The Cosmetic Record

“PPD-Free” Is Not the Same as Hypoallergenic

2026-08-06T00:00:00Z

Correction

“PPD-Free” Is Not the Same as Hypoallergenic

The claim: “If hair dye brings me out in a rash, a PPD-free box is the safe one.”
The short answer: Probably not. The EU files toluene-2,5-diamine — the usual PPD substitute — under the same phenylenediamine warning block, at the same 2% free-base cap after mixing. The regulation's own labelling choice contradicts the marketing.
Evidence status: Settled

The Cosmetic Record · Published 2026-08-06 · Last checked 2026-08-06 · 18 sources · Corrections: none

Who checks a box of hair dye before it reaches the shelf, and what does the warning on it commit the manufacturer to?
Market	The rule	What it requires	What it does not do
BR	ANVISA RDC 907/2024 Art. 24(II), Art. 34, Anexo I item II nº 45; substance limits in RDC 1.029/2026	Grau 2, exempt from registro and filed by comunicação prévia. The label must carry the touch-test warning; since 11 June 2026 the substance annex caps p-phenylenediamine at 2% free base after mixing and toluene-2,5-diamine at 2,0 %.	Notification is a filing, not an approval. ANVISA does not assess the formula before sale.
US	FD&C Act 21 U.S.C. 361(a) and (e); 21 U.S.C. 364d(b)	One sentence of prescribed caution text on the label, plus adequate directions for a preliminary skin test. That is the whole bargain.	No rule caps the concentration of a coal-tar hair dye substance, no colour-additive approval applies to it, and — uniquely among cosmetics after MoCRA — no duty to substantiate safety attaches to it.
EU	Regulation (EC) No 1223/2009, Annex III entries 8, 8a, 8b, 9 and 9a (CELEX 02009R1223-20260501)	A restricted list. Each dye substance has a named entry, a maximum concentration after mixing, and a warning block printed word for word, including the line that the product is not intended for anyone under 16.	Nothing in the entry obliges a consumer self-test, and no authority signs off an individual product before it is placed on the market.

1. What the allergy is: a delayed reaction to the amine and its oxidation intermediates

Allergic contact dermatitis from hair colour is a type IV reaction — cell-mediated and delayed, appearing hours to days after contact rather than within minutes. That lag is why the diagnostic test for it is a patch read at 48 and 72 hours rather than a prick test read immediately.

The sensitising material is not only the amine as supplied. In an oxidative colourant the primary intermediate is oxidised by alkaline hydrogen peroxide, and the product of that oxidation couples with a second molecule into the pigment that stays in the fibre. Which species does the work is still argued: the standard review of the chemistry lists both p-benzoquinone diimine and its conjugate acid, the diiminium ion, as proposals for p-phenylenediamine, and records that the self-coupling products called Bandrowski's bases form when no coupler is present but are believed not to form inside the hair.⁵ Sensitisation can be to the parent amine or to a reactive oxidation product, which is one reason a person who reacts to one member of the family often reacts to its relatives.

The European Commission's Scientific Committee on Consumer Safety has catalogued how strong each of these sensitisers is. Its 2013 memorandum sorts the classified hair dye substances into extreme, strong and moderate: p-phenylenediamine is extreme, with a local lymph node assay EC3 of 0.06 %, and toluene-2,5-diamine is extreme as well, at 0.31 %.² (An EC3 is the concentration estimated to triple lymph-node cell proliferation in a mouse assay — a laboratory potency ranking, not a human dose.) The committee concluded that substances meeting the classification criteria “may not be safe for consumers and that this is particularly so for hair dye substances categorised as extreme and strong sensitizers”. In a Polish series of 4,087 patients patch-tested for allergic contact dermatitis between 2006 and 2015, 166 — 4.1% — reacted to p-phenylenediamine or a related amine.⁴

2. What the EU box must say, quoted in full

Annex III of Regulation (EC) No 1223/2009 is a restricted list: a substance in it may be used only for the stated purpose, only up to the stated concentration, and only with the stated wording printed on the pack. Entry 8a covers p-phenylenediamine and its salts as a hair dye substance in oxidative products, and caps it — general and professional use alike — at 2% calculated as free base after mixing. This is the block that must appear:

Hair colourants can cause severe allergic reactions.
Read and follow instructions.
This product is not intended for use on persons under the age of 16.
Temporary “black henna” tattoos may increase your risk of allergy.
Do not colour your hair if:
— you have a rash on your face or sensitive, irritated and damaged scalp,
— you have ever experienced any reaction after colouring your hair,
— you have experienced a reaction to a temporary “black henna” tattoo in the past.
Contains phenylenediamines.
Do not use to dye eyelashes or eyebrows.

Regulation (EC) No 1223/2009, Annex III, entry 8a, conditions of use (a) general use — consolidated text of 1 May 2026, CELEX 02009R1223-20260501. The mixing ratio must be printed with it.¹

Two things in that block are worth holding on to. It is an unconditional warning that the product class can cause severe allergic reactions — not a warning triggered by a particular ingredient's presence above a threshold. And it does not tell anyone to test the product on themselves first.

3. PTD is under the same warning, and the regulation's own wording proves it

The substance put forward as the alternative to p-phenylenediamine is toluene-2,5-diamine, also written as 2,5-diaminotoluene or PTD, and listed in ingredient declarations as TOLUENE-2,5-DIAMINE or TOLUENE-2,5-DIAMINE SULFATE. The American Contact Dermatitis Society named its sulfate salt Allergen of the Year for 2025 on exactly that footing: a substance with a “dual role as both an allergen and an alternative for some PPD-allergic individuals”.⁶ It has its own Annex III entry, 9a, with CAS 95-70-5 and EC 202-442-1. After mixing under oxidative conditions it must not exceed 2,0 % calculated as free base, or 3,6 % calculated as the sulfate salt — the same ceiling, in free-base terms, as p-phenylenediamine.¹

The warning block on entry 9a is the entry-8a block, sentence for sentence: the same statement that hair colourants can cause severe allergic reactions, the same line that the product is not intended for anyone under 16, the same black-henna sentence, the same “Do not colour your hair if” list. Exactly two clauses differ. The first is a detail of scope: entry 8a's general-use block closes “Do not use to dye eyelashes or eyebrows.”, entry 9a's closes “Do not use to dye eyelashes.” — entry 9a carries a second limb permitting professional eyelash colourants, and its general-use limb says nothing about eyebrows either way. The second difference is the decisive one. Where an entry-8a product declares “Contains phenylenediamines.”, an entry-9a product declares:

Contains phenylenediamines (toluenediamines)

Regulation (EC) No 1223/2009, Annex III, entry 9a, conditions of use (a) — consolidated text of 1 May 2026, CELEX 02009R1223-20260501.

The drafting is not ambiguous: toluene-2,5-diamine is labelled a phenylenediamine, family name first and sub-family in brackets. The neighbouring entry 9 does the same for the whole class, restricting “Methylphenylenediamines, their N-substituted derivatives and their salts” with the identical parenthetical.

The clinical literature points the same way. In a Dutch series of 8,036 patients patch tested with the European baseline series, 251 were sensitised to p-phenylenediamine or toluene-2,5-diamine or both: 231 to PPD and 109 to TDA, which leaves 89 reacting to both. The authors concluded that cross-reactivity “was commonly found among individuals sensitized to PPD or TDA, and was strongly related to the strength of the patch test reaction”.³

Whether the claim is even true of the box has been measured. A New York University group ran 51 best-selling US hair dyes marketed as “PPD-free” through high-performance liquid chromatography coupled to high-resolution mass spectrometry. Seven contained a compound absent from the ingredient list; five contained p-phenylenediamine itself with no mention of it on the list, four of those carrying a “PPD-free” label, and one held more than 2% PPD by weight.⁷

An accurate “PPD-free” claim is still not a claim about allergy — it is a statement about one annex entry. Regulation (EU) No 655/2013, which sets the common criteria for cosmetic claims in the EU, requires claims to be supported by adequate and verifiable evidence and bars those conveying “the idea that a product has a specific benefit when this benefit is mere compliance with minimum legal requirements”.¹⁸ Nothing in it converts “PPD-free” into “safe for a person who reacts to hair dye”.

4. The American answer: the coal-tar carve-out and the exact caution it demands

The United States has no annex, no concentration limits for these substances and no positive list for them. It has one sentence. Section 601(a) of the Federal Food, Drug, and Cosmetic Act deems a cosmetic adulterated if it contains a poisonous or deleterious substance that may render it injurious — and then exempts coal-tar hair dye from that provision entirely, on two conditions:

Caution-This product contains ingredients which may cause skin irritation on certain individuals and a preliminary test according to accompanying directions should first be made. This product must not be used for dyeing the eyelashes or eyebrows; to do so may cause blindness.

Federal Food, Drug, and Cosmetic Act §601(a), 21 U.S.C. 361(a) — text in force as of 5 August 2026. The second condition is that “the labeling of which bears adequate directions for such preliminary testing”.⁸

“Coal-tar” is a historical term rather than a description of the raw material: FDA's page records that the colours were once by-products of the coal industry, are now mostly made from petroleum, and that the category covers permanent, semi-permanent and temporary hair dyes.¹¹ Section 601(e) then exempts coal-tar hair dyes from the colour-additive approval requirement that applies to every other colour in a cosmetic.⁸ The Modernization of Cosmetics Regulation Act of 2022 did not close the carve-out. It added a general duty to hold adequate substantiation of safety — and then, at FD&C §608(b), 21 U.S.C. 364d(b), disapplied that duty to “coal-tar hair dye that otherwise complies with the requirements of section 361(a)”, leaving only an obligation to keep safety records.⁹

Two American details are frequently misreported. The cancer warning at 21 CFR 740.18 has never taken effect: the effective-date note printed under the section records that at 47 FR 7829, 23 February 1982, it “was stayed until further notice”, and the stay is still there.¹⁰ And lead acetate was delisted by a rule published 31 October 2018 that took effect only on 6 January 2022, after an objection stayed it.¹⁴

As for the word in this article's headline: FDA states that “There are no Federal standards or definitions that govern the use of the term ‘hypoallergenic’”, that manufacturers need not submit substantiation for such claims, and that a 1977 appellate decision struck down the regulation that would have defined it.¹² A “PPD-free” claim is a narrower cousin of the same move.

5. Brazil: Grau 2, notified rather than registered, with the prova de toque mandated

Brazil classifies personal-care products by risk. Hair dye is Grau 2, the higher of the two grades, listed at item II nº 45 of Anexo I to RDC 907/2024 as “Tintura capilar temporária, progressiva ou permanente” — that annex as replaced by RDC 949/2024. Grau 2 is a risk grade, not a registration status. Art. 34 lists the nine groups that must go through registro; hair dye is not among them, and the closest item, IV, is “produto para alisar e tingir os cabelos”, a product that straightens and colours. Under Art. 35 everything off that list is exempt from registration and goes through comunicação prévia — so ordinary hair colour is notified, not registered.¹⁵ The gap between those two words is the single most misread thing in Brazilian cosmetic labelling, and it has its own article here.

What Brazil does require, and the EU does not, is an instruction to test:

CUIDADO. Pode causar irritação na pele ou reação alérgica. Antes de usar, fazer a prova de toque (descrever);
Em caso de contato com os olhos, lavar com água em abundância;
Manter fora do alcance das crianças;
Não aplicar se o couro cabeludo estiver irritado ou lesionado; e
Não usar nos cílios.

ANVISA RDC nº 907, de 19 de setembro de 2024, Art. 24, inciso II (“agentes clareadores de cabelos e tinturas capilares”) — consolidated text as amended by RDC nº 949, de 12 de dezembro de 2024. In the original the five warnings are lettered (a) to (e), and “(descrever)” sits outside the quotation marks: it is an instruction to the manufacturer, who must describe the touch test rather than merely name it. The opening line of Art. 24 also allows “uma redação que assegure a mesma interpretação”, so unlike the EU block this wording may be reworded to the same effect.¹⁵

Since 11 June 2026 Brazil also has a substance annex shaped like the European one. RDC 1.029/2026 internalises MERCOSUR resolution GMC nº 06/25, capping p-phenylenediamine at entry 8b at 2% as free base after oxidative mixing and toluene-2,5-diamine at entry 9b at 2,0 % free base or 3,6 % as the sulfate, with the label line “Contém fenilenodiaminas (diaminotoluenos)”. The entry numbers differ from the EU's; the limits and the warning do not. Art. 2 gives products already regularised 24 months to comply, so Brazilian stock before June 2028 may still be built to the previous list.¹⁶

6. Two regulators, two opposite answers on consumer patch testing

The United States makes the consumer self-test the price of the exemption: no adequate directions for a preliminary test, no carve-out. Brazil orders the same thing in its own words and requires the method to be described. The European Union, asked whether such a test should be put in consumers' hands, declined.

The Scientific Committee on Consumer Safety examined a standardised Allergy Alert Test in 2019 and found that its diagnostic performance showed “a very good specificity, but a potentially moderate sensitivity, leaving room for the possibility of false-negative AAT reactions in sensitised consumers who may then experience allergic contact dermatitis to a subsequently applied product”, and that it “cannot, at this stage, endorse the use of the AAT as an ‘alert’ for contact allergy to hair dyes in consumers”. It added a second objection: the test “carries the same basic risk of sensitisation as the use of a hair dye product by a consumer”, so if it “is used on a regular basis before hair dyeing, it will inevitably increase the number of exposures, and this may increase the risk of sensitisation”.¹⁷

Three jurisdictions looked at the same procedure and produced two opposite answers about whether to require it. That disagreement is itself a fact about the state of the evidence.

7. Black henna, and why the regulation names it

The EU warning block singles out one product that is not a hair dye at all: “Temporary ‘black henna’ tattoos may increase your risk of allergy.” Henna itself, from Lawsonia inermis, stains reddish-brown. To make a black, fast-developing skin stain, something else is added, and FDA states that the additive “is often a coal-tar hair dye containing p-phenylenediamine (PPD)”, that an artist may use a PPD-containing hair dye on its own, and that “By law, PPD is not permitted in cosmetics intended to be applied to the skin”. Henna is approved in the US only as a hair dye, so mehndi products containing it are adulterated.¹³

The relevance to a box of hair colour is exposure history. A holiday temporary tattoo is a high-concentration, long-contact, occluded application of the same chemistry, and it is a route by which someone may become sensitised years before opening a dye carton. That is why the sentence sits in the Annex III block, and why it appears in the Brazilian annex too.

8. What a patch test can and cannot rule out

A negative home test is evidence about one product, on one occasion, at one site, read once. It is not a property of the person. FDA's own hair dye page makes the limits explicit: “People can develop sensitivities with repeated exposure. In addition, formulations may change over time. So, it's possible to have a reaction even if you have dyed your hair in the past, without a problem.”¹¹ Sensitisation is acquired, and the acquiring event is usually an ordinary use of the product.

Add the SCCS finding on sensitivity and the failure mode is clear: a test with good specificity and moderate sensitivity produces few false alarms and a meaningful number of false reassurances. A positive result is informative. A negative one is weak evidence, and it expires when the formula changes.

Which leaves the narrow conclusion. “PPD-free” identifies one Annex III entry that is absent. It says nothing about entry 9a, nothing about entry 9, and nothing about the other primary intermediates the same committee classifies as strong sensitisers — among them N,N-bis(2-Hydroxyethyl)-p-phenylenediamine sulfate and Hydroxyethyl-p-phenylenediamine sulphate.² The ingredient declaration on the pack, read against the annex entries, carries information the front-of-pack claim does not. What a given person actually reacts to is a clinical question, and patch testing by a dermatologist is how it is answered.

What we could not verify

How often “PPD-free” means toluene-2,5-diamine. That toluene-2,5-diamine is used as the alternative to p-phenylenediamine is documented — it is the reason the American Contact Dermatitis Society named its sulfate salt Allergen of the Year for 2025. What we could not find is a market survey putting a number on the share of products carrying the claim that contain it. Needle et al. 2026 tested 51 US “PPD-free” dyes for six substances, toluene-2,5-diamine sulfate among them, but the abstract reports counts only for the undisclosed detections, so no proportion appears above.

The overlap figure of 89 patients. Vogel et al. 2017 report 251 patients sensitised to PPD and/or TDA, 231 to PPD and 109 to TDA. The 89 who reacted to both is our arithmetic on those three published figures, not a number the abstract states. We have not read the full text behind the paywall.

Whether any authority has ruled on the claim wording itself. Regulation (EU) No 655/2013 sets the common criteria; we did not locate a national enforcement decision, a Commission technical document annex, or a court judgment specifically addressing “PPD-free” as a claim. The paragraph in §3 states what the criteria require and stops there.

Brazilian enforcement of the new annex. RDC 1.029/2026 was published on 11 June 2026 with a 24-month adaptation period for products already regularised. We could not establish how many hair colourants on sale in Brazil today already meet it, and the entry numbers 8b and 9b are read off the annex as published in ANVISA's legislation system rather than off a consolidated version, because none yet exists.

Annex numbering moves. Every EU entry number above was read off the consolidated text of 1 May 2026 on 6 August 2026. Annex III renumbering follows amending regulations, so if this page is old when you read it, check the entry numbers against the then-current consolidation before quoting them.

Sources

European Parliament and Council. Regulation (EC) No 1223/2009 on cosmetic products, consolidated text of 1 May 2026, Annex III entries 8, 8a, 8b, 9 and 9a. CELEX 02009R1223-20260501 ↩
Scientific Committee on Consumer Safety. Memorandum on hair dye chemical sensitisation, adopted 26 February 2013, Table 1 (potency categories from LLNA and guinea pig assays). SCCS/1509/13 ↩
Vogel TA, Heijnen RW, Coenraads PJ, Schuttelaar MLA. “Two decades of p-phenylenediamine and toluene-2,5-diamine patch testing — focus on co-sensitizations in the European baseline series and cross-reactions with chemically related substances.” Contact Dermatitis 2017;76(2):81–88. 10.1111/cod.12619 (PMID 27555055) ↩
Bacharewicz-Szczerbicka J, Reduta T, Pawłoś A, Flisiak I. “Paraphenylenediamine and related chemicals as allergens responsible for allergic contact dermatitis.” Archives of Medical Science 2021;17(3):714–723. 10.5114/aoms.2019.86709 (PMID 34025842) ↩
Morel OJX, Christie RM. “Current trends in the chemistry of permanent hair dyeing.” Chemical Reviews 2011;111(4):2537–2561. 10.1021/cr1000145 (PMID 21265503) ↩
Atwater AR, Botto N. “Toluene-2,5-Diamine Sulfate: Allergen of the Year 2025.” Dermatitis 2025;36(1):3–11. 10.1089/derm.2024.0384 (PMID 39862176) ↩
Needle CD, Milam EC, Korman A, Flamm AF, Jones DR, Cohen DE, Karagounis TK. “Contact Allergens in ‘PPD-Free’ Hair Dyes.” Dermatitis 2026;37(3):421–427 (published online 24 June 2025). 10.1089/derm.2024.0542 (PMID 40552464) ↩
United States Code. Adulterated cosmetics (Federal Food, Drug, and Cosmetic Act §601), subsections (a), (e) and (g). 21 U.S.C. 361 ↩
United States Code. Safety substantiation (Federal Food, Drug, and Cosmetic Act §608, added by Pub. L. 117-328, div. FF, title III, §3502), subsection (b). 21 U.S.C. 364d ↩
Code of Federal Regulations. Coal tar hair dyes posing a risk of cancer, with the effective-date note: “At 47 FR 7829, Feb. 23, 1982, § 740.18 was stayed until further notice, effective Sept. 18, 1980.” 21 CFR 740.18 ↩
US Food and Drug Administration. Hair Dyes (Cosmetic Products). fda.gov/cosmetics/cosmetic-products/hair-dyes ↩
US Food and Drug Administration. “Hypoallergenic” Cosmetics (Cosmetics Labeling Claims), page dated 18 October 2000 with the 1978 FDA Consumer background article. fda.gov/cosmetics/cosmetics-labeling-claims/hypoallergenic-cosmetics ↩
US Food and Drug Administration. Temporary Tattoos, Henna/Mehndi, and “Black Henna”: Fact Sheet. fda.gov/cosmetics/cosmetic-products/temporary-tattoos-hennamehndi-and-black-henna-fact-sheet ↩
US Food and Drug Administration. Termination of Listing of Color Additives Exempt From Certification; Lead Acetate, published 8 October 2021, effective 6 January 2022 — the order that made effective the final rule published 31 October 2018 at 83 FR 54665. 86 FR 56183 ↩
Agência Nacional de Vigilância Sanitária. Resolução da Diretoria Colegiada RDC nº 907, de 19 de setembro de 2024, Art. 24(II), Art. 34, Art. 35 and Anexo I item II nº 45, consolidated text as amended by RDC nº 949, de 12 de dezembro de 2024. RDC 907/2024 ↩
Agência Nacional de Vigilância Sanitária. Resolução da Diretoria Colegiada Anvisa nº 1.029, de 11 de junho de 2026, internalising MERCOSUR resolution GMC nº 06/25; Anexo entries 8b and 9b, and Art. 2 (24-month adaptation period). Revokes RDC nº 530, de 4 de agosto de 2021. RDC 1.029/2026 ↩
Scientific Committee on Consumer Safety. Opinion on Allergy Alert Test (AAT) as a proof-of-concept study, final Opinion adopted by written procedure on 10 September 2019, §4 Conclusion. SCCS/1607/19 ↩
European Commission. Regulation (EU) No 655/2013 laying down common criteria for the justification of claims used in relation to cosmetic products, Annex, criteria 1(3) and 3(1). CELEX 32013R0655 ↩

Corrections to this article

None yet.

“Clean” Is a Retail Category, Not a Safety Standard

2026-08-06T00:00:00Z

Correction

“Clean” Is a Retail Category, Not a Safety Standard

The claim: “‘Clean’ and ‘non-toxic’ on a pack mean something a regulator checks.”
The short answer: In the US, no — FDA does not define “organic” for cosmetics and has no definition of “clean” or “non-toxic”; MoCRA created none. In the EU there is an enforceable rule, and “free from” claims routinely fail it.
Evidence status: Settled

The Cosmetic Record · Published 2026-08-06 · Last checked 2026-08-06 · 21 sources · Corrections: none

The same word on the same pack, judged in three places.
Market	The rule	What it requires	What it does not do
BR	RDC 907/2024 Art. 11 and Art. 12(I)¹⁹	Labelling “legível, clara, verdadeira e suficiente”, and no wording that misleads as to a product's properties, provenance, nature, origin, composition, admissible purpose or safety.	Defines no marketing term. The strings natural and orgânic do not occur anywhere in the resolution, and it contains no “free from” provision.
US	No rule	Nothing specific to these words. 21 U.S.C. 362(a) deems a cosmetic misbranded “if its labeling is false or misleading in any particular”; the FTC's Green Guides at 16 CFR 260.9 and 260.10 address “free-of” and “non-toxic” claims.⁵⁶	FDA defines neither “organic” nor “hypoallergenic”, and publishes no definition of “clean” or “non-toxic”. MoCRA created none. The Green Guides are guides, not regulations.
EU	Regulation (EC) No 1223/2009 Art. 20(1) and Regulation (EU) No 655/2013⁷⁸	Six binding common criteria — legal compliance, truthfulness, evidential support, honesty, fairness, informed decision-making — applied to every claim, with the substantiation held in the product information file.	Does not define “clean” or “natural” either. It regulates the claim rather than the vocabulary, which is why a “free from” claim can be unlawful without any word being banned.

1. What FDA says about “organic”, in FDA's own words

FDA answers this on a page of its own, in four questions. The first is “Does FDA have a definition for the term ‘organic’?” and the answer begins with the word No.

No. FDA regulates cosmetics under the authority of the Federal Food, Drug, and Cosmetic Act (FD&C Act) and the Fair Packaging and Labeling Act (FPLA). The term “organic” is not defined in either of these laws or the regulations that FDA enforces under their authority.

US Food and Drug Administration, “Organic” Cosmetics. Page dated 8 March 2010, updated 15 September 2010; content current as of 2022-02-25, read 2026-08-06.¹

The definition that does exist belongs to a different department. USDA's Agricultural Marketing Service runs the National Organic Program, whose regulations define “organic” for agricultural ingredients and set labelling tiers by percentage. A cosmetic carrying that seal has satisfied an agriculture rule about how its plant material was grown, not a cosmetic rule — there is no cosmetic rule for the word.

