Guidance

User guide to annual statistics of scientific procedures on living animals, Great Britain

Updated 11 September 2024

Applies to England, Scotland and Wales

1. Introduction

This user guide accompanies the Home Office collection ‘Statistics of scientific procedures on living animals, Great Britain’. The statistics in the collection relate to regulated scientific procedures performed using living animals under the Animals (Scientific Procedures) Act 1986 (ASPA).

1.1 Uses of the statistics

The expected uses of the statistics are listed using the standard categorisation for Official Statistics.

a) Informing the general public’s choices:

• about the state of the economy, society and the environment – figures on animal scientific procedures are used by the media
• about the performance of government and public bodies – figures on levels and trends in animal scientific procedures are requested via Parliament Questions and Freedom of Information Act request

b) Government policy making and monitoring

c) Resource allocation – typically by central and local government

2. Data coverage and source

2.1 Legislation and purpose of collection

The annual statistics publication relates to scientific procedures performed using living animals subject to the provisions of ASPA. Section 21A ASPA requires licensing and oversight of all places, projects, and people who want to conduct scientific procedures on living animals. The Act also states that annual statistics on protected animals (this release) should be published annually and laid in parliament.

Prior to 1986, figures were recorded for the number of ‘experiments’ on living animals, under the Cruelty to Animals Act 1876.

2.2 Data source

All scientific procedures carried out on living animals in the UK, subject to ASPA, must be authorised under a project licence, granted by the Home Secretary. The statistics are collected via online data returns submitted by project licence holders at the end of each calendar year, or within 28 days of the termination of the licence when this occurs during the year. Since the 2021 collection, this data is held in and extracted from the Animals in Scientific Procedures e-Licensing (ASPeL) system.

2.3 Data coverage

Within the UK, the Home Office collects and publishes information on regulated procedures for Great Britain, while in Northern Ireland the Department of Health separately collects and publishes this information under devolved arrangements.

Regulated procedures are defined in ASPA as any procedure applied to a protected animal for an experimental or other scientific purpose, or for an educational purpose, that may have the effect of causing an animal pain, suffering, distress or lasting harm equivalent to, or higher than, that caused by the introduction of a needle in accordance with good veterinary practice.

Protected animals are defined in the ASPA as any living vertebrate other than man and any living cephalopod. As ASPA indicates, the breeding of an animal is a regulated procedure if the animal is bred from, or is the descendant of, an animal whose genes have mutated or been modified.

The annual statistics publication predominately focuses on experimental procedures and procedures counted under the creation and breeding of genetically altered (GA) animals separately.

Experimental procedures include all animals used in basic research, regulatory use, translational/applied research, protection of the natural environment, higher education and training, preservation of species and forensic enquiries.

Creation and breeding of GA animals includes some animals that were bred with the intention of producing GA animals, but resulted in non-GA animals being born. In addition, some animals used for the creation of a new genetic line would also have been genetically normal animals, for example, those used for superovulation (production of eggs). These animals were not used in experimental procedures.

All establishments seeking to conduct regulated procedures on living animals must be licensed under ASPA. Information regarding establishment type is not collected as part of the return of procedures data used in this publication. Establishment type was previously reported as the result of a separate data collection exercise. The reported establishment type was not an indicator of the type of procedures carried out, and often establishments could be categorised as more than one type.

2.4 Counting rules

The figures provided in the statistical report refer to the numbers of procedures (completed) rather than the numbers of animals used (for the first time), unless indicated otherwise. However, in most cases, the number of procedures corresponds to the number of animals used. In instances where the number of procedures is higher than the number of animals used, this is due to a re-use of animals.

Each procedure (which may consist of several stages) is counted once for the year in which it was completed.

A study involving procedures on a number of animals is counted once for each animal. Where an animal that has previously been used in a completed procedure is used again for a further procedure (‘re-use’) it is counted as a separate procedure, but the animal itself is not recounted. The circumstances in which the re-use of an animal is permitted are limited (see section 5.19 of Guidance on the Operation of the Animals (Scientific Procedures) Act 1986).

Procedures on adult or free-living animals (including neonatal and juvenile mammals, and newly hatched birds) are counted. Details of procedures on immature forms (for example, larvae, embryos) are not counted unless they have reached the free-feeding stage (for example, zebrafish from 5 days post-fertilisation and tadpoles).

