Debated.
Start debating
All motions

Sources · Animal testing for medicine

This house would ban animal testing for medical research.

Fact bank v1 · 12 cards

  1. AT-01Mixed

    “About 90% of drugs that pass animal tests then fail in human trials, which proves animal testing doesn't work.”

    About 9 in 10 drugs that start human trials do fail. But they fail for many reasons, not only because animal tests got it wrong. A 2022 review said 40-50% fail because the drug doesn't work well enough, 30% because of harmful side effects, 10-15% because of how the body takes in and handles the drug, and 10% for business or planning reasons.

    Some of these failures may happen because animal tests did not show how people would react, but the 90% figure alone can't tell us how many. Before human trials, drugs are also checked with computer models and cells grown in the lab, not just animals. The review says only 10-15% of drugs that enter human trials succeed, and this has hardly changed in decades.

    “four possible reasons attributed to the 90% clinical failures of drug development: lack of clinical efficacy (40%–50%), unmanageable toxicity (30%)”
    Acta Pharmaceutica Sinica B (open-access article on PubMed Central)·Why 90% of clinical drug development fails and how to improve it? (Acta Pharmaceutica Sinica B, 2022)Government / parliament · 11 Feb 2022 ↗

    Further sources for this card

  2. AT-02True

    “Only about 1 in 20 treatments that are tested on animals ever gets approved for use in people.”

    Confirmed: a 2024 study that pulled together 122 earlier reviews, covering 367 treatments for 54 diseases, found that 50% of treatments tested in animals went on to be tested in people, 40% reached randomised controlled trials (the fairest kind of medical test), and just 5% were approved.

    The same review found that when a treatment worked in animals, human studies agreed 86% of the time. The authors say animal results may carry over to people better than is often claimed, and that the low approval rate points to weak design in both animal studies and early human trials. Going from animal studies to approval typically took about 10 years.

    “The overall proportion of therapies progressing from animal studies was 50% to human studies, 40% to RCTs, and 5% to regulatory approval.”
    PLOS Biology·Analysis of animal-to-human translation shows that only 5% of animal-tested therapeutic interventions obtain regulatory approval for human applications (PLOS Biology, 2024)Peer-reviewed research · 13 Jun 2024 ↗
  3. AT-03Mixed

    “Animal tests catch most of the harmful side effects a new drug will have in people.”

    Studies disagree. A 2000 survey by drug companies found animal tests had flagged 71% of the side effects later seen in people. But a larger 2017 industry study of 182 drugs found animal tests flagged only 48% of the problems seen in the first human trials.

    The 2017 study also found that when animals showed no harm to a part of the body, people usually had no problem there either (86% of the time). But only 43% of the harms seen in animals then showed up in people. Both studies used data from drug companies, and both looked only at drugs that went on to be tested in people.

    “Sensitivity was 48% with a 43% positive predictive value (PPV).”
    Toxicology and Applied Pharmacology (abstract via PubMed, US National Library of Medicine)·Current nonclinical testing paradigm enables safe entry to First-In-Human clinical trials: The IQ consortium nonclinical to clinical translational database (Toxicology and Applied Pharmacology, 2017)Government / parliament · 8 Sept 2017 ↗

    Further sources for this card

  4. AT-04Mixed

    “Mice share 99% of their genes with humans, so they are a good stand-in for people.”

    The 99% figure is about right: a major 2002 study of the mouse's DNA found about 99% of mouse genes have a related gene in humans, and about 80% have one clear human partner gene. But related is not the same as identical. Only about 40% of human DNA can be lined up with mouse DNA at all when the two are compared letter by letter.

    Matching genes still differ. In matched pairs of mouse and human proteins, 70.1% of the building blocks were the same. In the parts of genes that hold the recipe for proteins, the DNA is about 85% the same. Sharing genes does not show that a medicine will act the same way in mice and people.

