Home | Parabens


Dene Godfrey, Technical Sales Manager (Preservation Systems)

S. Black Ltd.


Just so we can't be accused of being biased by not publishing the views of the chemical industry, we have in all fairness to both sides, published this report by Dene Godfrey which refutes the notion that parabens actually accumulate in human tissue and, they are in fact, totally harmless.

Here's to YOUR health

Comment | Further Reading

Are the recent concerns over a possible link between parabens, underarm deodorant products and breast cancer justified?

Most people in the personal care industry will be aware of the media reports concerning the claims made by Dr. Phillippa Darbre, for a link between parabens in underarm deodorants and breast cancer (1). In this article I will address whether or not this concern is justified by looking at 4 main issues:

  1. i) Was the rationale behind the study correct?
  2. ii) Was the study carried out using sound experimental design?
  3. iii) Was the interpretation of the results accurate and logical?
  4. iv) Did the sensationalism promoted by the lead researcher help scientific credibility in the perception of the general public?

The answer to these questions is, quite simply "NO", but as this issue is too important to dismiss in a single word, let's examine each of the four points in turn.

Was the rationale behind the study correct?

Dr. Darbre continues to promote her belief that parabens are responsible for breast cancer, despite being informed several years ago that very few deodorant products contain any preservatives, let alone parabens. A recent survey (2) identified 3,324 new deodorant products launched in the preceding 6 months, only 12 of which contained parabens. This fact makes nonsense of the claimed adverse effects of parabens in deodorant products, although it cannot rule out a connection between parabens and breast cancer in isolation. In one interview (3) Dr. Darbre said "most deodorants no longer contain the parabens because many manufacturers removed them 3 years ago". This is simply not correct, as most deodorant products have never required preservation there was no sudden change in industry practise 3 years ago.

One of the studies that may have contributed significantly to this theory was the work of Routledge et al on parabens in the rat uterus (4). This study also gained significant media coverage around the time of its publication.

The Routledge et al study investigated the potential for oestrogenic activity of paraben esters using in vivo and in vitro methods. The in vitro study employed a well-established recombinant yeast oestrogen screening method and it identified weak oestrogenic activity in methyl, ethyl, propyl and butylparaben ranging between 10,000 times (butyl) and 1,000,000 times (methyl) weaker than oestradiol. In this study 4-hydroxybenzoic acid showed zero oestrogenic activity.

The in vivo study is significantly more relevant as in vivo studies reflect the human response better than in vitro testing. Oral and subcutaneous dosing methods were employed, using a rat uterotrophic assay to assess the results. This study tested methyl and butylparaben only, and neither ester exhibited any activity using the oral route. This can be attributed to hydrolysis of the esters in the stomach, forming inactive 4-hydroxybenzoic acid. The subcutaneous route found no oestrogenic activity for methylparaben, but butylparaben was claimed to be 100,000 times weaker than oestradiol. This is not particularly dramatic when it is considered that we regularly ingest phytoestrogens from dietary sources with activities ranging from 1,000 to 100,000 times weaker than oestradiol (5).

The lowest observed effect level of butylparaben was determined at 200mg/kg/day, administered by subcutaneous injection. Assuming that in normal, topical use, butylparaben is totally absorbed through the skin, this equates to a 60kg human applying 12kg of a product containing 0.1% butylparaben in a single application! Comparing this dosage with a more realistic (but still high) 5g of product containing 0.01% butylparaben, the actual effect may be scaled down by a further factor of 24,000. This gives an overall effect 2,400,000,000 times weaker than oestradiol under "normal" use conditions! Compare this again with dietary phytoestrogens. This does not take into consideration the significant degree of metabolic breakdown of butylparaben into non-oestrogenic by-products but, with a factor of 2.4 billion, it hardly seems worth including another safety margin.

The authors of this study state that "...because the doses employed in the majority of animal studies (including this study) were short term and greatly exceeded realistic doses in humans.......the application of these findings in terms of an assessment of a possible oestrogenic hazard are equivocal." Risk is the product of hazard and exposure and I question the use of the phrase "a possible oestrogenic hazard". I think they meant "a possible oestrogenic risk". This study identified a hazard of butylparaben but the authors did not assess the risk, so I have done this on their behalf. My conclusion, based on the factors quoted above, is that the use of butylparaben presents an infinitesimal oestrogenic risk and may continue to be used safely in personal care products.

Other organisations have expressed opinions on the subject of the Routledge et al study:

"....cosmetics and toiletries containing parabens present no risk to the consumer and may continue to be used with confidence." (6)

".....the study found also that no such effect resulted from oral intake as, in this case, there is rapid and complete biodegradation of parabens into non-hormonal breakdown products. The same complete biodegradation has been demonstrated to occur in human skin after topical application of parabens.

