The Federal Arbitration Numbers Insurers Would Rather You Not Read

Source: the Departments' Federal IDR bi-monthly reports, published every two months alongside the quarterly files; latest as of 30 September 2026. The first period of 2026 spans December 2025 to January 2026. These releases give volumes and closure reasons only — outcomes, award multiples, states, specialties and payers below come from the full supplemental tables for the second half of 2025, the most recent the Departments have published. CMS publishes both here. All of it is also available as JSON and CSV.

These percentages describe disputes that reached a payment determination, and they include default decisions, where one side failed to submit an offer or pay its fees. Separately, the share of closed disputes found ineligible was 18.7% in the second half of 2025 and has kept falling: 15.0% across January–September 2026, and 13.0% in August–September. Roughly one filing in seven still never reaches a decision, usually on eligibility or deadline grounds. That gap, not the win rate, is where most physician money is lost.

Why it matters for your practice: a headline “median award of 4–5× the QPA” is an artefact of very small claims, where a few dollars of difference reads as a large percentage. On the high-value claims that decide whether arbitration is worth your time, the realistic median is closer to 1.7–2.6×. We would rather you plan against the real number.

States marked “Yes” have a specified state law or All-Payer Model Agreement covering certain out-of-network disputes, which routes part of their volume through a state process instead of the federal one. Knowing which system a given claim belongs in is half the work — filing in the wrong forum is one of the ways a valid claim dies on eligibility.

42% of all federal IDR disputes have one of just two carriers on the other side. If your out-of-network denials cluster around one payer, that is not bad luck — it is a pattern the federal data already documents.

Plans and issuers initiated 177 disputes out of 1,372,563. Federal arbitration is, in the published data, almost entirely an instrument of the provider side. The plan you are disputing with is responding to your filing, not making its own.

Practices of 20 to 50 staff filed more disputes than any other known size band — about 2.7 times as many as groups over 500. Size was not recorded for 287,952 filings, so read the distribution as directional. It still says the same thing: this is not a process reserved for large systems. Figures cover non-air-ambulance disputes, which is how the Departments publish this table.

The last row is the one worth pausing on. In 35,730 filings — 2.6% of all disputes initiated in the second half of 2025 — the issuer did not respond and the plan type was never established. Since ERISA preemption generally keeps state arbitration away from self-funded plans, plan type is the fork that decides which process a dispute belongs in. Why disputes are found ineligible.

Of 1,145,039 determinations, 189,868 — 17% — were decided where only one party submitted an offer and paid its fees. Split decisions, where each side won part of a batched dispute, numbered 1,384: 0.12% of determinations. The process remains a choice between two offers.

The entity fee, not the administrative fee, is the money in this process: certified entities were paid 2.3 times what was collected in administrative fees. That fee falls on the non-prevailing party, which is why screening a claim before filing it matters more than the $15 at the door. What a dispute costs, in full.

• Win rates, QPA comparison and ineligibility shares: Supplemental Background on the Federal Independent Dispute Resolution Process, reports covering 2023 Q1–Q2, 2023 Q3–Q4 and 2025 Q3–Q4.
• Award multiples by claim size, initiations by state, and top disputing parties: Federal IDR Supplemental Tables, 2025 Q4 (Tables 7, 8, 9, 12 and 13).
• All of the above are published by the Departments of Health and Human Services, Labor, and the Treasury: cms.gov — Independent dispute resolution reports.

Figures describe the federal process nationwide. They are not Proprius Recovery case results and do not predict the outcome of any individual claim.