An adverse event is any unfavorable, unintended observation in a study animal after a product is used, whether or not anyone thinks the product caused it. The US definition of serious is in 21 CFR 514.3: fatal, life-threatening, requiring professional intervention, or causing an abortion, stillbirth, infertility, congenital anomaly, prolonged or permanent disability, or disfigurement. The investigator records every event and tells the sponsor promptly. What reaches FDA, and on which form, depends on the product. For an approved drug, the company that holds the approval files Form FDA 1932. For an investigational drug under an INAD, the sponsor reports significant hazards to FDA and to every investigator, on no form the rule names. For a compounded or other unapproved drug used outside an investigation, the route is Form FDA 1932a.
What counts as an adverse event
VICH GL9, which FDA publishes as Guidance #85, defines an adverse event (1.1) as "any observation in animals that is unfavorable and unintended and occurs after the use of a veterinary product or investigational veterinary product, whether or not considered to be product related." 21 CFR 514.3, the definitions behind FDA's reports on approved drugs, says the same and adds "whether or not the new animal drug was used in accordance with the approved labeling." Guidance #188, FDA's instructions for Form FDA 1932, counts a lack of expected effectiveness as an adverse event too.
So the first question is not whether the product did it. The event is recorded, then assessed. GL9 lists adverse events among a study's raw data (8.2.2.1) and asks the final study report to give every event and the measures taken, or to state that none was observed (7.3.9.4).
What counts as serious
In the US the test is 514.3, and Form FDA 1932 asks it as a yes or no against the same words (Guidance #188, B.3.6). It is wider than the definition VICH wrote for marketed products. VICH GL24 calls an event serious when it "results in death, is life-threatening, results in persistent or significant disability/incapacity, or a congenital anomaly or birth defect." That list does not name professional intervention, abortion, stillbirth, infertility or disfigurement. An event that requires professional intervention is serious under 514.3 even if the animal recovers fully. A protocol should say which definition it uses and who makes the call.
Three terms are easy to confuse:
- Serious is not severe. Form 1932 takes the severity of the response in the narrative and seriousness as a separate answer, so record both.
- Unexpected means "not listed in the current labeling for the new animal drug" (514.3). With an approved drug, serious and unexpected together make a 15-day report.
- Causality is an assessment of the link between product and event; Form 1932 carries the attending veterinarian's (B.5.1). It changes how an event is read, not whether it is recorded.
Who reports what, by study design
| Design | Investigator | Sponsor | Filing with FDA |
|---|---|---|---|
| Unapproved new animal drug under an INAD | Gives the sponsor "adequate and timely reports" (511.1(b)(7)(iii)); GL9 3.2.17 says to "promptly notify the sponsor" of adverse events | Must "promptly investigate and report to FDA and to all investigators" any findings that "may suggest significant hazards" (511.1(b)(8)(ii)) | 511.1 names no form and no deadline beyond "promptly" |
| Approved drug in the study, on label or off, as test or control | Records every event and tells the sponsor | Reports it to the company that holds the approval, as FDA's reporting page recommends, unless the sponsor is that company | The holder files each serious, unexpected event on Form FDA 1932 within 15 working days (514.80(b)(2)(i)); the rest go in periodic reports |
| Observational study or registry, nothing assigned | Records what the protocol asks | No filing because of the study | An event with a marketed drug can go to its maker; Form FDA 1932a takes reports on any animal drug |
When the sponsor holds the approval, 514.80 binds it directly. The 15-day clock runs from first receiving the information, "regardless of the source," and Guidance #188 files events from planned studies of an approved drug as "Safety Issue (Clinical Studies)," "regardless of how or why the VMP is used," with the study number in the narrative.
When the product is unapproved, 511.1 governs, and judging what is a significant hazard is the sponsor's job. A product compounded for the study sits here too: Guidance #256, FDA's policy on compounding from bulk drug substances, "does not apply to animal drugs compounded for use in investigations of new animal drugs." Outside an investigation, the same guidance asks the veterinarian or pharmacist who compounded a drug to report adverse events on Form FDA 1932a within 15 business days. FDA's reporting page says 1932a can be used "to report adverse events for any animal drug (FDA-approved or unapproved) or animal device."
