Animal Testing 101

The Physicians Committee

Animal Testing 101

The United States government does not require animal tests to assess the toxicity of most cosmetics and personal care products, but many companies still perform these tests. Often, companies sell their products in countries that have mandatory animal testing programs such as China and Brazil. Using animals is ineffective, expensive, and raises concerns for animal welfare. However, nonanimal methods exist, and are continually being developed, to address these problems.


Animal and Human Physiologies Often Don’t Overlap

There are physiological, anatomical, and genetic differences between humans and animals that make it difficult to translate data obtained from animal studies to human a response. For example, the metabolic pathway that a compound takes through a rat’s liver could be vastly different than how it makes its way through the human body. Tests could show that the compound has no effect in rats, but have the opposite response in humans. This can be extremely dangerous. In a study by Carnegie Mellon University, the lifetime rodent bioassay, which has been described as the "gold standard" for carcinogenicity testing, correctly identified only 50 percent of known human carcinogens.1 Another study found that in 150 human clinical trials on inflammatory diseases, 100 percent of drugs that were developed using mice failed.2 Clearly, the differences between humans and animals cannot be ignored.

Traditional Tests Are Poorly Structured

Traditional toxicity tests are conducted in a way that does not reflect the real-life environments in which humans use products. First, toxicity tests expose animals to doses of chemicals that are 100 to 1000 times higher than a human would be exposed to. This can overload the animal’s system, causing toxicities that may or may not be reflective of the true toxicity a human may experience—and often lead companies to perform more testing to prove the effects are irrelevant. The animals on which tests are performed lack genetic diversity. Generally, toxicity tests are performed on healthy, genetically inbred animals. Humans, of course, are often neither of those. For example, a product may be more dangerous to someone with asthma or a smoker, than a healthy person. Similarly, only single compounds are tested, but most products are mixtures of chemicals. These mixtures can interact with the body differently than they would separately, again making the results of animal tests of limited value.

Using Animals Is Resource Intensive

While the United States doesn’t have accurate statistics on the number of animals used for cosmetics testing annually, most scientists acknowledge the number is in the millions. Using animals is expensive: They require food, medical care, specialized facilities for housing, and more. Animal tests are also slow: It can take up to 3 years to prepare, conduct, and analyze a cancer study using rats for just one chemical. Then scientists often have to determine whether that data is relevant to humans using additional tests.

Using Animals Raises Welfare and Ethical Concerns

Even routine laboratory procedures, such as handling, blood collection, and separation from their friends and family members can cause animals marked psychological and physiological stress.3 The side effects of tests can range from moderate to severe pain and can include diarrhea, vomiting, seizures and convulsions, difficulty breathing, tumors, birth defects, and skin lesions, among others. Pain relief is never provided. After testing is done animals are killed, either by asphyxiation or by breaking the neck, and their organs may be harvested for further testing. That these tests are done for the sake of a new skin cream or fragrance defies common decency.


Alternatives to Animal Testing Are Effective and Humane

There are several available alternatives to animal testing. In vitro (cell culture) models are based on human cells and address the problem of not being able to translate animal data to human data. A study of EpiDerm, an alternative to the Draize (rabbit) skin irritation test, found that in vitro methods showed 76 percent accordance with human data compared to 56 percent with rabbit data.4

Many nonanimal tests have been validated by the Organisation for Economic Co-operation and Development. For example, the skin irritation and corrosion tests has multiple in vitro models: 431 In Vitro Skin Corrosion: Human Skin Model Test, 439 In Vitro Skin Irritation: Reconstructed Human Epidermis Test Method, and 435 In Vitro Membrane Barrier Test Method for Skin Corrosion.

Nonanimal methods to assess inhalation toxicity (or how toxic something is when you inhale it) are also available such as EpiAirway and Epithelix. These sophisticated in vitro models use human cells to assess the potential toxicity of a chemical.

Other types of cellular methods that replicate human organ systems using living human cells on polymer chips are also available. One of the benefits of this method is that both healthy and diseased cells can be tested. For example, a pair of healthy ‘lungs’ can undergo inhalation studies, as well as an asthmatic pair. Mixtures can also be tested, which is important since humans are likely exposed to chemicals in mixtures rather than separately. This means that data will be more inclusive. Scientists have replicated several organ systems, including lungs, kidney, bone marrow, and the gut. A short video from the Wyss Institute at Harvard explains the science behind organs-on-a-chip.

In vitro and in silico methods, as well as others, are outlined in the National Research Council’s report Toxicity Testing in the 21st Century: A Vision and a Strategy, you can read more about the vision here.

Nonanimal testing methods can be more cost effective and faster. Tests that would normally take months or years can take days or even hours, which will save companies money in the long run. And because alternative methods eliminate the use of animals they also eliminate the ethical concerns. It’s a win-win-win; the consumer, the companies, and the animals all benefit from this modern approach to toxicity testing.


1. Ennever, FK, Lave, LB. Implications of the lack of accuracy of the lifetime rodent bioassay for predicting human carcinogenicity. Regul Toxicol Pharmacol. 2003;38(1):52-57.
2. Seok J, et al. Genomic responses in mouse models poorly mimic human inflammatory diseases. PNAS. 2013;110(9):3507-3512.
3. Balcombe JP, Barnard ND, Sandusky C. Laboratory routines cause animal stress. Contemporary Topics. 2004;13:57-69.
4. Jirova D, et al. Comparison to human skin irritation patch test data with in vitro skin irritation assay and animal data. Contact Dermatitis. 2010;62(2):109-116.