Animals in the wild involved in rodenticide trials are also not counted. A rodenticide trial is the testing of any substance that is used to kill rats, mice, and other rodent pests. Information is collected on the number of project licences which undertake rodenticide trials.

2.5 Exclusions

The controls of ASPA, and therefore the statistics in the annual release, exclude:

• non-experimental agricultural practices
• non-experimental clinical veterinary practices
• practices undertaken for recognised animal husbandry
• the administration of any substance or article to an animal for research purposes in accordance with an animal test certificate granted under the Veterinary Medicines Regulations 2013 Sect 3 (2) (b)

The annual statistics also exclude:

• ‘genetically normal’ (that is, non-GA) animals that were bred for scientific procedures but were killed or died without being used in procedures
• animals that were subject to procedures for the purposes of genotyping (the process of investigating the genetic makeup of an animal)

The Home Office published additional statistics on breeding and genotyping of animals for scientific procedures for 2017 as experimental statistics.

3. Data quality and methodology

3.1 Data quality

The UK Statistics Authority has designated these statistics as Accredited Official Statistics, in accordance with the Statistics and Registration Service Act 2007, signifying compliance with the Code of Practice for Statistics.

Accredited Official Statistics status

Accredited Official Statistics status means that our statistics meet the highest standards of trustworthiness, quality, and value, and it is our responsibility to maintain compliance with these standards.

The Code of Practice for Statistics states that reports should be published at 9:30am on a weekday. It is not always possible to publish its statistical reports on Scientific Procedures on Living Animals at 9.30am because the laying of this document before Parliament is a statutory obligation laid down in ASPA, and it is not always possible to arrange a publication date that coincides with dates when the report can be laid before Parliament at 9.30am. The Office for Statistics Regulation has confirmed that in these circumstances it is acceptable for the Home Office to publish these statistics at 10am.

The designation of these statistics as Accredited Official Statistics was confirmed in 2023 following a compliance check by the Office for Statistics Regulation. The statistics last underwent a full assessment against the Code of Practice in 2012.

Since the latest review by the Office for Statistics Regulation, we have continued to comply with the Code of Practice for Statistics, and have made the following improvements:

• resumed relationship building with establishments conducting regulated procedures, including initiatives to improve understanding of project licence holder data collection and quality processes
• the statisticians producing this release accompanying ASRU on establishment inspections
• improved transparency of the data collection and quality assurance processes in this user guide, with intent to further expand on strengths and limitations of the data where possible
• provided context to the revisions and corrections table
• provided clarity on the responsibilities of other government departments in relation to these statistics
• conducting user engagement activities, including regular attendance at meetings with representatives from animal protection and welfare groups and the regulated community, and a user survey detailed in section 5.5 of this user guide
• investigating how to provide further commentary on potential causes for changes over time, with consideration as to the role of other government departments
• addition of a glossary in this user guide
• improved internal data quality assurance processes and tools

Any further plans for user engagement, improvements to the publication, and improvements to quality assurance processes will continue to be documented in this user guide.

The following improvements were also made in response to earlier reviews by the Office for Statistics Regulation:

• transitioned to a new online data collection system, with data validation to prevent invalid combinations of data being entered
• created a reproducible analytical pipeline (RAP) to automate quality assurance processes and the production of statistical outputs, in line with RAP principles
• for the 2017 data onwards, data collected on ‘other’ species is subject to data quality checks which have allowed for further analyses of the animals returned under ‘other’ (non-Schedule 2) species categories
• statisticians regularly consult with colleagues in the Animals in Science Regulation Unit to ensure the collection remains suitable for its purpose

3.2 Quality assurance

Before publication the data is subject to strict quality assurance (QA) processes. These QA processes are undertaken within licenced establishments, by the Animals in Science Regulation Unit (ASRU), and by the statisticians in the Home Office.

Within each establishment, project licence holders are responsible for the collection and submission of their data. Each individual conducting work under ASPA holds a personal licence and appropriate training for the species and procedures they are undertaking. Each establishment must hold appropriate animal accommodation and veterinary support. Researchers conducting procedures, veterinary and animal husbandry experts and the local Animal Welfare and Ethical Review Body all provide quality assurance on the numbers of animals used and the severity experienced. Establishments, including the internal records maintained on scientific procedures, are subject to audit (announced and unannounced) by ASRU officials.

