openFDA Adverse Events Scraper (FAERS)

---

🎯 What this scrapes

The FDA's Adverse Event Reporting System (FAERS) is the world's largest public feed of drug-safety signals, but the raw record is a deeply nested JSON blob that buries seriousness flags and patient data three levels down. This Actor pages api.fda.gov/drug/event.json on your search, flattens every report into one row per reaction, each carrying the report's full drug list, and hands pharmacovigilance and competitive-intel teams a table they can pivot on directly. openFDA's own guidance applies here: this is unvalidated, self-reported data, and it should never drive medical care decisions.

🔥 What we handle for you

• flattens each FAERS report into one row per reaction — each row carries the report's full drug list — instead of a deeply nested blob
• retries transient 429/5xx responses with backoff so a rate-limit blip does not kill the run
• stops cleanly at your maxResults row cap instead of over-fetching openFDA's full result set
• validates every row through a typed Pydantic model before it reaches your dataset

💡 Use cases

• Track adverse-event volume for a specific drug ahead of a competitive or regulatory review.
• Feed a pharmacovigilance signal-detection pipeline with flattened seriousness and outcome fields.
• Benchmark reported reaction severity across a drug class over a date range.
• Backfill a research dataset without hand-flattening FAERS' nested JSON.

⚙️ How to use it

1. Click Try for free at the top of the page.
2. Fill in the input form — most fields have sensible defaults.
3. Click Start. Output streams into the run's dataset.
4. Export from Storage → Dataset as JSON, CSV, or Excel — or fetch via the API.

📥 Input

Field	Type	Required	Default	Notes
searchQuery	string	no	'patient.drug.medicinalproduct:"ozempic"'	Raw openFDA Lucene-style search expression, passed through to the search query param unmodified, e.g.…
maxResults	integer	no	20	Hard cap on rows emitted. Each row is one billed result-row event.
pageSize	integer	no	20	openFDA limit per request. openFDA's keyless tier caps this at 100.
proxyConfiguration	object	no	{'useApifyProxy': False}	openFDA is a public government API and does not need a proxy. Leave this off unless your account requires egress…

Example input

{
  "searchQuery": "patient.drug.medicinalproduct:\"ozempic\"",
  "maxResults": 3,
  "pageSize": 3,
  "proxyConfiguration": {
    "useApifyProxy": false
  }
}

📤 Output

Every row is one dataset item.

Field	Type	Notes
safety_report_id	string	FDA safety report identifier (safetyreportid).
receive_date	string	Date FAERS received the report, ISO YYYY-MM-DD when parseable, else the raw receivedate value.
reporter_country	string	Reporter's country code (primarysource.reportercountry, falling back to occurcountry).
serious	boolean	Whether the report was flagged serious (serious == "1").
seriousness_death	boolean	Whether the report flagged death as an outcome (seriousnessdeath == "1").
seriousness_hospitalization	boolean	Whether the report flagged hospitalization (seriousnesshospitalization == "1").
seriousness_life_threatening	boolean	Whether the report flagged a life-threatening event (seriousnesslifethreatening == "1").
seriousness_disabling	boolean	Whether the report flagged a disabling event (seriousnessdisabling == "1").
patient_age	number	Patient age at onset, unit not normalized (patient.patientonsetage).
patient_sex	string	Patient sex mapped to male / female / unknown (patient.patientsex).
drug_names	string	Every implicated drug name for this report, from patient.drug[].medicinalproduct, joined with ; .
reaction_term	string	This row's adverse reaction term (patient.reaction[].reactionmeddrapt); empty string if the report listed no…
reaction_outcome	string	Reaction outcome mapped to text (e.g. fatal, recovered/resolved), or null if unmapped or missing…

Example output

{
  "safety_report_id": "10188271",
  "receive_date": "2014-05-22",
  "reporter_country": "US",
  "serious": true,
  "seriousness_death": false,
  "seriousness_hospitalization": true,
  "seriousness_life_threatening": false,
  "seriousness_disabling": true,
  "patient_age": 81.0,
  "patient_sex": "female",
  "drug_names": "LANTUS SOLOSTAR",
  "reaction_term": "Stenosis",
  "reaction_outcome": "unknown"
}

💰 Pricing

Pay-Per-Event — you pay only when these events fire:

Event	USD	What it is
actor-start	$0.2	One-off warm-up charge per run
result-row	$0.004	Per unique dataset item

Example: 1 000 results at the rates above ≈ $4.20. No subscription, no minimum, no card to start — Apify gives every new account $5 of free credit.

🚧 Limitations

• openFDA's own disclaimer applies in full: this is unvalidated, self-reported adverse-event data and must never be used to make medical care decisions.
• patient_age passes through openFDA's raw numeric value without normalizing the reporting unit (years vs. months vs. weeks).
• openFDA's keyless tier rate-limits aggressively (public guidance: ~240 requests/minute); a very large maxResults on a broad query may take longer to complete.

❓ FAQ

Do I need an API key?

No. openFDA serves FAERS keylessly; the Actor stays inside the anonymous rate limit and backs off automatically on 429 responses.

Can I use this data to evaluate a drug's safety for a patient?

No. openFDA states its own data is unvalidated and self-reported, and this Actor carries the same disclaimer — do not use these rows to make medical care decisions.

Why does one report produce multiple rows?

A report with three reported reactions becomes three rows, each carrying the full drug list for that report — the natural one-row-per-fact join for a reaction-level dataset.

💬 Your feedback

Spotted a bug, hit a weird edge case, or need a new field? Open an issue on the Actor's Issues tab on Apify Console — we ship fixes weekly and we read every report.

---