The fourth question is the one that matters most, and FDA also answers it with No: “An ingredient's source does not determine its safety.” Its example is plants, which “whether or not they are organically grown, contain substances that may be toxic or allergenic”; under section 601(a) of the FD&C Act every cosmetic product and ingredient is subject to the same safety requirement regardless of provenance.¹

2. “Clean” and “non-toxic”: no definition, and MoCRA did not create one

FDA's cosmetics site carries a small cluster of pages on labelling claims. The navigation block repeated on each of them lists five: “Alcohol Free”, “Cosmeceutical”, “Cruelty Free”/“Not Tested on Animals”, “Hypoallergenic” and “Organic”.¹² There is no page for “clean”, none for “natural” and none for “non-toxic” — an absence rather than a statement, and section 8 records the limits of the search behind it.

The “Hypoallergenic” page is worth reading precisely because that word looks the most technical of the set. FDA's position, unchanged since 18 October 2000, is flat: “There are no Federal standards or definitions that govern the use of the term ‘hypoallergenic.’ The term means whatever a particular company wants it to mean.” The page explains why: FDA issued a regulation defining the term in 1975, and the US Court of Appeals for the District of Columbia struck it down.²

The Modernization of Cosmetics Regulation Act of 2022 did not change this. MoCRA's definitions sit at 21 U.S.C. 364, and there are five: adverse event, cosmetic product, facility, responsible person and serious adverse event.⁴ Not one is a marketing adjective. What MoCRA added was facility registration, product listing, adverse-event reporting and a duty to hold safety substantiation records — and on that duty FDA's own summary is explicit: “Neither the law nor FDA regulations require specific tests to demonstrate the safety of individual products or ingredients.”³

Two American instruments do bite on the words themselves. 21 U.S.C. 362(a) deems a cosmetic misbranded “if its labeling is false or misleading in any particular”.⁵ And the Federal Trade Commission's Green Guides carry a section on “free-of” claims and a section on “non-toxic” claims — the latter four sentences of guidance and one worked example.

(a) It is deceptive to misrepresent, directly or by implication, that a product, package, or service is non-toxic. Non-toxic claims should be clearly and prominently qualified to the extent necessary to avoid deception. (b) A non-toxic claim likely conveys that a product, package, or service is non-toxic both for humans and for the environment generally. Therefore, marketers making non-toxic claims should have competent and reliable scientific evidence that the product, package, or service is non-toxic for humans and for the environment or should clearly and prominently qualify their claims to avoid deception.

16 CFR 260.10, “Non-toxic claims”, Guides for the Use of Environmental Marketing Claims. Code of Federal Regulations, title 16, volume 1, revised as of 1 January 2024, read 2026-08-06.⁶

Their status is stated in the guides themselves: they “do not confer any rights on any person and do not operate to bind the FTC or the public”, and in any enforcement action the Commission “must prove that the challenged act or practice is unfair or deceptive in violation of Section 5 of the FTC Act”.⁶ That is a long way from a definition a regulator checks before a pack ships.

Something does put criteria behind the word “clean”, though — a retailer, for its own shelves. Sephora's published chemicals policy describes its programme as one that “highlights brands that voluntarily comply with the program's standards to formulate products without certain ingredients”, a “formulated without” list “now in its third version having continually updated the … list since launch in 2018”.²¹ Voluntary, retailer-set, and revised whenever the retailer decides: that is a merchandising category, and it is an accurate description of the word rather than a criticism of it.

3. The EU does have a rule, and it is stricter than people assume

The European parent provision is one sentence in the Cosmetics Regulation.

In the labelling, making available on the market and advertising of cosmetic products, text, names, trade marks, pictures and figurative or other signs shall not be used to imply that these products have characteristics or functions which they do not have.

Regulation (EC) No 1223/2009, Article 20(1), “Product claims”. Consolidated text of 1 May 2026, CELEX 02009R1223-20260501, read 2026-08-06.⁷

Article 20(2) then instructed the Commission to adopt common criteria, and it did: Commission Regulation (EU) No 655/2013, applicable since 11 July 2013. Its annex sets six — legal compliance, truthfulness, evidential support, honesty, fairness, informed decision-making — and the responsible person must keep the substantiation in the product information file. The first of them is the one that catches a claim of absence:

Claims which convey the idea that a product has a specific benefit when this benefit is mere compliance with minimum legal requirements shall not be allowed.

Commission Regulation (EU) No 655/2013, Annex, point 1(3), “Legal compliance”. OJ L 190/31, 11.7.2013, CELEX 32013R0655, read 2026-08-06.⁸

How those criteria apply to the phrase “free from” was written up in 2017 by the Sub-Working Group on Claims, whose Technical document on cosmetic claims gained an Annex III on “free from” claims and an Annex IV on “hypoallergenic”, both applicable from 1 July 2019. The document is candid about its own weight, in capitals on its first page: it “is not a European Commission document” and the views in it “are not legally binding”; it calls itself a “tool” for case-by-case application by the Member States.⁹ It is guidance for national competent authorities rather than law — and it is the clearest published statement of how the six criteria bite, so the next three sections use its worked examples.

4. Worked example one: “paraben-free” does not tell you which paraben

“Paraben” is not one substance and the European annexes do not treat it as one. Five parabens are prohibited outright, at Annex II entries 1374 to 1378: isopropylparaben, isobutylparaben, phenylparaben, benzylparaben and pentylparaben, added by Commission Regulation (EU) No 358/2014. The reason given in recital 7 is not a finding of harm but an absence of a file — the SCCS “noted that limited or no information was submitted by industry” for their safety evaluation, so “the human risk cannot be evaluated”.¹⁰

Two more are capped rather than banned. Annex V entry 12 permits 4-hydroxybenzoic acid and its methyl and ethyl esters at 0,4 % (as acid) for a single ester and 0,8 % for mixtures of esters. Entry 12a covers butylparaben and propylparaben at 0,14 % (as acid) for the sum of the individual concentrations, and forbids them in leave-on products designed for the nappy area of children under three years of age.⁷ That 0.14% is a conversion, not a round number: Commission Regulation (EU) No 1004/2014 recorded the SCCS opinion SCCS/1514/13 concluding that use was safe as long as the sum did not exceed 0,19 % as esters, and recital 10 converted 0,19% as esters into “its equivalent as acid, 0,14 %”.¹¹

So a pack reading “paraben-free” collapses three different regulatory states — banned for want of data, capped at 0.4% as acid, and capped at 0.14% as acid with one product type excluded outright — into a single absence. The EU claims guidance goes further, and treats the phrase as unfair.

Certain parabens are safe when used in accordance to Regulation (EC) No 1223/2009. Considering the fact that all cosmetic products must be safe, the claim ‘free from parabens’ should not be accepted, because it is denigrating the entire group of parabens.

Sub-Working Group on Claims, Technical document on cosmetic claims, version of 3 July 2017, Annex III, “Fairness” row. Endorsed by the Working Group on Cosmetic Products; not legally binding. Read 2026-08-06.⁹

5. Where the paraben story came from, and what that paper actually did

Behind the phrase sits one 2004 paper — the paper the European Commission's consumer-products committee was asked to assess the following year¹⁷ — and it is worth describing by what it did rather than by what it is said to have shown. Darbre and colleagues measured individual parabens in tissue from 20 human breast tumours, supplied in collaboration with the Edinburgh Breast Unit Research Group,¹⁶ by high-pressure liquid chromatography followed by tandem mass spectrometry. The mean total was 20.6 ± 4.2 ng per gram of tissue; methylparaben — the paraben with the lowest oestrogenic activity in the in vitro (laboratory glassware) and in vivo (live animal) assays, since potency rises with the length of the alkyl side chain¹⁷ — was 62% of what was recovered. The stated purpose was to establish that parabens could be found intact in breast tissue at all.¹⁵ It ran in the same issue of the Journal of Applied Toxicology as a four-page commentary by Harvey and Everett on, among other things, “what can be inferred from this type of data (and what can not, such as the cause of these tumours)”.¹⁶

Four things the study did not contain are the whole of the matter: no healthy control tissue, no route of entry, no source for the parabens, and no test of causation. That is not our reading but the EU committee's, which reviewed the paper the following year and recorded both the criticisms and the authors' replies.

Darbre et al. acknowledge that the study on the concentrations of parabens in breast tumours could not identify either the route of entry or the source of the parabens. Carcinogenicity was not considered in the study and the presence of parabens was not claimed to cause the breast tumours.

Scientific Committee on Consumer Products, Extended Opinion on Parabens, underarm cosmetics and breast cancer, SCCP/0874/05, adopted by written procedure 28 January 2005, section 3.1. Read 2026-08-06.¹⁷

The committee's conclusion, in its own words, was “that, in the light of the present knowledge, there is no evidence of demonstrable risk for the development of breast cancer caused by the use of paraben-containing underarm cosmetics”.¹⁷ The later European restrictions on parabens are real, and section 4 above sets them out — but they were made for missing safety data and for infant exposure under a nappy, not for this. A marketing claim built on the 2004 paper is not reporting the regulatory history; it is reporting a headline about it.

6. Worked example two: preservatives are a trade-off the regulators keep adjusting

The second reason a “free from” list ages badly is that the underlying table moves. Take methylisothiazolinone, Annex V entry 57. Commission Regulation (EU) 2016/1198 removed it from leave-on products entirely, recording the SCCS conclusion that “the concentration of 100 ppm of methylisothiazolinone in cosmetic products is not safe for the consumer”; compliance was required from 12 February 2017.¹² Just under a year later Commission Regulation (EU) 2017/1224, citing opinion SCCS/1557/15, cut the surviving rinse-off ceiling from 0,01 % to 0,0015 %.¹³ The 3:1 mixture with methylchloroisothiazolinone sits at entry 39, rinse-off only, also at 0,0015 %, and the consolidated text notes that the two entries “are mutually exclusive”.⁷

Phenoxyethanol is Annex V entry 29, capped at 1,0 %, and it is in a large share of the products the word “clean” is attached to: the survey described in section 7 found it in 591 of 1,470 ingredient lists — 40.2 %, the second commonest potential allergen after fragrances and botanicals.¹⁸ The claims guidance treats a “free from” claim about it exactly as it treats parabens: phenoxyethanol and triclosan “are safe when used according to the Cosmetics Regulation. Hence the claim free from these substances should not be accepted because it is denigrating authorised substances.”⁹

Formaldehyde is the sharpest case, because the pack can be literally true and still fail. The substance itself is prohibited — Annex II entry 1577 — but ingredients that release it are not, and Commission Regulation (EU) 2022/1181 replaced point 2 of the Annex V preamble: the labelling threshold fell from 0,05 % to 0,001 % and the warning changed from “contains formaldehyde” to “releases formaldehyde”. Any finished product containing a listed substance that releases formaldehyde “shall be labelled with the warning ‘releases formaldehyde’ where the total concentration of formaldehyde released in the finished product exceeds 0,001 % (10 ppm)”.⁷¹⁴ The claims guidance draws the obvious conclusion under truthfulness: “The claim ‘free from Formaldehyde’ is not allowed, if the product contains a formaldehyde releasing ingredient (e.g. Diazolidinyl Urea).”⁹

7. What a “free from” label is worth

There is a published measurement of what the category contains. Tran, Comstock and Reeder queried the Target and Walgreens websites for products marketed as “clean” and analysed the ingredient lists of 1,470 of them. A potential contact allergen was present in 93.8% (1,379 of 1,470); fragrances and botanicals were the commonest at 82.9%, and the single commonest entry of all was the word fragrance, perfume or aroma itself, in 68.2% of the lists.¹⁸ That is a survey of labels rather than of patients, and it measures presence, not reactions — but it is the number, and it is not a small one.

The Green Guides reach the same structural problem from the other end. A free-of claim, they say, may be deceptive even when literally true, in two situations: where the product “contains or uses substances that pose the same or similar environmental risks as the substance that is not present”, or where “the substance has not been associated with the product category”.⁶ The first is substitution. The second is being free of something no product of that kind ever contained.

What a “free from” line reports is one decision a formulator made about one ingredient. It does not report what went in instead, at what concentration, or against which annex ceiling — none of which an ingredient list will give you either, for reasons set out in You Cannot Work Out Percentages From an Ingredient List. And the reasoning underneath it is hazard reasoning, applied one substance at a time and detached from dose, product type and site of use, which is the confusion taken apart in Hazard Is Not Risk. The word on the front of the pack is a shelf label. The document that decides what may lawfully be in the bottle is somewhere else, and it is public.

8. What we could not verify

Checked 2026-08-06.

Whether Brazil's organic framework reaches a finished cosmetic. Brazil does define “orgânico”, in Lei nº 10.831/2003, but that statute is about agricultura orgânica: Article 2 defines an organic product as one obtained in an organic agricultural production system or from sustainable extractivism, and Article 3 requires certification by an officially recognised body.²⁰ We searched the text of its implementing decree, Decreto 6.323/2007, on 2026-08-06: the strings cosmétic and higiene do not occur in it. We could not establish, from a primary document, whether or on what terms a finished cosmetic may carry the SisOrg seal, and we make no claim about MAPA certification practice.

FDA's labelling-claims index. Our statement that FDA's cosmetics site addresses exactly five labelling claims comes from the navigation block reproduced inside the two claim pages we did open — “Organic” and “Hypoallergenic”. The index page itself, /cosmetics/cosmetics-labeling-claims, returned HTTP 404 to our request on 2026-08-06, so we could not confirm the list against the index. We did not find, and do not claim to have exhaustively searched for, an FDA statement anywhere on the words “clean” or “non-toxic” in cosmetics.

ISO 16128. The international standard on definitions and criteria for natural and organic cosmetic ingredients is often cited as the definition the market lacks. We did not open it: iso.org returned a JavaScript challenge to our requests and the text is behind a purchase wall. We make no claim about what it defines or whether it sets any threshold.

The size of the retailer lists. Sephora's 2023 chemicals policy describes a “formulated without” list in its third version but does not print it, and we did not open the separate help page that itemises it. We give no count of excluded ingredients for that programme or any other, and no comparison between retailers.

Enforcement. We could not establish from a primary source whether the FTC has ever brought an action over a “clean” or “non-toxic” claim on a cosmetic, or whether any EU national authority has published a decision applying Annex III of the claims technical document. The rules quoted above are the rules as written; nothing here says how often they are enforced.

Sources

US Food and Drug Administration, “Organic” Cosmetics; page dated 8 March 2010, updated 15 September 2010, content current as of 25 February 2022. fda.gov/cosmetics/cosmetics-labeling-claims/organic-cosmetics. ↩
US Food and Drug Administration, “Hypoallergenic” Cosmetics; statement dated 18 October 2000, with the reprinted FDA Consumer article of April 1978; content current as of 25 February 2022. fda.gov/cosmetics/cosmetics-labeling-claims/hypoallergenic-cosmetics. ↩
US Food and Drug Administration, Modernization of Cosmetics Regulation Act of 2022 (MoCRA); content current as of 13 March 2026. fda.gov · MoCRA. ↩
United States Code, Federal Food, Drug, and Cosmetic Act section 604, Definitions — adverse event, cosmetic product, facility, responsible person, serious adverse event. 21 U.S.C. 364. ↩
United States Code, Federal Food, Drug, and Cosmetic Act section 602, Misbranded cosmetics, subsection (a). 21 U.S.C. 362. ↩
Federal Trade Commission, Guides for the Use of Environmental Marketing Claims — § 260.1 (status), § 260.9 (free-of claims), § 260.10 (non-toxic claims). Code of Federal Regulations title 16, volume 1, revised as of 1 January 2024. 16 CFR Part 260. ↩
European Parliament and Council, Regulation (EC) No 1223/2009 on cosmetic products, consolidated text of 1 May 2026 — Article 20; Annex II entries 1374–1378 and 1577; Annex V preamble point 2 and entries 12, 12a, 29, 39, 57. CELEX 02009R1223-20260501. ↩
European Commission, Commission Regulation (EU) No 655/2013 of 10 July 2013 laying down common criteria for the justification of claims used in relation to cosmetic products, OJ L 190/31, 11.7.2013. CELEX 32013R0655. ↩
Sub-Working Group on Claims, Technical document on cosmetic claims, version of 3 July 2017, endorsed by the Working Group on Cosmetic Products — Annex III (“free from” claims) and Annex IV (hypoallergenic claim), applicable from 1 July 2019. Not a European Commission document and not legally binding. The document itself (PDF) still resolves; the docsroom landing page for the same record now redirects to a CIRCABC URL that returns 404, checked 2026-08-06. ec.europa.eu/docsroom · document 24847, attachment 1 (PDF). ↩
European Commission, Commission Regulation (EU) No 358/2014 of 9 April 2014 amending Annexes II and V to Regulation (EC) No 1223/2009 — adds isopropylparaben, isobutylparaben, phenylparaben, benzylparaben and pentylparaben to Annex II as entries 1374–1378. OJ L 107/5, 10.4.2014. CELEX 32014R0358. ↩
European Commission, Commission Regulation (EU) No 1004/2014 of 18 September 2014 amending Annex V to Regulation (EC) No 1223/2009 — recitals 5, 7 and 10 (opinion SCCS/1514/13; 0,19 % as esters converted to 0,14 % as acid) and recital 12 (nappy area). OJ L 282/5, 26.9.2014. CELEX 32014R1004. ↩
European Commission, Commission Regulation (EU) 2016/1198 of 22 July 2016 amending Annex V to Regulation (EC) No 1223/2009 — methylisothiazolinone restricted to rinse-off products. OJ L 198/10, 23.7.2016. CELEX 32016R1198. ↩
European Commission, Commission Regulation (EU) 2017/1224 of 6 July 2017 amending Annex V to Regulation (EC) No 1223/2009 — rinse-off maximum for methylisothiazolinone reduced to 0,0015 %, citing SCCS/1557/15. OJ L 174/16, 7.7.2017. CELEX 32017R1224. ↩
European Commission, Commission Regulation (EU) 2022/1181 of 8 July 2022 amending the preamble of Annex V to Regulation (EC) No 1223/2009 — Article 1 replaces preamble point 2; recitals 3 and 4 record the move from “contains formaldehyde” above 0,05 % to “releases formaldehyde” above 0,001 % (10 ppm), on SCCS advice SCCS/1632/21 of 7 May 2021. OJ L 184/3, 11.7.2022. CELEX 32022R1181. ↩
Darbre PD, Aljarrah A, Miller WR, Coldham NG, Sauer MJ, Pope GS, “Concentrations of parabens in human breast tumours”, Journal of Applied Toxicology 2004;24(1):5–13. 10.1002/jat.958 · PMID 14745841. ↩
Harvey PW, Everett DJ, “Significance of the detection of esters of p-hydroxybenzoic acid (parabens) in human breast tumours”, Journal of Applied Toxicology 2004;24(1):1–4. 10.1002/jat.957 · PMID 14745840. ↩
Scientific Committee on Consumer Products, Extended Opinion on Parabens, underarm cosmetics and breast cancer (running title: Opinion on the safety evaluation of parabens, underarm cosmetics and breast cancer), adopted by written procedure on 28 January 2005 — background, section 3.1 and the conclusion. SCCP/0874/05. ↩
Tran JM, Comstock JR, Reeder MJ, “Natural Is Not Always Better: The Prevalence of Allergenic Ingredients in ‘Clean’ Beauty Products”, Dermatitis 2022;33(3):215–219. 10.1097/DER.0000000000000863 · PMID 35256558. ↩
Agência Nacional de Vigilância Sanitária, Resolução RDC nº 907, de 19 de setembro de 2024 — Art. 11 and Art. 12(I); consolidated text as amended, read 2026-08-06. Its Art. 51 revokes RDC nº 752/2022, which carried the same two articles word for word; a page citing 752/2022 for Brazilian labelling is citing a revoked text. RDC 907/2024. ↩
Presidência da República, Lei nº 10.831, de 23 de dezembro de 2003, dispõe sobre a agricultura orgânica — Art. 1º, Art. 2º and Art. 3º. Lei 10.831/2003. ↩
Sephora, Sephora Public Chemicals Policy 2023, page 3, on the “Clean at Sephora” programme and its “formulated without” list. inside-sephora.com · Public Chemicals Policy 2023. ↩

Corrections to this article

None yet.

You Cannot Work Out Percentages From an Ingredient List

2026-08-06T00:00:00Z

Correction

You Cannot Work Out Percentages From an Ingredient List

The claim: “If it is third on the list, you can work out roughly how much is in there.”
The short answer: The label cannot tell you. The order rule ranks ingredients by weight down to 1%, and below that line they may be listed in any order. You can locate the 1% line. You cannot derive a percentage from it.
Evidence status: Settled

The Cosmetic Record · Published 2026-08-06 · Last checked 2026-08-06 · 18 sources · Corrections: none

In what order must an ingredient list be written, and in what language?
Market	The rule	What it requires	What it does not do
BR	RDC 907/2024 Art. 13(II)(i) and Art. 20; RDC 898/2024 Art. 2¹¹¹²	The composition on the secondary packaging in INCI, plus the same composition in Brazilian Portuguese — either in brackets beside each INCI name or as a second list “na mesma ordem dos ingredientes em INCI”.	No rule fixes the order. The strings ordem, decrescente, proporção and 1% do not occur anywhere in the resolution.
US	21 CFR 701.3; 21 CFR 701.2(b)²³	“Descending order of predominance”, or — at the labeler's option — three groups: above 1% in descending order, then 1% and under in any order, then colour additives in any order. Label in English.	Says nothing on the pack about which of the two schemes was used. English is not required where a product is distributed solely in Puerto Rico or a Territory whose predominant language is not English.
EU	Regulation (EC) No 1223/2009 Art. 19(1)(g), 19(5), 19(6)¹	Descending order of weight at the time of adding; under 1 % in any order; non-hair colorants in any order at the end; the glossary name for every ingredient.	No rule puts the list in a national language. Art. 19(5) hands points (b), (c), (d) and (f) to Member State law and leaves point (g) out.

1. The rule, quoted from both jurisdictions

The European ordering obligation is two sentences long, and those two are the whole of it.

The list of ingredients shall be established in descending order of weight of the ingredients at the time they are added to the cosmetic product. Ingredients in concentrations of less than 1 % may be listed in any order after those in concentrations of more than 1 %.

Regulation (EC) No 1223/2009, Article 19(1)(g), fourth sub-paragraph. Consolidated text of 1 May 2026, CELEX 02009R1223-20260501, read 2026-08-06.¹

The American rule says the same thing in a different vocabulary, then offers something the European text does not. 21 CFR 701.3(a) requires “the name of each ingredient in descending order of predominance”. Paragraph (f) makes the 1% line explicit.

As an alternative to listing all ingredients in descending order of predominance, ingredients may be grouped and the groups listed in the following manner and order: (1) Ingredients, other than color additives, present at a concentration greater than 1 percent, in descending order of predominance; followed by (2) Ingredients, other than color additives, present at a concentration of not more than 1 percent, without respect to order of predominance; followed by (3) Color additives, without respect to order of predominance. Ingredients specified in paragraph (f)(2) of this section may be included with those specified in paragraph (f)(1) of this section and listed in descending order of predominance.

21 CFR 701.3(f), “Designation of ingredients”. Code of Federal Regulations, title 21, volume 7, revised as of 1 April 2024, read 2026-08-06 and checked the same day against the current e-CFR text, which is identical in every paragraph quoted here.²¹⁸

Nothing on an American pack tells a reader which of the two schemes the labeler chose. Both are lawful, both produce a list that looks identical, and the last sentence of (f)(3) lets a labeler move sub-1% ingredients back up into the sorted group as well. Only one of the resulting lists is sorted below the 1% line, and the pack does not say which.

Brazil is the case no English-language article covers, and the instrument usually cited for it is no longer in force: RDC 752/2022 was revoked by Art. 51(I) of RDC 907/2024, D.O.U. 23 September 2024. The successor carries the labelling article forward word for word. Among the data that must appear on secondary packaging, Art. 13(II)(i) requires “ingredientes ou composição (utilizando a codificação de substâncias INCI)”. That is the entire instruction, because it contains no instruction about order. We searched the consolidated text of RDC 907/2024 on 2026-08-06 for ordem, decrescente, proporção and 1%: none of those strings occurs anywhere in it.¹¹

2. “At the time they are added” — what that phrase quietly excludes

The European sentence does not rank ingredients by their weight in the product a reader holds. It ranks them by weight at the moment they went into the vessel. That is a fact about a batch record, and a batch record is not published.

