Report 01 · July 2026

The first year insurers had to show their denials

In March 2026, many health insurers had to publish their 2025 permission-before-care numbers. This is what we found when we went looking — insurer by insurer, posting by posting.

80.7%
of appealed Medicare Advantage denials reversed, 2024
11.5%
of denials appealed
4.1M
requests for care denied
1 of 10
reports included a file software could check

For patients, reporters, lawyers, employers, and regulators

What this report is for

Use this report when you need a sourced starting point: to see which numbers are public, compare like records, find the original posting, or ask an insurer why a required field is missing. It is useful context for a patient, a reporter, a lawyer, an employer, or a regulator. It is not a case decision, a legal finding, or a substitute for the notice and plan documents in front of you.

The appeal gap

Start with the national Medicare Advantage prior-authorization record. In 2024, insurers denied 4.1 million requests for care. When patients or their clinicians appealed, 80.7 percent of those denials were reversed. The reversal rate exceeded 80 percent in each year from 2019 through 2024.

But only 11.5 percent of denials were appealed. Of every 1,000 denials, 885 did not enter the formal appeal process. Of the 115 that did, 93 were reversed.

This does not prove that any individual denial was wrong. It does show that formal review frequently changes the result and that the appeal gap deserves scrutiny.

Source · KFF analysis of CMS Medicare Advantage data (2024)
Published
Jan. 28, 2026
Capture note
Insurer-reported to CMS; unaudited.
Public lineage
Original source linked below. A public immutable archive and provenance ID are not yet published.
Open original source ↗

What the new postings show

The disclosures required by rule CMS-0057-F went live around March 31, 2026, covering calendar 2025. They are self-reported and uneven — and they are still revealing.

The spread is enormous. Across UnitedHealthcare’s 75 Medicare contracts, denial rates run from 2.3 percent to 22.8 percent — a ten-fold difference inside one company, under one rule, in one half-year file. Kaiser Permanente’s flagship Southern California Medicare contract denied 19.5 percent of standard requests; of its 27,992 denials, 310 were appealed — one in ninety — and Kaiser reversed itself on only a quarter of those.

The sickest members see the most denials. Highmark’s dual-eligible special-needs contract shows triple the denial rate of its mainline Medicare contract. Industry-wide, insurers denied 65 percent of prior-authorization requests for long-term care hospitals and 54 percent for inpatient rehabilitation — against under 8 percent overall.

Source · UnitedHealthcare CMS Interoperability prior-authorization PDFs
Published
2026
Capture note
Accessible Medicare file covers Q1–Q2 2025 only. Captured Jul. 8, 2026.
Public lineage
Original source linked below. A public immutable archive and provenance ID are not yet published.
Open original source ↗
· Kaiser Permanente CMS prior-authorization reports (Southern California PDF)
Published
report dated Mar. 27, 2026
Capture note
Captured Jul. 8, 2026.
Public lineage
Original source linked below. A public immutable archive and provenance ID are not yet published.
Open original source ↗
· Highmark prior-authorization metrics report, CY2025
Published
2026
Capture note
Captured Jul. 8, 2026.
Public lineage
Original source linked below. A public immutable archive and provenance ID are not yet published.
Open original source ↗
· KFF, MA prior authorization for post-acute care
Published
Jul. 6, 2026
Public lineage
Original source linked below. A public immutable archive and provenance ID are not yet published.
Open original source ↗

The public-report check

The rule requires each insurer to post its metrics on its own public website. Of the ten major insurer organizations in the pilot, three published complete, usable data. Cigna disclosed percentages without a single request count. Several organizations scattered files across subsidiary websites with no corporate index, some on servers that reject archival tools. Highmark published detailed counts but repeated one table across three contracts. BCBS Michigan was initially missed by our search protocol; we corrected that finding on July 9, 2026.

Only one insurer of the ten, Kaiser Permanente, published a file software could open and check directly.

See the public-report check, insurer by insurer →

What happens next

The source postings refresh each March 31. We are checking the 2025 reports now, collecting the next year of activity during 2026, and building the source trail needed to compare changes over time. A public archive and version history are not live yet; we will not imply that they are.

Source · CMS WISeR model; WP Intelligence (Washington Post) Texas analysis via PYMNTS
Published
Jun. 25, 2026
Public lineage
Original source linked below. A public immutable archive and provenance ID are not yet published.
Open original source ↗

The first year of public denial reporting is not a clean answer. It is a new record — one that can finally be checked, compared, and challenged.