Skip to Content
Research Modernization NOW

Limited Predictive Value of Research Using Animals 

Many in the scientific community are aware of the flaws of experiments on animals. NIH reports that novel drugs fail “in about 95 percent of human studies,”1 even though they appeared safe and effective in preclinical experiments on animals. A 2014 analysis published in The BMJ found that animal studies largely have not furthered knowledge in the field of human health or led to the development of treatments for conditions affecting humans.2

Lack of Validity

Problems with internal and external validity contribute to the failure of experiments on animals in the translation of biomedical research from bench to bedside. The internal validity of experiments on animals is undermined by poor study design, including failure to implement processes to prevent bias, such as blinding, in which the individuals conducting the experiments or those analyzing the data do not know whether the animals or samples belong to the treatment or control group. Scientists have found that a lack of measures to reduce bias in experiments on animals likely results in overestimation of the benefits of the treatment studied, noting that this bias affects the trustworthiness of results, wastes resources, and should not be used to inform human clinical trials.34

Poor internal validity means that many experiments on animals cannot be reproduced, a critical aspect of the scientific process that speaks to the potential validity of a finding. It is unsurprising, therefore, that a 2015 investigation concluded that between 18% and 89% of all preclinical research, a large part of which involves animal testing, was irreproducible, resulting in billions per year spent on experimentation that is misleading for human health.5 Former NIH leadership has admitted, “[p]reclinical research, especially work that uses animal models, seems to be the area that is currently most susceptible to reproducibility issues.”6

However, the weaknesses of experiments on animals cannot be overcome simply by improving study design, because external validity, or the “extent to which research findings derived in one setting, population or species can be reliably applied to other settings, populations and species,”7 can never be achieved. Inherent species differences mean that other species cannot serve as analogs for understanding the biological mechanisms of disease and the effects of drugs on humans.

“On average, extrapolated results from studies using tens of millions of animals fail to accurately predict human responses.” 8

Therefore, experiments on animals lack internal and external validity. In other words, they are usually poorly executed, but even if the experimental methods were improved, the results would not translate to humans.

In a 2018 review published in the Journal of Translational Medicine, Pandora Pound and Merel Ritskes-Hoitinga discuss species differences as an insurmountable problem of external validity for preclinical animal models. Attempts to control for or correct species differences result in what the authors refer to as the “extrapolator’s circle.” They write, “if we want to determine whether a mechanism in animals is sufficiently similar to the mechanism in humans to justify extrapolation, we must know how the relevant mechanism in humans operates. But if we already know about the mechanism in humans then the initial animal study is likely to have been redundant.” 9

Inherent species differences mean that other animals cannot serve as analogs for understanding the biological effects of drugs and chemicals on humans.

They also discuss the concerning trend among those involved in experiments on animals to minimize the issue of species differences and the effects on external validity, a problem acknowledged by a number of researchers.1011 Pound and Ritskes-Hoitinga go on to state that it is unsurprising that the issue of species differences is downplayed, as not doing so would force experimenters to confront the “possibility that the preclinical animal research paradigm no longer has a great deal to offer.”12 There is growing scientific consensus that far more is to be gained from non-animal research methods that are better suited to solving human biomedical research questions.

The difficulties in applying data derived from one species to another are compounded by the confinement and unnatural conditions of laboratory life—including housing,1314 diet,151617 light cycles,18192021 noise,222324 and the temperature and humidity at which animal facilities are kept2526272829—which thwart animals’ ability to engage in natural behavior.303132 This deprivation contributes to their stress and alters their physiology and neurobiology, causing them to exhibit various morbidities and psychopathologies unrelated to the experiments at hand.333435363738394041 Importantly, the fact that animals in laboratories have altered physiology and neurobiology means that they would not even be good “models” for their counterparts in nature. A mouse in a laboratory will not respond to a drug in the same way a mouse in a field would. One then has to ask: How does this biologically distinct mouse reliably represent the biology of humans?

