openFDA Scraper: adverse events, recalls, labels, 510(k)

💊 What is openFDA Data?

Skip the query syntax: this Actor reads the US Food and Drug Administration's openFDA API for any brand, generic or manufacturer and returns one flat row per entry. Choose among half a dozen sources: FAERS adverse events, medicine recalls, package inserts, 510(k) clearances, device recalls and MAUDE events. Newest items come first.

• Pharmacovigilance and safety: adverse event reports for a drug with reactions, seriousness and patient age and sex
• Recall monitoring: new drug or device recalls for your products or suppliers since a date, on a schedule
• Regulatory and competitor research: 510(k) clearances by device type or company, with decision dates and product codes
• Medical writing and AI: label sections (indications, warnings, interactions) as text for analysis or an AI agent
• Supply chain and quality: recalling firm, reason, quantity, distribution and lot codes in one table

✨ Why use this Actor?

• ✅ Plain names work: no openFDA query syntax needed: a name is searched across generic, brand, product and firm fields
• ✅ Half a dozen sources, one Actor: medicine events, recalls, labels, 510(k) and equipment events
• ✅ Readable rows: FDA codes turned into words (seriousness, reporter, patient sex, drug role) and dates as YYYY-MM-DD
• ✅ Power users welcome: a query with a colon goes to openFDA as raw search syntax
• ✅ No credentials needed: works without an openFDA token; add your free one for heavy daily use
• ✅ Fair price: 1 USD per 1,000 records; names without a match are free

📦 What data can you get with openFDA Data?

🗂️ Dataset and your query	🆔 Record id (report, recall or K number)	📅 Report, decision or label date
🏷️ Title (product, reactions or device)	🚨 Recall class, status and reason	🏭 Recalling firm or applicant
📦 Product, quantity, lot codes	🗺️ Distribution and location	🤒 Reactions and serious outcomes
👤 Patient age and sex, reporter	💊 Drugs with role, dose, indication	📖 Label sections (warnings, dosage...)
✅ 510(k) decision and device class	🔧 Device problems and event text	🔗 FDA page URL

Each dataset has its own fields after dataset, query, id, date and title, which every row shares.

🚀 How to use openFDA Data

openFDA Data works in a few clicks, no coding needed:

1. Create a free Apify account (the free plan includes 5 USD of monthly credit).
2. Open openFDA Data and click Try for free.
3. Pick the dataset, enter your drug, device or firm names, and set a start date and how many records per name.
4. Click Start and wait for the run to finish (most runs take seconds).
5. Download your data as JSON, CSV, Excel, XML or HTML, or send it straight to Google Sheets, Make, Zapier or your AI agent.

⬇️ Input

Choose one dataset and give up to 100 drug, device or firm names. Everything else is optional.

Pick a dataset

drug-events (FAERS), drug-recalls, drug-labels, device-510k (premarket clearances), device-recalls and device-events (MAUDE reports). One dataset per run; start a second run for another.

Names or raw syntax

A plain name is searched in the dataset's naming fields: generic and brand for drugs, device and applicant for 510(k), product description and firm for recalls. A 510(k) number like K181956 works too. A query with a colon is passed on as raw openFDA syntax, e.g. classification:"Class I".

Dates and limits

since keeps records on or after a day (report, decision or publication stamp). Results come newest first; maxResultsPerQuery caps each name (up to 25,000). maxTextLength cuts long label sections and event narratives.

Field	Type	What it does	Default
Dataset (dataset)	string	Which FDA records to fetch.	"drug-recalls"
Drug or device names (queries)	array	Generic or brand drug names, device names, firms or 510(k) numbers (e.g.	["metformin"]
Since (since)	string	Keep records on or after this date (report, decision or label date), e.g.
Maximum records per query (maxResultsPerQuery)	integer	Cap per query, newest first.	100
Maximum text length (maxTextLength)	integer	Long label sections and event narratives are cut to this many characters.	2000
openFDA API key (optional) (apiKey)	string	Free key from open.fda.gov/apis/authentication: 120,000 requests a day instead of 1,000 per IP.
Maximum retries (maxRetries)	integer	Retries per request with exponential backoff on network errors, 429 and 5xx.	3

Input example

{
    "dataset": "drug-recalls",
    "queries": [
        "metformin",
        "valsartan"
    ],
    "since": "2018-01-01",
    "maxResultsPerQuery": 10
}

⬆️ Output

Every record becomes one flat row: dataset, query, id, date and title first, then the fields of that dataset. An entry matching nothing is no error: it adds no lines. The run summary in the key-value store (OUTPUT) lists every name with the count saved, the total at the FDA and the failure if one occurred.

