Every twelve seconds, an animal in a British laboratory is subjected to a scientific procedure. Burned. Blinded. Force-fed chemicals. Infected with engineered disease. Restrained in chairs until the stress causes physical injury so severe that their bodies begin to fail in ways that no euphemistic procedural classification can adequately describe. Every twelve seconds, for the entirety of the year 2025, amounting to 2.54 million procedures on living animals in the laboratories of Great Britain.

That figure, released in the new UK Government statistics on scientific procedures, represents a 4 percent reduction from the previous year. Animal Defenders International acknowledged the reduction and stated plainly that it is insufficient, given what the reduction was measured against and given what the scientific community has been saying about the predictive validity of animal experimentation with increasing clarity for years. The U.S. Food and Drug Administration has stated that animals do not provide adequate models of human health and disease. The evidence for that position has been accumulating for decades. And yet, in 2025, 2.54 million animals were still subjected to procedures in British laboratories because the systems that govern scientific research have not moved with any urgency commensurate with what the science demands.

At Sustainable Action Now, we have covered the case against animal testing from multiple angles because the issue does not resolve to a single argument. It is simultaneously an animal welfare crisis, a scientific credibility problem, and a regulatory failure. The 2025 UK statistics make each of those dimensions worse, not better, and they demand a response that goes beyond welcoming a 4 percent reduction that, at its current rate of improvement, would require many more decades to reach the level of change the government has promised.

This is what the numbers actually say.

The Headline Figure and What It Conceals

Two and a half million procedures sounds like a large number, and it is. What it conceals is even larger. The classification system that governs how animal experiments are described in official statistics uses categories of “mild,” “moderate,” and “severe” to characterize the suffering involved in individual procedures. These categories, while meaningful within their own framework, do not account for the cumulative experience of an animal whose entire existence in a laboratory context involves chronic stress, social deprivation, behavioral restriction, and the psychological impact of living in conditions that no domesticated or wild animal has any evolutionary preparation for.

The Animal Defenders International assessment is precise on this point: the numbers hide the true nature of what these animals experience. A procedure classified as mild does not mean an animal is living a comfortable life punctuated by one difficult event. It means that the specific procedure itself falls within the mild classification, while the animal’s overall circumstances, including how it was bred, how it arrived at the facility, what its housing conditions are, and how its daily existence is managed, are entirely separate considerations that the procedure statistics do not capture.

The 2025 statistics make this gap between numbers and reality particularly vivid in the specific figures they reveal. There were 28,205 procedures classified as severe, the highest category of suffering, which the classification system defines as including severe or prolonged pain, lasting disability, prolonged restriction of movement, and procedures that push animals to physical exhaustion. On average, every twenty minutes during the working year, an animal in a British laboratory is undergoing this level of recognized suffering. That is not a statistical abstraction. It is a recurring, predictable event happening at a pace that makes it difficult to process as anything other than institutional routine.

Dogs in Toxicity Tests: A Special Category of Preventable Harm

Among the most troubling specific findings in the 2025 figures is the continued and increased use of dogs in toxicity testing. There were 2,889 procedures performed on dogs, an increase from the previous year. The majority of these involved repeatedly dosing the animals with substances to establish toxicity thresholds, a process in which the dog’s response to escalating chemical exposure is monitored over time to determine what levels of a substance cause damage or death.

The use of beagles in toxicity testing has a specific history in both the United Kingdom and the United States that advocacy organizations have documented extensively. Beagles are chosen for research settings because their docile temperament makes them easier to handle in laboratory environments. The quality that makes them suitable for families and was selectively bred into them over generations, their trust of humans and their gentleness, is the same quality that makes them useful for research requiring repeated handling. It is one of the more direct examples available of a breed’s most valued characteristics being turned against its welfare.

The Monkey Crisis: 15 Percent More Primates, Imports Tied to Wild Capture

The 15 percent rise in primate procedures is the figure in the 2025 statistics that carries the most immediate urgency, both because of its direction and because of what the underlying data reveals about where these animals are coming from.

There were 2,236 procedures performed on monkeys in 2025, an increase of approximately 300 animals, representing an average of six procedures on monkeys every single day. The United Kingdom imported 1,676 primates from Africa and Asia, the highest import figure since 2021. Animal Defenders International believes these imports are substantially sourced from primate breeding facilities in Mauritius and Vietnam that ADI investigators have previously documented in detail, exposing conditions that included brutal handling of animals and management practices that fall far short of any reasonable animal welfare standard.

The regulatory gap that allows this sourcing to continue is both specific and troubling. The UK Home Office has acknowledged that it has no authority to regulate the conditions at overseas primate breeding facilities or to set standards for how the animals are treated before they arrive in the country. The regulatory framework that governs animal procedures in British laboratories does not extend to the supply chain that produces the animals used in those procedures. What happens in the breeding facilities before the animals board their transport is, from a regulatory standpoint, outside the jurisdiction of the rules that are supposed to protect them.

