Datadory notebook
Fda Drug Adverse Event Data: Dataset Structure and Field Coverage
Datadory delivers fda drug adverse event data covering comprehensive field definitions, entity mappings, and historical time series — structured for direct analytics and delivered on demand.
1,744 datasets. Pick your catch.
What fields does each FAERS report carry?
Reports follow the ICH E2b/M2 v2.1 standard, and each carries an 8-digit safetyreportid — seven digits identifying the report plus a checksum (sample record: 5801206-7). Dates arrive as YYYYMMDD strings (receivedate, receiptdate, transmissiondate), and seriousness flags mark death, hospitalization and life-threatening outcomes alongside disability and congenital anomaly. The patient block holds patientonsetage with unit codes (801 = years, 800 = decades), patientsex (1 = male, 2 = female) and a reaction array of MedDRA preferred terms such as OVERDOSE. The drug array carries medicinalproduct exactly as the reporter typed it — one verified example reads DURAGESIC-100 — plus drugcharacterization (1 = suspect, 2 = concomitant, 3 = interacting), drugindication, drugadministrationroute (041 = transdermal) and the manufacturer's own companynumb. primarysource.reportercountry records where the report originated, since submissions are global rather than US-only. The structural limitation outweighs any single field: when multiple drugs and reactions appear in one report, no causal link between a specific drug and a specific reaction can be drawn, and no personally identifiable information is included.
Which other FDA endpoints complement the adverse event file?
Adverse event reports earn their keep next to three neighbouring openFDA exports, all under the same commercial delivery terms terms:
- openFDA Drug Label API (Structured Product Labeling): 261,996 prescription and OTC structured product labeling records in a 14-partition JSON export (~1.77 GB, roughly 130-145 MB per partition), refreshed weekly, with the bulk of data from June 2009 onward. The
_tablefield variants enable full-text search across label sections — the quickest way to set a warning label beside reported reactions. - openFDA Drugs@FDA API: 29,273 applications in a single ~8.9 MB ZIP, refreshed daily Monday through Friday, documenting approval history since 1939 (documents mostly since 1998).
- NDC directory and Orange Book records: the pharmaceuticals slice credits openFDA Drug APIs with 137k NDC listings and 48.7k Orange Book records alongside its 262k label documents and 29.3k Drugs@FDA records.
Europe is the natural comparator, and EMA's contribution serves a different job: the EMA Medicines Catalogue lists 2,730 centrally authorised medicines (~900 KB XLSX, 6.7 MB JSON refreshed twice daily) covering marketing status, ATC codes, active substances and regulatory dates. Its companion tables — referrals, PIPs, orphan designations, PSUSAs, DHPCs, shortages — describe regulatory state, not case-level reaction reports.
Pick up where this leaves off
Every one of these ships with sample rows before you commit to anything.
openFDA Drug APIs
openFDA FAERS Adverse Event API
openFDA Drug Label API (Structured Product Labeling)
openFDA Drugs@FDA API
EMA Medicines Catalogue
name_of_medicine · ema_product_number
RxNorm / RxNav Browser and APIs
rxcui · name · synonym …+6 more
Want rows instead of a pitch? Name the datasets.
API, files, or your warehouse. Daily, weekly, or hourly.
Get a sampleQuestions worth asking
How current is the downloadable FAERS file?
Quarterly, with a lag that can run three months or more. The live index held 20,692,690 reports at last_updated 2026-07-30, and the bulk export was dated 2026-08-17. Records are append-only — new reports arrive rather than edits to existing cases — so refreshing a quarter adds cases without rewriting history.