The IDR submission, built from the record — acuity, complexity, training, all cited.
No Surprises Act IDR support prepares the offer submission from the file itself. The additional-circumstances factors at §149.510(c)(4)(iii)(C) are the entire argument, so we pull patient acuity, case complexity and the clinical detail behind them out of the record and the claim file, cited page by page, into a submission a certified IDR entity can read in one pass.
The rulebook is published. So is every determination.
We do not build for a case type until we can name the standard that defines a correct output and the person who grades the file against it. Federal IDR gives us both in writing, then publishes the scoreboard.
That makes this the cleanest feedback loop in claims work: a binary decision, on a published clock, against a factor list you can read in the regulation.
Twenty pages or three hundred, and millions of disputes behind them.
A single IDR dispute file runs 20 to 300 pages, with about 160 in the middle, and it rarely arrives as one production. The claim record comes out of the billing system, the clinical detail comes out of the chart, and the payer correspondence comes back separately. The acuity and complexity facts that decide the case are spread across all three.
Volume is the other half of the problem. Federal IDR now runs at roughly 2.6 million disputes a year and climbing: 1.2 million in the first half of 2025, twice the same period in 2024, and 4.8 million cumulative since 2022. CBO forecast 17,000 a year when the process was designed, off by a factor of 150. The backlog sits near 430,000.
The pages themselves are not clean. Faxed operative notes, handwritten anesthesia records, and the same discharge summary repeated in every production. Duplicates are free here, so a batched submission costs what its distinct pages cost.
One file first. The build comes after.
Send one real file
One dispute, as it already exists: the claim record, the clinical chart, the payer correspondence. In the demo case that is 342 pages across 2 productions, with a wrong-patient page quarantined before the read starts.
We return the read
Patient acuity, case complexity, level of training and the clinical detail behind each one, extracted and cited to the page it came from. You see the output before anything is scoped or signed.
We scope the build
If the read holds up, we scope the template against the factors at §149.510(c)(4)(iii)(C): your submission format, your fields, your batching conventions, your specialty's language.
You run it
Your team files. The platform organizes, cites and surfaces what the record says. The offer, the eligibility position and the argument stay with the people who sign them.
The rules the platform never breaks.
EVERY LINE CITED
HIPAA, under a signed BAA
Every file is handled under our Business Associate Agreement, from the first byte.
Never trains a model
Your records are never used to train any AI model — ours or anyone else's.
Every line cited
If we can't cite it, we don't say it. Every sentence links to its source page.
Deleted after delivery
Files are deleted 30 days after delivery, with a full audit log of every access.
Four capabilities behind every IDR submission.
IDR submission preparation is one deliverable of the same platform that reads, sorts and cites the whole record.
Claims Billing Ledger
Every billed line in the dispute traced back to the page that documents it, so the claim file and the chart tell the same story.
Medical Bill Review
Billed charges read against what the record documents, so the billed service and the documented service can be compared line by line.
Medical Chronology
The treatment timeline behind acuity and complexity builds itself from the chart, synced to every source page.
Custom Report Builder
Your submission template, filled in from the record — the same fields every time, in the format your reviewers expect.
No Surprises Act IDR support, answered.
Send one file. We'll tell you what we can read.
No obligation. If the case type is buildable we'll scope it; if it isn't, we'll say so.