Project licence holders are responsible for the completion of annual returns of procedures though the Animals in Scientific Procedures e-Licensing (ASPeL) system. It is an offence for a project licence holder to provide information they know to be false or misleading. ASPeL allows monitoring of incomplete or absent returns, prevents duplicate returns and assures a consistent data format. Data is extracted from ASPeL and are subject to multiple stages of QA checks by statisticians within the Home Office including compilation of data to highlight outliers and possible errors; variance checks such as investigating substantial changes in figures compared with the previous year; querying incomplete data or invalid data/combinations in data in a return; and investigating returns with unusual data or combination of data. These QA checks are carried out alongside veterinary experts within ASRU.

Through ASRU, any queries or issues identified are raised with the project licence holders to establish if the submitted data is correct and to prompt amendment of data if necessary. In some cases, corrections to the data may be made by the statisticians directly, for example if an unusual combination of data cannot be appropriately recorded in the system. Final summary data is compared against previous years to further check for variance that may be of concern.

Based on the above, we are satisfied that the data in this release is high quality and fit for purpose. As the data is sourced from a live administrative system, ASPeL, it is expected that there will be revisions each year.

Guidance provided for the completion of annual returns is available on GOV.UK.

3.3 Data quality issues

Changes to data collection from 2014

Following the European Directive 2010/63/EU being transposed into UK law (the animals in scientific procedures act 1986) in January 2013, the 2014 collection underwent substantial changes. As a result, some inconsistencies were expected in the reporting of the 2014 information.

Previously the Home Office counted ‘procedures started’ but post 2014 this changed to ‘procedures completed’. As a result, procedures started before 2014 but completed in 2014 should have been reported in both the pre-2014 and 2014 figures.

Additionally, there were discrepancies of a small number of duplicate returns, and misclassifications of reported severity in the 2014 collection. Home Office statisticians have since reviewed their processes and quality assurance checks in light of these issues.

Rounding

Data may be rounded to simplify the presentation of the figures. However, all numeric and percentage calculations are based on unrounded data. Where data is rounded, it may not sum to the totals shown, or, in the case of percentages, to 100%, because it has been rounded independently.

The rounding conventions used in the release, unless otherwise stated, are as follows:

• over 1 million - presented as millions and rounded to 2 decimal places, for example, 2,121,582 = 2.12 million
• 10,000 to 999,999 - rounded to the nearest thousand, for example, 343,465 = 343,000
• 1,000 to 9,999 - rounded to the nearest hundred, for example, 8,465 = 8,500
• 10 to 999 - rounded to the nearest 10, for example, 47 = 50
• less than 10 - unrounded, whole numbers
• percentages greater than 1% - rounded to the nearest per cent, for example, 1.43% = 1%
• percentages less than 1% - rounded to one significant figure, for example, 0.43% = 0.4%, and 0.043% = 0.04%

The data in the data tables is unrounded. All percentages in the data tables are rounded to the nearest 0.1%.

Corrections and revisions

It is standard practice across all Home Office statistical releases to incorporate revisions to previous years’ data in the latest release. Corrections and revisions follow the Home Office’s statement of compliance with the Code of Practice for Official Statistics.

As the data is sourced from a live administrative system, ASPeL, it is expected that there will be revisions each year as revised data is submitted in-year. There are a several reasons why counts of procedures may change between year, including amendments following a retrospective assessment or other intervention by an inspector, or self-reported corrections by project licence holders. It is an offence for a project licence holder to provide information they know to be false or misleading. Some changes reported in the publication will be a combination of small increases and/or decreases across multiple projects.

A correction to the statistics will be issued when there is a substantial change in number of animals or type of test, or change in reported values for areas of particular Home Office and public interest (for example, specially protected species), caused by incorrectly submitted data or issue in the analysis made for the publication. In some cases, these corrections will be identified during the development of the next year’s publication, in which case a separate correction may not be made and the change will be reported alongside the revisions.

4. Glossary

Applied research: attempts to address diseases through prevention and development of treatments. Within the data tables, this is shown as ‘Translational/Applied research’.

