Editorial & Corrections Policy

How pages are produced

Every breed, drug, reaction, species, and ranking page on PlainBreed is generated from one documented public dataset: the FDA Center for Veterinary Medicine's Animal & Veterinary Adverse Event reports (openFDA). We load the dataset into a structured database and render every page from that database. The figures you see - report counts, death-coded shares, top reactions and drugs, outcome breakdowns, and yearly trends, are counted directly from the FDA's records, not hand-typed and not estimated by us.

This is a data-publishing model: one reviewed template renders every breed page so coverage is consistent across all 506 breeds. We are transparent that these pages are produced programmatically from the source data. The editorial work goes into the pipeline, how data is sourced and computed, into the methodology, and into the written guides, not into hand-authoring hundreds of near-identical pages.

Sourcing standards

• One primary source. Every figure comes from the FDA CVM adverse-event extract. We do not blend in third-party datasets, estimates, or proprietary aggregators.
• Attribution in context. Each data page names the FDA CVM source and links to the methodology that explains how the voluntary reporting program works and what the counts can and cannot show.
• Derived values are labeled. Figures we compute, death-coded share, report rankings, are presented as our analysis of the FDA data, distinct from the raw report counts.
• No invented data. Where a value is unavailable for a breed, the page says so rather than filling the gap with an estimate.

Update cadence

The FDA refreshes the animal adverse-event extract periodically. We rebuild our database when a new release is published and recompute counts, shares, and trends. Between releases the figures are stable because the source itself does not change. The data window shown on each page tells you which release it is based on.

Corrections process

If a figure on PlainBreed looks wrong, please tell us. Because our pages are generated from the FDA dataset, a genuine error almost always traces back to either the source data or our processing of it:

1. Report. Use the contact page with the page URL and the number that looks off.
2. Verify. We compare the figure against the FDA CVM data for that breed, drug, or reaction.
3. Fix at the source. If the value is wrong on our side, we correct it in the database and pipeline that generate the page, not just on one page, so every affected page is fixed at once. If the figure faithfully reflects the FDA data, we explain that and, where useful, add context.
4. Note it. Material corrections are reflected the next time the affected pages rebuild.

We aim to acknowledge data-error reports within a few business days.

Independence & affiliation

PlainBreed is an independent project. It is not affiliated with, endorsed by, or operated by the U.S. Food and Drug Administration. We present the FDA's public-domain data in a more searchable form and link back to the official sources throughout.