Which plans deny the most, and overturn the fewest?
The plan-by-plan scoreboard, straight from CMS's own published numbers: which ACA marketplace issuers deny the most in-network claims, and overturn the fewest on appeal. Plus the real denial-code playbook you can use today.
CMS's own denial numbers, named
This ranks ACA marketplace issuers by the denial rate CMS publishes for them. The precision below is the point: it is the number and the source, with no adjective added.
The denial rate is CMS's own published number: in-network claims denied divided by in-network claims received, all reasons, for ACA marketplace issuers, dates of service CY2023 (CMS PY2025 Transparency in Coverage PUF). It is issuer-level, not per-CPT and not per-provider. It covers marketplace and individual ACA plans only, not an issuer's commercial, employer, or Medicare Advantage behavior.
In-network claims denied over in-network claims received, all reasons, for one issuer across CY2023. CMS's published figure.
Of the appeals actually filed, the share the issuer reversed. Shown only where appeals were filed, else unavailable.
CMS CCIIO Transparency in Coverage Public Use File, Plan Year 2025, reflecting CY2023 claims. data.healthcare.gov.
Presented as CMS's figures. Each rate is reproduced from CMS's own file with its citation. We add no claim about any issuer's intent. A withheld rate is shown as unavailable, never as a zero.
Which codes to fight for Private Practice
Editorial field notes for this segment, decoded from the published X12 CARC list. Not a measured frequency and not your practice. No dollars, no rates.
Across specialties, the highest-volume denial drivers are authorization, medical necessity, coordination of benefits, and bundling edits. Most are documentation or sequencing problems, which means most are worth working rather than writing off.
A required prior authorization, precertification, or notification was not on file.
Supply the auth number if one existed, or request a retro-authorization with clinical justification and any urgent or emergent exception.
Prior-auth denials are frequently retro-authorizable inside the payer's window; check the provider manual for the retro-auth exception before appealing.
The payer decided the service was not medically necessary as billed.
Attach clinical documentation and the applicable coverage policy or LCD showing the service met medical-necessity criteria, and correct any weak diagnosis linkage.
Attach the coverage policy or LCD alongside the clinical note; a medical-necessity appeal that cites the payer's own criteria back to it is the one that moves.
Under coordination of benefits, the payer believes another plan is primary.
Submit an updated COB record or the primary payer's EOB establishing this payer's correct order of liability.
Verify primary vs secondary and rebill to the correct primary with the EOB attached: usually a resubmission, not an appeal.
The payer sees this as an exact duplicate of a claim it already received.
If it is not a true duplicate (different date, site, or units), document the distinction and resubmit; otherwise confirm the original claim's status.
A true duplicate is a write-off; a distinct same-day service needs the right modifier and documentation to support it.
The benefit for this line is included in the allowance for another service that was already adjudicated.
If the services were separate and distinct, append the appropriate modifier (25, 59, or an X{EPSU} modifier) with documentation showing separately identifiable work.
Check the NCCI PTP edit and whether a modifier is genuinely supported by the note. Never append a bypass modifier that the documentation does not earn.
The claim was received after the payer's timely-filing deadline.
Provide proof of timely original submission (clearinghouse acceptance report) or document a valid good-cause exception.
Timely filing: proof of the original, accepted submission (clearinghouse acceptance report) is the only durable appeal.
The appeal clock. Appeal deadlines are set per payer and per plan: read the denial letter. The clock usually starts at the remit date, not the date of service, and a corrected claim is often faster than a formal appeal.
This board is issuer-level federal data. To see your practice's own denial rate per code and per payer, connect your 835 and ERA remits in the Clean Room. A denial rate for your practice can only be measured from your own claims, including a measured zero. It is never modeled or averaged from a segment.
Bring your own remits and we rank the fights worth having: rate gap, denial pattern, and renewal timing, in priority order. Documented, modeled, never guaranteed.