    “The proportion of mouse genes without any homologue currently detectable in the human genome (and vice versa) seems to be less than 1%.”
    Nature·Initial sequencing and comparative analysis of the mouse genome (Nature, 2002)Peer-reviewed research · 5 Dec 2002 ↗
  5. AT-05Mixed

    “Organ-on-a-chip tests can already predict human results better than animal tests, so they can replace them.”

    In a 2022 study, a 'liver-chip' made from human cells was tested on 27 drugs. It caught 87% of the drugs known to harm human livers, every one of which had passed animal tests, and it did not wrongly flag any safe drug. But it covered one organ and a small set of drugs, and the authors suggest using it alongside animal tests, not instead of them.

    Most of the study's authors worked for the company that makes the chip. A 2025 Nature Biotechnology news report said organ chips and similar new methods are 'not ready to fully replace animal testing just yet', partly because copying organs is hard. The UK government also says no suitable alternatives exist yet for some complex tests.

    “a blinded set of 27 known hepatotoxic and non-toxic drugs with a sensitivity of 87% and a specificity of 100%”
    Communications Medicine (Nature Portfolio)·Performance assessment and economic analysis of a human Liver-Chip for predictive toxicology (Communications Medicine, 2022)Peer-reviewed research · 6 Dec 2022 ↗

    Further sources for this card

  6. AT-06True

    “In the US, the law no longer says new medicines must be tested on animals.”

    Confirmed: a US law called the FDA Modernization Act 2.0, signed in December 2022, ended the rule that animal tests were required when companies ask the US medicines regulator, the Food and Drug Administration (FDA), to test or sell new medicines. They can now use other methods instead, such as tests on human cells, organ chips and computer models.

    The law allows other methods but does not ban animal testing. In April 2025 the FDA set out a plan to make animal testing 'the exception rather than the norm' within 3 to 5 years, starting with a type of medicine called monoclonal antibodies.

    “while animal testing had been the default standard, it is no longer a mandatory requirement for regulatory filings involving drugs and biological products”
    Circulation: Genomic and Precision Medicine (author manuscript on PubMed Central)·Precision Medicine and the FDA Modernization Act 2.0: Catalyzing Innovation in Cardiovascular Therapy (Circulation: Genomic and Precision Medicine, 2025)Government / parliament · 3 Nov 2025 ↗

    Further sources for this card

  7. AT-07True

    “In the UK, scientists are only allowed to use animals when a method without animals can't do the job.”

    Confirmed: the Animals (Scientific Procedures) Act 1986 says that, wherever possible, a scientifically suitable method that doesn't use animals must be used instead. Before a project gets a licence, the government must also weigh the harm to the animals against the expected benefits.

    This 'replacement' rule is one of the '3Rs': replace animals where possible, reduce how many are used, and refine methods to cause less pain. It was added to the Act in 2013. The law covers all animals with a backbone, plus octopuses, squid and their relatives. In November 2025 the government set a long-term aim of phasing out animal use 'in all but exceptional circumstances', but it did not ban animal research and says it will keep supporting it where good alternatives are not yet available.

    “wherever possible, a scientifically satisfactory method or testing strategy not entailing the use of protected animals must be used instead of a regulated procedure”
    legislation.gov.uk (The National Archives, UK Government)·Animals (Scientific Procedures) Act 1986, section 2A: Principles of replacement, reduction and refinementGovernment / parliament · 1 Jan 2013 ↗

    Further sources for this card

  8. AT-08Mostly false

    “Most animals used in experiments suffer severe pain.”

    In Great Britain in 2025, 1.3% of experiments on animals were rated 'severe'. About half (49%) were rated 'mild' and 28.9% 'moderate'. Another 17.2% caused less harm than a needle injection, and 4% were done fully under anaesthetic, with the animal never waking up.

    'Moderate' still means a clear, easy-to-spot upset to the animal's normal state, though not life-threatening. Most operations done under anaesthetic with good pain relief count as moderate. The ratings cover harm from the experiment itself, not from transport or housing. The severe share was higher in 2024, at 3.4%. These figures are for experiments; for breeding genetically altered animals, 1% were rated severe.

