Yeamyeam
← All posts
Market Research

The Payer Denial Playbook

JZ MalikAug 25, 20269 min read

Denial rates are high, appeals are almost nonexistent, and when a claim is contested a meaningful share of the money comes back. This is a tour of the public datasets that show all three at once.

Scope: ACA Marketplace (QHP) plus cross-market prior authorization. Primary sources: the CMS Transparency in Coverage Public Use File (PUF), KFF, the AMA, and California's Department of Managed Health Care. Every number below is drawn from public reporting, not internal data.

01

Headline numbers

The most-cited number set in this space comes from KFF's analysis of CMS's own Transparency in Coverage PUF, the mandatory issuer-level report every HealthCare.gov QHP issuer must file.

19%

of in-network claims denied on ACA Marketplace plans, 2024 (85M of 451M claims)

37%

out-of-network denial rate, 2024, roughly 2x the in-network rate

<1%

of denied claims were ever appealed by the member (263K of 85M)

34%

overturn rate when a member did appeal internally in 2024

What this means

Denials are common, appeals are rare, and roughly a third of contested denials get reversed. The gap between denials that are winnable on appeal and denials that are actually contested is enormous, and it is an operational gap, not a clinical one.

02

Payers ranked by denial rate

KFF's issuer-level breakdown of the CMS PUF, restricted to parent companies with 5M+ submitted claims on HealthCare.gov. 2024 is the most recent complete plan year; 2023 is shown alongside because it names a wider set of issuers and shows more spread.

RankPayer (parent company)Denial rateRelative
1Oscar Health25%
2Molina Healthcare22%
2GuideWell Mutual Holding22%
4Harris Health21%
4Cigna21%
4BCBS Tennessee21%
7BCBS North Carolina19%
7UnitedHealth Group19%
7BCBS Alabama19%
7IHC Group19%
-Market average19%
LowElevance Health8%
2024 plan year, in-network claims, ranked by denial rate.
Payer (parent company)Denial rateFootprint reported by KFF
Blue Cross Blue Shield of Alabama35%12 plans
UnitedHealth Group33%274 plans, 20 states
Health Care Service Corp. (BCBS IL/TX/OK/NM/MT)29%915 plans, 4 states
Molina Healthcare26%72 plans, 9 states
Elevance Health23%154 plans, 7 states
2023 plan year, a wider issuer set with more year-over-year spread.

The full observed range across individual issuer-and-state combinations in 2023 ran from about 1% to 54%. Texas alone spanned 12% to 36% across its own issuers in the 2024 data. Payer averages hide enormous state-level and plan-level variance.

Consumers rarely have enough information to know, in advance, whether their insurer is likely to deny a claim, and even less information about why.
KFF, Claims Denials and Appeals in ACA Marketplace Plans, 2024

03

Prior authorization, by market (2025 data)

KFF's newest release (published Aug 2026) is the first public, insurer-named comparison of prior-auth denial rates across Medicare Advantage, Medicaid managed care, and ACA Marketplace, mandated under CMS-0057-F. It covers roughly 71M enrollees across 14 insurers with 2.5%+ market share per segment.

12%

average prior-auth denial rate, Medicare Advantage

14%

average prior-auth denial rate, Medicaid MCO

18%

average prior-auth denial rate, ACA Marketplace (highest of the three)

MarketHighest denierRateLowest denierRate
Medicare AdvantageUnitedHealth Group17%Elevance Health5%
Medicaid MCOIndependence Health Group23%L.A. Care Health Plan2%
ACA MarketplaceCentene25%GuideWell (Florida Blue)3%
Highest and lowest prior-auth deniers by market, 2025.

A few more points from the same release: UnitedHealth Group's Medicaid MCO prior-auth denial rate was 11%, and on ACA Marketplace it denied 21% of standard prior-auth requests. Centene's Medicare Advantage expedited-request denial rate was 13%, against Elevance's 3%.

04

What claims actually get denied for

The CMS PUF lets issuers report denial reasons in five CMS-defined buckets plus a catch-all "other." This is the weakest part of the public dataset (see the data limitations below), but the pattern has held steady across both years KFF has analyzed it.

2024 in-network denials, by reason

"Other" / unspecified36%
Administrative reasons25%
Excluded service13%
Lack of prior authorization / referral9%
Medical necessity5%

2023 in-network denials, by reason

"Other" / unspecified34%
Administrative reasons18%
Excluded service16%
Exceeded benefit limit12%
Lack of prior authorization / referral9%
Medical necessity6%

What this means

"Administrative reasons" plus "Other" is 61% of all denials nationally: coding mismatches, missing modifiers, eligibility and coordination-of-benefits errors, timely-filing misses, and incomplete documentation. Medical necessity, the category people assume dominates, is only 5 to 6%. The bulk of denials are high-volume, fixable-at-submission errors, not clinical judgment calls.

