By Trevor Butterworth

March 22, 2009
An alarming new report by the Campaign for Safe Cosmetics claims bath products for babies contain carcinogens, but by the standards it used to measure risk from formaldehyde and 1,4-dioxane shouldn't we be even more worried about bathwater, tomatoes and fried chicken? And did the activist group actually measure exposure?

The Campaign For Safe Cosmetics (CSC), a coalition of activist groups that have been campaigning for years about chemical exposure in personal care products has released a new report — No More Toxic Tub — on the apparent cancer risks from baby bath products. The study was dutifully transcribed by news media outlets, including and USA Today, which headlined the piece Group finds carcinogens in kids bath products.

Many children's bath products contain chemicals that may cause cancer and skin allergies, according to a report released Thursday by the Campaign for Safe Cosmetics, wrote Liz Szabo in USA Today. Twenty-three of 28 products tested contained formaldehyde, the report says. Formaldehyde — considered a probable carcinogen by the Environmental Protection Agency — is released as preservatives break down over time in a container.

There followed a description of the study, comments from a scientist who worked with the cosmetics manufacturer saying there was no cause for concern and an environmental health pediatrician.

The story perhaps explains why only one percent of health journalists said health reporting in the U.S. was excellent.

First, the CSC is driven by a group of environmental activist groups with a long history of hyperbole. The study was self-published, it wasnt peer-reviewed; in fact, it wasnt even scientific — if one takes science to be formulating a hypothesis and testing it against the full range of data.

If USA Todays reporter had interviewed any cancer researchers on the studys content, or did any background research on the chemicals (instead of finding a source against the study and one for it — lazy, formulaic journalism which gets the story off the desk in super quick time), the result would not be the inevitable panic among consumers when confronted with the headline Group finds carcinogens in kids bath products.

First of all, some information crucial to understanding cancer risks from STATS survey of 401 randomly chosen members of the American Association for Cancer Research. Only one in four thought that cancer-causing agents were unsafe regardless of the dose (28 percent). This is important because if every carcinogen was carcinogenic at any level, one would have to start a Campaign for Safe Vegetables — which would have the unfortunate and silly goal of removing most vegetables from peoples diet.

The CSC is not interested these kind of distinctions. A carcinogen is a carcinogen is a carcinogen — and no risk, even the most wildly hypothetical, is justifiable (in cosmetics that is — members of the coalition presumably drink coffee, alcohol, eat carrots, smoke and so forth).

And mews organizations such as USA Today dont seem interested in the fact that this is not how science assesses cancer risks.

CSCs report focused on two chemicals: formaldehyde and 1,4-dioxane.


Formaldehyde is widely used in manufacturing for a vast array of purposes, including foam insulation material, plywood adhesives, explosives, and disinfectants. It is a nearly colorless gas with a pungent odor usually used and stored in a solution. It is lethal at very high doses by ingestion and considered a human carcinogen by the International Agency for Research on Cancer due to a weak association with nasal cancer incidence in workers with chronic exposure to formaldehyde.

The Carcinogenic Potency Project at U.C. Berkeley notes that formaldehyde damages DNA, kills cells and causes cell proliferation in both rodents and humans, but that the risk declines markedly at levels that do not kill cells.

Formaldehyde also occurs naturally in some fruits and food.

The risks of formaldehyde have been assessed as exposure from manufacturing in 1979 and long-term studies of workers exposed to high levels of the chemical (the National Cancer Institute explains why it is so difficult to prove that formaldehyde exposure in workers is linked to cancer); exposure from mobile home air (which caused a controversy in the wake of Hurricane Katrinas need for mobile homes), and normal home air.

The most substantial risk came from manufacturing through gas inhalation. Workers were exposed to levels so high that they were nearly equivalent to those used in tests that caused tumors in 10 rodents.

The average daily exposure to formaldehyde in conventional home air (over 12 hours per day) is 80 times less than the amount that produced tumors in 10 percent of rodents. This is close to that of coffee, where average daily exposure is 90 times less than the amount that produced tumors in 10 percent of rodents.

