NewMissing Records Detection: flags every visit, provider, and date missing from the file. See how →
HomeSolutionsMedical Record Review Vendors
WHITE-LABEL MEDICAL RECORD REVIEW SERVICES

Keep the client, the brand and the sign-off. Change what runs underneath.

White-label medical record review services let a chronology or summarization vendor keep its own brand, client relationship and sign-off while the reading, sorting, deduplication and page citation run on our platform. The rate is published: ten cents a deduplicated page, down to five cents at volume, or a flat annual license with no metering.

10¢ per deduplicated page·5¢ at volume·Your brand on every export

The rate is published, so you can price your own work against it.

Most of this market quotes. A vendor cannot build a margin on a quote it has to renegotiate every quarter, so all three routes below carry a number you can put in a spreadsheet before you talk to anyone.

Published rateMetering
Per page, self-service
Ten cents a deduplicated page, scaling to five cents at volume. Exact duplicate pages are detected and never charged.
No minimum
Flat annual, your cloud
Fifteen thousand a year plus a five thousand setup fee, uncapped pages, with model inference billed on your own AWS or Azure account.
No metering
Flat annual, your infrastructure
Forty-five thousand a year plus a five thousand deployment fee, the whole stack inside your own network, air-gap available.
No metering

Every figure above is the published rate, not an indication. The full detail and an ROI calculator sit on the plans page. For what the rest of the market publishes, the pricing study read 318 vendors and found 53 that name a number at all.

Three surfaces, no re-keying between them.

A vendor does not want a second workspace for its staff to log into. It wants the reading to happen inside the pipeline it already runs, and the output to come out wearing its own name.

  • Cases in, events outCreate a case over REST, subscribe to the events you care about, and your systems stay current without polling. Medical Data API →
  • Structured facts, not documentsQuery the evidence graph by case, entity type or date range. Extracted fact and inferred relationship are never returned as the same thing. Evidence Graph API →
  • Your letterhead on the way outLogo, letterhead and signature block set once, applied to every export, with the citations and the content untouched. Branding and export controls →

The line between the engine and your name.

White-label does not mean the software takes over the part your client is actually paying for. It reads and cites. Your reviewer decides, and your reviewer signs.

The platform

Does the reading

  • Reads every page, including the handwriting and the scans
  • Sorts the whole file into one chronology by date
  • Detects exact duplicate pages and does not bill them
  • Cites every output line back to its source page
  • Flags a page that does not belong to the patient
Your firm

Does the judging

  • Reviews what came back and decides what it means
  • Signs, certifies or attests, under your own credentials
  • Holds the client relationship and the engagement letter
  • Sets the deliverable format your client already expects
  • Carries the professional liability, as you do today

The same boundary holds on every page of this site: the platform drafts and cites, and a qualified person at your firm reviews, revises and signs. See how the record is handled and the HIPAA posture.

Who buys the engine rather than the deliverable.

Four shapes of business, one thing in common: the client relationship is already theirs and the reading is the cost centre.

Chronology and summarization vendors

You already sell the deliverable and already have the clients. What changes is the cost and the clock behind it, not the invoice your client sees.

Legal support services →

Legal nurse consulting practices

The nurse still reads what matters and still signs. The sorting, deduplication and citation stop being the part of the file that eats the week.

Legal nurse consultants →

IME and peer review companies

Panels get the record already assembled and cited, so the physician spends the hour on the opinion rather than on the paper.

IME companies →

Record retrieval companies adding review

Retrieval is a different business from review, and we do not compete with it. You keep retrieving; the review layer sits on top of what you already deliver.

TPAs and administrators →

Building for a specific case type instead? The case-type index lists every file we have built a named rulebook and a named referee for.

White-label medical record review, answered.

Yes. Your logo, letterhead and signature or attestation block are set once on your account and applied to every export, so the deliverable leaves in your name. Medrecords AI does not appear on the report and does not contact your client.

You do, or your reviewer does. The platform reads, sorts, deduplicates and cites every line back to its source page. It does not sign, certify, rate a case or state an opinion. The professional judgment and the signature stay inside your firm, which is what your client is buying.

Ten cents a deduplicated page to start, scaling to five cents with volume, and exact duplicates are never billed. Above roughly 15,600 pages a month the flat annual license usually costs less than metering, which is the crossover we publish on the plan page rather than quote case by case.

No. Medrecords AI does not retrieve records from providers or facilities. You bring the records you already have — review starts in minutes from upload. Retrieval vendors take days; you can keep yours and still cut the review to minutes.

Through the API rather than the workspace. Cases are created over REST, events push to your endpoint instead of you polling, and every extracted fact carries its page anchor and source document, so citations survive the API boundary into your own system.

No. Customer records never train a model, and that is contractual under the BAA rather than an account setting. On the flat annual plans the records and the model calls stay inside your own cloud tenancy or your own network.

Run one of your own client files through it.

Send a real file, in whatever shape it arrives. Compare what comes back against what that file costs you to produce today.

Test a file