FDA device adverse event reports
Medical device adverse event reports (MAUDE): report number, event type (death, injury, malfunction), dates, product problems, devices (brand, generic name, manufacturer, model, catalog and lot number, product code, operator, class), patients (age, sex, weight, outcomes, treatments, problems), the report narratives and every other report field. Names of the manufacturers' contact persons are left out. Newest first by default. Filters accept a comma list (any of the values). Pass next_page back as page; null on the last page.
With count_by=<field> the response gives facet counts of the matching records by that field instead of the records: items of {term, count} (top count_limit). Text fields count whole values on their .exact index (e.g. count_by=openfda.manufacturer_name.exact); date fields give a time series grouped by count_interval (items of {period, count}). The other filters narrow what is counted.
Every response includes credits_used: the credits this call cost.
Parameters
| name | type | description |
|---|---|---|
brand | string | Device brand name. Example: dexcom |
generic | string | Device generic name. |
manufacturer | string | Device manufacturer. |
model | string | Model number. |
product_code | string | Three-letter product code. |
event_type | death | injury | malfunction | other | no_answer | Event type. Example: injury |
problem | string | Product problem (e.g. Battery Problem). |
report_number | string | Report number. |
q | string | Free-text words; every word must appear somewhere in the record. |
from | string | Earliest date (YYYY-MM-DD) of the route's main date field. |
to | string | Latest date (YYYY-MM-DD). |
search | string | An openFDA query expression added to the filters with AND, for fields the named filters do not cover (e.g. openfda.route:"TOPICAL"). |
sort | date_received | date_of_event | Sort field (default: relevance, or the newest first where noted). |
order | desc | asc | Sort order (default desc). |
limit | integer | Results per page. |
page | integer | Page number (1-based). Pass next_page back as page; null on the last page. |
count_by | string | Count the matching records by this field (an openFDA field name, e.g. classification.exact, openfda.brand_name.exact, or a date field). |
count_limit | integer | Number of top terms for count_by (1-1000; default 100). |
count_interval | year | month | day | Grouping of a date count_by: year (default), month or day. |
Every endpoint also takes fresh, raw, fields (see lean responses and caching).
Example
curl "https://api.api.sh/openfda/v1/devices/events?brand=dexcom" \
-H "x-api-key: $APISH_KEY" import os, requests
res = requests.get(
"https://api.api.sh/openfda/v1/devices/events",
params={"brand": "dexcom"},
headers={"x-api-key": os.environ["APISH_KEY"]},
timeout=120,
)
res.raise_for_status()
data = res.json() const url = new URL("https://api.api.sh/openfda/v1/devices/events");
url.searchParams.set("brand", "dexcom");
const res = await fetch(url, { headers: { "x-api-key": process.env.APISH_KEY } });
if (!res.ok) throw new Error(`${res.status} ${await res.text()}`);
const data = await res.json(); Response
A JSON object with the fields described above, plus raw (the source's own data), cached, fetched_at and credits_used. Errors return {"success": false, "error": "…"} with the HTTP status (see errors). A lookup that ran but found nothing (404) costs the price above; refused calls and errors on our side are free (what's charged).