Lack of Clinical Success

X-ray of a human skull with an orange circle indicating the location of the brain.iStock.com/mr.suphachai praserdumrongchai

The failure of animal studies in basic and applied research is perhaps most evident in the stark litany of seemingly promising treatments that have not worked in humans. For example, stroke experiments on animals have been an outright failure: 30 years of animal testing have failed to result in any successful translation of drugs that protect against damage or repair the brain after stroke.42

Decades of experiments on mice and other animals have generated no new treatment or diagnostic technology for humans with sepsis.43 Oncology drugs, which undergo extensive animal testing, have a success rate of only 3.4%.44 This theme pervades many human disease areas.45 There is an abundance of literature documenting the failure of various animal models of neurodegenerative diseases, neuropsychiatric conditions, women’s health issues, and more. See the appendices for a comprehensive look at disease areas.

Misplaced Resources

Despite the growing evidence that experiments on animals are wasteful and can impede medical progress, a significant proportion of NIH research funding goes toward them.46 Federal funds available for biomedical research are a finite resource. In the fiscal year 2024, only 19% of research project grant applications submitted to NIH were awarded funding.47 Each decision to approve an application carries with it a refusal to fund other projects, leaving a large opportunity cost in terms of human-relevant research that has the potential to help patients.

“[I]f research conducted on animals continues to be unable to reasonably predict what can be expected in humans, the public’s continuing endorsement and funding of preclinical animal research seems misplaced.” 48

Funding for biomedical research is allocated into three categories: basic, translational, and clinical research. NIH defines basic research as that which is “directed toward greater knowledge or understanding of the fundamental aspects of phenomena and of observable facts without specific applications towards processes or products in mind.”49 A great deal of basic research involves experiments on animals.

NIH perceives basic research, including that which uses animals, as important because its intent is “ultimately aimed at protecting and improving human health”.50 In other words, the results of animal use in basic research should point the way toward translational and clinical research that should, in turn, benefit humans. However, the evidence shows that this is not the case. To assess whether the promises of basic biomedical research were being fulfilled, researchers identified 101 articles published in the most prestigious medical journals in which the authors explicitly stated that their research would lead to a new application with real potential for a clinical breakthrough. A majority of the articles analyzed (63%) described experiments on animals. The researchers’ investigation into the conversion of basic research into clinical applications found that fewer than 10% of these self-proclaimed “highly promising” discoveries entered routine clinical use within 20 years.51

Basic research is a critical step for generating foundational scientific knowledge, but when that knowledge produces no actionable benefits for humans—or the species harmed and killed for it—society’s continued investment in and support of it must be reassessed.

In the current system, bringing a new drug to market may cost more than $1 billion and takes an average of 14 years. 52

The Dangers of Misleading Results

Two medical professionals are pushing a patient in a hospital bed down a hospital hallway.iStock.com/AscentXMedia

Many novel drugs don’t simply fail, representing a huge loss in time and investment—they harm patients. In 2016, a drug intended to help with mood, anxiety, and motor problems related to neurodegenerative disease was administered orally to volunteers as part of a Phase I clinical trial. Six men, ages 28 to 49, experienced such adverse reactions that they had to be hospitalized. One participant was pronounced brain-dead and later died. A report on this incident revealed that the toxicity of the drug in humans “was not observed in animals despite administration of very high doses.”53

TGN1412 is another tragic example. “After [the] very first infusion of a dose 500 times smaller than that found safe in animal studies, all six human volunteers faced life-threatening conditions involving multiorgan failure for which they were moved to [the] intensive care unit.”54 Five of the six participants were hospitalized for three months after the initial dose, while the other was comatose. Even six months later, participants suffered from headaches and memory loss. One had to have toes and fingers amputated as a result of gangrene.55

The opposite is also true: Therapies that have not worked well in animals have sat useless on the shelf while patients have gone without lifesaving treatment. For example, aspirin is widely used in human medicine, but it may have never been approved if it were first tested in animals, for whom it has a wide range of toxic effects that are not observed in humans.56

Toxicologist Thomas Hartung noted a number of similar examples in his 2024 article, “The (Misleading) Role of Animal Models in Drug Development,” including the following: Severe liver injury and multiple deaths forced the termination of a hepatitis B drug trial despite earlier encouraging animal data. Differential species sensitivity to drugs like acetaminophen further highlights the pitfalls of reliance on animal models. Gene therapy vectors that have been safe in animal tests have caused liver failure and brain swelling in children. HIV vaccines, stroke treatments, inflammatory disease agents, and Alzheimer’s therapies have all elicited enthusiasm in animal models yet utterly failed in human trials.57

