KFF brief reports 2025 denial rates of 12% to 18% across three insurance markets

Insurers rejected between 12% and 18% of routine prior authorization requests in 2025, depending on which insurance market they operate in, and appeals reversed many of the rejections that patients or their doctors challenged, a KFF analysis published August 13, 2026 found ^[1][2].

The figures come from a first-time public reporting requirement. Under a 2024 CMS rule, Medicare Advantage, Medicaid managed care and ACA federally facilitated Marketplace insurers must post annual prior authorization metrics on their websites; the first batch, covering calendar 2025, was due March 31, 2026 ^[1]. KFF examined filings from 14 insurers that each held at least 2.5% of enrollment in their segment. Together they cover 25 million Medicare Advantage members (69% of that market), more than 35 million Medicaid managed care members (54%) and nearly 11 million Marketplace members, or about 71 million people ^[1][2].

The three markets did not behave the same way. Medicare Advantage insurers denied 12% of standard requests and 10% of expedited ones. Medicaid managed care plans denied 14% and 12%. Marketplace insurers denied 18% and 16% ^[1]. Approval rates ranged from 82% to 90% across the markets and request types ^[1].

Denials seldom get challenged, KFF wrote, though the brief does not put a number on how often patients or providers appeal ^[1]. When standard requests were appealed, the decisions went the other way 67% of the time in Medicare Advantage, 47% in Medicaid managed care and 43% in the Marketplace ^[1][2].

Speed was not the main obstacle. The median wait from submission to decision was about one day for standard requests in all three markets ^[1][2]. Expedited requests were decided in a median of roughly half a day in Medicare Advantage, just under a day in Medicaid managed care and one day in the Marketplace ^[1][2].

Averages hide wide differences among carriers. In Medicare Advantage, UnitedHealth Group had the highest denial share for standard requests, 17%, while Centene had the highest for expedited requests, 13%. Medicaid managed care's top rates came from Independence Health Group at 23% for standard requests and CareSource at 21% for expedited ones. Centene led the eight Marketplace insurers in both categories, at 25% and 23% ^[2].

The insurer reports exclude prescription drugs, so no medication-specific denial rate appears in them ^[1]. Carriers submit percentages rather than counts, and the data are not split by type of service, leaving readers unable to tell whether a given denial rate reflects a handful of requests or many thousands ^[1][2]. KFF also said the numbers are hard to find and interpret on insurer websites, limiting their usefulness to consumers directly ^[1][2].

Why it matters for patients

People who use GLP-1 drugs such as semaglutide (Ozempic, Wegovy, Rybelsus) or tirzepatide (Mounjaro, Zepbound) usually meet the prior authorization system when a pharmacy or clinic files paperwork on their behalf. Nothing in this data set shows how often coverage for those drugs is refused, or how often a refusal is reversed; the reporting requirement carves out prescriptions entirely ^[1]. Those drug-level rates have not been reported.

What the brief does establish is that outcomes vary by market and by insurer, and that appeals succeed more often than they fail in each market studied ^[1][2]. Because KFF did not report what share of denials are appealed, the reach of the overturn rates is unclear ^[1]. KFF noted that prior authorization can delay or block care and adds work for patients and providers ^[1]. Nearly 7 in 10 insured adults call prior authorization a burden ^[1], and a March 2025 KFF report found that one in four insured adults struggled to understand the terms their insurers use ^[2].

What happens next

Insurers post these metrics every year under the rule, but the brief does not give the next deadline ^[1]. CMS has revised its reporting requirements and proposed rules that would widen what insurers must disclose and make formats more consistent, which KFF says could allow comparisons between plans ^[2]. When those proposals take effect and what they will finally contain has not been disclosed in the brief ^[2].