    “The proportion of severe procedures decreased from 3.4% in 2024 to 1.3% in 2025.”
    Home Office (UK Government, accredited official statistics)·Annual statistics of scientific procedures on living animals, Great Britain 2025Government / parliament · 9 Jul 2026 ↗
  9. AT-09True

    “Nearly all animals used in research are mice, rats, fish and birds, not dogs, cats or monkeys.”

    Confirmed: in Great Britain in 2025, 95% of procedures used mice, fish, birds or rats (91% if you count only experiments, not breeding). Cats, dogs, horses and primates such as monkeys were used in 1% of experiments.

    There were 2.54 million procedures in 2025, the lowest number since the 1986 law. Mice alone were used in 56% of experiments, and rats in 9%. Dogs were used in 2,889 procedures, mostly for tests required by law. One animal can sometimes be used more than once, so the number of procedures is slightly higher than the number of animals.

    “the majority (95%) of procedures (both for experimental and breeding purposes) used mice, fish, birds or rats”
    Home Office (UK Government, accredited official statistics)·Annual statistics of scientific procedures on living animals, Great Britain 2025Government / parliament · 9 Jul 2026 ↗
  10. AT-10False

    “Most animal testing is done to make cosmetics and make-up.”

    In the UK, testing cosmetics on animals has been banned since 1998, and in Great Britain no animal procedures were done for cosmetics testing in 2025. Over half (54%) of animal experiments were for basic research, most often on the nervous system, the immune system and cancer.

    Another 19% were tests required by law before products can be approved, mostly for medicines for people and animals. The EU bans it too: testing finished cosmetics on animals has been banned there since 2004, and selling cosmetics tested on animals has been fully banned since 2013. This motion is about medical research, which is a separate question from cosmetics.

    “Animal testing for consumer safety of cosmetics and their ingredients has been banned in the UK since 1998.”
    Home Office (UK Government, accredited official statistics)·Annual statistics of scientific procedures on living animals, Great Britain 2025Government / parliament · 9 Jul 2026 ↗

    Further sources for this card

  11. AT-11Mostly false

    “Most people support using animals in scientific research.”

    In the US, opinion is split, with slightly more people against than in favour. In a 2018 Pew survey, 47% of adults favoured using animals in scientific research and 52% opposed it.

    Views differed a lot by gender: 58% of men were in favour but only 36% of women. People with high science knowledge were more likely to be in favour (63%). A 2014 Pew survey found much the same: 47% in favour and 50% against. This is one country, and the question asked about scientific research in general, not only medicine. Views in other countries may differ.

    “Some 47% favor the practice, while 52% oppose it, according to a new Pew Research Center survey”
    Pew Research Center·Americans are divided over the use of animals in scientific researchIndependent research institute · 16 Aug 2018 ↗
  12. AT-12True

    “Insulin, a life-saving diabetes medicine, was discovered through experiments on dogs.”

    Confirmed: in 1921 a team in Toronto made an extract from the pancreas and showed it lowered blood sugar in dogs whose pancreas had been removed. A colleague, Collip, then used rabbits to check how strong the extract was while he made it purer. In January 1922 a teenage boy with diabetes became the first person treated. The first dose didn't help, but Collip's purer extract worked.

    Banting and Macleod won the 1923 Nobel Prize for Physiology or Medicine. The 2021 article notes that other researchers had made similar extracts years before, and that the big step was making one safe enough for people. This is a historical example: it shows animal research helped in the past, but on its own it doesn't show whether animals are still needed today.

    “creating an extract of the pancreas and showing that it reduced blood glucose in depancreatized dogs”
    Canadian Medical Association Journal (on PubMed Central)·Rethinking the "discovery" of insulin (Canadian Medical Association Journal, 2021)Government / parliament · 25 Oct 2021 ↗

    Further sources for this card

The judge is an AI model. It can only rule on claims these cards cover. Everything else stays unverified, and it never rules on opinions. See every motion's cards.