References
  1. National Center for Advancing Translational Sciences. New Therapeutic Uses. April 19, 2024. Accessed September 25, 2024. https://ncats.nih.gov/research/research-activities/ntu ↩︎
  2. Pound P, Bracken MB. Is animal research sufficiently evidence based to be a cornerstone of biomedical research? BMJ. 2014;348:g3387. doi:10.1136/bmj.g3387 ↩︎
  3. Sena ES, van der Worp HB, Bath PMW, Howells DW, Macleod MR. Publication bias in reports of animal stroke studies leads to major overstatement of efficacy. PLoS Biol. 2010;8(3):e1000344. doi:10.1371/journal.pbio.1000344 ↩︎
  4. Hirst JA, Howick J, Aronson JK, et al. The need for randomization in animal trials: an overview of systematic reviews. PLoS One. 2014;9(6):e98856. doi:10.1371/journal.pone.0098856 ↩︎
  5. Freedman LP, Cockburn IM, Simcoe TS. The economics of reproducibility in preclinical research. PLoS Biol. 2015;13(6):e1002165. doi:10.1371/journal.pbio.1002165 ↩︎
  6. Collins FS, Tabak LA. Policy: NIH plans to enhance reproducibility. Nature. 2014;505(7485):612-613. doi:10.1038/505612a  ↩︎
  7. Pound P, Ritskes-Hoitinga M. Is it possible to overcome issues of external validity in preclinical animal research? Why most animal models are bound to fail. J Transl Med. 2018;16(1):304. doi:10.1186/s12967-018-1678-1 ↩︎
  8. Wall RJ, Shani M. Are animal models as good as we think? Theriogenology. 2008;69(1):2-9. doi:10.1016/j.theriogenology.2007.09.030 ↩︎
  9. Pound, Ritskes-Hoitinga., J Transl Med. 2018. ↩︎
  10. van der Worp HB, Howells DW, Sena ES, et al. Can animal models of disease reliably inform human studies? PLoS Med. 2010;7(3):e1000245. doi:10.1371/journal.pmed.1000245 ↩︎
  11. Bailoo JD, Reichlin TS, Würbel H. Refinement of experimental design and conduct in laboratory animal research. ILAR J. 2014;55(3):383-391. doi:10.1093/ilar/ilu037 ↩︎
  12. Pound, Ritskes-Hoitinga., J Transl Med. 2018. ↩︎
  13. Lahvis GP. Unbridle biomedical research from the laboratory cage. eLife. 2017;6. doi:10.7554/ELIFE.27438 ↩︎
  14. Cait J, Cait A, Scott RW, Winder CB, Mason GJ. Conventional laboratory housing increases morbidity and mortality in research rodents: results of a meta-analysis. BMC Biology. 2022;20(1):1-22. doi:10.1186/S12915-021-01184-0/TABLES/2 ↩︎
  15. Martin B, Ji S, Maudsley S, Mattson MP. “Control” laboratory rodents are metabolically morbid: why it matters. PNAS. 2010;107(14):6127-6133. doi:10.1073/pnas.0912955107 ↩︎
  16. Kurtz DM, Feeney WP. The influence of feed and drinking water on terrestrial animal research and study replicability. ILAR J. 2019;60(2):175-196. doi:10.1093/ilar/ilaa012 ↩︎
  17. Mesnage R, Defarge N, Rocque LM, Spiroux de Vendômois J, Séralini GE. Laboratory rodent diets contain toxic levels of environmental contaminants: implications for regulatory tests. PLoS One. 2015;10(7):e0128429. doi:10.1371/journal.pone.0128429 ↩︎
  18. Barabas AJ, Darbyshire AK, Schlegel SL, Gaskill BN. Evaluation of ambient sound, vibration, and light in rodent housing rooms. JAALAS. 2022;61(6):660-671. doi:10.30802/AALAS-JAALAS-22-000040 ↩︎
  19. Bell BA, Kaul C, Bonilha VL, Rayborn ME, Shadrach K, Hollyfield JG. The BALB/c mouse: effect of standard vivarium lighting on retinal pathology during aging. Exp Eye Res. 2015;135:192-205. doi:10.1016/j.exer.2015.04.009 ↩︎
  20. Dauchy RT, Blask DE. Vivarium lighting as an important extrinsic factor influencing animal-based research. JAALAS. 2023;62(1):3-25. doi:10.30802/AALAS-JAALAS-23-000003 ↩︎