Two further exclusions sit in the definition immediately above the ordering sentence. An ingredient means “any substance or mixture intentionally used in the cosmetic product during the process of manufacturing”, and the text then removes two categories from the count: “impurities in the raw materials used”, and “subsidiary technical materials used in the mixture but not present in the final product”.¹ Processing aids that do not survive into the finished product are absent by operation of the definition, not by anyone's omission; so is anything present as a contaminant of a raw material. Note also that the definition covers a mixture and not only a substance.

3. Finding the 1% line, and the three markers usually offered for it

The one thing the list genuinely encodes is a boundary. Two markers locate it, both inferences from published ceilings rather than from anything the pack states. A third is offered constantly and does not work.

A preservative at an annex ceiling of 1% or below. Phenoxyethanol is Annex V entry 29 of Regulation 1223/2009, capped at 1,0 % in the ready-for-use preparation.¹ It cannot lawfully be above 1%, and on a compliant European list everything above 1% precedes everything below it — so phenoxyethanol and every name printed after it lie in the unordered tail.

The colour block. Under Article 19(1)(g), colorants other than hair colorants “may be listed in any order after the other cosmetic ingredients”; under 21 CFR 701.3(f)(3), colour additives are listed “without respect to order of predominance”. By the CI numbers, the list has stopped being sorted in both jurisdictions.

The marker that fails: a named fragrance allergen. The Annex III entries carved out of “parfum” carry a disclosure floor, not a ceiling. The substance is named only when it exceeds 0,001 % in leave-on products or 0,01 % in rinse-off products, so the name means the substance is above 0.001% and nothing else — a floor three orders of magnitude beneath the 1% line, and no evidence about where that line falls.

What locating the line buys is exactly two bounds: everything below it is at or under 1%, everything above it is at or over 1%. It buys no value for any single ingredient, and it assumes the label is compliant, which is an assumption and not an observation.

4. What the list still cannot tell you, however carefully you read it

The gap between the third name and the fourth can be forty percentage points or four tenths of one; the notation is identical either way. The list does not state the grade or purity of a raw material, the pH of the finished product, whether an ingredient is encapsulated or free, or the ratio at which a plant extract was extracted. A single name covers all of those.

Nor do the other marks on the pack help. Annex VII of Regulation 1223/2009 is titled “Symbols used on packaging/container” and contains exactly three items: “1. Reference to enclosed or attached information”, “2. Period-after-opening”, “3. Date of minimum durability”. The open-jar symbol at point 2 appears where durability exceeds 30 months, followed by a period in months or years.¹ None of the three is a quantity.

One structural fact is worth having. Article 19(5) fixes the language of points (b), (c), (d) and (f) of paragraph 1 — and pointedly not point (g). The ingredient list is the one block of a European label that reads identically in every Member State, in a nomenclature that is nobody's national language.

5. “Parfum”: one word, and the allergens carved out of it

Perfume and aromatic compositions and their raw materials shall be referred to by the terms ‘parfum’ or ‘aroma’. Moreover, the presence of substances, the mention of which is required under the column ‘Other’ in Annex III, shall be indicated in the list of ingredients in addition to the terms parfum or aroma.

Regulation (EC) No 1223/2009, Article 19(1)(g), third sub-paragraph. Consolidated text of 1 May 2026, CELEX 02009R1223-20260501, read 2026-08-06.¹

One word therefore stands for a composition that may run to dozens of substances, and the carve-out is the only crack in it. The set of 26 named substances is older than it is usually credited: it was the SCCNFP, the predecessor committee, that “identified in 1999 a set of 26 fragrance allergens with a well-recognised potential to cause allergy”. The Scientific Committee on Consumer Safety reviewed that list thirteen years later, in opinion SCCS/1459/11, adopted at its 15th plenary of 26–27 June 2012, confirmed the 1999 set was still relevant, and went further: on clinical experience alone it classified 82 substances as established contact allergens in humans, 54 single chemicals and 28 natural extracts.⁶

The named substances sit in Annex III. Entry 67 is 2-Benzylideneheptanal, better known as Amyl cinnamal; entries 91 and 92 are Oak moss extract and Treemoss extract; entry 45 is Benzyl Alcohol. Entries 68 and 79 no longer exist — Commission Regulation (EU) 2017/1410 deleted 79, removing HICC from use altogether.⁷ The Commission counts what survives in recital 5 of the amending regulation below: “Currently, 24 fragrance allergens listed in entries 45 and 67 to 92 of Annex III … are to be mentioned in the list of ingredients (individually labelled).”

Commission Regulation (EU) 2023/1545, in the Official Journal of 27 July 2023, extends individual labelling considerably: its recital 6 records 56 further fragrance allergens identified by the SCCS beyond those 24, and its Annex adds new Annex III entries 327 to 371 while replacing a block of existing ones. The transition sits in a footnote to those entries: “Cosmetic products containing that substance that do not comply with the restrictions may be placed on the Union market until 31 July 2026 and made available on the Union market until 31 July 2028.”⁵ The first of those dates has now passed.

The United States has none of this yet. 21 CFR 701.3(a) permits fragrance to be listed as fragrance, full stop. MoCRA added section 609(b) to the Federal Food, Drug, and Cosmetic Act, requiring each fragrance allergen to be identified on the label, with the substances “determined by the Secretary by regulation” and a proposed rule due “not later than 18 months after December 29, 2022” — 29 June 2024.⁹ A Federal Register query restricted to FDA documents mentioning fragrance allergens, run on 2026-08-06, returns no such rule.¹⁶ The Unified Agenda entry, RIN 0910-AI90, “Disclosure of Fragrance Allergens in Cosmetic Labeling”, lists the NPRM at 11/00/2026.¹⁰

6. “+/-”, “may contain”, “and other ingredients”, “(nano)”

Four notations regularly get read as concentration information. None of them is.

“May contain” and “+/-” are a shade-range device. Article 19(1)(g) allows a decorative range marketed in several shades to list all its non-hair colorants together provided the words “may contain” or the symbol “+/-” are added, using CI nomenclature. In the United States the device is narrower: 21 CFR 701.3(g) permits an ingredient not in the product to be declared under “may contain” only where it is a colour additive, and (g)(2) attaches four conditions to the shared-declaration route.² In both cases the phrase means the substance may be absent entirely.

“And other ingredients” is a United States trade-secret ending. 21 CFR 701.3(a) allows it where an ingredient “is accepted by the Food and Drug Administration as exempt from public disclosure pursuant to the procedure established in § 720.8(a)”, and 21 CFR 720.8(b) sets out the six factors FDA weighs in deciding whether an ingredient's identity qualifies as a trade secret, beginning with “the extent to which the identity of the ingredient is known outside petitioner's business”.⁴ There is no European equivalent on the label: the words confidential and trade secret occur zero times in the consolidated text of Regulation 1223/2009, searched on 2026-08-06.¹

“(nano)” is the one qualifier in the whole list that describes physical form rather than identity: ingredients present as nanomaterials “shall be followed by the word ‘nano’ in brackets”. It is still not a quantity.

7. “Aqua”, thermal spring water and Latin binomials

Article 19(6) requires the ingredient list to be “expressed by using the common ingredient name set out in the glossary provided for in Article 33”. That glossary is reissued whole rather than amended, so the entry numbers move: Decision (EU) 2019/701 was repealed by Implementing Decision (EU) 2022/677, and that by Commission Implementing Decision (EU) 2025/1175 of 16 June 2025, applicable from 30 July 2026 and therefore the version quoted here. Its annex is one alphabetical table of 30,418 numbered entries.⁸ Entry 2112 is AQUA. Entry 16016 is MARIS AQUA and entry 16018 MARIS SAL — sea water and sea salt have names of their own.

Thermal spring water does not. We searched the whole annex on 2026-08-06: no entry anywhere in it contains the string SPRING, and all 33 entries containing THERM name a micro-organism or a derived peptide — STREPTOCOCCUS THERMOPHILUS FERMENT at entry 27941, for instance — not a water.

So the finding is flat, and it is the answer for anyone who arrived here from a thermal-water page. “Thermal spring water” has no INCI name. On a European label it is declared as Aqua, exactly like water from any other source. Thermal water is a sourcing and marketing category defined by origin and mineral content, and neither is a nomenclature entry. A pack reading “Aqua (Thermal Spring Water)” is the glossary name plus a parenthesis the glossary does not supply.

Latin binomials work the same way. A botanical entry names a species and a plant part, and the glossary carries them at extraordinary length — entry 4 is one flower-extract name listing a hundred species. A binomial fixes identity: which plant, which part. It fixes neither how much nor how concentrated, and an “extract” may be anything from a strong tincture to a trace in glycerin.

The American ladder is older and fixed at its editions. 21 CFR 701.3(c) sends a labeler first to the name specified in § 701.30, then to the CTFA Cosmetic Ingredient Dictionary, second edition, 1977 and four other compendia incorporated by reference, then to “the name generally recognized by consumers”, and only then to “the chemical or other technical name”.²

8. Having an INCI name is not approval

This is the misreading that follows the others. A name on a list is a name, and the two bodies that assign and adopt those names both say so in writing. It is the same mistake in a different costume as reading a notification as an approval, which is what “Grau 2” does not mean in Brazil.

The Commission shall compile and update a glossary of common ingredient names. To this end, the Commission shall take account of internationally recognised nomenclatures including the International Nomenclature of Cosmetic Ingredients (INCI). That glossary shall not constitute a list of the substances authorised for use in cosmetic products.

Regulation (EC) No 1223/2009, Article 33, first paragraph. Consolidated text of 1 May 2026, CELEX 02009R1223-20260501, read 2026-08-06.¹

The Commission repeats it on the CosIng glossary page — “This glossary does not constitute a list of ingredients authorised for use in cosmetic products” — and adds that it “does not establish an exhaustive list of ingredients and it is subject to updating”.¹⁵ The names themselves are developed by the International Nomenclature Committee; the Personal Care Products Council oversees the programme and publishes them in the International Cosmetic Ingredient Dictionary and Handbook. The Council's own page is unambiguous: “just because an ingredient has an INCI name does not mean that the ingredient has been approved for cosmetics”, and “the assignment of an INCI name to an ingredient also does not imply that the ingredient is safe, or that its use in a cosmetic product complies with the laws and regulations of the United States or other global regions”.¹⁴

An ingredient list is a nomenclature laid out in a partial order. It identifies, it ranks down to a line, and it stops. Everything a reader wants beyond that — the concentration, the grade, the permission — has to come from somewhere else.

What we could not verify

Checked 2026-08-06.

The Mercosur original. RDC 907/2024 Art. 2 incorporates MERCOSUR/GMC/RES. Nº 48/2021, the technical regulation on labelling for personal hygiene products, cosmetics and perfumes. We could not open that instrument at source: normas.mercosur.int returned HTTP 403 to our requests on 2026-08-06. We read the annexed technical regulation instead through Uruguay's Decreto 416/022, which reproduces it in full, where item 12 of clause 4.1 reads “Ingredientes/Composición (utilizando la codificación de sustancias INCI)” and no ordering or 1% clause appears anywhere in it.¹³ We state the Brazilian position from the Brazilian instrument, whose text we did open, and we cannot rule out wording in the Mercosur original that the Uruguayan reproduction does not carry.

The status of the US trade-secret route. 21 CFR 701.3(a) routes “and other ingredients” through § 720.8(a), which sits in a part titled “Voluntary filing of cosmetic product ingredient composition statements”. Part 720 is still in the Code of Federal Regulations, and FDA stopped accepting submissions to the Voluntary Cosmetic Registration Program “effective March 27, 2023”, in view of the mandatory listing regime MoCRA created.¹⁷ Whether an exemption can still be granted under a filing route that no longer accepts filings is a question FDA has not answered in any document available to us. We report the regulation as published and make no claim about current agency practice.

The EU fragrance-allergen count. We quote recitals 5 and 6 of Regulation (EU) 2023/1545 — 24 individually labelled, 56 further ones identified — and the entry ranges as printed. We publish no arithmetic total of our own, because new entries, replaced entries and deleted entries interact and we could not reconcile them against a Commission statement of the final count.

ANVISA's translation list. RDC 898/2024 Art. 3 requires the Portuguese composition to follow a list of translated ingredient names hosted on the Anvisa portal. We did not open that list and make no claim about what is in it.

Sources

European Parliament and Council, Regulation (EC) No 1223/2009 on cosmetic products, consolidated text of 1 May 2026 — Article 19, Article 33, Annex III entries 45 and 67–92 and 327–371, Annex V entry 29, Annex VII. CELEX 02009R1223-20260501. ↩
US Food and Drug Administration, Designation of ingredients, Code of Federal Regulations title 21, revised as of 1 April 2024. 21 CFR 701.3. ↩
US Food and Drug Administration, Form of stating labeling requirements, Code of Federal Regulations title 21, revised as of 1 April 2024. 21 CFR 701.2. ↩
US Food and Drug Administration, Confidentiality of statements, Code of Federal Regulations title 21, revised as of 1 April 2024. 21 CFR 720.8. ↩
European Commission, Commission Regulation (EU) 2023/1545 of 26 July 2023 amending Regulation (EC) No 1223/2009 as regards labelling of fragrance allergens in cosmetic products, OJ 27.7.2023. CELEX 32023R1545. ↩
Scientific Committee on Consumer Safety, Opinion on fragrance allergens in cosmetic products, adopted at the 15th plenary meeting of 26–27 June 2012; ISBN 978-92-79-30752-2. SCCS/1459/11. ↩
European Commission, Commission Regulation (EU) 2017/1410 of 2 August 2017 amending Annexes II and III to Regulation (EC) No 1223/2009 — deletes Annex III entry 79 (HICC). CELEX 32017R1410. ↩
European Commission, Commission Implementing Decision (EU) 2025/1175 of 16 June 2025 laying down rules for the application of Regulation (EC) No 1223/2009 … as regards the glossary of common ingredient names for use in the labelling of cosmetic products and repealing Commission Implementing Decision (EU) 2022/677, OJ 10.7.2025; applicable from 30 July 2026 — 30,418 annex entries, including 2112 (AQUA), 16016 (MARIS AQUA), 16018 (MARIS SAL), 27941 (STREPTOCOCCUS THERMOPHILUS FERMENT). CELEX 32025D1175. ↩
United States Code, Federal Food, Drug, and Cosmetic Act section 609, Labeling, subsection (b), fragrance allergens. 21 U.S.C. 364e. ↩
Office of Information and Regulatory Affairs, Unified Agenda entry, Disclosure of Fragrance Allergens in Cosmetic Labeling, HHS/FDA, proposed rule stage. RIN 0910-AI90. ↩
Agência Nacional de Vigilância Sanitária, Resolução RDC nº 907, de 19 de setembro de 2024, D.O.U. 23/09/2024 — Art. 2 (Mercosur incorporation), Art. 13(II)(i), Art. 20 and its sole paragraph, and Art. 51(I), which revokes RDC 752/2022. RDC 907/2024. ↩
Agência Nacional de Vigilância Sanitária, Resolução RDC nº 898, de 28 de agosto de 2024, D.O.U. 02/09/2024 — Art. 2 § 4, Art. 3, Art. 7. RDC 898/2024. ↩
República Oriental del Uruguay, Decreto 416/022, incorporating MERCOSUR/GMC/RES. Nº 48/21, Reglamento Técnico MERCOSUR sobre Rotulado para Productos de Higiene Personal, Cosméticos y Perfumes — clause 4.1, item 12. Decreto 416/022. ↩
Personal Care Products Council, International Nomenclature of Cosmetic Ingredients (INCI) — on the International Nomenclature Committee and on what an INCI name does not imply. personalcarecouncil.org/resources/inci. ↩
European Commission, CosIng — Glossary of ingredients, on the glossary established under Article 33 by Commission Decision (EU) 2019/701. single-market-economy.ec.europa.eu · CosIng glossary. ↩
Office of the Federal Register, documents API, query restricted to Food and Drug Administration documents matching “fragrance allergen”, run 2026-08-06; no proposed rule on fragrance allergen labelling for cosmetics is returned. federalregister.gov/api/v1/documents.json. ↩
US Food and Drug Administration, FDA Has Stopped Accepting Submissions to the Voluntary Cosmetic Registration Program (VCRP), Constituent Update, 27 March 2023. fda.gov · VCRP constituent update. ↩
Office of the Federal Register and Government Publishing Office, Electronic Code of Federal Regulations, title 21 current text of § 701.3, title 21 up to date as of 4 August 2026, read 2026-08-06. ecfr.gov · 21 CFR 701.3. ↩

Corrections to this article

None yet.

Where a Cosmetic Stops and a Drug Begins

2026-08-06T00:00:00Z

Explainer

Where a Cosmetic Stops and a Drug Begins

The claim: “If a cream is on the shelf saying it does that, somebody approved the claim.”
The short answer: Intended use decides the category, and the marketing is the evidence. The same molecule in the same jar is a cosmetic or a drug depending on the sentence on the pack. Nobody pre-approves a cosmetic; MoCRA registration is not approval.
Evidence status: Settled

The Cosmetic Record · Published 2026-08-06 · Last checked 2026-08-06 · 26 sources · Corrections: none

What legally counts as a cosmetic, and who says so?
Market	The rule	What it requires	What it does not do
BR	ANVISA RDC 907/2024 Art. 3 XVI–XVIII, Art. 12, Art. 34, Art. 35, Anexo I	A preparação for external use whose purpose is exclusively or mainly to clean, perfume, alter appearance, correct body odour or protect the body or keep it in good condition. Every product is graded Grau 1 or Grau 2 by its area of application, its intended public, specific formulation conditions or the sanitary impact of its declared purposes of use. Nine groups — among them protetor solar — need registro; everything else is filed by comunicação prévia.	Notification is a filing, not an assessment. Art. 12 bans therapeutic and disinfectant wording on the label, but no one reads the sentence before it reaches a shelf.
US	FD&C Act §201(i) and §201(g)(1), 21 U.S.C. 321; intended use per 21 CFR 201.128	Nothing before sale except colour-additive approval. Since MoCRA: facility registration and product listing (21 U.S.C. 364c), from which certain small businesses are exempt (21 U.S.C. 364h), and held records adequately substantiating safety (21 U.S.C. 364d).	No rule requires a cosmetic product or its ingredients, other than colour additives, to be approved before going on the market — and FDA states that registration and listing “is neither a cosmetic approval program nor a promotional tool”.
EU	Regulation (EC) No 1223/2009, Art. 2(1)(a), Art. 20 and Annexes II–VI (CELEX 02009R1223-20260501)	A substance or mixture for the external parts of the body, with an exclusive or main cosmetic purpose. A named responsible person, a product information file, a safety assessment, notification to the CPNP, and positive lists for colourants, preservatives and UV filters.	No authority signs off an individual product before it is placed on the market. Under CELEX 32013R0655 a claim that a product has been approved by a competent authority is itself prohibited.

1. Two definitions, one jar: intended use decides, and the marketing is the evidence

The United States defines both categories in a single section of one statute, and it defines them by purpose rather than by composition:

The term "cosmetic" means (1) articles intended to be rubbed, poured, sprinkled, or sprayed on, introduced into, or otherwise applied to the human body or any part thereof for cleansing, beautifying, promoting attractiveness, or altering the appearance, and (2) articles intended for use as a component of any such articles; except that such term shall not include soap.

The term "drug" means ... (B) articles intended for use in the diagnosis, cure, mitigation, treatment, or prevention of disease in man or other animals; and (C) articles (other than food) intended to affect the structure or any function of the body of man or other animals ...

Federal Food, Drug, and Cosmetic Act §201(i) and §201(g)(1), 21 U.S.C. 321(i) and (g)(1) — text in force as of 6 August 2026.¹

Neither definition mentions an ingredient, a concentration or a mechanism. Both hinge on the word intended, and that word has its own regulation. Under 21 CFR 201.128, intended use “refer[s] to the objective intent of the persons legally responsible for the labeling of an article”, and that intent “may, for example, be shown by labeling claims, advertising matter, or oral or written statements by such persons or their representatives”.² The marketing is not evidence of the category in a loose journalistic sense. It is the evidence the regulation names.

FDA gives three examples of how intended use is established: claims in labelling, advertising or on the internet; consumer perception, established through a product's reputation; and ingredients carrying a well-known therapeutic use, its example being fluoride in toothpaste. The same page gives the claims that flip a product across the line — “restore hair growth, reduce cellulite, treat varicose veins, increase or decrease the production of melanin (pigment) in the skin, or regenerate cells” — and disposes of a marketing word: the Act “does not recognize any such category as ‘cosmeceuticals’ … the term ‘cosmeceutical’ has no meaning under the law”.³

This is enforced on live copy. In a warning letter dated 14 May 2026, FDA told a Minnesota seller its skin creams were unapproved new drugs, citing as evidence of intended use “as defined in 21 CFR 201.128” the exact sentences on the seller's own product pages: “Reduce acne, freckles, dark spots and tighten pores.”⁴ The formula was not what moved the product into the drug category. The webpage was.

A product can also be both at once — FDA's example is an antidandruff shampoo, a cosmetic because it cleanses the hair and a drug because it treats dandruff.³ The regulations say it in one line: “Any cosmetic product which is also a drug or device or component thereof is also subject to the requirements of Chapter V of the act.”⁵

2. Nobody approves a cosmetic — and MoCRA registration is not approval

FDA's own summary of its authority opens with the point: “The law does not require cosmetic products and ingredients, other than color additives, to have FDA approval before they go on the market.”⁶ A drug takes the opposite route — an approved New Drug Application, or conformity with a monograph FDA has finalised for its category.³

The Modernization of Cosmetics Regulation Act of 2022 changed a great deal and did not change that. It added, at FD&C §607 (21 U.S.C. 364c), a duty on facilities manufacturing or processing cosmetic products for US distribution to register and renew biennially, and a duty on the responsible person to list each marketed product with its ingredients.⁸ Both duties fall away, at §612 (21 U.S.C. 364h), for responsible persons and facilities whose average annual US cosmetic sales over the previous three years are under $1,000,000, adjusted for inflation — unless they handle one of four listed product kinds: eye-mucosa contact, injected, internal use, or appearance-altering for more than 24 hours.²⁶ It added, at §608 (21 U.S.C. 364d), a duty to hold records showing “adequate substantiation of safety”, with an explicit carve-out for coal-tar hair dye.⁹ Neither is a licence.

The scale is the scale of a filing system, not a review system: FDA's published summary table records 16,398 unique, active facility registrations and 1,298,361 unique, active product listings, data as of 30 June 2026. The same page carries the sentence that settles the question:

Cosmetic product facility registration and product listing is neither a cosmetic approval program nor a promotional tool. FDA does not issue “certificates” for cosmetic product facility registrations or product listings.

US Food and Drug Administration, Registration & Listing of Cosmetic Product Facilities and Products, “Important Notes”, page read 6 August 2026.¹⁰

MoCRA also gave the phrase “cosmetic product” a statutory meaning that will matter in section 4 below: “a preparation of cosmetic ingredients with a qualitatively and quantitatively set composition for use in a finished product”.⁷

3. The same line in Europe and in Brazil

The European definition is drawn the same way — by purpose, with the word exclusively or mainly doing the work:

‘cosmetic product’ means any substance or mixture intended to be placed in contact with the external parts of the human body (epidermis, hair system, nails, lips and external genital organs) or with the teeth and the mucous membranes of the oral cavity with a view exclusively or mainly to cleaning them, perfuming them, changing their appearance, protecting them, keeping them in good condition or correcting body odours

Regulation (EC) No 1223/2009, Article 2(1)(a) — consolidated text of 1 May 2026, CELEX 02009R1223-20260501.¹¹

A product whose main purpose is to treat a disease is not covered by that definition and falls to the medicinal-products regime instead. Article 20(1) forbids text, names, trade marks, pictures and other signs “to imply that these products have characteristics or functions which they do not have”, and Regulation (EU) No 655/2013 sets the common criteria that give it teeth. Its first criterion is the one an English-language reader rarely expects: “Claims that indicate that the product has been authorised or approved by a competent authority within the Union shall not be allowed.”¹² In the EU it is not merely untrue to say a cosmetic was approved. It is prohibited to say it.