Hormonal effects from parabens used in cosmetics can, therefore, be excluded." (7)

A further study by Hossaini et al (8) repeated the subcutaneous application used by Routledge et al and found only a weak response with butylparaben at 400mg/kg/bw/day and a clear response at 600mg/kg/bw/day a dose much higher than that observed by Routledge. In the conclusion, Hossaini stated that "owing to .. rapid metabolism and excretion it is conceivable that concentrations high enough to produce an oestrogenic effect in target tissues will not be reached unless excessive doses are used". In the case of butylparaben, the excessive dose required to exhibit an oestrogenic effect was at least 72,000 times the highest typical application rate in humans surely an excellent margin of safety!

Another study quoted by Dr. Darbre was an in vivo fish assay (9) by Pederson et al that investigated ethyl, propyl and butylparaben and found oestrogenic activity at doses between 100 and 300mg/kg - another study that identified a hazard but failed to assess the risk. In order to be exposed to these doses the fish would have to swim in a neat bath product!

A hazard may be identified for any substance even water (inhalation hazard) given the appropriate test and sufficiently high dose levels. Risk assessment is essential in order to place the hazard in context. Ideally, all studies that identify a hazard should have to include a realistic risk assessment before the paper is published and publicised.

On the basis of the risk assessments above, I believe that the rationale behind Dr. Darbres theory is deeply flawed.

Was the study carried out using sound experimental design?

Benzylparaben is rarely used in personal care products, and the results were, unsurprisingly, all negative for this compound and may be discounted. All the blanks showed an apparent paraben presence and, on subtraction from the breast tissue reading, many gave a negative result leaving us with the novel concept of "negative parabens". This occurred in 26 out of 120 data points (over 20%) a significant number and sufficient to cause concern over the validity of the results. The presence of parabens in the blank samples was attributed to "the ubiquitous use of parabens as preservatives even in laboratory detergents and personal care products of the operators". Parabens are not normally used as preservatives in detergent products, so this source of contamination is unlikely, but it may well come from the operators. One of the blank samples contained higher concentrations of parabens than 12 of the 20 tissue samples tested and, therefore, is it not possible that all the parabens detected were resulting from the personal care products used by the operators? The ratio of the parabens detected in the blanks was broadly similar to the ratio found in the tissue samples, which suggests that the source of the parabens was the same in both

cases. It is difficult to imagine that the all the very different ratios and combinations of parabens used in the huge number of different personal care products should mysteriously average out to be similar in breast tumour tissue as it is in the contaminated blanks in this study. This suggestion is supported by the multitude of studies that have determined that parabens are completely absorbed, metabolised and excreted by mammals within approximately 24 48 hours. Furthermore, to assume that subtracting the blank data from the tissue sample gives an accurate paraben level in the breast tissue is not logical as it implies that parabens are present on the glassware used for the blank samples at exactly the same concentrations on the glassware used for the tissue samples. This cannot be assumed when one considers the wide variability of paraben concentrations within the blank samples themselves from between 19.6 61.4 ng/g. This is proven by the presence of so many negative values that are, of course, impossible. The mean total paraben concentration identified in the blank samples was 33.8ng/g and, whilst the mean concentration from tumour extracts was 54.8ng/g, the higher levels may be attributed to additional paraben contamination during handling and processing of the tissues that does not occur with the blank samples.

The use of only 20 tumour samples is extremely limiting in terms of statistical analysis, particularly in view of the enormity of what is being claimed for the results.

There was no control using healthy human breast tissue for comparison.

The procedure should have been refined such that all blank samples contained no detectable parabens, and using many more tumour samples, so it is my conclusion that the study was not carried out using sound experimental design.

Was the interpretation of the results accurate and logical?

The conclusions of the study claim proof of bioaccumulation of parabens in human breast tumour tissue. Leaving aside whether or not the parabens were actually present in the tumour tissue, it is simply not possible to establish bioaccumulation using only one data point. This can only be achieved by monitoring over a period of time and observing an increase in the concentrations present and this study uses only a single data point. If the parabens truly are present in the tumour tissue, they may be simply background levels present during the metabolic process. The report then draws a comparison between parabens and PCBs and OCPs, quoting mean levels of 20, 267 and 707ng/g tissue respectively as further evidence of bioaccumulation of parabens (and wrongly claiming that these levels are similar they differ by more than an order of magnitude). I believe that this proves precisely the opposite. Human exposure to parabens must be several orders of magnitude greater than exposure to PCBs or OCPs, yet the latter have significantly higher residual levels than

parabens. If parabens were bioaccumulative, this would be reflected in mean levels significantly higher than PCBs and OCPs.

