Prior Authorization Denial Rates: A Patient's Guide to CMS-0057-F Data

For the first time ever, health insurers are required to publicly report how often they deny prior authorization requests. Under CMS's Interoperability and Prior Authorization Final Rule (CMS-0057-F), insurers had to post their calendar year 2025 data on their websites by March 31, 2026. We reviewed the filings from the nation's largest insurers. The results confirm what patients and doctors have long suspected: denial rates vary wildly by insurer, and most denials are beatable on appeal.

Denial Rates by Insurer (CY2025)

This table is sourced directly from each insurer's own CMS-0057-F filing — not from any third-party aggregator. We reviewed the public filings from seven of the nation's largest insurers across multiple states and plan types.

What Insurers Must Now Disclose

CMS-0057-F requires MA, Medicaid, and ACA marketplace insurers to publicly post eight categories of PA data every year — including approval/denial rates, appeal outcomes, expedited request data, and average turnaround times. This is the first year this disclosure has been mandated.

Insurer-by-Insurer Detail

The comparison table above shows the full picture. Below are brief notes on each insurer's CY2025 filing, including what they reported, how to find it, and what stands out.

UnitedHealthcare

UnitedHealthcare's transparency page publishes denial rates broken out by plan type: Medicare Advantage members see a 4.6% denial rate while ACA marketplace members face 19.7%, with Medicaid (8.5%) and Employer & Individual (11.5%) in between.

Cigna

Cigna's IFP FFE disclosure covers ACA marketplace plans only and shows a 27% standard denial rate with only 16% of appeals overturned. Unlike most insurers where appeals routinely succeed, Cigna denials largely stick — making strong clinical documentation critical from the start.

Ambetter (Centene)

Ambetter's Florida filing (Issuer ID 49004) processed 419,452 standard PA requests — a massive volume that makes this data statistically robust. They denied 21% of standard requests, but unlike Cigna, 58% of appeals were overturned. If your Ambetter plan denies a PA, appealing is strongly worth it. Across 19 states, Ambetter's ACA marketplace denial rates range from 20.7% (North Carolina) to 32.9% (Texas) — a significant spread for the same insurer:

Aetna (CVS Health)

Aetna's Medicare Advantage filing (Contract H0523) shows the lowest denial rate we found: 2.39% standard, 2.94% expedited. A second contract (H0562) reported 100% approval with zero denials. When denials do occur, 53% are overturned on appeal.

Humana

Humana published contract-level MA reports showing a 6.17% standard denial rate with a strong 67% appeal overturn rate and a fast 5-hour average expedited turnaround. Their Kentucky Medicaid data tells a different story — 12.79% denied with only 29% appeal success. See the Medicaid gap below.

Anthem (Elevance Health)

Anthem published state-level Medicaid scorecards showing significant variation: Indiana at 5.3% denied with 42% appeal success vs. Virginia at 11.1% denied with only 28% appeal success. Notably, Anthem Indiana denies expedited requests at nearly double the standard rate (10.2% vs. 5.3%) — see the expedited paradox below. Anthem also publishes a separate commercial utilization report for Indiana showing a 20% denial rate on a smaller volume of requests — a notably higher rate than their Medicaid filing, though the two reports use different methodologies and scopes.

Blue Cross Blue Shield Affiliates

BCBS affiliates operate independently, and their CMS-0057-F data varies dramatically. The standout: Blue Cross Complete (Michigan's Medicaid managed care plan) denied 34.9% of standard PA requests — the highest rate we found from any insurer for any plan type — with a 5.9-day average turnaround. At the other extreme, Horizon BCBS of New Jersey denied just 0.5% of Braven Health (MA) requests with a 96.2% appeal overturn rate — the lowest denial rate and highest overturn rate we found anywhere. Other BCBS affiliates we reviewed include Florida Blue, Blue Cross NC, AZ Blue, Excellus (NY), CareFirst (MD), Highmark (DE), and BCBS Nebraska. Notably, HCSC — the parent of BCBS of Texas and BCBS of Illinois (the 4th-largest insurer in the US) — appears not to have published CMS-0057-F data in any publicly findable location.

Kaiser Permanente

Kaiser — the nation's largest integrated health system — published regional CMS-0057-F reports covering 8 regions. The surprise: Kaiser's Southern California MA plan denied 19.48% of standard requests — among the highest MA denial rates we found and far above the industry average. By contrast, their Georgia region denied just 3.13%. Other regions: Northern California (10.5%), Northwest/Oregon (12.6%), Colorado (9.6%), Hawaii (7.1%), Washington (6.1%), and Mid-Atlantic (5.5%).

