Methodology

How Branmoor sources, verifies, and presents data — and the limits of what we publish.

Primary sources only

Every figure Branmoor publishes is drawn directly from a primary source — a federal or state regulator, a court or agency filing, or an official disclosure. We do not buy aggregated third-party datasets and re-sell them, and we do not generate data with language models.

Where a value cannot be tied to a primary source, it is left blank, not estimated. Branmoor records and reports what the source says; it does not model, infer, or fill gaps with judgment. There is no subjective interpretation anywhere in the data — a value is either in the source or it is not.

Primary-source citations

Each data point carries its origin: the issuing source, the specific dataset or filing, and an as-of date. Diligence files and product pages label or link every figure back to that source so a reader can verify it independently. Branmoor's core healthcare sources include:

• FDA — openFDA drug shortages, labels, recalls, and FAERS adverse events
• CMS — HCRIS cost reports, Provider Data Catalog, Hospital Compare, Provider of Services, area wage index, and enrollment files
• HHS-OIG LEIE exclusions, NPPES, HRSA OPAIS and OPTN, and ClinicalTrials.gov
• MSRB EMMA municipal disclosures and IRS BMF / Form 990 filings
• OSHA ITA, the Federal Register, and state regulators and county assessors

Accuracy and corrections

Branmoor mirrors the primary source exactly, including its limitations: if the source is wrong, Branmoor will be wrong too, and we say so rather than silently “cleaning” the record. Normalization is limited to deduplication, formatting, and entity resolution — it never changes a reported value. Every page carries a dated snapshot. Spotted discrepancies go to corrections@branmoor.com.

Confidence and completeness

Diligence claims carry a confidence flag — H for an audited or filed primary source, M where there is a timing lag or partial coverage, L for a single-source or preliminary value — and a completeness note where a source is partial. A missing value is never shown as a zero. See it applied in the sample Provider Diligence Sheet (PDF).

Compliance and boundaries

Public pages expose small previews and methodology; complete products, exports, alerts, APIs, data feeds, and buyer-specific diligence are licensed. Branmoor diligence is outside-in, non-attest data work — not an audit, review, compilation, valuation opinion, tax opinion, or CPA-firm service. Any future licensed-CPA service would carry separate engagement terms, conflict checks, confidentiality controls, and limitation language.

Per-product methodology

The principles above apply to everything Branmoor publishes. Each product also documents its own source, cadence, and normalization rules — for example:

RXWatch: Drug Shortage Tracker

Primary source

All shortage records on Branmoor's Drug Shortage Tracker come from the U.S. Food and Drug Administration's Drug Shortages database, accessed via the openFDA Drug Shortages API (api.fda.gov/drug/shortages.json).

Update cadence

We re-pull the full shortage feed every 4 to 6 hours. The "Last refreshed" timestamp on each tracker page reflects the most recent successful pull. FDA's own posting cadence is irregular: most records are updated weekly, some daily during active shortages, others only when a manufacturer files new information.

Normalization rules

• Raw FDA records are deduplicated to one page per (substance, dosage form, brand) combination, keeping the record with the most recent update date.
• Substance and brand names are normalized to title case for display; original FDA strings are preserved in the page metadata.
• FDA's status field is used as the canonical state. Statuses include "Current" (active shortage), "To Be Discontinued" (manufacturer-announced discontinuation), and "Resolved" (FDA marked as resolved).
• Where FDA reports a shortage reason or related narrative, we reproduce it verbatim. We do not paraphrase or editorialize manufacturer-reported reasons.

What we do not do

• We do not generate clinical content with language models. Drug pages are template-driven directly from FDA data.
• We do not suggest therapeutic alternatives, dosing, or substitutions. Those decisions belong to a clinician with knowledge of an individual patient.
• We do not validate FDA records against secondary sources. If FDA's database is wrong, Branmoor will be wrong too.

Medical disclaimer

Content on Branmoor is for informational purposes only and is not a substitute for professional medical advice, diagnosis, or treatment. Always seek the advice of a licensed clinician or pharmacist with any questions about a medical condition or medication. Never disregard professional medical advice or delay seeking it because of something you read on this site.

Corrections

If you spot a normalization error or a stale record on Branmoor that differs from the current FDA database, please email corrections@branmoor.com with the page URL and the discrepancy.