Rabbits have historically been common research subjects due to specific anatomical and physiological traits that made them useful for certain tests. While calculating the total number of rabbits used in labs worldwide is difficult due to data collection limitations, available data from major jurisdictions offer a partial view of the scale of their use. This usage is now rapidly changing as non-animal testing methods gain scientific acceptance and regulatory approval.
Statistical Reporting and Data Gaps
Obtaining a single, accurate global count of rabbits that die during or after testing is impossible due to inconsistent international reporting standards. In the United States, data collected by the Department of Agriculture (USDA) under the Animal Welfare Act (AWA) focuses on the number of animals used. For example, the reported number of rabbits held or used in US research in 2024 was approximately 111,080.
The USDA reporting system does not explicitly track the number of animals that die during a procedure versus those euthanized afterward to collect tissue or prevent suffering. Instead, the focus is on the severity of the procedure, noting how many animals were subjected to unrelieved pain. A significant data gap exists because the AWA excludes rats, mice, and birds bred for research, which constitute the vast majority of animals used in labs.
The European Union employs a more detailed system, requiring member states to report animal use categorized by the severity of the procedure: mild, moderate, severe, or non-recovery. Non-recovery procedures are those conducted entirely under general anesthesia, where the animal is intentionally killed before regaining consciousness. This classification scheme offers insight into the level of suffering but does not provide a simple count of deaths, as many animals in the severe category are euthanized at the experiment’s conclusion. The absence of a centralized global registry means any worldwide figure for rabbit deaths remains an estimate.
Primary Uses of Rabbits in Testing
Rabbits became the preferred species for certain procedures due to unique biological characteristics that made them suitable models for specific toxicological endpoints. Historically, the most recognized procedure involving rabbits is the Draize Eye Irritancy Test, developed in the 1940s to assess a substance’s potential to cause eye damage. In this test, a substance is applied to the eye of a restrained albino rabbit, typically the New Zealand White strain.
Rabbits were selected for the Draize test largely due to their anatomy. They possess a less efficient tear-cleansing mechanism and a third eyelid, which prevents the test substance from being quickly washed away. The test involves observation for up to 14 days for signs of damage, such as corneal opacity, ulceration, or hemorrhage.
Another common application is Pyrogen Testing, which detects fever-inducing contaminants in injectable drugs or medical devices. Rabbits are used because their thermoregulatory response to pyrogens, specifically endotoxins, is similar to that of humans. The test involves injecting the substance intravenously and monitoring the rabbit’s rectal temperature for several hours; a temperature rise above a specific threshold indicates a positive result.
Rabbits are also a standard non-rodent model in Development and Reproductive Toxicity (DART) studies, often required for drug safety assessments. Their reproductive physiology, including the structure of their extraembryonic membranes and placenta, shares specific features with human gestation that are distinct from rodent models. This makes them a required second species, alongside a rodent, for evaluating a drug’s effect on fetal development and teratogenicity.
Global Regulatory Status of Rabbit Testing
Legal frameworks governing rabbit testing vary significantly across the globe, leading to a patchwork of regulations that influence their use. The European Union (EU) and the United Kingdom have implemented comprehensive bans on cosmetic testing. These bans prohibit the testing of finished cosmetic products and their ingredients on animals, including rabbits, and also ban the sale of products tested elsewhere. This regulatory environment has been a major driver in replacing the Draize test with non-animal methods.
In the United States, the regulatory landscape is more complex, with no federal ban on animal testing for cosmetics. While the use of rabbits for cosmetics has sharply declined, they are still required for certain testing mandated by agencies like the Environmental Protection Agency (EPA) for industrial chemicals and pesticides. However, the EPA encourages the use of non-animal methods to replace the Draize test for eye irritation. The Food and Drug Administration (FDA) requires animal studies for medical devices only when a suitable alternative is unavailable.
China, historically a major center for animal testing, has implemented significant changes, though not a complete ban. Since 2021, general cosmetics, such as shampoos and lipsticks, can be imported without mandatory animal testing if a safety assessment is provided and certain conditions are met. However, “special” cosmetics, including hair dyes and sunscreens, may still require animal testing for regulatory approval. Post-market testing, where a product is pulled from the shelf for re-testing, no longer automatically requires the use of animals.
Non-Animal Testing Methods
The scientific community has developed advanced non-animal testing methods, known as New Approach Methodologies (NAMs), that aim to replace, reduce, and refine animal use. For eye irritation testing, the Bovine Corneal Opacity and Permeability (BCOP) assay uses corneas obtained from cattle slaughtered for commercial purposes, avoiding the use of laboratory rabbits. The BCOP assay measures both the opacity (cloudiness) and permeability (leakage of dye) of the excised cornea after exposure to a test substance.
Other in vitro models use human-relevant cell structures, such as the EpiOcular Eye Irritation Test (EIT). The EIT employs a three-dimensional reconstructed human cornea-like epithelial model and measures the cytotoxicity of a substance. Cell viability below a certain threshold indicates an irritant classification.
For pyrogen testing, the Monocyte Activation Test (MAT) has replaced the rabbit test in many jurisdictions. The MAT uses human blood cells, specifically monocytes, and measures the release of pro-inflammatory signaling molecules called cytokines in response to a test substance. This process mimics the human immune system’s fever response.
A complementary approach is in silico toxicology, which uses computational models to predict a substance’s toxic potential based on its chemical structure. These methods include Quantitative Structure-Activity Relationship (QSAR) models, which employ machine learning and algorithms. QSAR models forecast a chemical’s properties by comparing it to the toxicological data of structurally similar compounds. These computational tools allow for the rapid, cost-effective screening of thousands of compounds, reducing the need for animal use early in product development.