New Medicare Reports Leave Gaps in Which Care Gets Denied

What an overall approval rate leaves out

A Medicare Advantage insurer can publish an approval rate above 90 percent without telling readers how often it rejected requests for the specific care they need after a hospital stay.

That limitation is visible in the government’s new transparency system. Medicare Advantage organizations had to begin publishing prior-authorization statistics this year, but the required figures combine different medical services. The rules provide a broader view of insurance decisions while leaving patients unable to isolate some of the categories federal investigators have flagged.

What’s the Scoop with B. Roach compared those requirements with a published Humana report and two June 2026 federal watchdog studies. The documents show a mismatch: the public reporting framework emphasizes aggregate outcomes, while investigators say meaningful oversight requires details about service types and the contractors making decisions.

The limits of Humana’s published figures

Humana’s disclosure for Medicare contract H0028 illustrates the difference. For calendar year 2025, it reports 609,174 standard, nonurgent requests, with 91.59 percent approved. A separate table records 917 approvals after appeal out of 1,341 appeals. Those are requests, not individual patients, and the document excludes drugs. It does not break those results into skilled nursing, rehabilitation or other service categories.

CMS guidance says overall approvals should include requests approved after appeal. Consequently, a headline approval percentage alone cannot distinguish immediate approval from approval obtained only after challenging a rejection. Separate appeal figures help, but combining services still prevents a reader from identifying where those disputes arose. This is a limitation of the required disclosure, not evidence that Humana violated it.

What the skilled-nursing investigation found

The stakes become clearer in the Health and Human Services inspector general’s skilled-nursing analysis. Among 19 Medicare Advantage organizations, investigators counted 13,458 initial denials of admission requests in June 2024. The records showed 2,313 appeal reversals and 132 upheld denials: approximately 95 percent of decided appeals favored the enrollee. Only about 18 percent of initial denials were appealed.

That does not mean 95 percent of all denials were wrong. Appealed cases are a selected group, and investigators did not review medical records to independently verify the submitted data. The study concerns one month’s admission requests, not an annual national patient count. Its results also cannot be treated as a year-to-year comparison with Humana’s separate 2025 disclosure.

The same investigation found that naviHealth processed half the skilled-nursing admission requests reviewed. Its denial rate was 14 percent, compared with 11 percent for requests processed within the insurers and 9 percent for other contractors. Those differences identify questions for investigation; they do not establish that comparable patients received different decisions.

A push for service and contractor details

A companion report examined long-term care hospitals and inpatient rehabilitation. It recommended that CMS routinely collect request-level data identifying the service and contractor involved. That recommendation concerns information collected by CMS; it does not itself require public service-by-service reports.

CMS has begun testing that approach. In its published response, the agency said a voluntary pilot launched in February 2026 collects service-level determinations and vendor involvement. CMS also described audits and possible enforcement action. It cautioned that additional documentation supplied during an appeal can justify reversing an initially appropriate denial.

Insurer commitments and proposed reporting changes

Humana makes a similar qualification in its disclosure: new information or a change in the member’s condition can explain a reversal. Separately, the insurer announced plans to remove approximately one-third of outpatient prior-authorization requirements by January 2026, arguing that review still protects patients receiving expensive or high-risk treatments. That announcement is a statement of the company’s commitments, not an independent assessment of their results.

CMS has also proposed adding numerical counts and other measures to public reports. Its September guidance still identifies that expansion as proposed.

For now, a high overall approval rate answers a broad statistical question. It does not tell a family how a plan handles the particular care their relative needs—or which outside company may be deciding.

Reporting method

This article analyzes publicly available federal records and an insurer disclosure. It does not include patient interviews or independent clinical reviews.