More than half of the imported primates, specifically 859 animals, were classified as F1, meaning they were born to at least one wild-caught parent. This classification is the core of what Animal Defenders International describes as the wild capture backdoor. A formal ban on the use of wild-caught primates in European and UK research has been in place since it was included in EU Directive 2010/63, which later became UK law. The Directive also included a commitment to phase out the use of F1 primates, recognizing that allowing the use of animals born to wild-caught parents creates an incentive to continue wild capture to stock the breeding colonies. Sixteen years after the Directive, that phaseout commitment has produced no discernible result. The F1 figures indicate that wild capture is continuing to supply the primate research pipeline through exactly the mechanism that the Directive attempted to close.

The experience of a primate entering this pipeline, as ADI has described it, begins before any experimental procedure is counted. Infant monkeys are separated from their mothers, handled for transport, tattooed for identification, shipped internationally in conditions whose welfare standards are unregulated by the importing country, and arrive in laboratories already carrying the physiological markers of chronic stress that begins in the breeding facility and is compounded by every subsequent stage of their journey. This is the context within which “mild” and “moderate” procedure classifications are applied, and it is the context that the procedure count alone does not convey.

The Forced Swim Test: A Promise Not Kept

One of the specific figures in the 2025 statistics deserves sustained attention because it documents a specific broken government commitment.

The forced swim test, also known as the Porsolt swim test or the behavioral despair test, is a procedure in which small mammals, most commonly mice or rats, are placed in a container of water from which they cannot escape and observed until they cease active swimming and float passively, a behavioral state that researchers in the test’s traditional framework have characterized as “despair.” The procedure is typically used in research on depression and antidepressant drugs.

The scientific validity of the forced swim test has been substantially challenged in recent years. Critics argue that the passive floating behavior that the test is designed to measure is not a reliable model of clinical depression in humans and does not predict antidepressant efficacy with the accuracy that would justify its continued use. Multiple pharmaceutical companies have voluntarily discontinued the test on scientific grounds. Journals have declined to publish results derived from it. The scientific community has been moving away from the forced swim test on the basis of the evidence.

The UK Government committed to ending the forced swim test. No date has been set for that commitment to take effect.

In 2025, 93 mice were subjected to forced swim tests in British laboratories. These were procedures conducted after the government commitment was made, under conditions that the government has declined to prohibit, using a methodology that the scientific community has been moving away from on its own initiative. The government’s position appears to be to wait for the test to become obsolete through voluntary discontinuation rather than to prohibit it, which is precisely the approach that Tim Phillips of Animal Defenders International identified as the critical failure of current UK strategy: the ambition to end tests that have already fizzled out rather than to drive actual change.

The same analysis applies to skin sensitization testing on animals. Thirty-one mice were used in skin sensitization tests in 2025, despite the existence of OECD-approved non-animal alternatives for exactly this kind of testing that are recognized as valid by the regulatory bodies that govern safety testing. The UK has committed to phasing out animal-based skin sensitization testing by the end of 2026. The alternatives exist now. The tests are still happening.

Genetically Altered Animals: The Hidden Numbers

The statistics on genetically altered animals represent a category of the total that most public discussions of animal experimentation do not adequately address.

In 2025, over 1.2 million procedures involved the creation and maintenance of genetically altered animal lines. More than one million of these were specifically for maintaining established lines, meaning keeping the existing population of genetically modified animals alive and breeding. Another 554,021 experimental procedures were performed on genetically altered animals. The number of genetically altered animals bred for experimental programs significantly exceeds the number actually used in experiments, with the surplus animals discarded.

The creation of animals with harmful phenotypes, meaning animals genetically engineered to develop painful or debilitating conditions such as cancer, neurological disease, or musculoskeletal disorder, rose to 93,322 in 2025. These animals are designed to suffer. The suffering is the point of their genetic modification, because the research they are created to support requires animals that display the disease or condition under study. Of these genetically altered animals with harmful phenotypes, 10,612 endured severe suffering as defined by the highest classification category.

The ethical framework that permits the creation of animals designed to be chronically ill is one of the most fundamental questions that animal experimentation raises, and it is one that tends to receive less public attention than the more visible categories of chemical testing or behavioral research. Creating a living being with the genetic instruction to develop a painful condition represents a category of human intervention in animal life that goes beyond using an existing animal for a purpose it was not designed for. It involves engineering an animal’s biology specifically to produce suffering that serves a research goal. The 2025 figures show this practice increasing.

The Alternatives That Already Exist

The most important context for every number in the 2025 UK statistics is the state of the alternative technologies that exist now, are currently in use, and consistently out-perform animal models on the specific measure that matters most: accurately predicting what happens in human biology.