The Medicaid Gap: Higher Denials, Lower Appeal Success

One of the most troubling patterns in the CY2025 data is the disparity between plan types within the same insurer. Medicaid enrollees — by definition lower-income — face both higher denial rates and lower appeal success rates:

Humana's Kentucky Medicaid members are denied at twice the rate of their Medicare Advantage members (12.79% vs. 6.17%) — and when they appeal, they win less than half as often (29% vs. 67%). The Trump administration's June 2025 decision to suspend CMS's health equity reporting requirements means this disparity will not be tracked at the demographic level — making data like this more important.

The Expedited Denial Paradox

You might expect urgent PA requests — filed when a patient's health is at immediate risk — would be approved more often. The CY2025 data shows the opposite at some insurers:

Anthem Indiana denies expedited requests at nearly double the rate of standard ones. If your urgent PA is denied, don't assume it was a well-considered clinical decision — the data suggests expedited reviews may receive less thorough evaluation, not more.

Why Most Patients Leave Healthcare on the Table

The appeal overturn rate is the most actionable number in these reports. When over 80% of appeals succeed, it means most initial denials don't hold up. But the CY2025 filings reveal exactly how few patients fight back:

Same Plan, Different State: Why Your Ambetter Coverage Varies by 12 Percentage Points

The Ambetter data reveals something remarkable: the same insurer, offering the same ACA marketplace plan type, denies PA requests at vastly different rates depending on the state. In North Carolina, Ambetter denies about 1 in 5 standard requests (20.7%). In Texas, it's nearly 1 in 3 (32.9%). That's a 12-percentage-point spread for the same company.

This pattern suggests that state regulatory environments may play a significant role in how aggressively insurers apply PA denials — though the relationship is not straightforward. Texas, for example, enacted a "gold card" law (TX Insurance Code 4201.653) in 2021 that exempts high-performing providers from PA requirements, yet still has the highest Ambetter denial rate at 32.9%. Iowa also has gold card legislation yet ranks second-worst at 31.1%. Gold card laws may be reactive — passed because denial rates were already high — and primarily benefit providers rather than all patients.

The Trend: Denials Are Rising (2019–2024)

Despite insurer pledges to streamline prior authorization, total PA volume in Medicare Advantage has grown 35% since 2019 and denial rates have trended upward:

How to Look Up Your Insurer's Report

Under CMS-0057-F, every MA, Medicaid, and ACA marketplace insurer must post their PA metrics publicly. Here's where to find them:

Who Actually Reviews Your Prior Authorization? The Vendors Behind the Denial

When patients hear "UnitedHealthcare denied my prior authorization," the decision often wasn't made by UnitedHealthcare staff at all. Most large insurers outsource clinical PA review to specialized vendors that evaluate medical necessity against the insurer's coverage criteria. Knowing which vendor handled your case is useful when you appeal — peer-to-peer reviews, escalation paths, and even the medical director who signs the denial all sit at the vendor.

Two vendors handle a large share of commercial and Medicare Advantage PA volume:

• eviCore by Evernorth (Cigna subsidiary, acquired through the 2018 Express Scripts deal) — the largest specialty PA vendor, performing medical necessity reviews for advanced imaging, radiation oncology, cardiology, musculoskeletal, oncology, laboratory management, sleep, and post-acute care. Per a 2024 ProPublica / Capitol Forum investigation, eviCore performs PA review for over 100 million people — about a third of insured Americans — across plans including major commercial insurers, BCBS affiliates, and state Medicaid programs. The investigation found eviCore's state-aggregate denial rate reached nearly 20% in Arkansas (versus ~7% baseline for Medicare Advantage), documented internal sales pitches promising a "3-to-1 return on investment" from denial volume, and detailed an internal algorithmic tool that staff call "the dial" — described as adjustable to raise denial rates.
• Cohere Health — an AI-driven PA platform used by Humana, starting with musculoskeletal (2021), expanding to cardiovascular and surgical (2023), and adding diagnostic imaging and sleep (2024). Cohere markets "AI-powered prior authorization" and claims "85% of prior authorizations approved in real-time", with the remainder routed to clinical reviewers. Useful to know if your Humana request is delayed — the workflow runs through Cohere, not Humana directly.

Other large insurers operate in-house or through proprietary subsidiaries for post-acute care review.

How to find out which vendor reviewed your case: read your denial letter — the reviewing entity is usually disclosed at the bottom ("reviewed by eviCore on behalf of [insurer]"). When you request a peer-to-peer review, the call is typically scheduled with the vendor's medical director, not your insurer's. Ask the vendor for the specific clinical guideline used (e.g., MCG, InterQual, or a proprietary criteria set) and request the citation in writing.

eviCore publishes its own utilization data — under state mandate (Arkansas SB 318 and DC Reasonable Health Insurance Ratemaking Act), eviCore posts procedure-level quarterly denial reports on its website at evicore.com/annual-utilization-statistics. Arkansas reports run from Q1 2018 through Q1 2025 (28 quarters of procedure-by-CPT-code denial data); DC reports include appeals/overturn data. eviCore also publishes marketing-level claims via Evernorth that "approximately two-thirds of medical decisions [are made] in real time" and "90% of approvals are completed in one business day". The state-mandated reports are the most useful — if your case involves a specific CPT code, you can look up the historical denial rate for that procedure under eviCore review.

