The allocation is only as defensible as the record behind it.
Medicare set-aside record review reads the full claim file, 300 to 3,000 pages, and returns every injury-related treatment, drug, dosage and frequency, cited to the source page, before the allocation is written. CMS-approved WCMSA and LMSA amounts ran 24% above proposal in FY2025, the highest variance in five years, and the record behind the number is where that gap closes.
The rulebook is published. So is the referee.
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. In Medicare set-aside allocation, both are already federal policy.
Three steps between the claim file and the allocation.
Send the claim file
The full claim file as you already have it, in any format. In the demo case, 342 pages across 2 packets, with 11 duplicate pages removed before the read starts.
We return every injury-related fact, cited
Treatment, drug, dosage and frequency for every visit tied to the accepted injury, each line linked to its source page. Non-injury treatment stays out of the extract, flagged as excluded rather than silently dropped.
You write the allocation
The allocator builds the WCMSA or LMSA proposal against CMS's Reference Guide methodology, using the extract as the treatment record. The AI never sets the dollar amount and never predicts CMS's approval.
A record review that knows it isn't the allocation.
We extract the injury-related treatment, drug, dosage and frequency, and cite every page. We do not set the allocation amount, do not predict what CMS's review contractor will approve, and do not give a legal or medical opinion. The WCMSA or LMSA proposal, and the number in it, belong to the allocator.
In the demo case, page 140 turned out to belong to a different patient. The claim-file review flags it and quarantines it in its own line, instead of folding it silently into the record CMS will review.
The rules the platform never breaks.
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 claim-file review.
The MSA claim-file review is one deliverable of the same platform that reads, sorts, and cites the whole record.
Medical Records OCR
Every claim file read in full — provider records, pharmacy logs, imaging reports — nothing skipped, nothing guessed.
Medical Chronology
The treatment timeline builds itself from the accepted injury forward, synced to every source page.
Verifiable AI Citations
If we can't cite it, we don't say it. Every treatment, drug, and frequency links to the page it came from.
Medical Summary Reports
The extracted treatment record formats straight into the allocator's WCMSA or LMSA worksheet — your template.
Medicare set-aside record review, 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.