Brazil arrives at the same boundary through different furniture. Art. 3 XVI of RDC 907/2024 defines the class as “preparações” for external use with an exclusive or main cosmetic purpose; Art. 3 XVII–XVIII then splits it into Grau 1 and Grau 2 by four criteria, of which the last is the sanitary impact of “das finalidades de uso declaradas” — the declared purposes of use. What the product says it is for is written into the risk class itself, as intended use is written into the American category. (Grau 2 is a risk grade and not a registration status, a distinction we take apart separately in In Brazil, “Grau 2” Does Not Mean “Registered”.) The label rule states the prohibition directly:

Art. 12. A rotulagem não deve conter nome comercial, marcas, imagens, links eletrônicos ou dizeres que:
[...]
II - representem alegações terapêuticas atribuídas ao uso do produto ou de seus ingredientes, como, por exemplo, prevenção ou tratamento de hematomas, feridas, rachaduras, dores, inflamações, câimbras, varizes, pediculose [...];
[...]
IV - mencionem a propriedade desinfetante ou que se refiram à eliminação completa de microrganismos

ANVISA, Resolução da Diretoria Colegiada RDC nº 907, de 19 de setembro de 2024, Art. 12, incisos II and IV — consolidated text as amended by RDC nº 949, de 12 de dezembro de 2024. Art. 51 I of the same resolution revoked RDC 752/2022.¹³

Nine groups require registro under Art. 34; everything else goes through comunicação prévia under Art. 35, which is a filing rather than an assessment.¹³

4. Applicators are not cosmetics

A brush, a sponge, a lash comb and a powder puff are not cosmetics in any of the three markets, and the reason is in the definitions already quoted. The European one reaches “any substance or mixture”; a brush is neither.¹¹ The American registration and listing scheme reaches a “cosmetic product”, which MoCRA defines as a preparation of cosmetic ingredients with a set composition.⁷ The Brazilian definition reaches “preparações”.¹³ An implement is not a preparation.

General consumer product law governs it instead. In the European Union that is the General Product Safety Regulation, which applies “to products that are placed or made available on the market insofar as there are no specific provisions with the same objective under Union law which regulate the safety of the products concerned” and imposes one obligation: “Economic operators shall place or make available on the market only safe products.”¹⁴ In the United States the Consumer Product Safety Act's definition of “consumer product” excludes “drugs, devices, or cosmetics”; an article that is none of the three is left inside the Consumer Product Safety Commission's remit.¹⁷

This matters most for one claim. A brush sold as antibacterial is not making a cosmetic claim, because the brush is not a cosmetic — it is making a biocidal or pesticidal one, and both the EU and the US have a separate regime waiting for it. In the EU an article treated with or intentionally incorporating a biocidal product is a “treated article” under Regulation (EU) No 528/2012. Article 58(2) bars placing it on the market unless every active substance is on the approved list for the relevant product-type, and Article 58(3) requires the label to say the article incorporates biocidal products and to name every active substance — triggered precisely when “a claim is made by the manufacturer of that treated article regarding the biocidal properties of the article”.¹⁵

The American rule turns on who the treatment protects. 40 CFR 152.25(a) exempts from the Federal Insecticide, Fungicide, and Rodenticide Act “[a]n article or substance treated with, or containing, a pesticide to protect the article or substance itself …, if the pesticide is registered for such use”.¹⁶ A brush treated to stop the bristles going mouldy is inside the exemption. A brush advertised as protecting the person using it is outside it, and outside it the claim is a pesticide claim. In Brazil, Art. 12 IV of RDC 907/2024 forbids disinfectant claims on cosmetics outright; which Brazilian regime governs an antimicrobial claim on the implement itself is one of the things we could not establish, and it is recorded below.

5. Sunscreen: the cleanest split in the subject

One molecule shows the whole argument. Bemotrizinol — Bis-Ethylhexyloxyphenol Methoxyphenyl Triazine, CAS 187393-00-6 — sits at entry 25 of Annex VI to Regulation (EC) No 1223/2009, the list of UV filters allowed in cosmetic products, at a maximum of 10 % in the ready-for-use preparation. It is in the base text of the Regulation, not an amendment.¹¹

In the United States the same molecule is a drug active, because a sunscreen is a drug: its purpose is to affect the structure or function of the body. On 10 June 2026 FDA announced the availability of final order OTC000039, issued under FD&C Act §505G(b)(1) (21 U.S.C. 355h(b)(1)), amending OTC Monograph M020 “to add bemotrizinol at concentrations up to 6 percent as a sunscreen active ingredient” — on a request filed by DSM Nutritional Products LLC on 23 September 2024.¹⁸ So the same substance is capped at 10% as a cosmetic ingredient in Europe and at 6% as a drug active in the United States, and the American ceiling arrived in 2026 against a European entry that has stood in the base text of a 2009 regulation.

Brazil takes the third position. Sunscreen is Grau 2, listed at item II nº 41 of Anexo I to RDC 907/2024, and it is one of the nine groups at Art. 34 that require registro rather than notification; permitted UV filters are set out in RDC 600/2022.¹⁹ A drug in one market, a cosmetic subject to a positive list in the second, a cosmetic subject to pre-market registration in the third.

6. What this does to the sentences on the pack

Once the category follows the claim, the copywriting becomes the regulated act. A cosmetic may say what it does to appearance; the moment it says what it does to a disease or to the structure or function of the body, it has asked to be judged as a drug, and in the United States as an unapproved one, because almost none of these products has an approved application or a finalised monograph behind it.

Two further prohibitions close the obvious escape routes. Criterion 1(3) of the EU common criteria bars claims “which convey the idea that a product has a specific benefit when this benefit is mere compliance with minimum legal requirements”, and criterion 3(6) requires a claim extrapolating an ingredient's properties to the finished product to be supported by evidence “such as by demonstrating the presence of the ingredient at an effective concentration”.¹² Brazil's Art. 12 II bans therapeutic claims by naming them — bruises, wounds, cracks, pain, inflammation, cramps, varicose veins, head lice.¹³ All of it is written down in advance; none of it is read by anybody before the product ships.

7. On the other side of the line: what the regulated version is, and what it costs

Hair reduction is a useful test case, because there is a cosmetic version and a regulated version and they claim different things. The regulated version is a device in all three markets. In the United States a laser for this purpose is classified at 21 CFR 878.4810, Class II, product code GEX, and reaches the market through a 510(k) clearance. The indication wording is precise, and it is a reduction claim rather than a removal claim. Clearance K180353 reads: “Permanent hair reduction is defined as the long-term, stable reduction in the number of hairs regrowing when measured at 6, 9, and 12 months after the completion of a treatment regime.”²²

The European Union reaches the same equipment from the other direction, by extending the Medical Devices Regulation to a listed group of products that have no medical purpose at all:

5. High intensity electromagnetic radiation (e.g. infra-red, visible light and ultra-violet) emitting equipment intended for use on the human body, including coherent and non-coherent sources, monochromatic and broad spectrum, such as lasers and intense pulsed light equipment, for skin resurfacing, tattoo or hair removal or other skin treatment.

Regulation (EU) 2017/745, Annex XVI, point 5, applied by Article 1(2) — consolidated text of 20 March 2023, CELEX 02017R0745-20230320.²⁰

Brazil says it in one sentence, at Art. 2 §6 of RDC 751/2022: “Os dispositivos ativos (equipamentos) indicados para correção estética e embelezamento são considerados dispositivos médicos.” The same article's §5 expressly excludes cosmetics.²¹

What that regime costs can be read off published price lists rather than estimated. Three were opened on 6 August 2026, one in each market. A course of six sessions on a single area is listed at US$599 to US$899 on the American list, which was showing no promotion²⁴. The London list prices its individual areas, also over six sessions, at £399 to £749, marked down on the day to £329 to £529, and prints per-session equivalents of £55 to £89 against the marked-down figures.²³ The Brazilian page sells a pacote completo with no session count stated, at list prices of R$899 to R$3.890, marked down on the day to R$99 to R$1.342,60.²⁵ The largest single package on each list is US$2,500, £2,069 — reduced to £1,449, or £242 a session — and R$3.890. No clinic is named in this section, deliberately. One of the three exhibits belongs to this site's publisher and is labelled as such where it is cited.

8. Why the price gap is the engine of the overstated claim

Set the two side by side. A device course that may lawfully claim long-term stable reduction, measured at 6, 9 and 12 months, is priced in the hundreds. A depilatory cream — in Brazil a Grau 2 cosmetic, item II nº 15 of Anexo I, Depilatório químico, carrying the mandated warnings at Art. 24 VII — is priced the way supermarket toiletries are priced.¹³ The cheaper product cannot make the more valuable claim, because that claim is a structure-or-function claim, and making it moves the product into a category with an application, a monograph or a conformity assessment attached.

That is the pressure that produces the sentences this site keeps correcting. Not “permanently reduces hair”, which is a device claim, but a word beside it carrying the same suggestion without triggering the definition. The regulators have named the manoeuvre: Article 20(1) bars implying characteristics or functions a product does not have, and criterion 4(1) of the common criteria bars presentations of performance going “beyond the available supporting evidence”.¹²

Be plain about the status of this section. The definitions, the categories and the prices are documented. That the price gap causes the overstatement is an argument about incentives, not a finding: we did not locate a study measuring claim inflation against price differentials, and none is cited above.

What we could not verify

Whether FDA has ever stated that a makeup brush is not a cosmetic. Section 4 reasons from the definitions — the EU's “substance or mixture”, MoCRA's “preparation”, ANVISA's “preparações” — and from the Consumer Product Safety Act's exclusion of cosmetics. We did not find an FDA document addressing applicators squarely, and the FD&C Act's own cosmetic definition at §201(i) speaks of articles “applied to the human body”, which is not on its face limited to preparations. Read section 4 as an argument from three definitions, not as a quoted agency position.

Which Brazilian regime governs an antimicrobial claim on an implement. Art. 12 IV of RDC 907/2024 forbids disinfectant claims on cosmetics, and a brush is not a cosmetic under Art. 3 XVI. We could not identify the ANVISA instrument that applies to a treated article of that kind, so no Brazilian rule is stated for the antibacterial-brush case.

An identifier the build document flagged. This article was specified against RDC 752/2022 Art. 3, Art. 12 and Anexo I. That resolution was revoked by Art. 51 I of RDC 907/2024, which carries the same subject matter with the same article numbers for definitions and for prohibited label wording. Every Brazilian citation above is to RDC 907/2024 as amended by RDC 949/2024.

A discrepancy on FDA's own page. FDA's explainer attributes the MoCRA definition of “cosmetic product” to “[FD&C Act, sec. 361]”. The definition is at FD&C §604, 21 U.S.C. 364(2); 21 U.S.C. 361 is §601, adulterated cosmetics. We quote the statute, and note the mismatch rather than resolve it.

When bemotrizinol was first permitted in the EU. Section 5 says only that the 10 % entry stands in the base text of Regulation (EC) No 1223/2009, of 2009. Secondary sources date the first European authorisation to around 2000, under Directive 76/768/EEC, but we did not open that consolidated Annex, so no figure is given for the gap between the European and American ceilings.

The price exhibits are not like-for-like. The American list states six sessions per package on its face, and the London list prints per-session equivalents consistent with six. The Brazilian page sells a “pacote completo” and states no session count anywhere we could find, and it paginates, so its figures are those of the twelve packages shown before the “Mostrar mais produtos” control. The London and Brazilian lists were both running promotions on 6 August 2026; both sets of figures are in the Sources entries, and the American figures are list prices with no promotion showing. We converted no currency and attempted no normalised per-session comparison. Section 8 also asserts a gap between a device course and a cosmetic depilatory without putting a number on the cosmetic: we found no published, dated, openable retail price list for depilatory creams that we were willing to cite, so no figure appears.

Numbering moves. The EU annex entries, ANVISA article numbers and CFR sections above were read on 6 August 2026 against the consolidations named in each citation. All three bodies renumber. Check them against the then-current text before quoting them from here.

Sources

United States Code. Definitions; generally (Federal Food, Drug, and Cosmetic Act §201), subsections (g)(1) and (i). 21 U.S.C. 321 ↩
Code of Federal Regulations. Meaning of “intended uses”, current text as amended at 86 FR 41401, 2 August 2021. 21 CFR 201.128 ↩
US Food and Drug Administration. Is It a Cosmetic, a Drug, or Both? (Or Is It Soap?), content current as of 11 September 2024. fda.gov/cosmetics/cosmetics-laws-regulations/it-cosmetic-drug-or-both-or-it-soap ↩
US Food and Drug Administration, Center for Drug Evaluation and Research. Warning Letter to Lus Essentials LLC, MARCS-CMS 728783, 14 May 2026. fda.gov/…/warning-letters/lus-essentials-llc-728783-05142026 ↩
Code of Federal Regulations. Definitions, paragraph (b), “cosmetic product”. 21 CFR 700.3 ↩
US Food and Drug Administration. FDA Authority Over Cosmetics: How Cosmetics Are Not FDA-Approved, but Are FDA-Regulated. fda.gov/cosmetics/cosmetics-laws-regulations/fda-authority-over-cosmetics-how-cosmetics-are-not-fda-approved-are-fda-regulated ↩
United States Code. Definitions (Federal Food, Drug, and Cosmetic Act §604, added by Pub. L. 117-328, div. FF, title III, §3502), paragraph (2). 21 U.S.C. 364 ↩
United States Code. Registration and product listing (Federal Food, Drug, and Cosmetic Act §607), subsections (a) and (c). 21 U.S.C. 364c ↩
United States Code. Safety substantiation (Federal Food, Drug, and Cosmetic Act §608), subsections (a), (b) and (c). 21 U.S.C. 364d ↩
US Food and Drug Administration. Registration & Listing of Cosmetic Product Facilities and Products, summary table (data as of 30 June 2026) and “Important Notes”. Read 6 August 2026. fda.gov/cosmetics/registration-listing-cosmetic-product-facilities-and-products ↩
European Parliament and Council. Regulation (EC) No 1223/2009 on cosmetic products, consolidated text of 1 May 2026: Article 2(1)(a), Article 20, and Annex VI entry 25. CELEX 02009R1223-20260501 ↩
European Commission. Regulation (EU) No 655/2013 laying down common criteria for the justification of claims used in relation to cosmetic products, Annex, criteria 1(1), 1(3), 3(6) and 4(1). CELEX 32013R0655 ↩
Agência Nacional de Vigilância Sanitária. Resolução da Diretoria Colegiada RDC nº 907, de 19 de setembro de 2024, Art. 3 XVI–XVIII, Art. 12, Art. 24 VII, Art. 34, Art. 35, Art. 51 and Anexo I — consolidated text as amended by RDC nº 949, de 12 de dezembro de 2024. RDC 907/2024 ↩
European Parliament and Council. Regulation (EU) 2023/988 on general product safety, Article 2(1), Article 3(1) and Article 5. CELEX 32023R0988 ↩
European Parliament and Council. Regulation (EU) No 528/2012 concerning the making available on the market and use of biocidal products, consolidated text of 15 April 2022: Article 3(1)(l), Article 58(1)–(3) and Annex V, product-types 1 and 2. CELEX 02012R0528-20220415 ↩
Code of Federal Regulations. Exemptions for pesticides of a character not requiring FIFRA regulation, paragraph (a), treated articles or substances. 40 CFR 152.25 ↩
United States Code. Definitions (Consumer Product Safety Act §3), subsection (a)(5), including exclusion (H). 15 U.S.C. 2052 ↩
US Food and Drug Administration. Amending Over-the-Counter Monograph M020: Sunscreen Drug Products for Over-the-Counter Human Use, and Related Information, notice of availability of final order OTC000039, 10 June 2026, Docket No. FDA-2025-N-6494. 91 FR 35220 ↩
Agência Nacional de Vigilância Sanitária. Resolução RDC nº 600, de 9 de fevereiro de 2022, on the list of ultraviolet filters permitted in personal hygiene products, cosmetics and perfumes; Art. 1, Art. 4 and Anexo. RDC 600/2022 ↩
European Parliament and Council. Regulation (EU) 2017/745 on medical devices, consolidated text of 20 March 2023: Article 1(2) and Annex XVI, point 5. CELEX 02017R0745-20230320 ↩
Agência Nacional de Vigilância Sanitária. Resolução RDC nº 751, de 15 de setembro de 2022, on risk classification, notification and registration of medical devices; Art. 2 §5 and §6. RDC 751/2022 ↩
US Food and Drug Administration, Center for Devices and Radiological Health. 510(k) substantial equivalence determination and Indications for Use, “Diode laser hair removal device”, 17 April 2018 — regulation 21 CFR 878.4810, Class II, product code GEX. K180353 ↩
Pulse Light Clinic (London, United Kingdom), “Laser Hair Removal Prices”, published price list of 6- and 8-session packages; the figures here are the 6-session ones. Read 6 August 2026, when the page was showing a “Was / Then / Now” promotion on every line: individual areas from £399 reduced to £329 (£55 a session) up to £749 reduced to £509; the women's Ultimate Full Body package at £2,299 reduced to £1,399 (£234); and the largest line on the page, the men's full-body package, at £2,069 reduced to £1,449 (£242). Cited as one of three national price exhibits for the figures given in section 7. pulselightclinic.co.uk/laser-hair-removal-prices ↩
Price table published by a med spa in Knoxville, Tennessee, United States, “Laser Hair Reduction Pricing (6-Session Packages)”. Read 6 August 2026: small area $599, medium area $799, large area $899 and full legs $2,500 per package, with per-area touch-ups at $49, $59, $69 and $115. The table states “All areas include 6 sessions per package”, and no promotional price was showing. Cited as one of three national price exhibits for the figures given in section 7. ↩
Dr. Laser (Brazil), “Depilação a laser”, published price list of pacote completo courses. Read 6 August 2026, when the page was showing a promotion: list prices from R$899,00 to R$3.890,00, reduced on the day to between R$99,00 and R$1.342,60. The listing page states no session count. Twelve packages were shown above a “Mostrar mais produtos” control; the figures are those twelve. Cited as one of three national price exhibits for the figures given in section 7. loja.drlaser.com.br/depilacao-a-laser ↩
United States Code. Small businesses (Federal Food, Drug, and Cosmetic Act §612), subsections (a) and (b). 21 U.S.C. 364h ↩

Corrections to this article

None yet.

There Is No Standard Behind the Hair-Colour Number Chart

2026-08-06T00:00:00Z

Correction

There Is No Standard Behind the Hair-Colour Number Chart

The claim: “A 7.4 is the same copper whichever brand you buy — the numbers are a standard code.”
The short answer: Because nobody owns the chart. The level/tone convention is industry practice, not a standard — no ISO, no annex, no legal definition — so the tone digits genuinely differ between manufacturers.
Evidence status: Settled

The Cosmetic Record · Published 2026-08-06 · Last checked 2026-08-06 · 15 sources · Corrections: none

1. What the two halves of the number are supposed to mean

Three manufacturers print the same shape on the tube: a digit, a separator, then one or two more digits. They also agree on what the positions are for. L'Oréal Professionnel's Majirel chart labels them Level of color, Primary tone of color and Secondary tone of color.¹ Wella's Koleston Perfect booklet sets out the same three positions against the worked example 4/71 as Depth of color, Major tone of color and Minor tone of color.² Schwarzkopf's IGORA ROYAL chart carries the same three positions as a labelled diagram — Base Level in front of the dash, Primary Tone Direction and Secondary Tone Direction after it — under the sentence “Our numbering system is based on a combination of numbers. The number(s) after the dash also have a corresponding letter that indicates the tone direction.”³

The depth digit is the half that behaves. Schwarzkopf prints the ladder under the heading Level of Depth: 1 black, 3 dark brown, 5 light brown, 7 medium blonde, 9 extra light blonde, 10 ultra blonde — no 2 in the list — with separate high-lift series at 10- and 12-.³ Majirel's rows run the same way, from 1/1N to 10/10N.¹ A reader who treats the first digit as a rough depth is not being misled. It is everything after the separator that fails.

2. Nobody owns this chart — we went looking for the standards body

Salon education calls the convention the “International Colour Chart”, usually abbreviated ICC, and the name does a lot of work: it implies an issuing body, a document number and a revision date. We looked for all three and found none, and this is the finding rather than a gap in it.

Cosmetics do have a technical committee at the International Organization for Standardization: ISO/TC 217. Its catalogue carries 52 published entries, 41 of them standalone documents and 11 amendments to those documents. They are microbiological methods, analytical methods, sun-protection test methods, packaging and labelling (ISO 22715:2006), good manufacturing practice (ISO 22716:2007) and guidelines on what may be called a natural or organic ingredient (ISO 16128-1 and -2). Not one concerns hair colour, shade nomenclature or colour classification.⁴ At European level, CEN/TC 392 “Cosmetics” divides its work between three working groups — analytical methods, microbiological methods, and efficacy including sun protection products. Its business plan does not enumerate work items, and says so; but neither its scope nor any working-group title touches shade naming.⁵

Nor is the numbering defined in law. Regulation (EC) No 1223/2009 restricts hair dye substances by name and concentration and prescribes label warnings word for word, and says nothing about what a digit means⁹; neither does Brazil's cosmetics labelling rule, nor the United States provision that exempts coal-tar hair dye from the adulteration section of the Federal Food, Drug, and Cosmetic Act.¹²¹⁴ What we are looking at is a trade convention, and conventions drift.

3. Why the tone digits diverge, and which diverge most

Each of the three companies publishes its own key. Set side by side, the keys disagree from the second digit onward, and they disagree most precisely where a reader is most likely to be shopping across brands — in the warm half of the wheel.

What a single tone digit means in each manufacturer's own published shade documentation — the L'Oréal chart file dated July 2025, the Schwarzkopf chart February 2026, the Wella booklet March 2015. Read across a row: the same digit, three different colours.
Tone digit	L'Oréal Professionnel Majirel	Wella Koleston Perfect	Schwarzkopf IGORA ROYAL
1	Blue (B)	Ash (blue/green)	Cendré
2	Violet (V)	Matt (green)	Ash (A)
3	Gold (G)	Gold (yellow)	Matte
4	Copper (C)	Red (warm)	Beige, muted gold (B)
5	Red-violet (Rv)	Mahogany (red/violet)	Gold (G)
6	Red (R)	Violet (cool)	Chocolate (Ch)
7	Green (Gr)	Brown (red, yellow, blue)	Copper, orange (C)
8	Mocha (M)	Pearl (blue)	Red (R)
9	Not listed in this key	Cendre (blue/violet)	Violet

Sources for the three columns, in order: the Majirel chart's own tone key¹; the Koleston Perfect tone numbering system²; the IGORA ROYAL number-and-letter tone key.³ The Wella column is the oldest of the three: its tone words come from a booklet file-dated 2015, and the leaflet Wella publishes now keeps the same depth/tone code without printing a word against each digit.¹⁵

Copper is the clearest case, and it is why this belief is worth correcting. A 7.4 in Majirel is a level-7 copper — the chart prints it as 7.4/7C, in the column headed Coppers.¹ The same digit in Koleston Perfect is warm red, and copper is not a discrete tone digit in that key at all.² In IGORA ROYAL, -4 is beige, in Schwarzkopf's own words muted gold, and copper has moved to -7 — the digit L'Oréal uses for green.³ One digit, three answers, every one of them printed by the company that sells the shade.

Notice also what L'Oréal does on its American chart: every shade carries two codes, the European number and a letter code, 7.4/7C. A convention that needs a translation printed beside it on the same page is not a standard.

4. Underlying pigment: the part of the system that is real

Underneath the numbering there is something that does not vary by brand. Hair colour comes from two pigments, and a 2023 review of mammalian pigmentation states the position plainly: “The diversity in hair pigmentation relies mainly on the quantity and ratio of the two types of melanin incorporated into the hair shafts, with other physical aspects of hair fibres intervening only as minor or segmental colour modifiers.”⁶ Eumelanin is the brown-black one; pheomelanin is the red-yellow one.

