The Veterinary Research Project in 2026: Structure, Ethics and RCVS Expectations

A veterinary research project is judged on evidence-based reasoning and on ethical governance, not only on findings. The decisive early question is whether your work is a regulated procedure under the Animals (Scientific Procedures) Act 1986 — because that determines your entire approval route and timeline.

Where the research project sits in a UK veterinary degree

UK veterinary degrees — BVetMed, BVSc, BVMS and BVM BVS — run five or six years and are accredited by the Royal College of Veterinary Surgeons. Research appears in one of three places, and which one you are doing changes the scale of what is expected.

All three connect to the same underlying expectation. The RCVS Day One Competences require new graduates to be able to find, appraise and apply evidence to clinical decisions. Your project is where that capability is developed and demonstrated, which is why appraisal quality is weighted heavily even in small projects.

The question that determines everything: is it a regulated procedure?

Work on protected animals that may cause pain, suffering, distress or lasting harm is regulated under the Animals (Scientific Procedures) Act 1986 and requires Home Office licensing — an establishment licence, a project licence and a personal licence. The government guidance on research using animals sets out the framework.

Undergraduate projects are almost never licensed under the Act, for the straightforward reason that project licences take substantial time to obtain and are held by established researchers. If your supervisor holds a project licence you may be able to work under it as a personal licensee, but that is a planned arrangement, not something arranged mid-project.

What this means practically is that most student veterinary research is deliberately designed to sit outside the Act. Common non-regulated designs:

• Retrospective analysis of clinical records from practice management systems or a referral hospital database.
• Clinical audit against an agreed standard, which is service evaluation rather than research.
• Owner or practitioner questionnaires, where the participants are people rather than animals.
• Observational studies of animals undergoing procedures already clinically indicated.
• Systematic review or meta-analysis of published veterinary literature.
• Analysis of existing surveillance datasets, such as the practice-derived records curated by VetCompass.

Non-regulated does not mean unreviewed. Your school’s ethics committee — often an Animal Welfare and Ethical Review Body or a veterinary ethics review committee — still reviews the project, and human participants bring their own consent and data protection requirements.

The 3Rs apply to your design, not just to licensed work

Replacement, Reduction and Refinement are embedded in UK law for regulated work and are expected as a matter of good practice everywhere else. The National Centre for the Replacement, Refinement and Reduction of Animals in Research publishes guidance and design tools.

Address each explicitly in your methodology:

• Replacement — could this question be answered from existing records, cadaveric material, or published data instead of live animals?
• Reduction — is your sample size justified by a power calculation rather than convenience? Under-powered animal studies waste animals and answer nothing.
• Refinement — how is discomfort minimised, and what are your humane endpoints?

Reviewers read for these terms. A methodology that addresses them by name reads as competent; one that does not invites questions you could have pre-empted.

Client-owned animals bring a second consent layer

Much veterinary student research involves patients belonging to clients, and that creates an obligation with no equivalent in most laboratory research.

Informed owner consent must be specific to the research use, separate from consent to clinical treatment, and freely given — with an explicit statement that declining will not affect their animal’s care. That last point matters: the power asymmetry between a practice and a client is real, and reviewers will look for how you handled it.

For retrospective record analysis, check whether your practice or school’s client agreement already covers anonymised research use. Where it does not, you may need a formal data-sharing arrangement, and the records must be anonymised before analysis — including the practice identifier, since a rare presentation in a named practice can identify a client.

Report against the right guideline

Veterinary and biomedical research has established reporting standards, and using the right one raises the quality of your write-up immediately.

The ARRIVE guidelines are the standard for reporting animal research and are endorsed by most veterinary journals. Even where your project is not in vivo, working through the relevant checklist before you write catches omissions — unreported exclusions, unstated randomisation, missing sample size justification — that examiners routinely flag.

Structuring the report

Give ethics and governance its own visible section rather than a sentence in the methods. In veterinary work this is assessed material, and a project that describes its regulatory position clearly demonstrates professional understanding as well as research competence.

In the discussion, resist overstating. Retrospective clinical data carry well-known biases — incomplete records, referral populations that do not represent first-opinion practice, and outcomes recorded inconsistently. Naming those limitations precisely is the mark of a strong project, and the framing in our guide to the research limitations section applies directly.

For the statistical reporting conventions, our guidance on effect sizes and confidence intervals is relevant, since veterinary student projects share the small-sample problem. Where your project is a review rather than a primary study, see desk-based versus primary-data designs.

Frequently asked questions

Do I need a Home Office licence for my veterinary research project?

Only if your work involves a regulated procedure on a protected animal under the Animals (Scientific Procedures) Act 1986. Most undergraduate projects are designed to fall outside the Act — using clinical records, audit, questionnaires or observation of procedures already clinically indicated. Your school’s ethics committee still reviews the project either way.

Can I use client-owned animals in my research?

Yes, with specific informed owner consent that is separate from consent to treatment and makes clear that declining will not affect the animal’s care. Retrospective use of records may be covered by your practice’s existing client agreement, but confirm this and anonymise the data before analysis.

What are the 3Rs and do they apply to student projects?

Replacement, Reduction and Refinement. They are legal requirements for regulated work and expected as good practice for all animal research, including student projects. Address each explicitly in your methodology — ethics reviewers look for them by name.

What is a clinical audit and does it count as research?

A clinical audit measures practice against an agreed standard to improve service, rather than generating new generalisable knowledge, so it is usually classified as service evaluation rather than research. That distinction often means a lighter governance route, but confirm the classification with your school rather than assuming it.

Which reporting guideline should I use?

ARRIVE for in vivo animal research, PRISMA for systematic reviews, STROBE for observational studies, and the REFLECT statement for randomised trials in livestock or food safety. Work through the checklist before writing, not after — it catches the omissions examiners most often flag.

Can I do a veterinary research project without animals?

Yes, and many students do. Systematic reviews, analyses of existing surveillance datasets such as VetCompass, owner and practitioner questionnaires, and retrospective record studies all produce publishable veterinary research without any new animal involvement.