Hospital Price Transparency MRF Normalizer

Every US hospital is federally required to publish a machine-readable file (MRF) of its standard charges. In practice those files arrive as tall CSV, wide CSV, or JSON, hundreds of columns wide, formatted differently by every hospital and EHR vendor. This Actor turns any of them into one clean, flat schema — so prices from different hospitals become directly comparable.

What you get

One row per (service, charge type, payer):

{
  "hospital_name": "Example General Hospital",
  "mrf_last_updated": "2026-07-01",
  "mrf_version": "2.2.0",
  "mrf_format": "wide-csv",
  "description": "Office visit established patient level 3",
  "code": "99213",
  "code_type": "CPT",
  "setting": "outpatient",
  "charge_type": "negotiated_dollar",
  "amount": 142.10,
  "payer_name": "Aetna",
  "plan_name": "PPO Broad",
  "methodology": "fee schedule"
}

All CMS template versions supported — including the new CY2026 v3.0 template (verified against CMS's official example files). Charge types: gross, discounted cash, payer-negotiated dollar / percentage / algorithm, de-identified min/max, estimated allowed amount (v2.x), and the v3.0 payer statistics (median, 10th/90th percentile).

Filters (pay only for what you need)

• codes — only these CPT/HCPCS/MS-DRG/NDC codes (secondary codes matched too)
• payerContains — only payers matching this text
• chargeTypes — only these charge types
• maxRowsPerFile — hard cap per file (default 100k; hospital files can hold millions of rows)

Typical uses

• Benefits consultants — compare negotiated rates for a service across the hospitals in a market
• Revenue-cycle teams — see competitors' negotiated rates by payer
• Health-tech builders — a normalization step you don't have to write or maintain
• Researchers & journalists — comparable price data without an ETL project

Where to find MRF URLs

CMS requires every hospital to link its MRF from its website footer ("Price Transparency"). Paste those URLs straight in. CSV files are streamed, so large files are fine; JSON is capped at 400 MB per file.

Notes

• April 1, 2026: CMS began enforcing the CY2026 (v3.0) MRF template — hospitals are actively reformatting, which is exactly why a normalizer beats hand-written parsers. This Actor reads v3.0 files (including the spaced-pipe tall CSV headers many parsers choke on) as well as legacy v2.x files.
• This data is public by federal mandate (45 CFR 180.50). No logins, no ToS friction.
• Output is provided as-is and is not pricing, billing, or legal advice. Hospitals occasionally publish erroneous data; mrf_last_updated tells you what vintage you're looking at.

Pricing

Event	Price
Run start	$0.05
Per file normalized	$1.00
Per 1,000 output rows	$0.02

Normalizing one hospital's file to a filtered set of 5,000 comparable rows: ~$1.15 — versus hours of custom parsing per hospital format, or an enterprise platform subscription.

FAQ

Which file formats does it support? All CMS machine-readable templates: v2.0, v2.2, and the CY2026 v3.0 — in tall CSV, wide CSV, and JSON. Verified against CMS's official example files, including the v3.0 tall format's spaced-pipe headers that break naive parsers.

Where do I find a hospital's MRF URL? Every US hospital is required to link its machine-readable file from its website footer (look for "Price Transparency"). Copy that link directly into mrfUrls.

How big a file can it handle? CSVs are streamed, so hundreds of MB is fine. JSON is capped at 400 MB. Use maxRowsPerFile and the code/payer filters to control output size and cost.

Why do some rows have an algorithm instead of a dollar amount? Hospitals may express negotiated charges as a percentage or contractual algorithm when no dollar figure exists. Those come through as negotiated_percentage / negotiated_algorithm rows with the text preserved — that's the data as filed, not a parsing gap.

Is this data legal to use? Yes — these files are public by federal mandate (45 CFR 180.50). No logins, no terms-of-service issues.

Can I compare hospitals? That's the point: every file normalizes to the same flat schema, so rows from different hospitals join directly on code + payer.

Changelog

• 2026-08-22 — CMS v3.0 (CY2026) template support: spaced-pipe tall headers, location_name, payer statistics (median, 10th/90th percentile). Verified against official CMS example files.
• 2026-08-20 — Launch: v2.0/v2.2 tall CSV, wide CSV, and JSON with code/payer/charge-type filters.

---

More tools by Businessquik

Verified data tools built on official and public-mandate sources — no logins, no gray areas, no fragile scrapers:

• Bank Statement Extractor — PDF statements to reconciled CSV/JSON
• FDA Recall Monitor — new-recalls-only alerts for drugs, devices & food
• Patent Signals — filing velocity, new entrants & expiry intel
• US Location Profile — economy, labor & weather in one call

Support: open an issue on this Actor's Issues tab — layout/format reports are usually fixed within days. Describe the shape of your file or query; never post real bank statements or sensitive data.