Summary
The statistics on animal research in the UK for 2025 were released on 9 July 2026. The significant reduction in experimental procedures causing ‘severe’ suffering, was a positive development, with figures falling to 17,593 in 2025. This follows the 48,380 procedures recorded in 2024, and contributes to a significant 88% overall reduction since 2014 (the first year for which data were collected and published).
Species
Mice were most commonly used in severe procedures, followed by fishes (including zebrafish), and rats. Other species that experienced severe suffering include chickens, guinea pigs, frogs and other amphibians, hamsters, rabbits, other birds, ferrets, pigs, cattle, gerbils, dogs, sheep, and monkeys.
Purpose of procedures
During 2025, the primary purpose of procedures causing severe suffering was basic research (54%), translational or applied research (26%), and regulatory testing (19%). This represents a major shift from the 2024 data, where regulatory testing was the main purpose (74%).
Sub-categories of severe use: Basic research
Of the 9,664 severe procedures conducted for basic research purposes, the greatest number of procedures involved studies of the immune system, nervous system, and cardiovascular blood and lymphatic systems. These data reveal a 21% rise in immune-related procedures and an 8% increase in neurological research, contrasted by a 21% reduction in cardiovascular studies, compared to the 2024 figures.
Sub-categories of severe use: Translational research
In 2025, a total of 4,375 severe procedures were conducted for translational research - a 26% increase compared to 2024. In this sub-category, non-regulatory toxicology and ecotoxicology, human infectious disorders, and human cancer were the leading causes of severe procedures. Notably, the number of procedures in these three sub-categories increased compared to 2024 (80%, 42%, and 13%, respectively).
Sub-categories of severe use: Regulatory use
Throughout 2025, there were 3,552 severe procedures carried out for regulatory purposes, representing a substantial 90% reduction from the previous year. Within this specific sub-category, chronic toxicity, batch potency testing, and acute toxicity were the three main causes of severe suffering. Interestingly, despite a 12% increase in chronic and a 38% increase in acute toxicity testing compared to 2024, batch potency testing saw a significant 98% decrease. Furthermore, the use of lethal LD50 and LC50 methods, which previously represented one of the main causes of suffering, decreased from 5,476 procedures in 2024 to 91 in 2025, a 98% reduction.
The significant decrease in the overall number of experimental procedures undertaken in the UK causing severe suffering to the animals involved - falling from 3.4% in 2024 to 1.3% in 2025 - is primarily driven by a significant reduction within the category of regulatory testing. This is especially evident in the declining use of batch potency testing (quality control) and toxicity and safety testing (lethal LD50 and LC50 methods).
The development and quality control of biological therapies and vaccines have traditionally involved batch testing based on assays that can cause severe suffering. A notable example of this involves the batch potency testing of widely popular ‘botox-type’ products. While botulinum toxin (Bt) has therapeutic uses, the use of Bt for cosmetic purposes has greatly increased demand, with the International Society of Aesthetic Plastic Surgery reporting that 7.8 million botox treatments were administered globally in 2024. Due to the high risk of batch-to-batch variability and the highly toxic nature of the product, all batches of Bt require potency testing for quality control. For this, the mouse LD50 assay has historically been used to determine the lethal dose of the product in 50% of the mice involved in the testing.
With companies developing and then transitioning to validated and reliable non-animal methods accepted as ‘alternatives’ in the UK, such as in vitro cell-based assays, the requirement for thousands of regulatory tests is declining. In fact, the Medicines and Healthcare products Regulatory Agency (MHRA) now accepts a non-animal alternative method for the testing of the most common and specific strengths of Type A botulinum toxin drug substances. Moreover, replacing the use of animals in Bt batch potency testing has been prioritised in ‘Basket 1’ of the recent UK government’s strategy to phase out the use of animals in research and testing, with a commitment to achieve complete replacement by 2027. This category includes tests or animal models for which replacement technologies already exist and are validated, and could be applied to phase out animals in all but exceptional circumstances. Other tests in this Basket include the rabbit pyrogen test and the skin and eye irritation tests, among others.
It is likely that procedural refinements have also contributed to these reductions. Conducting refined welfare assessment and using advanced monitoring techniques help inform and refine humane endpoints, consequently reducing severity levels e.g. from severe to moderate. Crucially, while the refinement of procedures is vital to reduce animal suffering, other aspects of the animal’s life, such as housing, husbandry, and care, also warrant attention, requiring their own targeted refinements. Real-life examples of refinements that successfully reduced severe suffering can be found here.
To help individuals working with and caring for laboratory animals reduce animal suffering, we developed the Roadmap to Reducing Suffering. This free and easy to use resource enables you to review your own protocols, identify potential sources of harm, and implement targeted refinements.
Importantly, the UK Government has recently committed to publishing better statistics on animals used in science, including the number of animals bred for research but not used in experiments, and those subjected to tissue sampling for genotyping. Greater transparency about the full scale of animal use, and capturing more of the individuals impacted on by science, is essential for identifying further opportunities to reduce the numbers of animals bred and used in research.