Cosmetic Safety: Public Perception and Scientific Reality

Editor’s note: This article is the basis for the Dec. 12, 2013, Frontiers of Science Award Lecture sponsored by Cosmetics & Toiletries and presented at the Society of Cosmetic Chemists’ Annual Meeting in New York. Here, Joe Schwarcz, director of the McGill University Office for Science and Society, discusses scrutiny and scientific challenges faced by the cosmetics industry.

Cosmetics are under attack, although this is not the first time. Back in 1770, the English Parliament passed an act1 declaring that marriages could be pronounced null and void if the man had been “led into matrimony by false pretenses through the use of scents, paints, cosmetic washes, artificial teeth, false hair, bolstered hips, high heels or iron stays.” It is not quite clear what iron stays were, although likely they were used to steady features of the female anatomy beginning to droop. It remains a mystery whether anyone sought divorce after being disappointed that the “goods” were not as advertised, but it is safe to assume that cosmetic manufacturers were not happy with the situation. They also were probably not thrilled when Queen Victoria publicly declared makeup to be improper, vulgar and acceptable only for use by actors.2

As is well-known to the industry, cosmetics are being assaulted again today, but for a different reason: harboring potentially toxic ingredients. Regulatory authorities are being challenged for not doing enough to protect public health. Finger-pointers range from faceless composers of inane e-mails to various activist organizations that bolster their crusade for “safer cosmetics” with references to scientific literature. Some accusations, such as the assertion that certain mascaras or lipsticks contain toxic amounts of lead, are unrealistic because toxicity is a function of exposure, and the exposure in lipstick is well below toxic levels.3 However, allegations that some cosmetics may contain hidden carcinogens or hormone-disrupting substances merit scrutiny. Hormones are active at extremely low concentrations, and some “endocrine disruptors” can be found in blood and urine samples at concentrations comparable to naturally occurring hormones.4

The cosmetics industry is huge; the U.S. market alone nets $55 billion a year.5 Unlike pharmaceuticals in the United States, no pre-marketing testing for the safety of cosmetics is required—a fact often vociferously pointed out by cosmetic critics who infer that such a lack in regulations puts consumer health at risk. Of course, governments do not exactly maintain a “hands-off” policy. Canada has a “hot list” of some 500 chemicals that cannot be used in cosmetics,6 and before any item is marketed, its list of ingredients must be submitted to Health Canada for approval. Furthermore, Health Canada has the power to order the removal of products from stores if it decides there is any risk involved. Regulations are less stringent in the United States, where the U.S. Food and Drug Administration (FDA) must prove that a product is dangerous before removing it from store shelves.7

One reason that governments have not taken a heavy-handed approach and required the pre-market testing of cosmetics is that the cosmetics industry has an effective self-regulating program. The U.S.-based Cosmetic Ingredient Review panel is an industry-sponsored group of experts that includes representatives from the FDA as well as consumer organizations. It is charged with the responsibility of compiling and scrutinizing research that is relevant to cosmetic ingredients. The panel’s in-depth reports are used by industry to make decisions about product formulation.

Although some may claim, “the fox is in charge of the hen house,” no industry wants to harm its customers. At the very least, the inclusion of ingredients that turn out to be harmful is bad for business, especially in the United States, with its litigious society and plethora of “personal injury” lawyers seeking prospective clients. Cosmetic companies know the best way to make money is by selling products that are safe and effective. Admittedly, there is a fly in the ointment here, in that safety and efficacy are open to interpretation, as is the degree of acceptable risk associated with any consumer product. It is also true, however, that no matter how much care is taken, there is always some risk. Cosmetics can be responsible for some acute adverse effects but these are usually readily recognized; it is the hypothetical link of some chemicals in cosmetics to cancer, or to endocrine disruption, that is difficult to evaluate and this provides grist for the alarmist mill.

Hormonal Activity

Undoubtedly, endocrine disruptors are a legitimate concern. By definition, these are chemicals that can alter the way that hormones function. Hormones are crucial to life, acting as chemical messengers that govern a myriad of biological processes ranging from food metabolism to cell division. Obviously, then, any substance that somehow disrupts hormonal activity has the potential to affect health.

Modern laboratory techniques can reveal whether a given substance is capable of mimicking hormones by studying the effects on certain yeasts or cancer cells. However, finding that a chemical has hormonal activity is easier than interpreting what that information means. Hundreds of compounds, both natural and synthetic, have been found to have such activity. Consumers live in a world full of potential endocrine disruptors, starting with one’s own urine, which contains a variety of natural hormones that are not removed by water treatment and can end up in drinking water. Add to this remnants of birth control pills and drugs used to alleviate menopausal symptoms. In addition, milk, soybeans, flaxseed and numerous vegetables contain natural estrogenic compounds. Then, of course, there are the much talked-about synthetics like phthalates, bisphenol A and oxybenzone.

Most studies that have attempted to find a link between environmental hormone mimics and human disease have relied on data derived from a single urine specimen. In contrast, recent research examining samples taken over a three-year period from the same subjects show great variability—indicating the unreliability of spot urine samples.8 A further complication is that studies generally look for individual hormone-like compounds, whereas there are numerous such compounds cruising through the human body at any given time with mixed agonist-antagonist activities. What happens when weak estrogens interact? Whether they produce a synergistic effect or cancel out each other’s effects, no one really knows. Neither can it be assumed that all hormone-mimicking activity is negative; after all, estrogens and androgens have a variety of medical uses.