Output example

One row of the dataset (drug recalls for metformin since 2018):

{
    "dataset": "drug-recalls",
    "query": "metformin",
    "id": "D-0657-2026",
    "date": "2026-07-08",
    "title": "Synjardy XR Tablets (empagliflozin and metformin hydrochloride extended-release tablets), 12.5/1000 mg, 60 Tablets per Bottle, Rx only, Distributed by: Boehringer Ingelheim Pharmaceuticals, Inc., Ridg...",
    "classification": "Class II",
    "status": "Ongoing",
    "recallingFirm": "Boehringer Ingelheim Pharmaceuticals, Inc.",
    "location": "Ridgefield, CT, United States",
    "reason": "CGMP Deviations: An incorrect detergent was used and the required swab sampling was not performed on a production hopper that had previously been used for a different product.",
    "product": "Synjardy XR Tablets (empagliflozin and metformin hydrochloride extended-release tablets), 12.5/1000 mg, 60 Tablets per Bottle, Rx only, Distributed by: Boehringer Ingelheim Pharmaceuticals, Inc., Ridgefield, CT 06877 USA.  NDC: 00597-0300-45",
    "quantity": "560 bottles",
    "distribution": "OH",
    "initiatedBy": "Voluntary: Firm initiated",
    "initiationDate": "2026-06-12",
    "classificationDate": "2026-06-30",
    "terminationDate": null,
    "codeInfo": "Lot 3244413; Exp. 12/31/2028",
    "eventId": "99233",
    "brandNames": [],
    "genericNames": [],
    "url": "https://www.accessdata.fda.gov/scripts/ires/index.cfm?Event=99233"
}

Drug adverse event row (drug-events)

One FAERS report for Ozempic from a real run:

{
    "dataset": "drug-events",
    "query": "Ozempic",
    "id": "26940200",
    "date": "2026-06-30",
    "title": "Monoplegia, Back pain, Product use in unapproved indication",
    "serious": true,
    "seriousOutcomes": [
        "hospitalization",
        "other serious"
    ],
    "country": "PK",
    "reporter": "consumer",
    "patientSex": "female",
    "patientAge": "45 years",
    "reactions": [
        "Monoplegia",
        "Back pain",
        "Product use in unapproved indication"
    ],
    "drugs": [
        {
            "name": "OZEMPIC",
            "activeSubstance": "SEMAGLUTIDE",
            "role": "suspect",
            "indication": "Obesity",
            "dose": null
        }
    ],
    "company": "PK-NOVOPROD-3616663",
    "url": null
}

510(k) clearance row (device-510k)

One clearance for a pulse oximeter from a real run:

{
    "dataset": "device-510k",
    "query": "pulse oximeter",
    "id": "K252448",
    "date": "2026-02-27",
    "title": "AViTA Pulse Oximeter (SP61)",
    "applicant": "Avita Corporation",
    "decision": "Substantially Equivalent",
    "clearanceType": "Traditional",
    "dateReceived": "2025-08-04",
    "productCode": "DQA",
    "deviceClass": "2",
    "regulationNumber": "870.2700",
    "specialty": "Cardiovascular",
    "location": "New Taipei City, CN",
    "url": "https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn.cfm?ID=K252448"
}

View the results as a table or JSON in the Output tab and download them as JSON, CSV, Excel, XML, HTML or RSS, all fields or only those you pick.

💰 How much does openFDA Data cost?

openFDA Data is priced per event: 1 USD per 1,000 records (0.001 USD each), with no monthly rental. Each record saved to the dataset is billed once; a name without a match is free.

You get	You pay
100 records	0.1 USD
1,000 records	1 USD
10,000 records	10 USD
100,000 records	100 USD

💡 Apify's free plan gives you 5 USD of credit every month, enough for about 5,000 records at no cost. Set a spending limit per run and the Actor stops there.