One of the most questionable aspects of the interpretation of the results is the use of a "corrected average level of parabens" suggested on the basis that 4 of the 20 tumours contained more than twice the true average level of total parabens and that only a 50% recovery of parabens was achieved during the analytical procedure. The "corrected average level" of 100ng/g was calculated using only the 4 highest results (and is, therefore, a gross distortion of the true figures), then doubling the figure based on the claim of 50% recovery. The recovery figure is based on spiking samples with benzylparaben and detecting 48.5% of the material on analysis. Benzylparaben was used because it was not otherwise detected in any of the samples. It is strange logic to base the % recovery on the one compound that was not detected and then to assume that all the esters have the same recovery factor. Benzylparaben is an aryl ester the remainder all being alkyl esters and the least representative of the group. It should not have been used to determine the recovery rate. It is also likely to have the lowest % recovery, leading to a greater distortion of the data. As methylparaben was present at 62%, this would have been best used to indicate the true recovery rate, but I would expect a difference in recovery for all esters as their solubility profiles are substantially different.

The distorted corrected average level of 100ng/g was then used as a comparator against studies (10, 11, 12, 13) that found levels of c. 150ng/ml of n-propyl, n-butyl and iso-butylparaben stimulated growth of oestrogen-dependant MCF7 human breast cancer (HBC) cells. This results in yet further distortion as the distorted average is then being compared with results for, essentially, different compounds as 62% of the total paraben level was methylparaben and a further 10% was ethylparaben, for which there are no data on their effect on MCF7 HBC cells. Therefore, only 28% of the comparator figure is an acceptable comparison as, based on many other studies, methyl and ethylparaben are significantly less likely to exhibit any oestrogenic effect than the higher esters. A more reasonable comparison would be to take the propyl and butylparaben components from the "uncorrected" average to give a figure of c. 5.6ng/g, which is significantly lower than the 150ng/g level claimed to stimulate growth of MCF7 HBC cells.

In the light of all the above factors I have to conclude that the interpretation of the results of the study was neither accurate, nor logical and grossly distorted.

Did the sensationalism promoted by the lead researcher help scientific credibility in the perception of the general public?

The basic, incontrovertible fact that more than 99% of deodorant products do not contain parabens makes this a very easy question to answer! This is a fact that the general public can easily understand, but it is quickly lost beneath much of the media hype surrounding the issue, although many of the articles I read did

contain a reasonable balance of views. Unfortunately, very few of the headlines were balanced.

This study was deeply flawed, both in design and interpretation and I detect a large element of evangelism on the part of the researchers who are determined to find the cause of breast cancer at any cost. Their cause is highly laudable but, unfortunately, their methods are highly questionable. The study, and the publicity surrounding it has caused unnecessary concern for both the general public and the manufacturers of deodorant products in particular, and there is likely to be a "knock-on" effect on paraben-containing products in general.

There is no basis for any of the harmful claims made for parabens in this study as it is by no means conclusive that parabens are present in breast tumour tissue, and I believe that the researchers should be heavily censured for the use of poor experimental procedure, flawed logic and gross distortion of the results. I would also call into question the peer review process in this instance.

I began this article by asking 4 questions, and I will finish with just one:

How was this study ever accepted for publication?

My thanks go to Jim Bootman of Bootman Chemical Safety Ltd. for his invaluable contribution to my understanding of the Routledge et al study.


1. Darbre, P.D, Aljarrah, A, Miller, W. R, Coldham, N. G, Sauer, M. J. and Pope,

G. S, J. Appl. Toxicol. 24, 5 13 (2004)

2. Formulators Post Bag, 3 Feb 2004 (Formulator@stepex.co.uk)

3. Daily Mail, Jan. 13, 2004 p. 45

4. Routledge, E.J, Parker, J, Odum, J, Ashby, J and Sumpter, J.P. Tox. &

Appl. Pharm. 153, 12 - 19 (1998)

5. Thomas, Env. & Nut. Interactions 1, 1997

6. CTPA press statement, October 1998

7. COLIPA press statement, October 1998

8. Hossaini, A, Larsen, J-J. and Larsen, J. C, Food & Chem. Tox. 38, 319 323 (2000)

9. Pedersen, K. L, Pedersen, S. N, Christiansen, L. B, Korsgaard, B, and Bjerregaard, P, Pharm. & Toxicol. 86, 110 113 (2000)