Permanent colour works on both at once. A 2025 review of hair dyes describes the reaction as mixing a precursor and a coupling agent with hydrogen peroxide in an alkaline medium at pH 8 to 10, and notes that “in an alkaline medium, the oxidizing agent lightens melanin, the natural pigment responsible for hair color in the hair shaft”.⁷ Morel and Christie's Chemical Reviews survey of permanent hair dyeing is the standing review of that chemistry; it is paywalled, and we cite it for scope, not for a sentence.⁸ The consequence is that a tone digit is never applied to a blank surface. It is applied to whatever warmth the lift has exposed.

Salon education attaches a fixed sequence to this — red, red-orange, orange, orange-gold, gold, yellow, pale yellow as the hair lightens — and pairs each stage with a depth level. We could not verify this sequence against a peer-reviewed source, and we say so again below.

5. Ruivo dourado, worked

Golden copper is the shade this address was asked about for years, so take it as the worked example. In Majirel's grammar, 7.43 reads as depth 7, primary tone copper, secondary tone gold, and the chart prints it as 7.43/7CG.¹ Reverse the two reflects and you should get 7.34, gold-primary with a copper secondary. That shade does not appear in the level-7 row of the 2025 chart, although 6.34 and 8.34 both do.¹

That absence is the second half of the correction. The code is an index into a catalogue of shades a company actually manufactures, not a coordinate system in which every combination exists. You cannot derive a shade from the notation; you can only look it up.

The depth digit carries the same caution. It describes the intended result, not the hair in front of it, and a permanent colour lightens only so far: in the European Union, hydrogen peroxide in hair products is capped at 12 % present or released — the developer strength sold as 40 volumes.¹⁰ A level-7 golden copper is a description of a level-7 outcome.

6. Where the law does touch the box

The numbering is unregulated. The contents are heavily regulated, and confusing the two is what makes the belief sound plausible.

In the European Union, Annex III of Regulation (EC) No 1223/2009 lists the permitted oxidative dye substances individually. Entry 8a covers p-phenylenediamine and its salts, and its condition of use is quoted here in full:

After mixing under oxidative conditions the maximum concentration applied to hair must not exceed 2 % calculated as free base

Regulation (EC) No 1223/2009, Annex III, entry 8a (p-Phenylenediamine and its salts), consolidated text CELEX 02009R1223-20260501, consolidation date 1 May 2026.

Entries 8 and 9a impose the same architecture on other phenylenediamines and on toluene-2,5-diamine, and entry 8b covers eyelash products.⁹ That shared architecture is why a box marked “PPD-free” is not the same as hypoallergenic: the usual substitute sits under the same entry block, at the same limit, carrying the same printed warning. Annex II bans named direct dyes outright — 1236 HC Yellow No 11, 1237 HC Orange No 3, 1238 HC Green No 1 and 1239 HC Red No 8 and its salts.¹¹ Every one of those entries regulates a molecule. None regulates a number.

Brazil is the same shape. Hair colourant is a Grau 2 product — Anexo I, item II, nº 45 of RDC 907/2024 reads Tintura capilar temporária, progressiva ou permanente — and Art. 24, II of the same resolution prescribes the warnings the label must carry, beginning with “CUIDADO. Pode causar irritação na pele ou reação alérgica. Antes de usar, fazer a prova de toque”.¹² The colourant list that governs every other cosmetic explicitly steps around hair:¹³

Art. 9º A lista do Anexo desta Resolução não inclui as substâncias corantes destinadas exclusivamente a tingir os cabelos.

ANVISA, Resolução da Diretoria Colegiada RDC nº 628, de 10 de março de 2022, Art. 9º; text as published and unamended, carrying the status Vigente on ANVISA's legislation database, consulted 2026-08-06. In English: the list annexed to this Resolution does not include colouring substances intended exclusively for dyeing hair.

In the United States, coal-tar hair dye is carved out of the adulteration provision of the Federal Food, Drug, and Cosmetic Act altogether, provided the label bears the statutory caution and directions for a preliminary test.¹⁴ Three regimes, three different mechanisms, and not one of them says what 7.4 means.

What we could not verify

That no standard exists anywhere. A negative drawn from a catalogue is weaker than a positive drawn from a document. We enumerated the published catalogue of ISO/TC 217 and read the business plan of CEN/TC 392, and found nothing on shade nomenclature; we did not search every national body, and we did not search ABNT or ASTM. If an ISO, CEN, ABNT or ASTM document governing hair-shade numbering exists, tell us and we will log a correction.

How we read the ISO catalogue. On 2026-08-06, iso.org returned HTTP 403 to our automated requests, so the published list was read from the same committee record served under the /sites/isoorg/ path of the Standards Pavilion site. The titles are ISO's; the retrieval route was not the canonical one, and we did not see the page as iso.org renders it. The split we print — 41 standalone documents and 11 amendments inside 52 published entries — is our count of that table, not a figure ISO publishes.

The age of the CEN document. The CEN/TC 392 business plan we opened is dated 2016-03-14. It is the current published business plan at that address, but it is ten years old and a newer work item could exist without appearing in it.

The “International Colour Chart”. We could not locate an issuing body, a document number, a publication date or a revision history for the ICC under that name. We could not establish who first published it or when.

The underlying-pigment sequence. The seven-stage ladder printed in section 4 is taught in effectively every colour course we could find, and we could not locate a peer-reviewed source for it. That eumelanin and pheomelanin behave differently is documented; that sequence, and its mapping onto depth levels, is not. A citation would be welcome.

The age of the Wella tone key. The only Wella document we found that prints a word against each tone digit is a booklet file-dated 2015-03-23, still published on wella.com. We could not find a later Wella document that says in words what /4 is.

Three manufacturers is not a census. We compared these three because all three publish a downloadable tone key that can be checked against the shade rows on the same page. We cannot say which convention is the most common, nor that the charts at those addresses are the current retail edition in every country.

Sources

L'Oréal Professionnel. Majirel colour chart, 2025 edition (United States). Tone key, depth ladder and shade rows including 7.4/7C, 7.43/7CG, 6.34/6GC and 8.34/8GC. us.lorealprofessionnel.com — majirel-colorchart-2025 ↩
Wella Professionals. Koleston Perfect technical folder / usage booklet, PDF file dated 2015-03-23: shade numbering system and tone numbering system (/1 to /9), worked on the example 4/71. Still published at this address, consulted 2026-08-06. wella.com — koleston-perfect_usagebooklet.pdf ↩
Schwarzkopf Professional. IGORA ROYAL shade chart, PDF dated February 2026, published by Henkel: numbering system, level of depth, and the number/letter/tone-direction key. publisher.henkel-dam.com — IGORA ROYAL shade chart ↩
International Organization for Standardization. ISO/TC 217 Cosmetics — standards and projects. Catalogue of 52 published entries (41 standalone documents and 11 amendments), plus 15 withdrawn and 10 under development, consulted 2026-08-06. ISO/TC 217. iso.org returned 403 to our automated requests, so the list was read from the identical committee page served on ISO's own publishing platform at the Standards Pavilion site (an IEC/ISO/UL initiative), whose path /sites/isoorg/contents/data/committee/05/49/54974/ renders the same ISO record: standardspavilion.org. ↩
CEN. Business Plan CEN/TC 392 Cosmetics, dated 2016-03-14. Scope and working-group structure (WG 1 analytical methods, WG 2 microbiological methods, WG 4 efficacy including sun protection products). The plan states that “the active projects in each WG are not described here but are provided in the work program”. standards.cencenelec.eu/BPCEN/679535.pdf ↩
Fernandes B, Cavaco-Paulo A, Matamá T. “A Comprehensive Review of Mammalian Pigmentation: Paving the Way for Innovative Hair Colour-Changing Cosmetics.” Biology 2023;12(2):290. Open access. 10.3390/biology12020290 ↩
de Souza JC, Ferraz Avelino ERA, Bessegato GG, et al. “Comprehensive Review of Hair Dyes: Physicochemical Aspects, Classification, Toxicity, Detection, and Treatment Methods.” ACS Omega 2025;10(27):28567–28586. Open access. 10.1021/acsomega.5c01576 ↩
Morel OJX, Christie RM. “Current Trends in the Chemistry of Permanent Hair Dyeing.” Chemical Reviews 2011;111(4):2537–2561. Cited for the chemistry of oxidative hair dyeing as a field; the full text is behind a paywall and we read the indexed record, not the article. 10.1021/cr1000145 · PMID 21265503 ↩
Regulation (EC) No 1223/2009 on cosmetic products, Annex III, entries 8, 8a, 8b and 9a (hair dye substances in oxidative hair dye products). Consolidated text, consolidation date 1 May 2026. CELEX 02009R1223-20260501 ↩
Regulation (EC) No 1223/2009, Annex III, entry 12 (hydrogen peroxide and other compounds or mixtures that release hydrogen peroxide): hair products, 12 % of H2O2 (40 volumes), present or released. Consolidated text, consolidation date 1 May 2026. CELEX 02009R1223-20260501 ↩
Regulation (EC) No 1223/2009, Annex II, entries 1236 (HC Yellow No 11), 1237 (HC Orange No 3), 1238 (HC Green No 1) and 1239 (HC Red No 8 and its salts). Consolidated text, consolidation date 1 May 2026. CELEX 02009R1223-20260501 ↩
ANVISA. Resolução da Diretoria Colegiada RDC nº 907, de 19 de setembro de 2024: Art. 24, II (warnings for hair lightening agents and hair colourants) and Anexo I, item II, nº 45 (Tintura capilar temporária, progressiva ou permanente in the Grau 2 list). Status Vigente com Alterações, consulted 2026-08-06; this resolution revoked RDC nº 752/2022, which carried the same two provisions at Art. 24, II and Anexo I, item II, nº 56. RDC nº 907/2024 ↩
ANVISA. Resolução da Diretoria Colegiada RDC nº 628, de 10 de março de 2022 (permitted colouring substances), Art. 9º. RDC nº 628/2022 ↩
Federal Food, Drug, and Cosmetic Act, section 601(a), codified at 21 U.S.C. 361(a): adulterated cosmetics, with the coal-tar hair dye proviso and the statutory caution statement. 21 U.S.C. 361 ↩
Wella Professionals. Koleston Perfect usage guidelines, PDF file dated 2023-11-24: the “Numbering System” panel, depth 2/ to 10/ and tone /0 to /9, with no tone word printed against each digit. wella.com — Wella-Professionals_Koleston-Perfect_Leaflet.pdf ↩

Corrections to this article

None yet.

Record: The Cosmetic Record

2026-08-06T00:00:00Z

6 August 2026 · Record

This site replaced an unrelated Portuguese-language blog at this address

Date: 6 August 2026
Page affected: Every address on this domain.
What this address used to say: Until December 2016, testandoprodutoscosmeticos.com carried a Brazilian Portuguese beauty blog of the same name, which published personal accounts of using cosmetic products from 2008 onward and then stopped. Between 2019 and 2026 the same addresses were used by two unrelated occupants: an automated content generator, and later an online gambling operator.
What was wrong with it: Nothing, on the blog's own terms — this is a record, not a correction of anyone's error and not an update to a rule. It exists because a reader arriving from a 2011 link is owed an explanation on the page rather than a silent substitution. The blog is not continued here. We are not its author, we have no connection to whoever wrote it, and none of its writing is reproduced anywhere on this site. The two later occupants have nothing to do with this site either.
What it says now: The Cosmetic Record, an English-language site that does not test products and attaches a document to every factual claim. Addresses from the 2008–2016 blog are forwarded to whichever article here answers the same underlying question; where there is no close match they go to the article index, which is why that index keeps the Portuguese path /artigos/. Addresses created by the two later occupants deliberately do not resolve. The full account is on About. The table below lists where the forwarded addresses go.

Which legacy addresses were forwarded where, at launch on 2026-08-06.
Addresses from the 2008–2016 blog	Where they go now	Why that destination
/2010/10/desamarelando-o-loiro… and the rest of the de-yellowing, toning and colour-remover posts	Purple Shampoo Does Not Remove Brass. It Covers It.	The largest cluster of inbound links on the domain asks one question — why bleached blonde keeps going yellow — and that article answers it.
The copper and shade-code posts: /2012/03/entendendo-os-acobreados-da-tintura.html, the Igora and Wella shade-number posts, the multi-year red-hair series	There Is No Standard Behind the Hair-Colour Number Chart	They are all about what a shade number means, which turns out to be a question about whether a standard exists.
Masks, oils, reconstruction and cronograma capilar posts, including the February 2011 argan-oil post	Argan Oil Does Not Repair Hair — and “Repair” Is Not a Regulated Word	Same underlying question: what a treatment can and cannot do to a fibre, and what “repair” commits a manufacturer to.
Brush and applicator posts; thermal-water posts	The named sections Applicators are not cosmetics and “Aqua”, thermal spring water and Latin binomials	A section anchor that a redirect points at is an address, not a heading, and those two are treated as permanent.
/category/cabelos and /search/label/cabelos	Hair colour and hair treatment	A topic archive deserves a topic index, not the general one.
/p/sobre-o-blog.html, /p/autora.html, and the old first-person Testei archive	About	The archive of first-person product write-ups resolves to the page explaining that this site does not test products. It is the most honest mapping in the file.
/p/faq.html	How we source	The nearest thing here to an explanation of how the site works.
/p/contato_30.html, /p/participe.html	Contact	Like for like. Note that the address /contact-us is not ours and does not resolve.
/p/anuncie-no-testando.html, the old “advertise here” page	About	Someone following the blog's old business page is looking for whoever runs the site now, and About is where that is recorded.
/feeds/*, /rss.xml, /atom.xml	The Atom feed	A feed reader that has been polling since 2012 gets a working feed back.
Every other genuine post address dated 2008 to 2016, and the month archives	The article index	Roughly nine hundred addresses have no close match here. The index is grouped hair first, which is what the archive was mostly about.
Addresses created by the two later occupants, 2019–2026 — the automated content, its tag and author trees, and the gambling paths	Nothing. They return 404.	They were never this blog. Forwarding roughly ten thousand addresses that were never written by anyone would be the abuse pattern itself, not a favour to a reader.

Recorded 6 August 2026. Anchor: /corrections/#2026-08-06-launch

Purple Shampoo Does Not Remove Brass. It Covers It.

2026-08-06T00:00:00Z

Correction

Purple Shampoo Does Not Remove Brass. It Covers It.

The claim: “Purple shampoo strips the yellow out of bleached blonde hair.”
The short answer: Lightening does not remove warmth, it exposes it. A violet shampoo deposits a direct dye that sits on the fibre and washes out; it never removes the yellow underneath. Only further lifting or an oxidative toner changes the pigment itself.
Evidence status: Settled

The Cosmetic Record · Published 2026-08-06 · Last checked 2026-08-06 · 10 sources · Corrections: none

1. Where the yellow comes from: what lightening leaves behind

The International Agency for Research on Cancer’s 1993 survey of hairdressing exposures states that “all of the bleaching methods used currently are oxidation processes”, that hydrogen peroxide is the commonest oxidant, “used usually as a 6 or 9% solution or, rarely, as a 12% solution”, and, describing permanent dyes, that hydrogen peroxide “is an excellent decolorizing agent for melanin, the hair’s natural colouring matter”.²

Decolorising melanin is not a switch. A 2017 study in Free Radical Biology and Medicine separated the two oxidising species alkaline peroxide supplies. On soluble Sepia melanin, the hydroxyl radical and the hydroperoxide ion each bleached melanin on their own, and with both present bleaching “is increased dramatically in both rate and extent”; on melanosomes isolated from human hair, hydroxyl radical generated in the surrounding solution “does not contribute significantly to bleaching”. The authors proposed that the radical pre-oxidises melanin units to o-quinones so that the more nucleophilic hydroperoxide ion can open the rings.³ That is a pigment dismantled in stages, and the colour a fibre shows at any moment is the part not yet dismantled. Nothing yellow is being added.

The far end has been photographed: electron microscopy of excessively bleached hair found torn cuticle scales, fissures along the exposed cortex, and pores where melanin granules had dissolved, with no granules detected at all.⁵ A redox proteomics study reported “extensive melanin granule degradation… after the mildest bleach treatment”.⁴ So the brass is not a film resting on the hair. It is the hair’s own pigment part-way through being destroyed, and no wash step reaches it.

2. Two different jobs — taking pigment out, and putting pigment on

Two operations get discussed as though they were one. The subtractive operation degrades melanin by oxidation; it is the only thing that reduces the quantity of warm pigment in the fibre. The additive operation puts a coloured molecule onto or into the fibre, which changes what you see without changing what is underneath.

A preformed dye cannot do the first job, and the IARC monograph says so flatly: “because the dyes in semipermanent hair colours are preformed and do not require added oxidant, such products cannot lighten the natural hair colour.”² That sentence is the whole correction. It is also why the subtractive route has a price the additive route does not: peroxide treatments “quickly access the cortex, causing untargeted oxidative damage across the fibre in addition to the desired loss of melanin”, with cortical intermediate filaments oxidised at their cystine disulphide bonds.⁴ In the European Union the tool is capped: Annex III entry 12 permits hydrogen peroxide and compounds releasing it in hair products at 12 % of H₂O₂ (40 volumes), present or released.¹ That ceiling is the reason the lift a box promises has a hard limit, whatever the number on the front of it says — see There Is No Standard Behind the Hair-Colour Number Chart.

3. What is actually in a violet shampoo

Read the INCI list and the violet is a named direct dye. Annex III of Regulation (EC) No 1223/2009 — the restricted-substances list — clears several violets by name for exactly this job, each with its own ceiling. Acid Violet 43 is reference number 291 (CAS 4430-18-6, EC 224-618-7), permitted as a “hair dye substance in non-oxidative hair dye products” at a maximum of 0,5 %; Disperse Violet 1 is entry 265 on the same condition at 0,5 %; HC Violet No 2 is entry 224 at 2,0 %; and Basic Violet 2 (CI 42520) is entry 278, capped at 0,5 % in non-oxidative products and at 1,0 % in the mixture applied to hair under oxidative conditions.¹ Every one of those is a ceiling on a deposit, not on a lift.

The annex a violet shampoo is judged against is not the colourant annex most people would guess.

… cosmetic products shall not contain … (c) colorants (i) colorants other than those listed in Annex IV …, except for hair colouring products referred to in paragraph 2; … 2. Subject to a decision of the Commission to extend the scope of Annex IV to hair colouring products, such products shall not contain colorants intended to colour the hair, other than those listed in Annex IV …

Regulation (EC) No 1223/2009 on cosmetic products, Article 14(1)(c)(i) and Article 14(2). Consolidated text of 1 May 2026.

Acid Violet 43 sits in Annex IV too — entry 90, colour index 60730, “not to be used in products applied on mucous membranes” — but Article 14 excepts hair colouring products from that list, so the operative limit for a toning shampoo is the Annex III hair-dye entry.¹

“Semi-permanent” is not a synonym for “unregulated”. Annex II, the prohibited list, carries four consecutive direct dyes at entries 1236 to 1239 — HC Yellow No 11, HC Orange No 3, HC Green No 1, HC Red No 8 and its salts — and, unlike entries 1230 to 1233 just above them, none of the four carries the qualifier “when used as a substance in hair dye products”. They are prohibited in any cosmetic product at all.¹

Brazil splits the same way. Article 9 of ANVISA’s RDC 628/2022 says its list of permitted colourants “não inclui as substâncias corantes destinadas exclusivamente a tingir os cabelos” — does not include colourants intended exclusively to dye hair.⁸ And Brazil does not leave a violet shampoo to be placed by inference from a list. Since December 2024 the operative definition of a hair dye names the toning shampoo itself:⁷

XXXIV - tintura de cabelo temporária, progressiva ou permanente: produto para tingir os cabelos, incluindo xampu e enxaguatório capilar tonalizante ou matizador. Maquiagem capilar não está incluída nesta categoria.

ANVISA, Resolução da Diretoria Colegiada RDC nº 907, de 19 de setembro de 2024, Art. 3, inciso XXXIV, as included by RDC nº 949, de 12 de dezembro de 2024. Text read 6 August 2026. In English: a temporary, progressive or permanent hair dye is a product for dyeing the hair, including a toning or shade-correcting (matizador) shampoo or hair rinse; hair make-up is not in this category.

So a violet shampoo is not a shampoo that happens to tint. In Brazilian law it is a hair dye that happens to lather — and the two sit in different halves of Anexo I. Plain “Xampu” is item 70 of the Grau 1 list, with “Xampu condicionador” at item 71; “Tintura capilar temporária, progressiva ou permanente” is item 45 of the Grau 2 list, filed there beside “Clareador ou descolorante para cabelos e/ou pelos do corpo” at item 9.⁷ The regulator has already decided which family a violet shampoo belongs to, and it is not the cleansing one.

4. Why the effect washes out, and why more of it is not more permanent

No reaction fixes the dye. The IARC monograph describes temporary colourants as water-soluble acid or basic dyes of the kind used for wool, “which have a molecular size too great to penetrate the cortex of the hair”, so “the dye is deposited on the surface, from which it is easily removed by washing”; semi-permanents it defines as products that “last through 6–12 washings and do not involve the use of an oxidizing agent in colour development”, their wash-fastness coming from dyes small enough to enter the cortex.² A violet shampoo lives between those two descriptions, and neither of them is permanent.

That is also why doubling the dose does not buy durability. The thing that ends the effect is the next wash, and the ceiling on what may be deposited is legal as well as physical — 0,5 % for Acid Violet 43 in a non-oxidative hair dye product in the EU.¹ Leaving it on longer changes how much dye goes on. It does not change what removes it.

5. What a toner or a demi-permanent does that a shampoo cannot

An oxidative toner is doing both jobs at once, and that is the entire difference. In oxidative dyeing the colour does not exist in the bottle: precursors penetrate the fibre and react with hydrogen peroxide to form indo dyes in place, and because the same peroxide is decolorising melanin at the same time, manufacturers “can balance the amounts of hydrogen peroxide and of dye precursors in such a way as to produce lightening, darkening or matching of the natural colour of the hair”.² A shampoo has no oxidant and no precursors, so it has no access to that balance.

Published developer strengths for the gentler oxidative products are thin. A 2015 review in the International Journal of Trichology puts demi-permanent products at 2% hydrogen peroxide against 6% for permanent, and their colour at “10–15 shampoos, or even earlier”.⁶ Neither figure carries a primary measurement, and that review also describes demi-permanents as not reaching the cortex, which is not how the trade uses the word. The only number with a document behind it is the regulatory ceiling.¹

Because these products act on the fibre, Brazil requires them to say so on the pack:

II - agentes clareadores de cabelos e tinturas capilares: a) “CUIDADO. Pode causar irritação na pele ou reação alérgica. Antes de usar, fazer a prova de toque” (descrever); b) “Em caso de contato com os olhos, lavar com água em abundância”; c) “Manter fora do alcance das crianças”; d) “Não aplicar se o couro cabeludo estiver irritado ou lesionado”; e e) “Não usar nos cílios”.

ANVISA, Resolução da Diretoria Colegiada RDC nº 907, de 19 de setembro de 2024, Art. 24, inciso II. Text in force as consulted on 6 August 2026. In English: hair lightening agents and hair dyes must warn that the product can irritate skin or cause an allergic reaction and that a patch test comes first, plus instructions on eye contact, children, broken scalp and eyelashes.

6. Colour removers: what they undo and what they cannot

There are two documented routes back. “In order to remove permanent and semi-permanent hair colourings,” the IARC monograph records, “hairdressers use reducing agents (sodium hydrosulfite or sodium or zinc formaldehyde sulfoxylate dissolved just before use in acidified water) or high-strength oxidants such as those mentioned above” — that is, the same peroxide and persulfate chemistry used for bleaching.² One route reduces a formed dye; the other oxidises it away.

Neither route puts melanin back. The electron microscopy is unambiguous that heavy bleaching leaves a cortex with no melanin granules in it,⁵ and no cosmetic regenerates them. So stripping a toner off does not return the hair to where it started; it returns it to whatever stage the lift reached, which is the warm stage the toner was covering. The remover undoes the deposit. It cannot undo the lift.

What we could not verify

Which chromophores make heavily bleached hair yellow. We could not verify this. The microscopy study cited above found no melanin granules in the cortex after excessive bleaching but did not measure colour,⁵ and we found no paper naming the residual absorbing species. The claim made above is narrower: what you see part-way through a lift is undegraded pigment, not a deposit.

