# VICH GL9 (FDA Guidance #85) in plain words

Good clinical practice asks a field study for a signed protocol, owner consent, attributable raw data and a monitor who is not the investigator.

Canonical: https://signalment.app/guides/vich-gl9-in-plain-words/
Updated: 2026-10-14

Good clinical practice, as VICH GL9 sets it out, asks a veterinary field study for five things. A protocol signed and dated by the sponsor and the investigator before the study starts. Documented informed consent from each owner before their animal takes part. Raw data that are attributable, original, accurate, contemporaneous and legible, with every correction dated, signed and explained. A monitor, appointed by the sponsor, who is not the investigator. And study documentation kept for as long as the regulator requires. GL9 is guidance, not law, written for studies whose data will go to a regulator. In the United States the rule that gives it force is 21 CFR 514.117, which requires a study offered in support of approval to follow "a standard of conduct acceptable to FDA".

## The guidance and the rules around it

VICH is the International Cooperation on Harmonization of Technical Requirements for Registration of Veterinary Medicinal Products, which recommended GL9 in June 2000. FDA's Center for Veterinary Medicine (CVM) issued it as Guidance for Industry #85 in May 2001, and the European Union gave it effect in July 2001. Its introduction says it "should be followed when developing clinical study data that are intended to be submitted to regulatory authorities", and section 2.6 puts studies "not intended to be used for regulatory support" outside its scope.

Three rules sit around it:

- 21 CFR 514.117 describes an adequate and well-controlled study. Paragraph (b)(2) requires a standard of conduct covering "study conduct, study personnel, study facilities, and study documentation", and a protocol statement of intent to follow it. Paragraph (c)(2) says a field study must "use an appropriate control that permits comparison, employ procedures to minimize bias", and record observations "in sufficient detail to permit evaluation of the study".
- 21 CFR 511.1 is the investigational new animal drug (INAD) rule. An unapproved new animal drug may be shipped for a clinical investigation only under the exemption it describes. It names no ethics committee, no owner consent and no adverse event form.
- 21 CFR Part 11 governs records kept electronically under an FDA records requirement.

## The three roles

The duties that decide a field study:

- The investigator signs the protocol (3.2.2). They record any deviation "in a signed and dated statement" (3.2.5), obtain informed consent "from each owner, or owner's agent, before their animal(s) participate" (3.2.13), notify the sponsor promptly of adverse events (3.2.17), and account for the product from receipt to reconciliation and disposal (3.2.19 to 3.2.25).
- The sponsor submits any notice the regulator requires (4.2.2), appoints monitors (4.2.4), prepares and signs the protocol (4.2.6, 4.2.7), reports adverse events (4.2.10) and keeps the documentation (4.2.14). A contract research organization may take on duties, but "the ultimate responsibility for the quality and integrity of the study data always resides with the sponsor" (4.3.1).
- The monitor is appointed by the sponsor and visits "with sufficient frequency before, during and after the study" (5.2.6). They ensure consent "is obtained and recorded" (5.2.8) and that anything illegible, missing or corrected "is fully explained" (5.2.10). Section 3.1.5: "An individual should not serve as both the investigator and the monitor of any one study."

## The protocol

The protocol is "a document signed and dated by the investigator and the sponsor" (1.29), and amendments are signed the same way (1.30). From the section 6.3 list, the items a field study cannot skip:

1. A unique identifier: protocol number, status (draft, final, amended) and version date, on the title page (6.3.2).
2. The objective and design, with the randomization method, "the extent and methods of blinding" and who may access treatment codes (6.3.5, 6.3.8).
3. Animal selection, with inclusion, exclusion and removal criteria (6.3.9, 6.3.10).
4. The products, the dose and the dosing regimen (6.3.13, 6.3.14).
5. The effectiveness endpoints, how they are measured and when (6.3.16).
6. The statistical methods and a justified sample size (6.3.17).
7. Procedures for "recording, processing, handling, and retaining raw data" (6.3.18).
8. Adverse event procedures: observing, acting, recording, reporting (6.3.19).
9. Every data capture form, and the information given to owners (6.3.20).
10. How amendments are made and deviations reported (6.3.21).