10. Okubo, T, Yokoyama, Y, Kano, K, Kano,I, Food Chem. Toxicol. 39, 1225 1232 (2001)

11. Byford, J. R, Shaw, L. E, Drew, M. G. B, Pope, G. S, Sauer, M. J, Darbre, P. D, J. Steroid Biochem. Mol. 80, 49 60 (2002)

12. Darbre, P. D, Byford, J. R, Shaw, L. E, Horton, R. A, Pope, G. S, Sauer, M. J, J. Appl. Toxicol. 22, 219 226 (2002)

13. Darbre, P. D, Byford, J. R, Shaw, L. E, Hall, S, Coldham, N. G, Pope, G. S, Sauer, M. J, J. Appl. Toxicol. 23, 43 51 (2003)

This article was originally published in SPC, April 2004 (Wilmington Publishing).

Health-Report Comment:

Well it is great to get feedback like this and also excellent to see the research that Dene Godfrey has carried out, certainly seems to refute the idea that parabens could have any possible link with cancer. Interesting! I wonder who actually paid for this research? Obviously the paper was written to offset the bad publicity that parabens have attracted in recent press articles and news items of late.

I find it interesting that Dene couldn't find many examples of deodorants that contained parabens.  "A recent survey (2) identified 3,324 new deodorant products launched in the preceding 6 months, only 12 of which contained parabens." Perhaps he was looking at the paraben free variety which is becoming so much more readily available in recent times since the adverse publicity. When I took the time to look through a number of fairly common products over 12 months ago, around 80% of them listed parabens in some form in them. Then again I live in Australia and things may be different here compared to the UK or USA

If it was so hard to find deodorants with parabens contained in them when the above paper was written, perhaps it was the adverse publicity from Dr Philippa Darbre's research, which has forced manufacturers to remove these chemicals due to people actually reading labels and resisting buying products containing parabens. One must question the validity of any research just as Dene Godfrey is refuting the research carried out by Dr Dabre.

For your reference here are some of the products I found which contained parabens. Anyone can view the ingredients on hundreds or even thousands of skin "care" and body "care" products which contain, at the best, questionable chemicals by visiting the www.ewg.org website and specifically this page to view the deodorants and the chemicals they contain, including parabens in the more toxic varieties listed. http://www.ewg.org/reports/skindeep2/search.php?nperpage=10&stype=brands&category=Antiperspirant%2FDeodorant&sort=&showall=1

Bear in mind, up until relatively recently, parabens were in the majority of skin "care" products and most deodorants on the market. Perhaps since Dr Dabre's report there may have been 3,212 deodorants placed on the market which are totally free of parabens in all their forms. Trouble is I can't find a great number of them in our neck of the woods.

As I am fond of saying - YOUR Health... YOUR decision!

Geoff Goldie www.health-report.co.uk

Product Name Ingredients: (from retailer or packaging)
Este Lauder Youth Dew Roll-On Anti-Perspirant Deodorant

From the EWG Website - Score 4.2

where 5 is the highest toxicity


Aluminum Chlorohydrate; Water (Aqua Purificata) Purified; Steareth-2 Alcohol; Denat.; Ppg-11 Stearyl Ether; Steareth-20; Youth-Dew Fragrance (Parfum); Myristalkonium Chloride; Quaternium-14; Benzyl Salicylate; Linalool; Hydroxycitronellal; Geraniol; Citronellol; Eugenol; Limonene; Cinnamyl Alcohol; Coumarin; Hexyl Cinnamal; Evernia; Furfuracea (Treemoss) Extract; Benzyl Alcohol; Benzyl Benzoate; Trisodium Edta; Sorbic Acid; Methylparaben; Propylparaben; Butylparaben
Fresh Roll-on Sugar
Deodorant and Antiperspirant

Score 3.2 out of 5

Active Ingredients: Aluminum Chlorohydrate 15% Inactive Ingredients: Water (Aqua), PPG-15 Stearyl Ether, Steareth-2, Steareth-20, Hamamelis Virginiana Distillate (Witch Hazel), Citrus Aurantium Bergamia Fruit Extract (Bergamot), Cymbopogon Schoenanthus Extract, Cupressus Sempervirens Cone Extract, Avena Sativa Kernel Extract (Oat), Aloe Barbadensis Leaf Extract, Fragrance (Parfum), Glyceryl Stearate, Propylene Glycol, BHT, Phenoxyethanol, Propylparaben, Ethylparaben, Methylparaben, Butylparaben, Isobutylparaben
Fresh Umbrain Clay Roll-On Deodorant