Basic research: aims to expand our knowledge of the structure, functioning and behaviour of living organisms and the environment.

Breeding: the production of GA animals of an established line that has been bred for at least 2 generations. Breeding procedures also include other techniques applied to the animal after birth, for example, genotyping but not any techniques applied as part of an experiment or study.

Creation: includes the natural breeding of different strains to produce a new strain and procedures that use standard techniques such as vasectomy for the generation of novel transgenic or mutant lines of GA animals. The birth of a GA animal counts as creation when the line is new and before it is ‘established’ (stable and characterised).

Experimental procedures: involve using animals in scientific studies for purposes such as: basic research and the development of treatments, safety testing of pharmaceuticals and other substances, education, specific surgical training and education, environmental research and species protection.

Other efficacy and tolerance testing: efficacy testing of biocides and pesticides is covered under this category as well as the tolerance testing of additives in animal nutrition.

Procedures for creation and breeding: involve the breeding of animals whose genes have mutated or have been modified. These animals are used to produce genetically altered offspring for use in experimental procedures but are not themselves used in experimental procedures.

Protected animals: Any living vertebrate, other than man, and any living cephalopod. This includes embryos after two-thirds of gestation (although these are not included as countable procedures), and fish and amphibian larvae after they become capable of free feeding.

Quality control: the testing of quality control parameters of a product, and any controls carried out during the manufacturing process for registration purposes, to satisfy any other national or international requirements or to satisfy the in-house policy of the manufacturer.

Regulated procedures: Any procedure applied to a protected animal for an experimental or other scientific purpose, or for an educational purpose, that may have the effect of causing an animal pain, suffering, distress or lasting harm equivalent to, or higher than, that caused by the introduction of a needle in accordance with good veterinary practice.

Regulatory testing: procedures carried out to satisfy legal requirements, including: ensuring substances are produced to legal specification; evaluating the safety or effectiveness of pharmaceuticals and other substances.

Routine production: covers studies carried out for manufacturing processes requiring regulatory approval.

Self-sustaining colony:

• marmosets, tamarins, and other new-world primates: a self-sustaining colony is a colony that contains no wild caught animals, is kept in a way that ensures animals are used to humans and is sustained using animals from within or from other self-sustaining colonies
• macaques and other old-world primates: a self-sustaining colony is a colony that no longer sources animals from the wild (it may contain some existing wild caught animals) and is sustained using only captive bred animals

Severity: The severity (pain, distress or suffering) experienced by animals in procedures has been recorded since 2014.

There are 5 severity assessments:

• sub-threshold: when a procedure was authorised under a project licence but did not actually cause suffering above the threshold of regulation, that is, was less than the level of pain, suffering, distress or lasting harm that is caused by inserting a hypodermic needle according to good veterinary practice
• non-recovery (under general anaesthesia): when the entire procedure was carried out under general anaesthesia from which the animal shall not recover consciousness; it includes unintended death of animals on recovery protocols while under anaesthesia, provided that no regulated procedure had been carried out prior to the induction of anaesthesia
• mild: any pain or suffering experienced by an animal was, at worst, only slight or transitory and minor so that the animal returns to its normal state within a short period of time
• moderate: the procedure caused a significant and easily detectable disturbance to an animal’s normal state, but this was not life threatening; most surgical procedures carried out under general anaesthesia and with good post-operative analgesia (pain relief) would be classed as moderate
• severe: the procedure caused a major departure from the animal’s usual state of health and well-being; this would usually include long-term disease processes where assistance with normal activities such as feeding and drinking were required, or where significant deficits in behaviours/activities persist; it includes animals found dead unless an informed decision can be made that the animal did not suffer severely prior to death

Severity assessments measure harms to an animal during a procedure and generally reflect the peak or cumulative severity of the entire procedure; they do not include harms caused to animals as a result of non-procedural events such as transport and housing.

Specially protected species: Cats, dogs, horses and non-human primates.

Toxicity and other safety testing: studies for safety evaluation of products and devices for human medicine, dentistry, veterinary medicine and other chemicals.