The 2026–2027 CMS-0057-F Timeline

The rule that produced the disclosure data above also rewrites how PA decisions must be made. Beyond the annual denial-rate disclosure (which the data above is built on), CMS-0057-F phases in operational requirements:

• Faster decisions: 7 calendar days for standard PA requests and 72 hours for expedited requests, effective January 1, 2026 (down from the 14-day standard under 42 CFR 422.568 for Medicare Advantage). Applies to MA, Medicaid fee-for-service, Medicaid managed care, and CHIP — but not to ACA marketplace QHPs on the federally-facilitated exchanges, which were explicitly excluded from this provision (CMS fact sheet).
• Specific denial reasons: beginning in 2026, impacted payers must provide a specific reason for each denied prior authorization, regardless of how the request was submitted — not generic boilerplate.
• Electronic PA API (FHIR-based): impacted payers must implement an HL7 FHIR-based Prior Authorization API beginning January 1, 2027. The API must publish covered services, identify documentation requirements, and support PA request/response transactions. Drug PAs are excluded from this requirement.

For patients, the practical implication is straightforward: if your PA decision takes longer than 7 days (standard) or 72 hours (expedited), or if your denial letter doesn't cite a specific reason, that is itself a basis for appeal and a complaint to your state insurance department. (For ACA marketplace plans on healthcare.gov, the federal timeframe rule does not yet apply; check your state insurance regulator for any state-level standard.)

What to Do If Your Prior Authorization Is Denied

For detailed step-by-step instructions, see our complete prior authorization appeal guide. For help with the bills that result from delayed or denied care, you can upload your medical bill to check for billing errors and generate professional appeal letters.

Methodology

This article presents original research. We reviewed the CMS-0057-F prior authorization transparency filings published directly by each insurer on their public websites, covering calendar year 2025 (January 1 – December 31, 2025). We extracted data from UnitedHealthcare, Cigna, Ambetter/Centene (19 states), Humana, Aetna, Anthem/Elevance (6 states), Kaiser Permanente (8 regions), and nine BCBS affiliates — covering Medicare Advantage, Medicaid, and ACA marketplace plan types across more than 25 states.

Data methodology varies between insurers in how they count post-appeal approvals and how they aggregate by plan type. Cigna's report covers only their IFP FFE (ACA marketplace) plans. Anthem reports at the state level for Medicaid. Humana reports at the MA contract level. Ambetter reports at the issuer level per state. We note these differences throughout the article and show the plan type for each data point in our comparison table.

Missing data: HCSC, the parent company of BCBS of Texas and BCBS of Illinois (the 4th-largest insurer in the US), has no publicly findable CMS-0057-F filing as of publication. We will update this article if and when their data becomes available.

Historical trend data (2019–2024) uses CMS aggregate Medicare Advantage data.

Resources

• CMS — Interoperability and Prior Authorization Final Rule (CMS-0057-F)
• UnitedHealthcare — CMS-0057-F Prior Authorization Transparency (CY2025)
• Cigna — IFP FFE Prior Authorization Disclosure Report (CY2025)
• Aetna — Medicare Advantage Prior Authorization Metrics (CY2025)
• Ambetter (Centene) — ACA Marketplace Prior Authorization Metrics by state (CY2025): FL, TX, IA, OK, AL, OH, DE, AZ, KS, IN, LA, NH, MO, TN, MI, NE, MS, SC, NC
• Anthem (Elevance) — Indiana Medicaid Prior Authorization Metrics (CY2025)
• Anthem (Elevance) — Virginia Medicaid & CHIP Prior Authorization Metrics (CY2025)
• Anthem (Elevance) — Nevada Medicaid & CHIP Prior Authorization Metrics (CY2025)
• Anthem (Elevance) — New York Medicaid Prior Authorization Metrics (CY2025)
• Anthem (Elevance) — California Medicaid Prior Authorization Metrics (CY2025)
• Anthem (Elevance) — Wisconsin Medicaid (BadgerCare) Prior Authorization Metrics (CY2025)
• BCBS Affiliates — Prior Authorization Metrics (CY2025): Blue Cross Complete MI, BCBS MI, Florida Blue, Blue Cross NC, Horizon NJ, Excellus NY, CareFirst MD, AZ Blue, Highmark DE, BCBS NE
• Kaiser Permanente — Regional Prior Authorization Metrics (CY2025): SoCal, NorCal, CO, GA, HI, Mid-Atlantic, Northwest, WA
• KFF — Medicare Advantage Insurer Prior Authorization Data (CY2024)
• HealthCare.gov — How to Appeal an Insurance Company Decision