05

Appeals and overturn rates

Two different appeal systems produce two different numbers: standard claim denials appealed after the fact (CMS PUF), and prior-authorization denials appealed before service (CMS-0057-F reporting, 2025). Both point the same direction: appeals win more often than providers assume, but almost nobody files them.

Metric20232024
Denied claims appealed internally~1%<1% (263K of 85M)
Internal appeal, denial upheld56%66%
Internal appeal, overturned44%34%
External review filings (post-internal)rare5,881+ (4% of upheld internal appeals)
Post-claim appeals, ACA Marketplace, CMS PUF.
MarketAvg. overturned on appealNotes
Medicare Advantage~67%Range 40-93% by insurer; Centene 93%, Kaiser Permanente 40%
Medicaid MCO~50%UnitedHealth Group 81%, Molina 48%, CVS 22%
ACA Marketplace43%Range 16-54%; Centene 54% (only plan >50%), HCSC 16%
Prior-auth appeal success rate by market, 2025 (CMS-0057-F).

What this means

On Medicare Advantage, roughly two of every three appealed prior-auth denials get overturned, meaning the majority of MA denials that get contested were arguably wrong the first time. Yet post-claim appeal rates sit under 1%. The revenue is sitting on the table because appeals are labor-intensive to draft, not because the underlying denials are unbeatable.

06

State-level depth: California's IMR data

No national source publishes category-by-payer overturn rates. California's Department of Managed Health Care comes closest with its Independent Medical Review (IMR) program: every denial a member escalates to external review since 2001 is logged by health plan, treatment category, and outcome, in a public dataset that is updated regularly.

10.2%

of CA external-review denials overturned in 2025 (down from 12.7% in 2024)

201K

IMR applications processed in 2025, up 0.7% year over year

30.6%

of all IMR treatment requests were pharmacy-related

Behavioral and mental-health services, evaluations, and program services had the highest overturn rates of any category in the 2025 report: the categories most likely to have been wrongly denied in the first place. The underlying trend dataset is payer-level and category-level and downloadable, which makes it the best candidate for a follow-up deep-dive on plan-by-plan overturn benchmarks in California.

07

The administrative burden this creates

The AMA's 2024 physician survey quantifies the staff-time cost that denials and prior auth impose: the operational pain behind every one of the numbers above.

43

prior authorizations completed per physician, per week (average)

12 hrs

physician plus staff time spent on prior auth per week

35%

of physicians employ staff solely for prior auth

27%

of physicians say prior-auth requests are 'often or always' denied

94% of physicians say prior auth delays access to necessary care and 78% say patients abandon treatment because of it. 93% report a negative impact on clinical outcomes and 24% report that prior auth led to a serious adverse event. 95% say prior auth increases physician burnout, which makes it a retention problem, not only a revenue one.

08

Data limitations, read before quoting

CMS PUF / KFF caveats

The Transparency in Coverage PUF only covers HealthCare.gov (the federally facilitated marketplace). It excludes state-based marketplaces, Medicare, Medicaid, and employer-sponsored coverage. Denial reason codes cannot be linked to the specific service denied; a claim initially denied and later paid on appeal is still counted as denied; multiple reason codes per claim are counted separately, which can distort the category percentages; and issuer self-reporting quality varies, with CMS itself flagging data suppression where volumes are too low to report reliably.

2025 prior-auth dataset caveats

The CMS-0057-F dataset reports percentages only, not underlying request volumes, for most insurers. So a "low" denial rate at a huge insurer can still represent more denied patients than a "high" rate at a small one. Reporting formats were inconsistent across insurers in this first reporting cycle.

Sources

  • KFF, Claims Denials and Appeals in ACA Marketplace Plans in 2024
  • KFF, HealthCare.gov Insurers Denied Nearly 1 in 5 In-Network Claims in 2023
  • KFF, Prior Authorization Metrics Provide New Insights Into Insurer Practices (2025 data, published Aug 2026)
  • Healthcare Dive, Prior authorization denials vary widely among insurers
  • Becker's Payer Issues, 10 ACA insurers with the highest claim denial rates
  • CMS, Health Insurance Exchange Public Use Files (incl. Transparency in Coverage)
  • CA DMHC / CHHS Open Data, Independent Medical Review (IMR) Determinations
  • AMA, 2024 Prior Authorization Physician Survey

Figures compiled Aug 2026 from public datasets. CMS and KFF refresh these annually, so verify current-year numbers before citing them.

Yeam deploys AI medical employees into clinics.

Reception, documentation, coding, and billing, handled by agents so staff can focus on patients.