The British Union for the Abolition of Vivisection (BUAV), R (on the application of) v Secretary of State for the Home Department

The severity limit for an experimental protocol must be assessed by an objective standard. A “substantial” limit is required where the protocol may cause a serious, significant or important departure from an animal’s health or well-being. The assessment must consider the worst potential outcome for any animal, including outcomes controlled only by killing, and must take account of actual experience under an existing licence. Expert judgment receives proper respect, but cannot stand where it applies the wrong legal test or is clearly wrong. An inspector’s report may be challenged only where its error is so obvious that it would be irrational for the Secretary of State to rely on it. Death is not itself an “adverse effect” under section 5(4) of the Animals (Scientific Procedures) Act 1986. Unpublished guidance does not vary a statutory code of practice unless the statutory procedure is followed.

The claimant challenged the Secretary of State’s decision to take no action after an investigation into alleged miscategorisation of procedures involving marmosets at a Cambridge research facility and alleged breaches of licence conditions. It also challenged the treatment of animal deaths as adverse effects and the publication of guidance on restricting food and water without consultation or laying it before Parliament.

The central questions were whether the procedures had been correctly classified as moderate rather than substantial, whether the Chief Inspector’s report was irrational, whether death was an adverse effect under section 5(4) of the Animals (Scientific Procedures) Act 1986, and whether the guidance note unlawfully varied a code of practice.

1. Severity classification. The claim succeeded on the first issue. The guidance required an objective assessment of the upper limit of expected adverse effects, taking account of the measures specified to control them, the worst potential outcome for any animal, humane endpoints and actual experience. The “single animal” test applied: one animal suffering a major departure from health or well-being could require a substantial severity limit.
2. The expression “major departure” denoted a serious, significant or important departure, not merely the most severe departure that could lawfully be licensed. The Chief Inspector erred by failing to recognise that adverse effects requiring killing could indicate substantial severity and by failing to apply the single-animal test to the outcomes actually recorded. His conclusion was therefore legally erroneous and clearly wrong.
3. Inspector’s report. The challenge based on individual alleged deficiencies in care failed. The inspector was entitled to accept staff evidence about overnight monitoring, and minor errors concerning a small number of animals would not vitiate the overall conclusion. The Secretary of State could rely on the report unless it was so obviously erroneous that reliance would itself be irrational.
4. Death and adverse effects. Death was not an adverse effect for section 5(4). The statutory expression was shorthand for pain, suffering, distress or lasting harm, all of which concern living animals. Death could nevertheless be relevant to assessing the severity of a procedure.
5. Guidance note. The fourth issue failed. A document published without the consultation and parliamentary procedure required for a variation of a code was not a variation. It was non-statutory guidance.
6. The claim accordingly succeeded on the first issue and failed on the other three. No order for costs was made. Permission to appeal on the first issue was granted, subject to a £20,000 costs condition; permission on the other grounds was refused.

The judgment records that permission-stage proceedings and a hearing before Bean J had preceded the substantive hearing. The High Court granted permission to appeal on the first issue, subject to a costs condition, and refused permission on the other grounds.