When nothing is assigned, nothing is filed because of the study. GL9 covers studies meant for regulatory submission (2.6), so an observational registry follows its own protocol. The events still belong in the record, and one involving a marketed drug can still go to its maker or to FDA. For a pet food or treat, FDA takes complaints through its Safety Reporting Portal.
What the adverse event form should hold
A form that serves every route holds, from GL9 and Guidance #188:
- The study number, the animal's study code, species, breed, sex, reproductive status, age and weight.
- Each product the animal received: name or study code, lot, dose, route, interval, first and last dates, who gave it (clinic or owner), and whether it was used as labeled.
- Concomitant treatments and relevant history.
- The event as first described, the date of onset, the duration and the clinical signs. Reports to FDA code the signs as VeDDRA lowest level terms (B.3.2).
- Serious, yes or no, with the 514.3 criterion met; severity recorded on its own.
- What was done: treatment, whether the product was stopped, and whether the blind was broken, by whom and why. GL9 lets a code be broken only as the protocol allows and with the sponsor's knowledge and consent (3.2.18).
- Whether the event abated when the product stopped and returned when it restarted.
- The outcome: ongoing, recovered, recovered with sequela, died, euthanized or unknown, with necropsy findings if there were any.
- The causality assessment, with who made it and when.
- The dates that start clocks: when the site learned of the event, when the sponsor was told, and when anything was filed. For Form 1932 the clock starts when the holder first has an identifiable reporter, animal, product and event (A.4.2).
GL9 also asks the protocol to describe how animals are observed often enough to detect events, what is done in response, how events are recorded and how they reach the sponsor (6.3.19), and to attach a copy of every event record form (6.3.20.2).
What stays with the sponsor and the investigator
The protocol's adverse event procedures, the judgement on seriousness and causality, any code break, the reports to the sponsor, to the approval holder and to FDA, the follow-up, and the account of events in the final study report belong to the sponsor and the investigator. So do the ethics review and the consent text, including what owners are told about risks and how to report a change they see at home. No software decides that an event is serious or files it.
Where Signalment fits
Signalment is a data system. Its adverse event form carries the 514.3 serious criteria and an ABON causality assessment. Owners record what they see from their phones in the eDiary, with photos. Every entry and change is on an append-only, hash-chained audit trail, and a signature re-checks the signer's password and binds to what was signed. It does not file with FDA or with an approval holder. VeDDRA coding is not built yet, so the terms on a Form 1932 or 1932a are coded outside it for now. There is more on the sponsor page. When an unapproved product needs a notice to FDA before it ships, see When a pet study needs an INAD.
Sources
- 21 CFR 514.3, definitions: https://www.ecfr.gov/current/title-21/section-514.3
- 21 CFR 514.80, reports on approved new animal drugs: https://www.ecfr.gov/current/title-21/section-514.80
- 21 CFR 511.1, new animal drugs for investigational use: https://www.ecfr.gov/current/title-21/section-511.1
- FDA Guidance #85, VICH GL9, Good Clinical Practice: https://www.fda.gov/media/70333/download
- FDA Guidance #188, data elements for veterinary adverse event reports (January 2024): https://www.fda.gov/media/79105/download
- VICH GL24, management of adverse event reports: https://vichsec.org/wp-content/uploads/2024/10/GL24-st7.pdf
- FDA Guidance #256, compounding animal drugs from bulk drug substances (August 2022): https://www.fda.gov/media/132567/download
- FDA, How to Report Animal Drug and Device Side Effects and Product Problems: https://www.fda.gov/animal-veterinary/report-problem/how-report-animal-drug-and-device-side-effects-and-product-problems
- FDA, How to Report a Pet Food Complaint: https://www.fda.gov/animal-veterinary/report-problem/how-report-pet-food-complaint