  21. Greenman DL, Bryant P, Kodell RL, Sheldon W. Influence of cage shelf level on retinal atrophy in mice. Lab Anim Sci. 1982;32(4):353-356 ↩︎
  22. Barabas, Darbyshire, Schlegel, Gaskill., JAALAS. 2022. ↩︎
  23. Povroznik JM, Faith RE, Kessler MJ, et al. Locomotor effects of a low-frequency fire alarm on C57BL/6 male mice: a preliminary study. Lab Anim. 2017;51(6):647-651. doi:10.1177/0023677217711966 ↩︎
  24. Corbani TL, Martin JE, Healy SD. The impact of acute loud noise on the behavior of laboratory birds. Front Vet Sci. 2021;7. doi:10.3389/fvets.2020.607632 ↩︎
  25. James CM, Olejniczak SH, Repasky EA. How murine models of human disease and immunity are influenced by housing temperature and mild thermal stress. Temperature. 2023;10(2):166-178. doi:10.1080/23328940.2022.2093561 ↩︎
  26. Han A, Hudson-Paz C, Robinson BG, et al. Temperature-dependent differences in mouse gut motility are mediated by stress. Lab Anim. 2024;53(6):148-159. doi:10.1038/s41684-024-01376-5 ↩︎
  27. Gaskill BN, Rohr SA, Pajor EA, Lucas JR, Garner JP. Some like it hot: mouse temperature preferences in laboratory housing. Appl Anim Behav Sci. 2009;116(2):279-285. doi:10.1016/j.applanim.2008.10.002 ↩︎
  28. Kasza I, Cuncannan C, Michaud J, et al. “Humanizing” mouse environments: humidity, diurnal cycles and thermoneutrality. Biochimie. 2023;210:82-98. doi:10.1016/j.biochi.2022.10.015 ↩︎
  29. Hylander BL, Eng JWL, Repasky EA. The impact of housing temperature-induced chronic stress on preclinical mouse tumor models and therapeutic responses: an important role for the nervous system. Adv Exp Med Biol. 2017;1036:173-189. doi:10.1007/978-3-319-67577-0_12 ↩︎
  30. Gozalo AS, Elkins WR. A review of the effects of some extrinsic factors on mice used in research. Comp Med. 2023;73(6):413-431. doi:10.30802/AALAS-CM-23-000028 ↩︎
  31. Mieske P, Hobbiesiefken U, Fischer-Tenhagen C, et al. Bored at home?—a systematic review on the effect of environmental enrichment on the welfare of laboratory rats and mice. Front Vet Sci. 2022;9:899219. doi:10.3389/fvets.2022.899219 ↩︎
  32. Ratuski AS, Améndola L, Makowska IJ, Weary DM. Effects of temporary access to environmental enrichment on measures of laboratory mouse welfare. Sci Rep. 2024;14(1):15143. doi:10.1038/s41598-024-65480-9 ↩︎
  33. Martin, Ji, Maudsley, Mattson., PNAS. 2010. ↩︎
  34. Bell, Kaul, Bonilha, Rayborn, Shadrach, Hollyfield., Exp Eye Res. 2015. ↩︎
  35. Mieske, Hobbiesiefken, Fischer-Tenhagen, et al., 2022. ↩︎
  36. Gaskill BN, Garner JP. Stressed out: providing laboratory animals with behavioral control to reduce the physiological effects of stress. Lab Anim. 2017;46(4):142-145. doi:10.1038/laban.1218 ↩︎
  37. Hylander BL, Repasky EA, Sexton S. Using mice to model human disease: understanding the roles of baseline housing-induced and experimentally imposed stresses in animal welfare and experimental reproducibility. Animals. 2022;12(3):371. doi:10.3390/ani12030371 ↩︎
  38. Müller K, Lengheimer T, Kral-Pointner JB, et al. Exposure to soiled bedding reduces abnormal repetitive behaviors in mice. Front Behav Neurosci. 2022;16. doi:10.3389/fnbeh.2022.1062864 ↩︎