Where does our exposure to formaldehyde come from? As the eleventh edition of the Report on Carcinogens by the National Toxicology Program notes:

"The general population may be exposed to formaldehyde through its use in construction materials, wood products, textiles, home furnishings, paper, cosmetics, cigarette smoke, and pharmaceuticals. Formaldehyde released to indoor air from construction materials, furnishings, and cigarettes are major sources of exposure. Cigarettes may contribute as much as 10% to 25% of the indoor exposure. Automobile exhaust is a major source of formaldehyde in ambient air. In addition, formaldehyde can be absorbed through the skin from cosmetics or contact with other consumer products containing formaldehyde."

Formaldehyde is used in some cosmetics as a preservative, and the Campaign for Safe Cosmetics claims that 23 out of 28 products they tested contained formaldehyde at levels of between 54 and 610 parts per million.

The European Union has stipulated that the total content of formaldehyde in the finished cosmetic product must not exceed 0.2% — which is equivalent to 2000 parts per million. In other words, the products assessed by the CSC were well within European safety margins. There is no indication that this could pose a risk. CSC did find one item which released levels above the EU limits — Baby Magic Baby Lotion.

But here we come to the fundamental methodological flaw in the report — one that underscores the need for real, peer-reviewed scientific analysis of chemical exposures and health and not activist reports designed to maximize media attention through sensationalism: The Campaign measured how much formaldehyde was in the product, but not how much a child would actually be exposed to or absorb in the course of using that product.

Heres the CSC testing protocol:

Formaldehyde: 1 to 2 grams of product sample was weighed to the nearest milligram, and placed into a glass vial. 20 milliliters of an aqueous buffer (pH=5 acetic acid) was added volumetrically. The sample vial was sealed with a teflon lined cap, vortexed and then placed on a shaker table for 12 hours. 0.5 to 1 milliliter of the aqueous buffered extract was volumetrically transferred to a 250 bottle to which 100 milliliters of organic free dionized water had been added; 4 milliliters of acetic acid buffer was added to the bottle to maintain a pH=5 and then 6 milliliters of a 2,4-Dinitrophenyl hydrazine (DNPH) solution was added. The 250 ml bottles were placed in a shaking water bath with the temperature maintained at 40 degrees C for 1 hour to complete the derivatization. The aqueous derivatized sample was transferred to a separatory funnel and extracted with methylene chloride. The methylene chloride extract was concentrated and exchanged to a final 5 milliliter volume of acetonitrile (ACN). 10 microliters of the acetonitrile extract was introduced into an Agilent HPLC. The HPLC column was a Restek Ultra Aqueous C18 (150mm x 4.6mm). A 70%/30% ACN/water to 100% ACN eluent gradient program was used to elute the derivatized formaldehyde which was detected with an ultraviolet detector set to 365 nanometers. Standards of formaldehyde were derivatized and extracted similarly. Method blanks were used to assess background contamination from formaldehyde.

In other words, not only did all but one of the products meet the EU safety guidelines, they may or may not expose infants and children to formaldehyde at the levels found. It is unlikely that a parent is going to place an infant in a shaking bath filled with water at 104 degrees Fahrenheit for an hour.

This isnt meant to be flippant. If formaldehyde in bath lotions is as dangerous to children as the Campaign for Safe Cosmetics claims, why didnt it measure actual exposure? And why didnt the media spot — by simply reading the report — that it failed to do this?

This is the ongoing problem with activist studies: they arent really scientific and, of course, they always manage to confirm what the activist group is lobbying against.


The other chemical highlighted by the Campaign for Safe Cosmetics is 1,4-dioxane, which occurs in small amounts in cosmetics as a byproduct in manufacturing, and which cosmetics companies strive to remove. The Campaign found that 32 out of 48 products tested (67%) contained 1,4-dioxane at levels ranging from 0.27 to 35 ppm — and noted that the chemical was banned by the EU in cosmetics (although it isnt banned in the U.S. and the FDA is not, apparently, concerned about the trace amounts in cosmetics.)

The IARC has declared 1,4-dioxane as possibly carcinogenic to humans. The National Toxicology Program has described it as reasonably anticipated to be a human carcinogen based on studies in rodents where the chemical was consumed through drinking water and at very high levels. The FDAs position on the chemical is that: The 1,4-dioxane levels we have seen in our monitoring of cosmetics do not present a hazard to consumers.