5. Related statistics and reports

5.1 Great Britain

On 8 November 2018 the Home Office published ‘Additional statistics on breeding and genotyping of animals for scientific procedures, Great Britain 2017’. These experimental statistics were collected by the Home Office under the EU Directive (2010/63/EU) and cover the number of ‘genetically normal’ (that is, non-genetically altered (non-GA)) animals that were bred for scientific procedures but were killed or died without being used in procedures, and the number of animals (GA and non-GA) subject to tissue sampling for the purposes of genotyping.

The statistics presented in Statistics on animals in scientific procedures have been collected since 2001 and are available online. Data was collected under different categories prior to 2014, meaning some of the data (for example, for species, purpose, genetic status, severity) is not comparable or does not exist.

Older editions of this statistical release may be found in the national archives. However, not all previous publications are available online. Annual publications giving detailed figures for scientific procedures under ASPA have been published as ‘Statistics of Scientific Procedures on Living Animals’ since 1987. Detailed figures for experiments on living animals under the Cruelty to Animals Act 1876 were published as ‘Statistics of experiments on living animals’ between 1977 and 1986. Less detailed information about experiments on living animals for the years prior to 1977 was published in the form of a ‘Return to an Address of the Honourable the House of Commons’.

Forthcoming publications are pre-announced on the ‘Statistics: release calendar’ on the GOV.UK website.

5.2 Northern Ireland

Statistics for scientific procedures in Northern Ireland can be found on the Northern Ireland Department for Health website. The data for Northern Ireland is collected in the same manner and undergoes similar quality assurance processes as the data for Great Britain, meaning the 2 datasets can be used for comparison.

5.3 European Union

In 2018, ASRU made arrangements for EU exit by preparing legislation, known as a Statutory Instrument, to amend ASPA and deliver EU exit. This Statutory Instrument, UK Statutory Instrument 2019 No.72 The Animals (Scientific Procedures) Act 1986 (EU Exit) Regulations 2019, was enacted on 31 January 2020, the day of EU exit. The legislation had the result of removing the requirements for mandatory reporting of information on scientific procedures to the European Commission (EC).

The UK will no longer submit information on scientific procedures on animals to the EC. Therefore, UK data will no longer be incorporated into EC published reports on procedures on living animals in the EU.

5.4 Other related information

A consolidated version of ASPA can be found online, as well as guidance on the operation of ASPA.

The Animals in Science Regulation Unit (ASRU) releases an annual report describing their role and performance in regulating work under the Animals (Scientific Procedures) Act 1986.

ASRU also publishes non-technical summaries of projects granted under the Animals (Scientific Procedures) Act 1986. Publication of non-technical summaries is a legal requirement under ASPA. They are a statement, in non-technical language of the proposed programme of work and state the objectives, predicted harm, benefits of the programme and the number and types of animals to be used in the programme. They also demonstrate that the proposed programme of work will be carried out in compliance with the principles of replacement, reduction and refinement. The non-technical summaries of projects granted by ASRU includes those that should be retrospectively assessed.

ASRU’s publications are available on GOV.UK.

Information about the Animals in Science Committee.

Information about the National Centre for the Replacement, Refinement and Reduction of Animals in Research (NC3Rs) can be found on their website.

5.5 Feedback and enquiries

A user engagement survey was conducted in 2024 with representatives from organisations in the regulated community, and the animal welfare and protection community. The survey findings were broadly positive and provided insight as to how organisations may use the statistics, and also identified areas for improvement. Survey respondents provided multiple topics not currently reported in the statistics that they deemed of public interest, and views on the previously published Additional statistics on breeding and genotyping of animals for scientific procedures. These topics and views are being taken into consideration, though they are out of scope of the current statistics production process.

We have made improvements to this publication following the conclusion of this survey, including improved ordering of the data tables; increased visibility of this user guide; improvements to the content of the user guide; improved clarity in areas of text (for example, the “Techniques of special interest”).

We are reviewing future changes including our policy on the rounding of values in the publication and further improvements to the publication and user guide text. Some requests are not possible to implement as they would require changes that do not align with accessibility requirements or best practice under the Code of Practice for Statistics.

We welcome feedback on these statistics. If you have any feedback or enquiries about this publication, please contact the Statistical Transformation team, the Home Office Unit which produced the statistics, via HOAIStatisticalTransformation@homeoffice.gov.uk.