  39. Garner JP. Stereotypies and other abnormal repetitive behaviors: potential impact on validity, reliability, and replicability of scientific outcomes. ILAR J. 2005;46(2):106-117. doi:10.1093/ilar.46.2.106with deficits in progenitor proliferation and cortical neuron content. Nat Commun. 2022;13(1):2387. doi:10.1038/s41467-022-29942-w ↩︎
  40. Balcombe JP. Laboratory environments and rodents’ behavioural needs: a review. Lab Anim. 2006;40(3):217-235. doi:10.1258/002367706777611488 ↩︎
  41. Cannon TH, Heistermann M, Hankison SJ, Hockings KJ, McLennan MR. Tailored enrichment strategies and stereotypic behavior in captive individually housed macaques (Macaca spp.). J Appl Anim Welf Sci. 2016;19(2):171-182. doi:10.1080/10888705.2015.1126786. Zhou Z, Lu J, Liu WW, et al. Advances in stroke pharmacology. Pharmacol Ther. 2018;191:23-42. doi:10.1016/j.pharmthera.2018.05.012 ↩︎
  42. Zhou Z, Lu J, Liu WW, et al. Advances in stroke pharmacology. Pharmacol Ther. 2018;191:23-42. doi:10.1016/j.pharmthera.2018.05.012 ↩︎
  43. National Institute of General Medical Sciences. NAGMSC Working Group on Sepsis. NIH; 2019:31. Accessed March 22, 2026. ↩︎
  44. Wong CH, Siah KW, Lo AW. Estimation of clinical trial success rates and related parameters. Biostatistics. 2019;20(2):273-286. doi:10.1093/biostatistics/kxx069  ↩︎
  45. Marshall LJ, Bailey J, Cassotta M, Herrmann K, Pistollato F. Poor translatability of biomedical research using animals—a narrative review. Altern Lab Anim. 2023;51(2):102-135. doi:10.1177/02611929231157756 ↩︎
  46. United States Government Accountability Office. National Institutes of Health. Assessing efforts to improve animal research could lead to greater human health benefits. December 2024, Revised 2026. GAO-25-107140. Accessed February 6, 2026. https://www.gao.gov/assets/gao-25-107140.pdf ↩︎
  47. National Institutes of Health. NIH Data Book. Updated January 2025. Accessed February, 2026. https://report.nih.gov/nihdatabook/category/10 ↩︎
  48. Sena, van der Worp, Bath, Howells, Macleod., PLoS Biol. 2010. ↩︎
  49. NIH. Glossary. Grants & Funding. July 10, 2024. Accessed March 22, 2026. https://grants.nih.gov/grants/glossary.htm ↩︎
  50. NIH. NIH-Wide Strategic Plan for Fiscal Years 2021-2025. July 202. Accessed March 22, 2026. https://www.nih.gov/sites/default/files/2025-01/strategic-plan-fy2021-2025.pdf ↩︎
  51. Contopoulos-Ioannidis DG, Ntzani EE, Ioannidis JPA. Translation of highly promising basic science research into clinical applications. Am J Med. 2003;114(6):477-484. doi:10.1016/S0002-9343(03)00013-5 ↩︎
  52. National Center for Advancing Translational Sciences. New Therapeutic Uses. April 19, 2024. Accessed September 25, 2024. https://ncats.nih.gov/research/research-activities/ntu ↩︎
  53. Temporary Specialist Scientific Committee. “FAAH (Fatty Acid Amide Hydrolase)”, on the Causes of the Accident During a Phase 1 Clinical Trial in Rennes in January 2016.; 2016. Accessed November 10, 2024. https://archive.ansm.sante.fr/var/ansm_site/storage/original/application/744c7c6daf96b141bc9509e2f85c227e.pdf  ↩︎
  54. Attarwala H. TGN1412: From discovery to disaster. J Young Pharm. 2010;2(3):332-336. doi:10.4103/0975-1483.66810 ↩︎
  55. Ferguson P. The TGN1412 drug disaster. SciTech. 2009;5(4):12-13. doi:10.4103/0975-1483.66810 ↩︎
  56. Hartung T. Per aspirin ad astra…. Altern Lab Anim. 2009;37(2_suppl):45-47. doi:10.1177/026119290903702S10 ↩︎
  57. Hartung T. The (misleading) role of animal models in drug development. Front Drug Discov. 2024;4. doi:10.3389/fddsv.2024.1355044 ↩︎