How effectively the compound binds to estrogen receptors is also critical. Exposure to estrogenic substances around the age of puberty, for example, protects against future breast cancer but hormone replacement therapy at menopause slightly increases the risk—yet such therapy also decreases the risk of osteoporosis, a major health issue. Basically, coming to conclusions about environmental hormone mimics is difficult, especially when the possibility of hormesis is thrown into the mix. This phenomenon refers to substances that may be more dangerous at a smaller dose than at a higher one. Hormesis seems to fly in the face of the basic tenet of toxicology, namely that effects increase with dose, but there is accumulating evidence that dose-response relationships are not always linear.9 This is a complex situation. Nevertheless, some claim that just removing this or that compound from the environment will remove a significant health hazard. As Mencken once said “For every complex problem, there is an answer that is clear, simple and wrong.”10

If one studies any chemical in depth, they will find some effect and then, placing integrity aside, can portray that effect in whatever way suits their purpose. Take resveratrol, for example, the current darling of the supplement industry. This compound is found in the skin of red grapes and is abundantly present in red wine. It has been touted as the substance responsible for red wine’s supposed ability to reduce the risk of both cardiovascular disease and cancer. Well, guess what—resveratrol is an endocrine disruptor.11 This comes as no great surprise to any chemist, who may notice that its molecular structure (see Figure 1) bears a striking resemblance to diethylstilbesterol (DES), the notorious synthetic estrogen that was introduced in the 1940s to prevent miscarriages. DES was later linked with an increased risk of a rare form of vaginal cancer in the female children of its users.12

In cultured human cells, resveratrol has been shown to activate the same genes as natural estradiol, and to a greater extent.13 It can also support the proliferation of certain breast cancer cells that require estrogen for growth. By making selective references to the scientific literature, it would not be difficult to portray resveratrol as a devilish substance that tinkers with hormones. Of course, there is absolutely no evidence that resveratrol is dangerous to humans; in fact, evidence suggests it may be beneficial, and the benefits are possibly due to its hormone-like properties.

Irritation and Allergies

If there is to be a safety issue with cosmetics, it is in the area of skin irritation and allergies. Sodium hydroxide in some hair straighteners, or methacrylic acid in artificial nail products can be potent skin irritants if improperly used. Also, a host of chemicals ranging from fragrance components and preservatives, to emulsifiers and colorants can cause allergic dermatitis. While the vast majority of consumers never encounter such problems, the significant number of people who have has resulted in the promotion of “hypoallergenic” products that avoid the most obvious sensitizers such as lanolin, formaldehyde-releasing preservatives and fragrance components like cinnamic alcohol, geraniol, limonene or linalool.


Without a doubt, the scariest allegation is that cosmetics may contain carcinogens. Indeed, some do. It is important to realize, though, that the definition of a carcinogen is a substance that is capable of causing cancer in some animal at some dose. It does not mean that it is known to cause cancer in humans. Dioxane, for example, is an impurity found in some cosmetics and is listed as a carcinogen because it triggers the disease when fed to rodents. Amounts in cosmetics, however, are vanishingly small, known to be present only because of the availability of extremely sensitive detection techniques. Furthermore, application of a chemical to the skin is not the same as ingestion. Still, elimination of any carcinogen is desirable, and methods to remove the dioxane impurity have been developed by major cosmetic manufacturers.

Continuous Process

Cosmetic formulation is a continuously evolving process that must keep up with a massive amount of research. Unfortunately, as far as the public is concerned, repetition of a hypothetical risk turns it into reality and when it comes to marketing, the customer is right even when science indicates they are likely wrong. There is no simple solution to offer here because the fact is: science is not black or white, it is various shades of gray.

Of course, when a true risk emerges, the industry must move to address it. After all, consumer confidence is what puts money in the bank. Recent research that links parabens to aging in skin cells, or links some moisturizing creams with the promotion of skin cancer in mice after UV exposure, merits further investigation; as do the possible hormonal effect of phthalates, chemicals found in some fragrances, and nail polishes. Whether such chemicals should be removed from the market comes down to a proper risk-benefit analysis. Bacterial growth in a cream due to the absence of an effective preservative such as parabens is not an attractive proposition.

While there is no compelling evidence to support the alleged risk posed by parabens or other cosmetic ingredients, there is one major established cosmetic risk that is avoidable: applying mascara in a moving vehicle. This causes loads of eye injuries. Maybe a law should be passed to prohibit it.


  1. R Corson, Fashions in Makeup: From Ancient to Modern Times, London, PeterOwen (1972) p 245
  2. L Woodhead, War Paint: Madame Helena Rubinstein and Miss Elizabeth Arden: Their Lives, ISBN 0-471-48778-3 (2004) p 1905
  3. NM Hepp et al, Determination of total lead in lipstick: Development and validation of a microwave-assisted digestion, inductively coupled plasma–mass spectrometric method, J Cosmet Sci 405–414 (Jul/Aug 2009)
  4. LN Vandenberg et al, Hormones and endocrine-disrupting chemicals: Low-dose effects and nonmonotonic dose responses, Endocr Rev 33(3) 378-455 (Jun 1, 2012)
  5. (Accessed Sep 30, 2013)
  6. (Accessed Sep 30, 2013)
  7. (Accessed Sep 30, 2013)
  8. (Accessed Sep 30, 2013)
  9. (Accessed Sep 30, 2013)
  10. (Accessed Sep 30, 2013)
  11., Int J Mol Med 18(6) 1165-8 (Dec 2006, Accessed Sep 30, 2013)
  12. (Accessed Sep 30, 2013)
  13. (Accessed Sep 30, 2013)
More in Claims/Labeling