Score 1.9 out of 5

Mentha Piperita Leaf Water (Peppermint), Lavandula Angustifolia (Lavender) Flower Water, PEG-8, Cyclopentasiloxane, Methyl Methacrylate Crosspolymer, Glycerin, Polyacrylamide, Fuller's Earth (Solum Fullonum), Citrus Medica Limonum (Lemon) Peel Oil, Rosmarinus Officinalis (Rosemary) Leaf Extract, Helianthus Annuus (Sunflower) Seed Oil, Sodium Hyaluronate, Ethylhexylglycerin, C13-14 Isoparaffin, Fragrance (Parfum), Laureth 7, Glyceryl Polyacrylate, Chitosan Succinamide, Phenoxyethanol, Methylparaben, Tetrasodium EDTA
Bliss Under-Army Mint Antiperspirant Gel

Score 2.5 out of 5

Active Ingredients: Aluminum Sesquichlorohydrate 16% (antiperspirant) Inactive Ingredients: Water, Propylene Glycol, Cyclopentasiloxane, PEG-12 Dimethicone Crosspolymer, Glycerin, Cyclomethicone, Dimethicone, Alcohol Denat, Fragrance, Phenyltrimethicone, Olea Europaea Fruit Extract (Olive), Prunus Amygdalus Dulcis Extract (Sweet Almond), Phenoxyethanol, Chlorphenesin, Methylparaben, Benzoic Acid
Karl Lagerfeld Lagerfeld Deodorant Stick

Score 2.1 out of 5

Propylene Glycol, Denatured Alcohol, Water, Sodium Stearate, Fragrance, Triclosan, FD&C Blue No. 1, FD&C Red No. 4, FD&C Yellow No. 5.

Note: NO parabens but what about triclosan



The Synonyms for Parabens
Isobutylparaben Benzoic acid, 4-hydroxy-, 2-methylpropyl ester
4-Hydroxybenzoate d'isobutyle
isobutyl 4-hydroxybenzoate
Isobutyl-4-hydroxybenzoat 4-hidroxibenzoato de isobutilo
2-Methylpropyl p-hydroxybenzoate
Benzoic acid, p-hydroxy-, isobutyl ester
iso-Butyl p-hydroxybenzoate
Isobutyl p-hydroxybenzoate
p-Hydroxybenzoic acid isobutyl ester
Butylparaben Benzoic acid, 4-hydroxy-, butyl ester
4-Hydroxybenzoate de butyle butyl 4-hydroxybenzoate
4-hidroxibenzoato de butilo
4-Hydroxybenzoic acid butyl ester
benzoate, 4-hydroxy-, butyl

Aseptoform Butyl

Benzoic acid, p-hydroxy-, butyl ester
Butyl p-hydroxybenzoate
n-Butyl 4-hydroxybenzoate
n-Butyl p-hydroxybenzoate
n-Butylparabenp-Hydroxybenzoic acid butyl ester
n-Propylparaben Benzoic acid, 4-hydroxy-, propyl ester
4-Hydroxybenzoate de propyle propyl 4-hydroxybenzoate Propyl-4-hydroxybenzoat
4-hidroxibenzoato de propilo
4-Hydroxybenzoic acid propyl ester
4-hydroxybenzoic acid propylester
propyl p-hydroxybenzoate
propyl paraben
benzoate, 4-hydroxy-, propyl

Benzoic acid, p-hydroxy-, propyl ester
n-Propyl 4-hydroxybenzoate
p-Hydroxybenzoic acid propyl ester
p-Hydroxybenzoic acid, propyl ester
p-Hydroxybenzoic propyl ester
Ethylparaben Benzoic acid, 4-hydroxy-, ethyl ester
4-Hydroxybenzoate d'ethyle ethyl 4-hydroxybenzoate
4-hidroxibenzoato de etilo
benzoate, 4-hydroxy-, ethyl
ethyl parasept
4-Hydroxybenzoic acid ethyl ester
Benzoic acid, p-hydroxy-, ethyl ester
Ethyl p-hydroxybenzoate
p-Hydroxybenzoate ethyl ester
p-Hydroxybenzoic acid ethyl ester
Methylparaben Benzoic acid, 4-hydroxy-, methyl ester
4-Hydroxybenzoate de methyle
methyl 4-hydroxybenzoate
4-Hidroxibenzoato de metilo
4-Hydroxybenzoic acid methyl ester
, 4-hydroxy-, methyl methyl p-hydroxybenzoate
p-hydroxybenzoic acid
methyl ester
methyl paraben
methyl ester of p-hydroxy benzoic acid
Benzoic acid, p-hydroxy-, methyl ester


Hit Counter