The “1–3% developer” range, and every “fades after N shampoos” figure. The developer range circulates in trade writing with no textbook or regulation behind it, so the regulatory ceiling is printed above instead. The 6–12 washings for semi-permanents² and the 10–15 for demi-permanents⁶ are published without a stated wash protocol, and neither was measured on a toning shampoo.

Four sources were not read in full. The 2017 melanin bleaching study³ and the 2018 redox proteomics study⁴ sit behind publisher paywalls; the sentences citing them state only what their author abstracts state. The publisher’s copy of the 2011 Chemical Reviews survey of permanent hair dyeing⁹ returns HTTP 403 to us; its citation was confirmed against the PubMed record only. Robbins’s monograph¹⁰ was confirmed against the publisher and Crossref records only. No sentence above rests on either of those two.

Whether the Commission has ever adopted the decision under Article 14(2). The consolidated text of 1 May 2026 still states that extension conditionally, which is why Annex III is read here as operative for hair dye substances. The whole Official Journal was not searched. If such a decision exists, write in and it will be logged as a correction.

Whether a damaged fibre takes up more direct dye. The microscopy and proteomics studies establish cuticle loss, fissuring and protein loss;⁴⁵ neither measured dye uptake. The link is plausible and is not asserted above.

Sources

European Parliament and Council. Regulation (EC) No 1223/2009 on cosmetic products — Article 14; Annex II entries 1230–1239; Annex III entries 12, 224, 265, 278 and 291; Annex IV entry 90. Consolidated text of 1 May 2026. CELEX 02009R1223-20260501 ↩
International Agency for Research on Cancer. Occupational Exposures of Hairdressers and Barbers and Personal Use of Hair Colourants; Some Hair Dyes, Cosmetic Colourants, Industrial Dyestuffs and Aromatic Amines. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Volume 57. Lyon, 1993; section 1.2.1 (Hair preparations), subsection (a) Hair colouring preparations. NBK493259 ↩
Smith RAW, Garrett B, Naqvi KR, Fülöp A, Godfrey SP, Marsh JM, Chechik V. Mechanistic insights into the bleaching of melanin by alkaline hydrogen peroxide. Free Radical Biology and Medicine 2017;108:110–117. 10.1016/j.freeradbiomed.2017.03.014 (PMID 28323131) ↩
Grosvenor AJ, Deb-Choudhury S, Middlewood PG, Thomas A, Lee E, Vernon JA, Woods JL, Taylor C, Bell FI, Clerens S. The physical and chemical disruption of human hair after bleaching — studies by transmission electron microscopy and redox proteomics. International Journal of Cosmetic Science 2018;40(6):536–548. 10.1111/ics.12495 (PMID 30229956) ↩
Kim DH, Oh SH, Chang BS. Effects of excessive bleaching on hair: comparative analysis of external morphology and internal microstructure. Applied Microscopy 2024;54(1):11. 10.1186/s42649-024-00104-0 (PMC11652465) ↩
Gavazzoni Dias MFR. Hair cosmetics: an overview. International Journal of Trichology 2015;7(1):2–15. 10.4103/0974-7753.153450 (PMC4387693) ↩
ANVISA. Resolução da Diretoria Colegiada — RDC nº 907, de 19 de setembro de 2024; Art. 3, inciso XXXIV (definition of temporary, progressive or permanent hair dye, included by RDC nº 949, de 12 de dezembro de 2024), Art. 24, inciso II, and Anexo I (Grau 1 list items 70 and 71; Grau 2 list items 9 and 45, in the wording given by RDC nº 949/2024). Art. 51, inciso I of this resolution revoked the since-revoked RDC 752/2022, the predecessor this page cited until 6 August 2026; the consolidation did not change the technical content. Text read 6 August 2026. RDC 907/2024 ↩
ANVISA. Resolução da Diretoria Colegiada — RDC nº 628, de 10 de março de 2022 (lista de substâncias corantes permitidas; internaliza a Resolução GMC MERCOSUL nº 16/2012); Art. 9 and the annex entry for colour index 60730. RDC 628/2022 ↩
Morel OJX, Christie RM. Current trends in the chemistry of permanent hair dyeing. Chemical Reviews 2011;111(4):2537–2561. Citation confirmed against the PubMed record; full text not readable to us. PMID 21265503 ↩
Robbins CR. Chemical and Physical Behavior of Human Hair. 5th edition. Springer Berlin Heidelberg, 2012. ISBN 978-3-642-25610-3. Record verified; text not read. 10.1007/978-3-642-25611-0 ↩

Corrections to this article

None yet.

In Brazil, “Grau 2” Does Not Mean “Registered”

2026-08-06T00:00:00Z

Explainer

In Brazil, “Grau 2” Does Not Mean “Registered”

The claim: “Grau 2 means ANVISA registered the product and approved it.”
The short answer: Grau 1 and Grau 2 are a risk classification decided by what a product claims. Registro versus notificação is a separate axis. Only nine product groups need registration; most Grau 2 products, including ordinary permanent hair dye, are merely notified.
Evidence status: Settled

The Cosmetic Record · Published 2026-08-06 · Last checked 2026-08-06 · 20 sources · Corrections: none

What does a regulator do before a cosmetic reaches a shelf?
Market	The rule	What it requires	What it does not do
BR	ANVISA RDC 907/2024, Arts. 32, 34, 35 and 40, as amended by RDC 949/2024	Every product must be regularised before sale. Nine product groups go through registro: the file is examined, the decision is published in the Diário Oficial da União and lasts ten years. Everything else goes through comunicação prévia — an electronic filing on ANVISA's portal.	For a notified product, nobody reads the dossier before it is sold. Being Grau 2 does not by itself trigger any assessment: 46 product groups are Grau 2 and only nine are registered.¹
US	FD&C Act as amended by MoCRA: 21 U.S.C. 364c, 21 U.S.C. 364d, 21 U.S.C. 379e	Facility registration and product listing, and a responsible person who must “ensure, and maintain records supporting, that there is adequate substantiation of safety”. Colour additives are separately listed by regulation.	No rule requires premarket approval of a cosmetic product or of any ingredient other than a colour additive. Listing is a filing: nothing in 21 U.S.C. 364c makes it an authorisation.¹⁸¹⁹²⁰
EU	Regulation (EC) No 1223/2009, Arts. 4, 10, 11 and 13 (CELEX 02009R1223-20260501)	A named responsible person inside the Union, a safety assessment and a cosmetic product safety report before placing on the market, a product information file kept for ten years, and an electronic notification to the Commission.	No authority authorises an individual product. Art. 13 is a notification duty, and the safety assessment under Art. 10 is done by the responsible person, not by a regulator.¹⁷

1. One resolution, and when it came into force

Brazil puts cosmetics, personal-hygiene products and perfumes in one instrument. Since September 2024 that instrument is RDC nº 907, de 19 de setembro de 2024, which sets, in its own words, "a definição, a classificação, os requisitos técnicos para rotulagem e embalagem, os parâmetros para controle microbiológico, bem como os requisitos técnicos e procedimentos para a regularização de produtos de higiene pessoal, cosméticos e perfumes". Art. 51 revokes its predecessor RDC nº 752/2022, revokes RDC nº 841/2023, and revokes Art. 13 of RDC nº 814/2023.¹

Three months later RDC nº 949, de 12 de dezembro de 2024 amended it. That amendment is the one that matters here: it rewrote Art. 2, expanded the definitions in Art. 3, and replaced Anexo I outright with two lists — 71 Grau 1 product groups and 46 Grau 2 product groups. Its Art. 5, in the wording given by RDC nº 966, de 26 de fevereiro de 2025, reads "Esta Resolução entra em vigor em 7 de abril de 2025".²³

Two consequences follow immediately. Any page describing Brazilian cosmetic classification by reference to RDC 752/2022 is describing a revoked text,⁴ and any page reproducing Anexo I without RDC 949/2024 is reproducing a superseded annex. ANVISA's own consumer FAQ is in the first category: the page headed Conceitos e definições, last updated 05/04/2023, still defines both grades by reference to RDC 752/2022.⁶

2. A Mercosur transposition, not a purely national rulebook

The grade system is not Brazilian in origin, and Art. 2 says so:

Esta Resolução incorpora ao Ordenamento Jurídico Nacional as Resoluções GMC MERCOSUL nº 110/1994 "Definição de Produto Cosméticos", 48/2021 "Regulamento Técnico Mercosul sobre Rotulagem para Produtos de Higiene Pessoal, Cosméticos e Perfumes", 44/2018 "Requisitos Técnicos para Produtos de Higiene Pessoal, Cosméticos e Perfumes", 51/1998 "Parâmetros para Controle Microbiológico de Produtos de Higiene Pessoal, Cosméticos e Perfumes" e 18/2023 "Regulamento Técnico Mercosul sobre Classificação de Produtos de Higiene Pessoal, Cosméticos e Perfumes".

ANVISA RDC nº 907, de 19 de setembro de 2024, Art. 2 — text as amended by RDC nº 949, de 12 de dezembro de 2024, read on ANVISA's legislation system on 6 August 2026.¹

The classification rule is therefore a regional one, GMC 18/2023, which Brazil internalised through RDC 949/2024. The substance lists behave the same way: the permitted-preservative list internalises GMC 35/20, the prohibited list internalises GMC 62/14 as amended by GMC 37/20, and the restricted list in force since June 2026 internalises GMC 06/25.⁹¹⁰¹² A reader looking for "the Brazilian rule" is usually looking at a Mercosur text with a Brazilian number on it.

3. Grau 1 and Grau 2: a risk class defined by what the product claims

The two definitions sit at Art. 3, incisos XVII and XVIII. They are one sentence written twice, and the only difference is inicialmente menor against maior, plus which half of Anexo I the inciso points at. Note what does the sorting: the declared purpose of use. Brazil grades risk by what a product claims, which is the same lever the United States uses to decide whether a thing is a cosmetic or a drug at all.

XVII - produtos Grau 1: produtos de higiene pessoal, cosméticos e perfumes que cumprem com a definição adotada no inciso XVI deste artigo e que, por sua área de aplicação, público a que está destinado, condição específica de formulação ou impacto sanitário das finalidades de uso declaradas, requerem inicialmente menor grau de vigilância sanitária, conforme mencionado na lista indicativa "LISTA DE GRUPOS DE PRODUTOS DE GRAU 1" estabelecida no item "I" do Anexo I;

XVIII - produtos Grau 2: [...] requerem maior grau de vigilância sanitária, conforme mencionado na lista indicativa "LISTA DE GRUPOS DE PRODUTOS DE GRAU 2" estabelecida no item "II" do Anexo I;

ANVISA RDC nº 907, de 19 de setembro de 2024, Art. 3, incisos XVII and XVIII, in the wording given by RDC nº 949, de 12 de dezembro de 2024 — consolidated text read 6 August 2026. The bracketed ellipsis stands for the words repeated verbatim from inciso XVII.¹

Four things decide the grade: where the product is applied, who it is aimed at, a specific formulation condition, and "o impacto sanitário das finalidades de uso declaradas" — the health impact of the declared purposes. The last of those is the sentence on the pack. A grade is not a property of a molecule; it is a property of a claim, and Anexo I says so before it lists anything: item 1 provides that every product on the Grau 1 list is reclassified as Grau 2 the moment it declares one of eleven indications, among them "anticaspa" and "para tingimento, clareamento ou descoloração de cabelos e pelos corporais". The formula need not change.¹

The pairs are easy to read side by side. Item I nº 12, "Condicionador ou creme com enxágue ou enxaguatório capilar", is Grau 1. Item II nº 29 and nº 30, "Produtos capilares anticaspa com enxágue" and "sem enxágue", are Grau 2. The same rinse-off hair product changes grade because the label says anti-dandruff. Item II nº 33 is "Produtos destinados para pele sensível, desde que não se proclame como agente de tratamento" — a group whose very definition contains the limit of the claim that may be made. Item II nº 31 is everything "destinados ao público infantil", which is the audience criterion doing the work rather than the formula. Formulation can do it too: Art. 7 provides that a Grau 1 product must not contain the UV filters listed in RDC 600/2022, because their presence moves it up a grade.¹

4. Registro versus notificação is a different axis

Here is the whole of the registration list. It is nine groups:

I - bronzeador; II - gel antisséptico para as mãos; III - produto para alisar os cabelos; IV - produto para alisar e tingir os cabelos; V - produto para ondular os cabelos; VI - protetor solar; VII - protetor solar infantil; VIII - repelente de insetos; e IX - repelente de insetos infantil.

ANVISA RDC nº 907, de 19 de setembro de 2024, Art. 34 — consolidated text read on ANVISA's legislation system, 6 August 2026.¹

And everything else:

Os produtos dos grupos que não estão elencados no art. 34 desta Resolução são isentos de registro e estão sujeitos ao procedimento de comunicação prévia à Anvisa.

ANVISA RDC nº 907, de 19 de setembro de 2024, Art. 35 — consolidated text read 6 August 2026. Art. 3, inciso II defines comunicação prévia as "o procedimento administrativo a ser aplicado para informar à Anvisa a intenção de comercialização de um produto isento de registro por meio de notificação".¹

Compare the two lists and the answer falls out. Seven of the nine registered groups appear by name in Anexo I item II, the Grau 2 list — bronzeador at nº 7, protetor solar at nº 41, repelente de insetos at nº 42, the three hair-structure groups at nº 34, 35 and 38, and hand antiseptics at nº 22. The other two, protetor solar infantil and repelente de insetos infantil, reach Grau 2 by the audience route: nº 31 of the same list is "Produtos destinados ao público infantil", and item 1(a) of Anexo I moves any Grau 1 product into Grau 2 once it is declared for children. Registration is therefore a subset of Grau 2 — nine groups against forty-six. Grau 2 is the larger set, and it is not the set that gets examined.

Permanent hair colour is the cleanest worked example. Anexo I item II nº 45 is "Tintura capilar temporária, progressiva ou permanente" — Grau 2, the higher grade. It does not appear in Art. 34. The nearest entry is IV, "produto para alisar e tingir os cabelos", a product that straightens and colours. An ordinary box of permanent dye is notified, not registered, and no Brazilian official reads its dossier before it goes on sale.¹

5. Ten years, not five

English-language summaries routinely give the validity of a Brazilian cosmetic regularisation as five years. That number was correct once and is not correct now. The current text reads:

O registro tem validade por 10 (dez) anos, contados a partir do dia da sua publicação no Diário Oficial da União, podendo ser revalidado sucessivamente por igual período.

ANVISA RDC nº 907, de 19 de setembro de 2024, Art. 40 — consolidated text read 6 August 2026. Art. 41 dispenses products exempt from registration from revalidation altogether.¹

The five-year figure comes from the resolution two generations back, RDC nº 7, de 10 de fevereiro de 2015, whose Art. 22 read: "A regularização de produtos de higiene pessoal, cosméticos e perfumes Grau 1 e Grau 2 tem validade de 5 (cinco) anos e poderá ser revalidado por períodos iguais e sucessivos."⁵ That resolution was replaced by RDC 752/2022, which was replaced by RDC 907/2024. The ten-year term is not a 2024 novelty either: RDC 752/2022 already carried it, at its own Art. 40, in the same words the current Art. 40 uses, and that resolution took effect on 3 October 2022. Anything still citing the five-year period is citing a text revoked twice over and superseded on the point since 2022.⁴

The same 2015 resolution is, we think, where the "Grau 2 equals registered" shorthand was born. Its registration list was Anexo VIII, and Anexo VIII was headed "Produtos Grau 2 sujeitos a Registro" — Grau 2 products subject to registration. It ran to 23 items, most of them children's products: infant shampoo, infant conditioner, infant toothpaste, infant lip gloss, infant nail polish. A heading that named a subset was read as though it named the whole class, and the reading survived the annex. Of those 23 items, today's nine-group list keeps one child-specific entry — protetor solar infantil — and adds a second, repelente de insetos infantil; the shampoos, toothpastes and nail polishes are gone from it.⁵

6. Where Brazil keeps its annexes

The European Union puts its positive and negative lists in five annexes to one regulation — Annexes II to VI of Regulation (EC) No 1223/2009.¹⁷ Brazil puts them in six separate resolutions, cross-referenced from Art. 6 of RDC 907/2024: the actives permitted for straightening or waving hair (IN 220/2023), permitted UV filters (RDC 600/2022), permitted colourants (RDC 628/2022), permitted preservatives (RDC 528/2021), substances that may not be used at all (RDC 529/2021), and substances that may be used only under stated restrictions (RDC 530/2021, with RDC 645/2022).¹

Each moves on its own timetable, which is the practical cost of the design. IN 220/2023 revoked IN 124/2022 and took effect on 2 May 2023.⁷ The restricted-substances list was replaced in June 2026 by RDC nº 1.029, de 11 de junho de 2026, which internalises GMC 06/25, revokes RDC 530/2021 at its Art. 4 and gives products already regularised with ANVISA a 24-month adaptation period.¹² The preservative list runs to 60 numbered entries; the prohibited list runs to entry nº 1.419; the restricted list revoked in June ran to 105.⁹¹⁰¹¹ There is no single document to check, and no consolidated Brazilian equivalent of CELEX 02009R1223-20260501.

7. Labelling and the mandatory warnings

Art. 13 sets what must appear on the pack. On secondary packaging that includes the ingredients "utilizando a codificação de substâncias INCI", the "número do processo de regularização do produto", the holder's company authorisation (AFE) number, its CNPJ, the country of origin and the shelf life. The regularisation process number on the carton is the hook that makes §9 of this article possible.¹

Art. 24 then adds warnings for named product groups — aerosols, depilatories, antiperspirants, intimate-hygiene products, hair fixatives — and inciso II covers hair bleaches and colourants. Its first alínea is the line a Brazilian box of dye must carry:

a) "CUIDADO. Pode causar irritação na pele ou reação alérgica. Antes de usar, fazer a prova de toque" (descrever);

ANVISA RDC nº 907, de 19 de setembro de 2024, Art. 24, inciso II ("agentes clareadores de cabelos e tinturas capilares"), alínea a — consolidated text read 6 August 2026. The trailing "(descrever)" is an instruction to the manufacturer: the touch test must be spelled out on the pack. Alíneas b to e add rinsing on eye contact, keeping the product away from children, not applying to an irritated or damaged scalp, and not using it on eyelashes.¹

Note what this pairs with. A Brazilian hair colourant carries a mandated warning and a mandated self-test, and is notified rather than registered. The warning is the regulator's instrument here, not the dossier review.

8. Formaldehyde and escova progressiva

Products that alter hair structure are among the nine. RDC nº 906, de 19 de setembro de 2024 deals with them separately and its Art. 2 opens with one line: "Os produtos cosméticos para alisar ou ondular os cabelos sujeitam-se a registro." Its §1 closes the door: only products whose actives appear in IN 220/2023 may be registered, and §2 says applications that fail that test are refused. It revokes RDC 409/2020 and RDC 765/2022.⁸ The permitted actives are a closed list in IN 220/2023: thioglycolic acid and its salts, esters of thioglycolic acid, sodium or potassium hydroxide, lithium hydroxide, calcium hydroxide, inorganic sulfites and bisulfites, pyrogallol and thiolactic acid, each with a maximum concentration and, for several, a pH window and a professional-use restriction.⁷ Formaldehyde is not on that list.

ANVISA states the position for consumers in one word. Asked whether formol may be used to straighten hair, its FAQ answers: "Não. Tanto o formol quanto o glutaraldeído (devido à semelhança química com o formol) não podem ser utilizados como alisantes capilares."¹⁶ On 7 July 2025 the agency issued both a public alert on irregular straighteners and Informe de Segurança GGMON nº 03/2025, which records that adding formol to another product is a sanitary infraction and a crime hediondo under Art. 273 of the Penal Code, and that products containing formaldehyde and glyoxylic acid circulate through salons, informal markets and online platforms.¹⁴¹⁵

One boundary is where the Brazilian salon vocabulary and the regulation stop matching. ANVISA regulates products, not procedures. Its campaign page on straighteners, updated 27 April 2026, puts it in two sentences: "Procedimentos e métodos para alisamento capilar, como escovas inteligente, progressivas, BTOX, definitiva e de chocolate, não são regulamentados pela Anvisa. A Agência registra somente os produtos que serão utilizados." Salons are licensed by the local vigilância sanitária, not by ANVISA, and the straighteners FAQ says the same thing. What is regulated is each product in the bowl, and each of those must be registered, not merely notified.¹⁵¹⁶

The concentrations usually quoted alongside this — formol permitted at 0.2% as a preservative and 5% as a nail hardener — we could not confirm against any substance list currently in force, and the section below says exactly how far we got.

9. How to look a product up yourself

Art. 32 provides that "a regularização de produtos de higiene pessoal, cosméticos e perfumes é realizada por meio de procedimento eletrônico por meio do portal da Anvisa", and Art. 45 provides that failing to comply with the resolution cancels the regularisation and the cancellation is published on the agency's site.¹ The consumer-facing route runs through the process number that Art. 13 already requires on the carton. ANVISA's guidance page states it plainly: "Verifique se o produto está registrado na Anvisa, acessando o link: https://consultas.anvisa.gov.br/#/cosmeticos/registrados/", that the search needs "o número do processo, número de registro ou nome do produto, além do CNPJ da empresa detentora do registro", and — the useful part — "Dica: o número do processo sempre inicia com '25351'."¹⁶

Two limits on what that lookup tells you. A hit for a registered product means a file was examined and a decision published; a hit for a notified product means a filing exists. And the ten-year term at Art. 40 means the examination behind a current registration can be a decade old: a decision published today stands until 2036 unless ANVISA cancels it under Art. 45. A regularisation number is a fact about a process, not a verdict on the contents of the bottle in front of you.

What we could not verify

The 0.2% and 5% formaldehyde figures. We searched the full text of the substance lists cross-referenced by Art. 6 of RDC 907/2024, plus the restricted list that has replaced one of them, and found no formaldehyde entry in any of them: not among the 60 permitted preservatives of RDC 528/2021, which lists only formaldehyde releasers (DMDM hydantoin, imidazolidinyl urea, diazolidinyl urea, methenamine); not among the entries of RDC 529/2021, which run to nº 1.419; not among the 105 restricted entries of the revoked RDC 530/2021; and not in the annex of RDC 1.029/2026, the restricted list now in force. The 0,2 % preservative figure does appear at item 5 of RDC nº 162, de 11 de setembro de 2001 — "0,1% (em produtos de higiene oral) 0,2% (outros produtos não destinados à higiene oral)", with a "Contém formaldeído" label line above 0.05% — but that item was itself revoked by RDC nº 15, de 26/03/2013, and the whole resolution is marked REVOGADO on ANVISA's own system, revoked by RDC nº 29, de 01/06/2012. It contains no nail-hardener entry at all. We therefore could not locate the instrument currently authorising either figure.¹³

ANVISA disagrees with itself on the preservative limit. The 7 July 2025 news alert and Informe de Segurança GGMON nº 03/2025 both say formol is permitted "apenas como conservante (em concentrações de até 0,2%) e como endurecedor de unhas (até 5%)". The agency's straighteners FAQ and its campaign page say "em concentrações muito baixas, inferiores a 1%" for the same preservative use, while giving the same 5% for nail hardeners. All four are ANVISA publications; we print both figures and take neither as the operative limit.

The Mercosur texts themselves. Every GMC number and title above is read off Art. 2 of RDC 907/2024 and off the ementa of each Brazilian list. We did not open the MERCOSUR originals, and we did not find a public consolidated address for them.

RDC 966/2025. ANVISA's system rendered its ementa but not its body. The date "7 de abril de 2025" is read from the editorial note attached to Art. 5 of RDC 949/2024 on the same system, not from the amending text.

RDC 7/2015. ANVISA's legislation system returned only a "Este ato está sendo processado" placeholder for this act, and the Ministry of Health's Saúde Legis host reset the connection twice. Art. 22 and Anexo VIII are quoted from a PDF copy of the published resolution hosted elsewhere, cited below as what it is.

The consultation portal. consultas.anvisa.gov.br refused our request with an HTTP 403, so the search fields, the address and the 25351 prefix are quoted from ANVISA's own guidance page rather than observed in the portal.

How many products are notified rather than registered. The ratio of nine registered groups to forty-six Grau 2 groups is a count of groups in Anexo I, not of products on the market. We found no published figure for the number of individual products in either regime.