Organoids, which are miniature three-dimensional tissue structures grown from human stem cells that replicate the architecture and function of specific organs, allow researchers to study how human tissue responds to drugs, toxins, and disease in ways that animal models cannot approach. Organs-on-chips, the microfluidic devices lined with living human cells that replicate organ-level physiology, have demonstrated in multiple comparative studies that they produce more accurate human response predictions than animal testing for the same substances. High-throughput screening using human cell cultures allows simultaneous testing of thousands of compounds against human biological targets. Computational modeling and artificial intelligence tools can predict the behavior of chemical compounds in human systems from existing molecular data at speeds that biological testing of any kind cannot match.

These are operational technologies. They are being used in research and regulatory settings now. The rate at which they have replaced animal testing is, as the 2025 figures document, insufficient relative to both their capability and the number of animals still being subjected to procedures. The gap between what the alternative science can do and what regulatory systems require is not a scientific gap. It is a regulatory and institutional gap, one that reflects the inertia of systems built around animal models, the financial infrastructure of the testing industry, and the conservatism of regulatory agencies that have historically required animal data and have been slow to accept validated alternatives as substitutes.

Tim Phillips of Animal Defenders International framed this directly: there is a disconnect, and it is up to governments and regulators to force the change, because it is simply not going to happen otherwise. The voluntary adoption of alternatives by individual researchers and companies, while genuinely valuable, has not produced the pace of change that 2.54 million procedures per year demands. The regulatory requirement for animal data is what sustains the market for animal testing, and the regulatory requirement is what governments have the power to change.

What the UK Roadmap Promises and What the Numbers Say

The UK Government’s 2025 roadmap, titled “Replacing Animals in Science: A Strategy to Support the Development, Validation, and Uptake of Alternative Methods,” set out a framework with specific commitments to end certain categories of animal testing. The document was welcomed by animal welfare organizations as a positive signal of governmental intent.

The 2025 statistics raise the question of whether the commitments in the roadmap are calibrated to the actual scale of the problem. The Home Office has stated that it aims to replace rabbit pyrogen tests and certain skin and eye tests by the end of 2026. The 2025 figures reveal that no rabbit pyrogen tests and none of the targeted skin and eye tests were performed in 2025. These procedures have already effectively ended in practice. Committing to prohibit them by 2026 is not driving change. It is documenting an absence that has already occurred.

The roadmap’s stated ambition is a world where the use of animals in science is eliminated in all but exceptional circumstances. If the annual reduction in animal procedures continues at the pace documented in recent years, that ambition represents a timeline measured in generations, not in the years that the urgency of the science and the scale of the suffering demand.

What Needs to Change, and What You Can Do

The argument against animal experimentation has never been stronger on purely scientific grounds. The FDA’s own position that animal models do not adequately predict human outcomes is not the position of a fringe advocacy group. It is the regulatory assessment of the agency responsible for approving every drug sold to American consumers. The growing body of evidence on alternative methodologies is not speculative. It is peer-reviewed, validated, and increasingly incorporated into regulatory guidance in the United States, in the European Union, and elsewhere.

What is needed is the conversion of that scientific consensus into regulatory requirement, at a pace that reflects the urgency of 2.54 million procedures per year rather than the comfort of incremental reduction. Specific policy actions that would make that conversion real include mandatory timelines for the replacement of specific animal test categories where validated alternatives already exist, stronger regulatory authority over the international primate supply chain that currently allows UK laboratories to source animals from facilities whose conditions the UK government has no power to inspect or regulate, and funding frameworks that actively incentivize the adoption of alternative methodologies rather than simply making them available as options alongside the existing animal model standard.

For individuals who want to contribute to this change, supporting organizations doing the regulatory advocacy and scientific validation work is the most direct available action. Animal Defenders International, the Physicians Committee for Responsible Medicine, the National Centre for the Replacement, Refinement and Reduction of Animals in Research in the UK, and parallel organizations in the United States and Europe are conducting the day-to-day work of pushing regulatory systems toward the change that the science supports.

Choosing cruelty-free certified products in every consumer category where the option exists sends market signals that have real influence on corporate investment in alternative testing methodologies. Engaging with elected representatives on the specific regulatory changes that would accelerate the transition, particularly the replacement of animal data requirements with validated alternative requirements in drug approval pathways, connects the scientific argument to the political process where the necessary decisions are made.

Every twelve seconds, another animal in a British laboratory is subjected to a procedure. The alternatives to that procedure often already exist. The regulatory system that could require their use instead is the variable that remains to be changed, and changing it is the work that the science and the suffering demand.

Sustainable Action Now will continue covering animal testing legislation, the development of human-relevant research alternatives, and the global campaign to end animal experimentation.