Drug Label Change Records — NLM DailyMed (SPL)

Official source. No API key. Pay per record.

This actor turns the U.S. National Library of Medicine's DailyMed — the official distribution point for FDA Structured Product Labeling (SPL) — into clean, per-label records. Each run reads the keyless DailyMed v2 web service and delivers one row per drug label: the SPL set id, the product/drug name, the labeler organization, the SPL version, and the published date — the facts you track to know when a drug label changed and who changed it. Product and organization grain, public-domain metadata, no person fields.

What this is, in one paragraph

Every prescription and OTC drug marketed in the U.S. has a Structured Product Label, and NLM redistributes all of them through DailyMed. A label is versioned: each time the labeler updates it, the SPL version increments and a new published date is stamped. This actor reads DailyMed's SPL list newest-first, parses each into a single analysis-ready record — set id, product, labeler organization, version, date — and optionally attaches the NDC package codes. Values are verbatim. You are billed only for records actually delivered.

Who buys this and for what job

• Pharma regulatory and competitive-intelligence teams watching label versions across their own and competitors' products to catch label changes as they publish.
• Pharmacovigilance and safety teams tracking when labels are revised (a version bump is the signal).
• Health-tech and formulary products ingesting a structured, keyless drug-label index keyed by set id and NDC.
• Data engineers who want a one-call, no-key feed of the U.S. drug-label universe reduced to product/org fields.

Public domain — metadata only

DailyMed is distributed by NLM, and NLM's works are U.S. Government works. Per NLM: "Works produced by the U.S. government are not subject to copyright protection in the United States" and "may be freely used or reproduced without permission in the U.S." — with attribution: "Source: National Library of Medicine." One honest boundary travels on every record: an individual product label's content is submitted by the labeler and can include third-party material, so this actor carries label metadata only (set id, product, labeler, version, date, NDC) — never the label body text. These are metadata records for change-tracking, not medical or regulatory advice.

Product / organization grain — no person data

This actor works at product and organization grain: the drug product and its labeler company. SPL label metadata carries no natural-person fields, and this actor emits none — a person-key guard runs on every record as a backstop.

Pricing

Event	Free plan	Bronze	Silver	Gold / Platinum / Diamond
label_change_record	$0.05	$0.045	$0.04	$0.0335

Delivered before charged. Blocked runs and empty results cost nothing.

Input

Field	Type	Default	Description
drugName	string	—	Optional. Keep only SPLs whose drug name matches (e.g. metformin).
includeNdc	boolean	false	If enabled, fetch each SPL's NDC package codes (one extra request per record, slower).
maxRecords	integer	500	Ceiling on records delivered and billed.

The actor reads SPLs newest-first and follows the service's next_page_url pagination.

Output

One JSON object per label. Real example (values verbatim):

{
  "record_id": "0e043431-9d87-478d-87f5-5aa8b0a932eb",
  "setid": "0e043431-9d87-478d-87f5-5aa8b0a932eb",
  "product_name": "AUTOBRUSH MINT (SODIUM FLUORIDE) LIQUID",
  "labeler_organization": "LANDER ENTERPRISES, LLC",
  "spl_version": 1,
  "published_date": "Aug 19, 2026",
  "title": "AUTOBRUSH MINT (SODIUM FLUORIDE) LIQUID [LANDER ENTERPRISES, LLC]",
  "ndcs": [],
  "source": "U.S. NLM DailyMed (SPL v2 web service)",
  "source_url": "https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=0e043431-9d87-478d-87f5-5aa8b0a932eb",
  "licence": "U.S. NLM DailyMed ... may be freely used or reproduced ... label METADATA only ...",
  "attribution": "Source: National Library of Medicine (DailyMed).",
  "observed_at": "2026-08-19T00:00:00Z"
}

An unbilled RUN_RECEIPT records the robots check, the filters, SPLs seen and delivered, and whether charged equals delivered.

Field reference

Each record is a flat set of label-tracking fields. Identity: record_id / setid (the stable SPL set id that keys a product's label across versions). What and who: product_name and labeler_organization, both parsed from the SPL title (which is also carried verbatim as title). The change signal: spl_version (increments on each label revision) and published_date. Optional: ndcs, the NDC package codes, populated only when includeNdc is enabled. Provenance: source, source_url (a direct DailyMed label link), licence, attribution, disclaimer, and observed_at.

How a run works

A run first re-reads the source's robots posture from your runtime. DailyMed serves no robots file — /robots.txt redirects to the site homepage, which carries no directives — a no-policy posture; the run stops at the door only if that ever changes to a real disallow. It then reads the SPL list newest-first and flattens each label to one record, parsing the product and labeler out of the title. If you enabled NDC codes, it fetches each label's package codes with one extra paced request. Each record is delivered before it is charged, so a blocked or empty run costs you nothing; the actor follows the service's own next_page_url cursor, pauses briefly between pages, and stops at your maxRecords ceiling. The charge_equals_delivered flag in the receipt confirms you were billed for exactly what you received.

Cost guidance

One label_change_record per delivered label: cost = labels delivered × your plan's per-record price. DailyMed carries about 158,681 SPLs. Scope with drugName to a single molecule (typically a few hundred to a couple thousand labels) to keep runs bounded. 1,000 labels is $50 (Free) / $33.50 (Gold). Enabling includeNdc adds one request per record — leave it off for the fastest, cheapest runs.

Freshness

DailyMed publishes new and revised labels continuously, and this actor reads the list newest-first, so a small run returns the most recently published or updated labels across the entire drug universe — the freshest edge of the label stream. Because the SPL version and published date are carried on every record, you can diff two runs of the same drugName over time to detect exactly which products had a label version bump between them, which is the core change-tracking workflow this feed is built for. The db_published_date the service reports is also surfaced in the run receipt so you know how current the underlying database snapshot is.

Honest limitations

• Metadata only. The label body text is not included — this is the SPL index (set id, product, labeler, version, date) plus optional NDC codes.
• Product/labeler are parsed from the SPL title, which follows DailyMed's PRODUCT [LABELER] convention; the full title is always carried verbatim so you can re-parse if needed.
• A version bump is the change signal — this actor surfaces the current version and date, not a field-level diff of what changed inside the label.
• NDC codes are opt-in (one extra request per record).
• No person data is emitted.

Differentiation

Distinct from fda-recall-risk-records in this fleet: a recall is a product being pulled from the market, not a label revision. It is also distinct from fleet-1's fda-drug-approvals: an approval is a market-entry decision, not the ongoing label content/version that DailyMed distributes. Both are named so buyers pick the right one.

The NexGen Signal family

Part of the NexGen Signal family of official-source, pay-per-record products:

• FDA Device Clearance Records — 510(k)
• FDA Recall Risk Records — openFDA
• Federal Rulemaking Records — US Federal Register
• Grant Opportunity Records — US Grants.gov
• EU Regulatory Change Records — EUR-Lex

Source: U.S. National Library of Medicine (DailyMed) SPL v2 web service. NLM works are U.S. Government works, free to use with attribution; individual label content may include third-party material, so this actor carries label metadata only. Reformatted to product/org records; values verbatim; no person data emitted.