Sources

Agência Nacional de Vigilância Sanitária. Resolução da Diretoria Colegiada RDC nº 907, de 19 de setembro de 2024, Arts. 1, 2, 3 (II, XVI, XVII, XVIII), 6, 7, 13, 24, 32, 34, 35, 40, 41, 45 and 51, and Anexo I — items I (71 groups) and II (46 groups) and the reclassification rule at item 1 — consolidated text as amended, read 6 August 2026. RDC 907/2024 ↩
Agência Nacional de Vigilância Sanitária. Resolução da Diretoria Colegiada RDC nº 949, de 12 de dezembro de 2024 — amends RDC nº 907/2024, internalises GMC MERCOSUL nº 18/2023, replaces Anexo I with the Grau 1 (71 groups) and Grau 2 (46 groups) lists; Art. 5 as amended. RDC 949/2024 ↩
Agência Nacional de Vigilância Sanitária. Resolução da Diretoria Colegiada RDC nº 966, de 26 de fevereiro de 2025 — "Altera a Resolução da Diretoria Colegiada - RDC nº 949, de 12 de dezembro de 2024." Ementa only; see the note above. RDC 966/2025 ↩
Agência Nacional de Vigilância Sanitária. Resolução da Diretoria Colegiada RDC nº 752, de 19 de setembro de 2022, Arts. 40 and 51 — in force 3 October 2022, already carrying the ten-year registration term; the revoked predecessor still cited by most English-language summaries and by ANVISA's own consumer FAQ. RDC 752/2022 ↩
Agência Nacional de Vigilância Sanitária. Resolução da Diretoria Colegiada RDC nº 7, de 10 de fevereiro de 2015, Art. 22 (five-year validity), Art. 25 and Anexo VIII ("Produtos Grau 2 sujeitos a Registro", 23 items). PDF of the published resolution, consulted 6 August 2026 because ANVISA's own system would not render this act. cosmeticsonline.com.br — RDC nº 7 de 10 de fevereiro de 2015 (PDF) ↩
Agência Nacional de Vigilância Sanitária. Conceitos e definições (Perguntas frequentes — Cosméticos), page last updated 05/04/2023, defining Grau 1 and Grau 2 by reference to the since-revoked RDC nº 752/2022. gov.br/anvisa — perguntasfrequentes/cosmeticos/conceitos-e-definicoes ↩
Agência Nacional de Vigilância Sanitária. Instrução Normativa IN nº 220, de 13 de abril de 2023 — "Lista de ativos permitidos em produtos cosméticos para alisar ou ondular os cabelos", eight numbered actives; Art. 2 revokes IN nº 124, de 24 de março de 2022; Art. 3 sets entry into force at 2 May 2023. IN 220/2023 ↩
Agência Nacional de Vigilância Sanitária. Resolução da Diretoria Colegiada RDC nº 906, de 19 de setembro de 2024, Arts. 1, 2 and 24 — regularisation of cosmetic products for straightening or waving hair; revokes RDC nº 409/2020 and RDC nº 765/2022. RDC 906/2024 ↩
Agência Nacional de Vigilância Sanitária. Resolução da Diretoria Colegiada RDC nº 528, de 4 de agosto de 2021 — permitted preservatives, 60 entries; internalises GMC MERCOSUL nº 35/20; revokes RDC nº 29/2012. RDC 528/2021 ↩
Agência Nacional de Vigilância Sanitária. Resolução da Diretoria Colegiada RDC nº 529, de 4 de agosto de 2021 — substances that may not be used; the annex is numbered to entry nº 1.419 (a handful of numbers are unused), the last being N,N-dimetil-p-toluidina. Internalises GMC MERCOSUL nº 62/14 as amended by GMC nº 37/20. Formaldehyde appears nowhere in it. RDC 529/2021 ↩
Agência Nacional de Vigilância Sanitária. Resolução da Diretoria Colegiada RDC nº 530, de 4 de agosto de 2021 — restricted substances, 105 entries in Anexo I plus the fragrance-component annex; revoked by RDC nº 1.029/2026. RDC 530/2021 ↩
Agência Nacional de Vigilância Sanitária. Resolução da Diretoria Colegiada Anvisa nº 1.029, de 11 de junho de 2026 — restricted substances; republished in the Diário Oficial da União nº 117 of 25 June 2026. Art. 1 internalises MERCOSUL/GMC nº 06/25, Art. 2 gives a 24-month adaptation period, Art. 4 revokes RDC nº 530/2021. RDC 1.029/2026 ↩
Agência Nacional de Vigilância Sanitária. Resolução da Diretoria Colegiada RDC nº 162, de 11 de setembro de 2001, item 5 (Formaldeído e paraformaldeído) — the item carries the note "Revogado pela RDC nº 15, de 26/03/2013"; the resolution as a whole is marked REVOGADO, revoked by RDC nº 29, de 01/06/2012. RDC 162/2001 ↩
Agência Nacional de Vigilância Sanitária. Informe de Segurança GGMON nº 03/2025 (Cosmetovigilância) — Alisantes Capilares: riscos, cuidados e alternativas seguras, 7 July 2025. gov.br/anvisa — Informe de Segurança GGMON nº 03/2025 ↩
Agência Nacional de Vigilância Sanitária. Anvisa alerta para riscos à saúde associados ao uso de alisantes capilares irregulares, published 07/07/2025 12h18; and the agency's campaign page Produtos alisantes e ondulantes para cabelo (published 24/06/2025, updated 27/04/2026), which states that named salon treatments "não são regulamentados pela Anvisa". gov.br/anvisa — alerta, 07/07/2025 · gov.br/anvisa — comunicacao/campanhas/estetica/produtos-alisantes-e-ondulantes-para-cabelo ↩
Agência Nacional de Vigilância Sanitária. Orientações sobre alisantes (Perguntas frequentes — Cosméticos), pointing to the permitted-actives list at IN 220/2023, answering the question on formol and glutaraldeído, stating that named salon procedures are not regulated, and giving the consultation address and the 25351 process-number prefix. gov.br/anvisa — perguntasfrequentes/cosmeticos/alisantes ↩
European Parliament and Council. Regulation (EC) No 1223/2009 on cosmetic products, Arts. 4, 10(1), 11(1) and 13(1), and the list of Annexes I to VIII — consolidated text of 1 May 2026. CELEX 02009R1223-20260501 ↩
United States Code. Registration and product listing (Federal Food, Drug, and Cosmetic Act §607, added by Pub. L. 117-328, div. FF, title III, §3502). 21 U.S.C. 364c ↩
United States Code. Safety substantiation (Federal Food, Drug, and Cosmetic Act §608), subsection (a). 21 U.S.C. 364d ↩
United States Code. Listing and certification of color additives for foods, drugs, devices, and cosmetics, subsection (a). 21 U.S.C. 379e ↩

Corrections to this article

None yet.

Hazard Is Not Risk: How to Read Any Ingredient Score

2026-08-06T00:00:00Z

Correction

Hazard Is Not Risk: How to Read Any Ingredient Score

The claim: “A high score in an ingredient checker means the product is dangerous.”
The short answer: No. It measures hazard — whether a substance could cause harm under any conditions. Risk depends on dose, formulation, rinse-off versus leave-on, exposure site and frequency. A score that ignores all five is an index into the literature, not a verdict.
Evidence status: Settled — the distinction is written into the European safety-evaluation procedure, and EWG's Skin Deep states on its own methodology page that its rating does not account for exposure.

The Cosmetic Record · Published 2026-08-06 · Last checked 2026-08-06 · 14 sources · Corrections: none

1. Two words that are not synonyms

Hazard is a property of a substance. Risk is a property of a situation. The European Commission's Scientific Committee on Consumer Safety builds its procedure on that gap: a safety evaluation comprises four elements, the first being hazard identification, “carried out to identify the intrinsic toxicological properties of the substance, i.e. whether it has the potential to damage human health”. The other three are exposure assessment, dose–response assessment, and risk characterisation, in which a Margin of Safety is a point of departure divided by the Systemic Exposure Dose.¹ Exposure sits in the denominator; identity is not in the equation at all.

The Committee's own worked example: “A cosmetic substance that has the intrinsic property to be corrosive is not necessarily excluded for use in cosmetics. An example is potassium hydroxide KOH, the corrosivity of which depends on the final concentration, the pH, the presence of ‘neutralising’ substances, the excipient used, the exposure route, etc.”¹ KOH is a hazard in every jar it has ever been in. Whether it is a risk is settled by the five variables that sentence names, plus an “etc.”, and no ingredient name carries any of them. A number from step one alone has done a quarter of the work.

2. Dose makes the poison — and what that does and does not excuse

The maxim is load-bearing, and it is the thing toxicologists say the press handles worst: in a 2009 survey of Society of Toxicology members, 96% rated media performance as poor at explaining that “the dose makes the poison”, against 3% rating it well.¹⁰

What it does not excuse is treating a limit, once set, as a permanent verdict. Until 11 February 2017, Annex V entry 57 allowed methylisothiazolinone at 0.01% in any cosmetic product, leave-on included. A regulation of 22 July 2016 cut leave-on out and kept 0.01% for rinse-off, applying from 12 February 2017.⁴ Less than twelve months after that regulation, on 6 July 2017, the Commission cut the rinse-off figure again by nearly seven-fold, on an SCCS opinion of 15 December 2015 concluding “that for rinse-off cosmetic products, a concentration up to 0,0015 % (15 ppm) Methylisothiazolinone is considered safe for the consumer from the point of view of induction of contact allergy”:³

In Annex V to Regulation (EC) No 1223/2009, entry 57 is replaced by the following:

Reference number — 57
Chemical name/INN — 2-Methyl-2H-isothiazol-3-one
Name of Common Ingredients Glossary — Methylisothiazolinone
CAS number — 2682-20-4
EC number — 220-239-6
Product type, Body parts — Rinse-off products
Maximum concentration in ready for use preparation — 0,0015 %

Commission Regulation (EU) 2017/1224 of 6 July 2017 amending Annex V to Regulation (EC) No 1223/2009 on cosmetic products, Annex, entry 57. OJ L 174, 7.7.2017, pp. 16–18. Columns a to g of a nine-column table, each set here under its own heading; h and i are empty for this entry. Still the text in force: the consolidated Regulation of 1 May 2026, CELEX 02009R1223-20260501, carries entry 57 in these terms, read 2026-08-06.¹³

Three positions in under three years — 0.01% anywhere, 0.01% rinse-off only, 0.0015% rinse-off — on chemistry that never changed. What changed was the estimated dose at which induction of allergy occurs. (Entry 39 carries the same 0,0015 % figure for a 3:1 mixture with methylchloroisothiazolinone, and the footnote to entry 57 makes the two entries mutually exclusive: a product may use one or the other, never both.¹²) Nor is “more is always worse” universal: the SCCS records non-monotonic dose–response curves — meaning the effect does not rise steadily with the dose but can peak and fall away again, tracing a U or an inverted U — as an observed phenomenon.¹ The maxim licenses attention to the dose, not the conclusion that a permitted ingredient is settled for good.

Which raises the obvious question, since dose is what decides this: can you read the dose off the pack? Almost never. The order rule on an ingredient list tells you less than it appears to, and what it can and cannot tell you is worth knowing before you try to score anything from a label.

3. The variables a score throws away

Two rows of the SCCS exposure table. Shampoo: 10.46 g applied per day, retention factor 0.01, calculated daily exposure 0.11 g. Body lotion: 7.82 g per day, retention factor 1.00, calculated daily exposure 7.82 g.² The retention factor exists “to take into account rinsing off and dilution of finished products by application on wet skin or hair”. The same ingredient at the same percentage therefore differs about seventy-fold between those two products in the amount left on skin per day — and a score attached to a name cannot move between them, because the name has not moved.

EWG's Skin Deep — a database of ingredients in over 100,000 products, built by comparing labels against “nearly 60 toxicity and regulatory databases” — says so in its own methodology: the hazard rating “considers potential health hazards but does not account for exposure or individual susceptibility, factors which will drive health risks, if any, but which are generally not available for assessment”.⁸ The disclosure is on the page. The failure is in the reading.

There is an inversion worth naming too. In that methodology the hazard category “use restrictions” — “prohibited for use in cosmetics, or subject to concentration, use, or manufacturing method restrictions, according to industry safety guidelines and government requirements and guidance from the U.S., E.U., Japan, and Canada” — carries a weighting factor of 0.9, behind only the six categories weighted 1.0.⁸ Writing an exposure limit is risk management, and the act reduces risk. Fed into a hazard score it raises the number. Entry 57 above is exactly such a restriction.

4. What a data-gap penalty does to a new, well-characterised molecule

“Every product and ingredient in Skin Deep® gets a two-part score – one for hazard and one for data availability.”⁹ The second is reported as none, limited, fair, good or robust, and half of it is arithmetic on a literature search: “The remaining 50 percent reflects the number of studies listed for that ingredient in the government's PubMed scientific research index.” The band table runs 0 or 1 study → 0; 2–100 → 1; over 100 → 2; over 1,000 → 3; over 10,000 → 4, and the ceiling is anchored by example — “well-studied ingredients are set at 4 (formaldehyde, benzene and other ingredients with higher ratings than these two)”.⁸ A molecule introduced in the last decade with a complete modern dossier — read-across, in silico prediction (computer modelling from structure), an OECD-guideline test battery, an SCCS opinion — can return a hundred indexed papers or fewer under its INCI name and score 1, “limited”, on that half of the rating. Not because less is known, but because less has been printed under that name.

The Cosmetic Ingredient Review refuses to compress the same problem. Its procedures define safety such that “a lack of information about an ingredient shall not be sufficient to justify a determination of safety”, and one of the Panel's four conclusions is insufficient data, on which it “does not state whether the ingredient is safe or unsafe”; the other three are safe as used, safe with qualifications and unsafe.⁵⁶ Where use is reported and the data never arrive, the classification becomes after two years “Use Not Supported by the Data and Information Submitted to the Expert Panel”.⁵ That is a statement about a file rather than about a molecule, and a 1–10 axis cannot make it. On who pays, CIR's about page reads: “Although funded by the Council, CIR, the Expert Panel for Cosmetic Ingredient Safety, and the review process are independent from the Council and the cosmetics industry.”⁷

5. The survey everyone cites, and nobody reads properly

“79% of toxicologists say the Environmental Working Group overstates the risks of chemicals” is an accurate sentence and the least interesting line in its table. Here is the whole column, in the order printed: Greenpeace 96%, Environmental Defense Fund 85%, PETA 80%, Environmental Working Group 79%, Natural Resources Defense Council 79%, Center for Science in the Public Interest 79%, EPA 41%, CPSC 36%, WHO 31%, NIEHS 29%, NIOSH 29%, OSHA 28%, the International Agency for Research on Cancer 28%, AMA 28%, FDA 22%, National Cancer Institute 21%, CDC 17%, USDA 12%, National Science Foundation 11%, FASEB 10%, the Society of Toxicology itself 8%, ASPET 7%, PhRMA 2% and the American Chemistry Council 2%.¹⁰ A quotation that keeps the 79% and drops the 17% has converted a spread into a verdict.

The denominator matters more than the spread. Those percentages exclude respondents who answered “not sure”, because familiarity with the organisations varied so widely that leaving them in would, the authors wrote, stop the comparisons being “commensurable”; the newsletter reprints that caveat in a footnote and records that the full report gives the calculation both ways.¹⁰ It does, in a second chart, and the second chart reorders the list: with “not sure” counted in the denominator, Greenpeace falls to 83%, the Environmental Defense Fund to 61%, and the Environmental Working Group to 40% — one point above the EPA's 39%.¹⁴ The 79% that gets quoted is the figure among the minority of toxicologists who had an opinion about the group at all.

The rest of the methodology deserves the same treatment. Between 27 January and 2 March 2009 the Center for Media and Public Affairs, STATS and the Center for Health and Risk Communication at George Mason University contacted 3,562 full members of the Society of Toxicology, using an online instrument administered by Harris Interactive; 1,136 responded, a 32% return rate, and the published percentages rest on the 937 who answered every question.¹¹ It is a survey of professional opinion about organisations, not a measurement of any ingredient, and it was fielded seventeen years ago.

6. How to use a score properly

Treat it as an index. A high number means something has been written about the molecule in one of the nearly sixty databases behind the score; follow it to the finding and read four things — the endpoint, the species, the dose, the route. Then ask what the score cannot: does this product hold the ingredient against skin for sixteen hours, or send it down a drain in ninety seconds? EWG's own advice is to read both axes at once, buying “products with lower hazard ratings AND at least ‘fair’ data availability”.⁸ Used as a verdict, a score is step one of four wearing a number.

7. Why this site publishes no score of its own

Any single digit does one of two things. It encodes exposure assumptions — how much, how often, on which body site, rinsed or left on — which for products we have not opened we would have to invent; or it drops them, at which point it is a hazard score again. A rating also implies a comparison across products, and this site does not test products. And numbers outlive their reasoning: entry 57 read 0,01 % and then 0,0015 %, and only the citation makes the change visible. What we publish instead is the limit, the conclusion and the study, with the document attached — the method is at How we source.

8. Four questions to ask of any score, including ours

What is measured — a property of the molecule, or of the exposure? If the method never mentions concentration, product type or frequency, it is hazard.

Can you reach the underlying finding in one click, with dose, species and route? A source named but not linked cannot be checked, and that failure is as available to this site as to any other.

What does the system do when data are missing? It can emit a low number and warn separately, decline to conclude, or say the file is incomplete. Watch which.

When was the source data last refreshed, and is that date printed? Ours is at the top of this page as “Last checked”. Where a source carries no such date, it is recorded immediately below.

What we could not verify

The live EWG methodology pages. ewg.org/skindeep/contents/about-page/ and ewg.org/skindeep/understanding_skin_deep_ratings/ returned HTTP 403 to every plain fetch attempted on 6 August 2026. Every quotation attributed to them above comes from Internet Archive captures — 26 February 2024 and 17 December 2023 — and we could not verify that the live text is identical today.

When the Skin Deep database was last refreshed. The methodology page dates its own listed sources from 1987 to 2012 — most of them 2005 to 2007, and including IARC Monographs volumes 1–88, the CIR Compendium and a 2012 PubMed pull — and dates the data availability rating to “the April 2011 release”. It carries no revision date we could find, so we cannot say when the underlying hazard data were last updated, and the 2024 capture we read is the newest text we have.⁸

Where the score bands sit. The same page states that hazard ratings appear in “low”, “moderate” and “high” bands on a 1-to-10 scale, and that 10 goes to the top 5% most hazardous entries. It does not print the boundaries between the three bands, and we could not find them stated anywhere else, so we cannot tell you which number turns “moderate” into “high”.⁸

Sources

Scientific Committee on Consumer Safety. The SCCS Notes of Guidance for the Testing of Cosmetic Ingredients and their Safety Evaluation — 12th revision, Corrigendum 2. Adopted 15 May 2023; corrigendum 1, 26 October 2023; corrigendum 2, 21 December 2023. Section 3-1, pp. 11–13 (the four elements of a safety evaluation, and the Margin of Safety equation on p. 13); section 3-1, p. 12 (non-monotonic dose–response, Figure 2); section 3-4.5.1, p. 56 (potassium hydroxide). Read 2026-08-06. SCCS/1647/22 ↩
Scientific Committee on Consumer Safety, Notes of Guidance, 12th revision, section 3-3.4.2.1, Table 3A, p. 27 — daily exposure levels for cosmetic product categories in Europe. Shampoo: q 10.46 g/d, f_ret 0.01, E_product 0.11 g/d. Body lotion: 7.82 g/d, 1.00, 7.82 g/d. Amounts are P90 values from Hall et al., 2007 and 2011; retention factors derive from SCCNFP/0321/00. SCCS/1647/22 ↩
Commission Regulation (EU) 2017/1224 of 6 July 2017 amending Annex V to Regulation (EC) No 1223/2009 of the European Parliament and of the Council on cosmetic products. Recital (2) and Annex, entry 57. OJ L 174, 7.7.2017, pp. 16–18. CELEX 32017R1224 ↩
Commission Regulation (EU) 2016/1198 of 22 July 2016 amending Annex V to Regulation (EC) No 1223/2009 — entry 57, Methylisothiazolinone, rinse-off products, 0,01 %; compliance from 12 February 2017. OJ L 198, 23.7.2016, pp. 10–12. CELEX 32016R1198 ↩
Cosmetic Ingredient Review. Cosmetic Ingredient Review Procedures & Support to the Expert Panel for Cosmetic Ingredient Safety, December 2025. Section 1(m) (definition of “safe”), section 2 (purpose), section 42 (determinations), section 46(a)(1)–(3) (classification of ingredients with insufficient data). cir-safety.org · CIR Procedures, December 2025 ↩
Cosmetic Ingredient Review, “How Does CIR Work?”, consulted 6 August 2026 — the four conclusions available to the Expert Panel, in the page's own words. Safe: “Ingredients safe in the practices of use (product categories) and concentrations of use for each product category as documented in the safety assessment.” Safe with qualifications: “The Panel may reach the conclusion that an ingredient can be used safely, but only under certain conditions.” Unsafe: “These are ingredients with specific adverse effects that make them unsuitable for use in cosmetics.” Insufficient data: “If the Panel reaches an ‘insufficient data’ conclusion, it does not state whether the ingredient is safe or unsafe.” cir-safety.org/how-does-cir-work ↩
Cosmetic Ingredient Review, “About”, consulted 6 August 2026 — the funding sentence quoted above. “Council” is defined in the Procedures, section 1(j), as the Personal Care Products Council. cir-safety.org/about ↩
Environmental Working Group, “The science behind the database” — the Skin Deep methodology page at ewg.org/skindeep/contents/about-page/. Sections “Skin Deep's dual rating factors”, “Hazard ratings”, Table 1 (hazard categories and weighting factors), Table 10 (data availability from the open literature) and “Data availability ratings”. The live page returned HTTP 403; quoted from the Internet Archive capture of 26 February 2024. web.archive.org/web/20240226142749 ↩
Environmental Working Group, “Understanding Skin Deep ratings”, at ewg.org/skindeep/understanding_skin_deep_ratings/ — the two-part score and the 1-to-10 hazard scale. Live page returned HTTP 403; quoted from the Internet Archive capture of 17 December 2023. web.archive.org/web/20231217110403 ↩
Center for Media and Public Affairs. “The Media and Chemical Risk: Toxicologists' Opinions on Chemical Risk and Media Coverage.” Media Monitor, Volume XXIII, Number 2, Summer 2009. Page 4 (the full chart of organisations toxicologists believe overstate chemical risk, and the footnote recording that “don't know” responses were excluded and that the full report gives both calculations); page 5 (media accuracy, “dose makes the poison” — 96% poor against 3% well); page 6 (survey methodology). Read 2026-08-06. cmpa.gmu.edu · Media Monitor XXIII(2), Summer 2009 ↩
Lichter SR. “Are chemicals killing us?” Statistical Assessment Service (STATS) and the Center for Health and Risk Communication, George Mason University, 21 May 2009 — the sampling frame of 3,562 SOT full members, 1,136 respondents, 32% return rate, the 937 who answered every question, and the exclusion of “not sure” responses from Table 3 “to insure that the comparisons are commensurable”. STATS no longer operates; read in the Internet Archive capture of 2 January 2011. web.archive.org/web/20110102211813 ↩
Commission Regulation (EU) No 1003/2014 of 18 September 2014 amending Annex V to Regulation (EC) No 1223/2009 — entry 39, Methylchloroisothiazolinone (and) Methylisothiazolinone, rinse-off products, 0,0015 % of a mixture in the ratio 3:1 of 5-chloro-2-methylisothiazol-3(2H)-one and 2-methylisothiazol-3(2H)-one. OJ L 282, 26.9.2014, pp. 1–4. Cited to keep entry 39 and entry 57 apart: the two carry the same figure, are mutually exclusive under the footnote to entry 57, and are not the same restriction. CELEX 32014R1003 ↩
Regulation (EC) No 1223/2009 of the European Parliament and of the Council of 30 November 2009 on cosmetic products, Annex V, entries 39 and 57 and their footnotes. Consolidated text of 1 May 2026, CELEX 02009R1223-20260501 (version 040.001), read 2026-08-06 — entry 57 still reads Rinse-off products, 0,0015 %, marked as amended by Regulation (EU) 2017/1224. CELEX 02009R1223-20260501 ↩
Toxicologists' Opinions on Chemical Risk: A Survey of the Society of Toxicology, 21 May 2009 — the full report behind the Media Monitor summary. Title page: “Conducted by Statistical Assessment Service (STATS) and Center for Health and Risk Communication at George Mason University. Administered by Harris Interactive.” No individual author is named on the report. The chart on report page 20 excludes “not sure” responses; the chart on page 21 includes them and is the source of the Greenpeace 83%, Environmental Defense Fund 61%, Environmental Working Group 40% and EPA 39% figures. Retrieved from the Internet Archive capture of 2 January 2011 of stats.org, which the 2009 article linked as “Download a PDF of the full report”. web.archive.org · Toxicology Survey.pdf ↩

Corrections to this article

None yet. Errors of fact are fixed in public and logged with the date, what we said and what it says now. The corrections log · Report an error.