⚖️ openFDA Data vs. the alternatives

	openFDA Data	openFDA API directly	FDA website search
Plain drug or device names	✅	query syntax per dataset	✅
Six datasets with one input	✅	✅ six endpoints	separate tools
Readable fields instead of FDA codes	✅	❌ raw codes	partly
Paging and retries handled	✅	your code	manual
Export to CSV, Excel, JSON	✅	your code	limited
Schedules and MCP for AI agents	✅	❌	❌

🔌 Integrations, API and MCP

Connect openFDA Data to your AI agent (MCP)

Give Claude, ChatGPT, Cursor or any MCP client direct access to openFDA Data through the Apify MCP server. Add this URL to its settings:

https://mcp.apify.com/?actors=bepolabs/openfda

For Claude Desktop, add it under mcpServers in claude_desktop_config.json:

{
    "mcpServers": {
        "openfda": {
            "command": "npx",
            "args": [
                "-y",
                "@apify/actors-mcp-server",
                "--actors",
                "bepolabs/openfda"
            ],
            "env": {
                "APIFY_TOKEN": "<your Apify API token>"
            }
        }
    }
}

Then ask: "Show me all Class I drug recalls from the last 12 months and the reason for each."

Run it from your code

Python (pip install apify-client):

from apify_client import ApifyClient

client = ApifyClient("<YOUR_APIFY_TOKEN>")
run = client.actor("bepolabs/openfda").call(run_input={"dataset":"drug-recalls","queries":["metformin","valsartan"],"since":"2018-01-01","maxResultsPerQuery":10})
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item)

JavaScript (npm install apify-client):

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: '<YOUR_APIFY_TOKEN>' });
const run = await client.actor('bepolabs/openfda').call({"dataset":"drug-recalls","queries":["metformin","valsartan"],"since":"2018-01-01","maxResultsPerQuery":10});
const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items);

No-code integrations

Send results to Google Sheets, Slack, Airtable, HubSpot, Make, Zapier, n8n and more from the Integrations tab, start runs on a schedule, or trigger your own endpoint with a webhook when a run finishes. The Apify API gives you full control over runs and datasets.

❓ FAQ

Where does the data come from?

From the official openFDA API of the US Food and Drug Administration. Every run asks it live.

How fresh is the data?

As fresh as openFDA: recalls are updated about weekly, adverse event reports about quarterly, labels and 510(k) clearances as the FDA publishes them.

Do I need an openFDA API key?

No. Without one openFDA allows 1,000 requests daily per IP; for heavy use add your free key from open.fda.gov/apis/authentication in apiKey (120,000 allowed).

Does an adverse event report prove the drug caused it?

No. FAERS and MAUDE entries are unverified notes sent to the FDA; they do not prove that a drug or device caused the event. Use them as signals, not as conclusions.

How many records can I get?

Up to 25,000 per name (the paging cap of openFDA) and up to 100 names per run. Narrow with since to stay below it.

Is it legal to use openFDA Data?

Yes. openFDA data is public domain (CC0) and the Actor reads it through the FDA's official API. The FDA asks users not to imply FDA endorsement and to remember that adverse event reports alone do not establish causation. The data is not medical advice.

What if something is missing or does not work?

Open an issue on the Issues tab; we read each one and reply. Feature requests are welcome too.

💬 Your feedback

Tell us what openFDA Data should add on the Issues tab or leave a review; your notes decide what we work on next.

🔗 Related Actors

Actor	What it does
🏛️ SEC EDGAR Filings	Tickers or CIK numbers to SEC filings: 10-K, 10-Q, 8-K, Form 4 and any other form, with document links.
📄 PDF to Text and Markdown	PDFs to clean text and Markdown, per page, per document or as RAG chunks, with columns in reading order and headers and footers removed.
🌦️ US Weather Forecast	Official National Weather Service forecasts, hourly and 7-day, plus active weather alerts for any US point or state.
📈 Wikipedia Pageviews	Daily or monthly views of any Wikipedia article in any language, with totals and averages, from the official Wikimedia API.
⭐ App Store Reviews	iOS App Store reviews with stars, text, app version and date, plus the app's rating and details, for any store country.