Concerns initially were raised in the 1970s, when studies at the National Cancer Institute found an association between 1,4-dioxane and cancer in animals when 1,4-dioxane was administered in high levels in the animal feed. However, the levels in cosmetic products are far lower than those found to be harmful in feeding studies and, for the most part, the types of products in which it is found are only in contact with the skin for a short time.

As a precaution, FDA followed up with skin absorption studies, which showed that 1,4-dioxane can penetrate animal and human skin when applied in certain preparations, such as lotions. However, further research by FDA determined that 1,4-dioxane evaporates readily, further diminishing the already small amount available for skin absorption, even in products that remain on the skin for hours. (Robert L. Bronaugh, "Percutaneous Absorption of Cosmetic Ingredients," in Principles of Cosmetics for the Dermatologist, Philip Frost, M.D., and Steven Horwitz, M.D., Eds. St. Louis: The C.V. Mosby Company, 1982)

The other confounding problem with 1,4-dioxane is that we are exposed to it routinely in tap water, either by drinking or when we shower (as a volatilized compound), and in seafood, cooked meat, fried chicken, deep fry oil, ripe tomatoes, tomato paste, peppers, coffee, herbs and spices (within the range of 2-15ppm). In fact, given that 1,4-dioxane is more easily absorbed by ingestion and inhalation rather than absorption (due to its propensity to evaporate), the route of exposure is much more likely from the water we wash and shower in than through skin absorption from a cosmetic lotion.

Again, this underscores the need for exposure to be measured, and not simply to examine whether a solution contains 1,4-dioxane or not. Is a child actually exposed to more 1,4-dioxane through eating a regular meal and drinking water than through using a lotion? If 1,4-dioxane is as dangerous as CFSC claims, surely this is more alarming?

But there is considerable controversy over how dangerous 1,4-dioxane really is. The studies all focus on inhalation and ingestion rather than absorption, and required substantial quantities of the chemical to produce cancer in the lab animals. This underscores the important of distinguishing the presence of a chemical from actual exposure.

Should you take the Campaign for Safe Cosmetics report seriously?

It will be obvious by now that we are exposed to formaldehyde and 1,4-dioxane from a multiplicity of routes, and in each case absorption through cosmetics is, arguably, the least significant route of exposure. There is much greater concern in the regulatory and scientific research about environmental release and exposure from other sources.

The problem with the Campaign for Safe Cosmetics is that its report isnt scientific. It is an argument that any exposure to a chemical that can be shown to cause cancer in any possible way should be banned. This is not consonant with mainstream scientific or regulatory thinking.

Second, the Campaign equates presence with exposure. But if you cant show that a chemical is being absorbed, you cant measure its effect. A tank full of gasoline is lethal if you drink it, but the fact that its sitting in your car doesnt pose the same degree of risk.

Third, the bodys metabolism — even in an infant — is capable of processing many chemicals quickly and without any negative consequence. Even chemicals that are dangerous at high levels can have no negative effect whatsoever. If this wasnt the case, almost every known food would be toxic.

There is also increasing disquiet among toxicologists about the reliance on high doses of a chemical in animals to determine human cancer risks. As the Carcinogenic Potency Project notes in its latest evaluation of cancer risks: The chronic, high dose rodent cancer test is not much use in understanding human cancer risk. Tumor development is likely due to high dose effects or processes that are not relevant to humans.

In pursuing its cause, the Campaign for Safe Cosmetics systematically avoids addressing the science in a scientific way, and acknowledges no scientific limitations to its claims despite failing to develop the kinds of tests that would actually support its conclusions.

We need real science. And we need the media to be able to distinguish what counts as real science and what is merely self-serving activism.


Trevor Butterworth is a Senior Fellow at STATS, a research organization affiliated with the George Mason University. He received his BA and M.Phil from Trinity College Dublin and did graduate work in philosophy and intellectual history at Georgetown University. He received an M.S. from Columbia University's Graduate School of Journalism, where he was awarded the Sevellon Brown Prize for outstanding knowledge of the history of the American press.