about the pledge?
In an AMA survey released in May 2026, 33% of physicians said they believe the pledge will make a meaningful difference, and 24% said medical necessity denials are consistently reviewed by appropriately qualified clinicians.