## Raw data and the audit trail

Raw data are the first-hand observations; transcribed data are not raw data (1.24). Section 8.3.1 says raw data "whether handwritten or electronic, should be attributable, original, accurate, contemporaneous and legible", and that in automated systems "the individual(s) responsible for direct data input should record their name along with the date at the time of data input". A correction keeps the original legible and carries initials, a date and a reason (8.3.5). Section 8.3.6 then says: "if data are entered directly into a computer system, the electronic record is considered the raw data", each entry "including any change, should be made under the electronic signature of the individual making the entry", and changes "should be maintained in an audit trail". Consent, treatment assignment, case report forms and adverse events are all raw data (8.2.2.1).

Two later FDA guidances add to this. The October 2024 guidance on electronic systems in clinical investigations, co-issued by CVM, says audit trail components "must include" the date and time, the individual making the change, "and the old value and the new value" (Q12). Guidance #287 (2025) says "the audit trail should also be submitted if raw data were collected by an electronic data capture system". Part 11 adds that signed records show the signer's name, time and meaning (11.50), and that an organization certifies to FDA before it uses electronic signatures (11.100(c)).

Signalment is built to section 8.3.6. Electronic signatures re-check the signer's password and bind to the hash of what was signed, and the audit trail is append-only, hash-chained and exported so a monitor can verify it offline. It does not yet keep a drug accountability ledger; the receipt, inventory and reconciliation of 3.2.19 to 3.2.25 stay on the investigator's own record.

## Owner consent

GL9 defines informed consent as "a documented process by which an owner, or owner's agent, voluntarily confirms the owner's willingness to allow their animal(s) to participate" (1.15). CVM's draft Guidance #282 (September 2023) says what the form contains. It recommends a consent form for all studies with client-owned companion animals, a copy to the owner 12 to 24 hours before enrollment where possible, a statement that the product "is investigational, and safety and effectiveness have not been demonstrated", no wording that suggests the product is "safe" or "effective", a note that FDA may inspect the records, and written authority for any agent who signs for the owner. It is a draft, not a requirement. There is more in [Informed consent for client-owned animal studies](/guides/informed-consent-for-client-owned-animal-studies/).

## What changes with the design?

| Design | Filing and reporting | GL9 | Electronic records |
|---|---|---|---|
| Observational, nothing assigned | None | Out of scope unless the data go to a regulator | Part 11 only if kept under an FDA records requirement |
| Approved product used as labelled, data for FDA | No INAD; adverse drug experiences go to the applicant under 514.80, a nonapplicant forwarding them "within 3 working days" | Applies through 514.117 | Part 11 applies |
| Unapproved new animal drug | INAD notice under 511.1 before shipment; records kept two years; the sponsor monitors, reports hazards "promptly" to FDA and all investigators, and never calls the drug "safe or effective" | Applies through 514.117 | Part 11 applies |

## What stays the sponsor's and the investigator's job?

The protocol and its signatures. The INAD notice. The ethics review: GL9 names no committee, but requires both parties to "comply with applicable regulatory requirements governing the humane care of study animals" (3.2.12, 4.2.17), and whether a committee must review the study depends on the funder and the institution. The consent content. Adverse event reporting and the final study report. Signalment is a data system: it holds the records and shows who did what. The decisions above are made by people. The sponsor's view of it is at [signalment.app/sponsors](https://signalment.app/sponsors/).

## Sources

- FDA Guidance #85 (VICH GL9), Good Clinical Practice, May 2001: https://www.fda.gov/media/70333/download
- VICH, GL9 Good Clinical Practice: https://vichsec.org/guidelines/pharmaceuticals-pharma-efficacy-good-clinical-practice/
- European Medicines Agency, VICH GL9 Good clinical practices, Step 7: https://www.ema.europa.eu/en/vich-gl9-good-clinical-practices-scientific-guideline
- 21 CFR 514.117, adequate and well-controlled studies: https://www.ecfr.gov/current/title-21/section-514.117
- 21 CFR 511.1, investigational new animal drugs: https://www.ecfr.gov/current/title-21/section-511.1
- 21 CFR 514.80, adverse event reports for approved new animal drugs: https://www.ecfr.gov/current/title-21/section-514.80
- 21 CFR Part 11, electronic records and signatures: https://www.ecfr.gov/current/title-21/part-11
- FDA draft Guidance #282, Informed Consent Forms for Studies that Enroll Client-Owned Companion Animals, September 2023: https://www.fda.gov/media/172056/download
- FDA Guidance #287, Raw Data for Safety and Effectiveness Studies, June 2025: https://www.fda.gov/media/177901/download
- FDA, Electronic Systems, Electronic Records, and Electronic Signatures in Clinical Investigations: Questions and Answers, October 2024: https://www.fda.gov/media/166215/download