Hair Dye and Cancer: Four Findings Usually Quoted as One

2026-08-06T00:00:00Z

Correction

Hair Dye and Cancer: Four Findings Usually Quoted as One

The claim: “Hair dye causes cancer. The WHO classified it.”
The short answer: Four different studies asked four different questions and headlines fuse them. IARC classes working as a hairdresser as Group 2A and personal use of hair colourants as Group 3, not classifiable. The straightener finding is not a dye finding.
Evidence status: Contested

The Cosmetic Record · Published 2026-08-06 · Last checked 2026-08-06 · 9 sources · Corrections: none

1. Four separate findings, and how headlines fuse them

“Hair dye causes cancer, the WHO classified it” is one sentence assembled out of four separate bodies of work, no one of which says it. In publication order:

The IARC evaluations. A Working Group that met in Lyon in February 2008 published two separate verdicts in Volume 99 of the IARC Monographs (2010): one on occupational exposure as a hairdresser or barber, one on personal use of hair colourants. The second is the one that applies to a box of dye at home.¹
Zhang and colleagues, BMJ, 2020. A prospective cohort (exposure recorded before the disease, then followed forward) of 117,200 women in the Nurses' Health Study, over 36 years.⁵
Eberle and colleagues, Int J Cancer, 2020. 46,709 women in the Sister Study, breast cancer as the outcome, with a difference between Black and white participants that outside statisticians immediately disputed.⁶
Chang and colleagues, JNCI, 2022. 33,947 women from the same cohort, uterine cancer as the outcome, and an association with chemical straighteners — explicitly not with dyes.⁸

The fusion happens in two moves. IARC's occupational verdict is quoted and its personal-use verdict dropped, so a classification about a career in a salon is attached to a consumer. Then the 2022 straightener result is reported under the phrase “hair products”, which by the next news cycle has become “hair dye”. Both moves are avoidable by reading the abstract.

2. IARC: working as a hairdresser is not the same exposure as dyeing your own hair

Volume 99 of the IARC Monographs states its conclusions in five sentences, and they are worth reading in full because the halves are usually separated:

6.1 Cancer in humans

There is limited evidence in humans for the carcinogenicity of occupational exposures as a hairdresser or barber.

There is inadequate evidence in humans for the carcinogenicity of personal use of hair colourants.

6.2 Cancer in experimental animals

There is limited evidence in experimental animals for the carcinogenicity of hair colourants.

6.3 Overall evaluation

Occupational exposures as a hairdresser or barber are probably carcinogenic to humans (Group 2A).

Personal use of hair colourants is not classifiable as to its carcinogenicity to humans (Group 3).

IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Volume 99, Some Aromatic Amines, Organic Dyes, and Related Exposures (IARC, Lyon, 2010), the monograph Occupational Exposures of Hairdressers and Barbers and Personal Use of Hair Colourants, §6 Evaluation, p. 646. Text of the volume as published, 2010; read 2026-08-06.¹

The evidence behind the occupational verdict is specific, and it is not about breast or ovarian cancer. It is bladder cancer in men. Volume 99's summary of the human data reports that cohort studies showed an increased risk of bladder cancer among male hairdressers but not female ones; that in a Scandinavian cohort identified in the 1970 census and followed twenty years there was a significant 50% increase in men and a non-significant 10% decrease in women; and that most of more than twenty case–control studies reported increased risks in the range of 1.3–1.7 in male hairdressers.³

The rest of that section is rarely quoted: a pooled analysis of eleven studies in six European countries found no significant increase in either sex; smoking could generally not be adjusted for; and on breast cancer, many studies including the largest case–control and cohort studies showed no increased risk from professional use of hair colourants. One sentence governs all of it — the Working Group noted that the evidence mainly concerned exposures occurring before the 1980s, and often much earlier.³

None of this is new. IARC reached both conclusions in Volume 57 in 1993, and Volume 99 re-examined the question and arrived at the same two.⁴ The classification that circulates as breaking news is thirty-three years old.

3. What “Group 2A” and “Group 3” mean, and what they explicitly do not mean

The Monographs are graded on the strength of the evidence, not the size of the effect. The Preamble printed at the front of Volume 99 puts the distinction first:

A cancer ‘hazard’ is an agent that is capable of causing cancer under some circumstances, while a cancer ‘risk’ is an estimate of the carcinogenic effects expected from exposure to a cancer hazard. The Monographs are an exercise in evaluating cancer hazards, despite the historical presence of the word ‘risks’ in the title.

IARC Monographs Volume 99 (IARC, Lyon, 2010), Preamble as printed in that volume, p. 10. Read 2026-08-06.²

Group 2A therefore does not mean “more dangerous than Group 2B”. The Preamble states that probably carcinogenic and possibly carcinogenic “have no quantitative significance and are used simply as descriptors of different levels of evidence of human carcinogenicity”. Limited evidence is a positive association for which a causal interpretation is credible “but chance, bias or confounding could not be ruled out with reasonable confidence”. Inadequate evidence means the studies are of insufficient quality, consistency or statistical power to permit a conclusion about the presence or the absence of a causal association.²

That last clause decides how Group 3 should be read, and IARC says so itself:

An evaluation in Group 3 is not a determination of non-carcinogenicity or overall safety. It often means that further research is needed, especially when exposures are widespread or the cancer data are consistent with differing interpretations.

IARC Monographs Volume 99 (IARC, Lyon, 2010), Preamble as printed in that volume, p. 36. Read 2026-08-06.²

Group 3 is not a clearance and Group 2A is not a dose. Anyone quoting either label without the sentence that defines it is quoting a colour, not a conclusion — the same reasoning as Hazard Is Not Risk: How to Read Any Ingredient Score.

4. The largest prospective cohort, and both halves of what it found

The largest study of personal permanent hair-dye use is a prospective cohort of 117,200 women in the Nurses' Health Study, free of cancer at baseline and followed for 36 years, reported by Zhang and colleagues in the BMJ in 2020.⁵ It has two halves and coverage generally carried one of them.

The first half: ever-users of permanent hair dyes had a hazard ratio of 0.98 (95% confidence interval 0.96 to 1.01) for solid cancers excluding non-melanoma skin cancers, and 1.00 (0.91 to 1.10) for haematopoietic cancers overall, against non-users. Cancer-related death came out at 0.96 (0.91 to 1.02). Bladder cancer, melanoma, brain, colorectal, kidney and lung cancer, and the hormone-receptor-positive breast cancers, all showed no increase for ever-use.

The second half, in the paper's own words: “Basal cell carcinoma risk was slightly increased for ever users (n=22 560; 1.05, 1.02 to 1.08). Cumulative dose was positively associated with risk of estrogen receptor negative breast cancer, progesterone receptor negative breast cancer, hormone receptor negative breast cancer, and ovarian cancer.” The conclusion prints both halves in one paragraph: no positive association for most cancers or for cancer-related mortality, and the increases they did see “warrant further investigation”.⁵

The dose bands behind that second half are worth printing, because they are not a straight line. These are the paper's own Table 3 figures, for women of any natural hair colour: hazard ratios against non-users, with 95% confidence intervals, at three levels of lifetime use.⁵

Zhang et al. 2020, Table 3 — cumulative lifetime uses of permanent dye. A hazard ratio of 1.00 is no difference from a non-user; an interval that includes 1.00 has not separated from the null.
Cancer	1–99 uses	100–199 uses	≥200 uses	P for trend
ER-negative breast	0.99 (0.87–1.13)	0.91 (0.76–1.09)	1.20 (1.02–1.41)	0.09
PR-negative breast	1.06 (0.95–1.17)	0.98 (0.86–1.13)	1.12 (0.98–1.28)	0.16
ER-negative / PR-negative breast	0.97 (0.83–1.12)	0.91 (0.75–1.10)	1.28 (1.08–1.52)	0.03
Ovarian	1.00 (0.86–1.16)	1.20 (1.01–1.44)	1.15 (0.96–1.37)	0.046

Read down the columns rather than across the rows and the shape is the point. Only the heaviest band separates from the null in the breast-cancer rows; the middle band sits below the lightest in three of the four; and the ovarian row peaks in the middle and falls again. A dose that mattered would climb. This climbs, dips and climbs — a shape compatible with a dose effect and also with noise.

Two design features bear on how much weight the second half can take. More than 96% of the women had European ancestry, which the paper states and calls a limit on the generalisability of its findings. And it tested many cancer sites and subtypes across five exposure scales — the situation in which a few associations cross significance even under the null. The authors take the point up rather than ignore it, arguing that their five scales were complementary and their adjustment conservative.

5. The Sister Study: a difference between groups that is real and disputed

Eberle and colleagues analysed 46,709 Sister Study participants, aged 35–74 at enrolment between 2003 and 2009, over a mean 8.3 years of follow-up during which 2,794 breast cancers were identified.⁶ Permanent dye use in the previous twelve months carried a hazard ratio of 1.45 (95% CI 1.10–1.90) in Black women and 1.07 (0.99–1.16) in white women, heterogeneity p = 0.04. Personal straightener use carried 1.18 (0.99–1.41), with risk rising with frequency (p for trend = 0.02).

The design constraint is printed in the paper's own limitations: “By design, all study participants have a family history of breast cancer, which may limit the generalizability of these findings.” Every participant was recruited because a sister had been diagnosed. The authors argue this does not bias the internal comparison — a defensible position — but it is not a general-population cohort and the paper does not claim to be one.

The dispute was immediate and on the record. The Science Media Centre's expert reaction, published the morning the paper appeared, carries three named statisticians.⁹ Dr Michael Jones of the Institute of Cancer Research noted that “only just over 200 of the breast cancer patients were African American, limiting the power to draw conclusions about this group”. Dr Stephen Burgess of the MRC Biostatistics Unit wrote that the confidence intervals “were generally wide, and a zero association was always close to the range of the most likely values”. Professor Paul Pharoah of Cambridge went furthest: “The researchers carried out many analyses and the results highlighted in the abstract have been rather cherry picked.” That is commentary, not evidence, and is labelled as such here — but dropping it would misrepresent how the finding was received.

A published comment on the paper exists in the same journal, by Alipour; we could not read it, so we do not characterise its argument.⁷

6. Straighteners are not dyes — the finding most coverage dropped

The 2022 paper behind the largest wave of “hair products cause cancer” headlines is about a different product category. Chang and colleagues followed 33,947 Sister Study participants who had a uterus at enrolment, over an average of 10.9 years, during which 378 uterine cancers were identified.⁸ Ever-use of straightening products in the previous twelve months carried a hazard ratio of 1.80 (95% CI 1.12 to 2.88); use more than four times in that year carried 2.55 (1.46 to 4.45), p for trend = .002.

And then the sentence most coverage left out, from the same abstract: “Use of other hair products, including dyes and permanents or body waves, was not associated with incident uterine cancer.” The questionnaire separately recorded permanent, semi-permanent and temporary dyes, bleach, highlights, straighteners and perms. Straighteners were the category that moved. The authors call their result “the first epidemiologic evidence of association between use of straightening products and uterine cancer” and say replication is warranted.

The two categories are separated for a mechanistic reason: the straightener papers point at formaldehyde and formaldehyde-releasing chemicals, the dye papers at aromatic amines. Eberle's discussion notes that older straighteners used sodium hydroxide and thioglycolic acid salts and that formaldehyde-containing keratin treatments entered global markets in the early 2000s — which is also why an older null result and a newer positive one need not be in conflict.⁶

7. What a hazard ratio is, and what an absolute risk would have to look like

A hazard ratio is a relative rate, not an absolute chance. “A 45% higher risk” means the rate in the exposed group was about 1.45 times the rate in the unexposed group over the same period. It says nothing about how large either rate was, and if the underlying rate is small, a large ratio is still a small number of people. A ratio quoted without its baseline cannot be acted on.

Only one of these four papers publishes the baseline. Chang and colleagues estimated that among women who never used straighteners in the twelve months before baseline, approximately 1.64% were predicted to develop uterine cancer by age 70. The estimated risk was 1.18 percentage points higher for ever-users (95% CI for the risk difference 0.15% to 2.54%) and 2.41 points higher for frequent users (0.52% to 4.80%). They give the numbers needed to harm as 85 and 42 — one additional uterine cancer for every 85 ever-users, or every 42 frequent users.⁸ That is what an absolute risk looks like, and it is conditional on the association being causal, which the paper does not claim to have established.

Neither Zhang nor Eberle publishes an equivalent figure. For hair dye, there is no absolute-risk estimate in these papers to quote, only hazard ratios — and a hazard ratio of 1.07 with an interval running from 0.99 to 1.16 is compatible with no effect at all.

8. What would change our reading

This page is marked contested rather than settled. Four things would move it.

A new IARC evaluation using post-2000 exposure data. Volume 99's Working Group met in 2008 and recorded that its human evidence mainly concerned exposures before the 1980s; the same volume notes that in 2007, 135 individual ingredients were no longer allowed in the European Union for use in hair dyes.³ A classification resting on pre-1980 salon exposure describes products that in large part no longer exist.
Replication of the straightener–uterine-cancer association in an independent cohort. One prospective finding, called the first by its own authors, is a hypothesis with a confidence interval attached.
An absolute-risk estimate for the dye findings, of the kind Chang produced for straighteners. Until one exists, the receptor-negative breast-cancer and ovarian signals cannot be weighed against anything.
A subgroup analysis in Black women with adequate statistical power. The Eberle difference rests on just over 200 cases — the criticism the Science Media Centre respondents made, and the gap the authors themselves flag: “Future studies that are powered to robustly evaluate the possibility of differences by race and tumor subtype are needed.”⁶

What would not change it: another headline built by quoting IARC's occupational verdict and omitting its personal-use verdict from the same page of the same document.

What we could not verify

Checked 2026-08-06.

The published comment on Eberle et al. The PubMed record for PMID 31904111 (Alipour S, Int J Cancer 2020;146(9):2651, 10.1002/ijc.32852) is real and we opened it, but it carries no abstract, and the publisher's full text at Wiley returned HTTP 403 to us rather than the article. We could not read it, and therefore say nothing about what it argues.
Why the Zhang abstract lists PR-negative breast cancer. We read the per-band hazard ratios in Table 3 and printed them in section 4. The abstract names progesterone-receptor-negative breast cancer among the endpoints positively associated with cumulative dose, but the trend across dose bands in that table, for women of any natural hair colour, is P = 0.16. The paper's own results text is hedged where the abstract is not — it says “higher or potentially higher risks” — and it refers the reader to supplementary tables 1–28, which we did not open. We cannot say which estimate the abstract sentence rests on, so we have printed the table numbers and left the discrepancy visible rather than resolve it for the authors.
The version of Eberle et al. we read. The publisher's version of record at Wiley returned HTTP 403. The text quoted here is from the free author manuscript deposited at PMC7246134, which is the accepted manuscript and may differ in wording from the published article.
Whether IARC has revisited this exposure since 2010. Volume 99 is the most recent IARC monograph on hairdressing and hair colourants that we located, but IARC's consolidated Agents Classified by the IARC Monographs list did not render as readable text when we opened it, so we could not confirm from IARC's own index that no later volume revises either entry.
Any absolute risk for hair dye. We could not find an absolute-risk or number-needed-to-harm figure for permanent hair dye in any of these papers. Where a number of that kind appears in coverage, it did not come from these four documents.

Sources

International Agency for Research on Cancer. Occupational Exposures of Hairdressers and Barbers and Personal Use of Hair Colourants, in IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Volume 99: Some Aromatic Amines, Organic Dyes, and Related Exposures. IARC, Lyon, 2010. Section 6, Evaluation, p. 646. IARC Monographs Vol. 99 (mono99.pdf); evaluation wording cross-checked against the IARC/NLM electronic edition, section 6, NBK385456. ↩
International Agency for Research on Cancer. Preamble, in IARC Monographs … Volume 99. IARC, Lyon, 2010. Hazard versus risk, p. 10; limited evidence and inadequate evidence definitions, pp. 31–32; Group 2A and Group 3 definitions and the sentence that Group 3 is not a determination of non-carcinogenicity, pp. 35–36. IARC Monographs Vol. 99, Preamble (mono99.pdf). ↩
International Agency for Research on Cancer. Section 5, Summary of Data Reported, in the hairdressers and barbers monograph, IARC Monographs … Volume 99, pp. 641–646: exposure data and the 135 ingredients no longer permitted in the European Union in 2007 (§5.1, p. 642); bladder cancer in male hairdressers, the Scandinavian 1970-census cohort, the pooled analysis of eleven studies in six European countries, breast cancer, and the note that the evidence mainly concerned exposures before the 1980s (§5.2.1, pp. 642–643). Electronic edition of section 5, NBK385436. ↩
International Agency for Research on Cancer. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Volume 57: Occupational Exposures of Hairdressers and Barbers and Personal Use of Hair Colourants; Some Hair Dyes, Cosmetic Colourants, Industrial Dyestuffs and Aromatic Amines. IARC, Lyon, 1993 — the volume in which the Group 2A and Group 3 evaluations were first made. The hairdressers and barbers monograph begins at p. 43; the evaluations are at §5.5 and Overall evaluations, p. 107, where the 1993 wording is “Occupation as a hairdresser or barber entails exposures that are probably carcinogenic (Group 2A)” and “Personal use of hair colourants cannot be evaluated as to its carcinogenicity (Group 3)”. IARC Monographs Vol. 57; evaluation wording read from the IPCS INCHEM reproduction at inchem.org/iarc/vol57/01-occh. ↩
Zhang Y, Birmann BM, Han J, Giovannucci EL, Speizer FE, Stampfer MJ, Rosner BA, Schernhammer ES. Personal use of permanent hair dyes and cancer risk and mortality in US women: prospective cohort study. BMJ 2020;370:m2942. Prospective cohort, 117,200 women, Nurses' Health Study, 36 years' follow-up. Dose-band hazard ratios read from Table 3 (cumulative dose, women with any natural hair colour). 10.1136/bmj.m2942; free full text at PMC7463170; PMID 32878860. ↩
Eberle CE, Sandler DP, Taylor KW, White AJ. Hair dye and chemical straightener use and breast cancer risk in a large US population of black and white women. International Journal of Cancer 2020;147(2):383–391. Sister Study, 46,709 women, mean 8.3 years' follow-up, 2,794 breast cancers. 10.1002/ijc.32738; author manuscript at PMC7246134; PMID 31797377. ↩
Alipour S. Comments on: Hair dye and chemical straightener use and breast cancer risk in a large US population of black and white women. International Journal of Cancer 2020;146(9):2651. Record opened; full text not available to us — see What we could not verify. PMID 31904111. ↩
Chang C-J, O'Brien KM, Keil AP, Gaston SA, Jackson CL, Sandler DP, White AJ. Use of Straighteners and Other Hair Products and Incident Uterine Cancer. Journal of the National Cancer Institute 2022;114(12):1636–1645. Sister Study, 33,947 women with a uterus at enrolment, mean 10.9 years' follow-up, 378 uterine cancers. 10.1093/jnci/djac165; free full text at PMC9949582; PMID 36245087. ↩
Science Media Centre. Expert reaction to study looking at permanent hair dyes, chemical hair straighteners and risk of breast cancer, 4 December 2019. Named comments from Dr Michael Jones (Institute of Cancer Research), Dr Stephen Burgess (MRC Biostatistics Unit, Cambridge) and Prof Paul Pharoah (University of Cambridge), each with a declared-interests statement. Cited as commentary on source 6, not as evidence. sciencemediacentre.org — 2019-12-04. ↩

Corrections to this article

None yet.

Argan Oil Does Not Repair Hair — and “Repair” Is Not a Regulated Word

2026-08-06T00:00:00Z

Correction

Argan Oil Does Not Repair Hair — and “Repair” Is Not a Regulated Word

The claim: “A reconstruction treatment or an argan-oil mask repairs damaged hair.”
The short answer: No. Cuticle loss and cortical protein loss do not grow back. An oil can lubricate, reduce swelling on wetting and lay a surface film — real effects, all of them cosmetic and temporary. “Repair” and “nutrition” are marketing words, not regulated ones.
Evidence status: Contested — the structural point is not in dispute; what an oil does to a fibre's mechanics is. Two papers published in 2024 point opposite ways, and only one of them put argan in the experiment.

The Cosmetic Record · Published 2026-08-06 · Last checked 2026-08-06 · 14 sources · Corrections: none

1. What damage actually is: cuticle loss, cortical protein loss, and what is not coming back

The shaft you comb is not living tissue in the sense the rest of you is. It leaves the follicle already keratinised — no blood supply, no dividing cells, no maintenance crew — and what happens to it after it clears the scalp stays happened.

Two things go missing. The first is cuticle. Clarence Robbins' monograph, the standard reference work on the fibre, puts ordinary handling at the centre of it: “Aging of individual hair fibers on one's head by everyday grooming actions generally results in a gradual degradation of the scales through cuticle fragmentation,” a process tied “to the actual age or time the individual hair has been on the scalp (or its residence time) rather than to the chronological age of the individual.”¹ Length is exposure, and those scales are not re-laid.

The second is protein, and bleaching is the clearest case. Comparing extensively bleached hair with unaltered hair from the same head, Robbins reports “lower cystine content, a higher cysteic acid content, and lower amounts of tyrosine and methionine in the bleached hair” — the disulfide crosslink oxidised through to cysteic acid — with fewer total amino acids per gram, partly through “solubilization of protein or protein derived species into the bleach bath.”¹ Protein that left in the bleach bath is gone.

Hence the trade's own measure of damage: protein loss, meaning how much material washes out of a tress, quantified by a colorimetric assay — the Lowry method in the argan work cited below.⁵⁶ A treatment that lowers that number has prevented a loss. It has not restored one. The two are not the same claim, and a “repair” pack routinely fuses them.

2. What an oil can do — lubrication, reduced hygral swelling, surface film

The list of real effects is short, and rather less exciting than the box. An oil lowers friction between fibres and between fibre and comb. A conditioner deposit does something adjacent: Gavazzoni Dias' review describes conditioners as acting “by neutralizing the electrical negative charge of the hair fiber by adding positive charges and by lubricating the cuticle.”² Both effects live on the outside, and both come off in the wash.

Penetration is real too. A 2024 study in the International Journal of Cosmetic Science used LC–MS to identify the triglycerides in coconut, Camellia oleifera and safflower seed oils, then NanoSIMS imaging of fibre cross-sections treated with 13C-labelled triolein to locate them. The oils partitioned into the lipid-rich cell membrane complex — the mortar between the cells, not the cells — and every oil tested improved single-fibre fatigue strength.³ That is a mechanical benefit of occupying a lipid compartment, not of rebuilding keratin.

Then the disagreement. Also in 2024, a group at the University of Campinas and the Symrise hair-care laboratory in Cotia ran tensile and fatigue tests plus Raman spectroscopy on virgin and bleached Caucasian tresses treated with coconut, avocado and argan oils for 24 hours at 25 °C. All three diffused into the cortical region, but the mechanical outcome split by oil: coconut and avocado reinforced the hydrophobic barrier of the cell membrane complex, while argan — “due to the high degree of unsaturation of its fatty acid chains” — increased water absorption, “resulting in losses in hair resistance”. In bleached hair, already more hydrophilic, affinity for argan rose and so did fragility. The authors' own summary: “the analyzed vegetable oils are not always beneficial for hair care.”⁴

So the heading above needs a caveat. Swelling of the fibre in water is measurable, and the reduction of it by an oil is the mechanism the review literature offers for the popular term hygral fatigue — glossed there, in one clause, as “repeated swelling and drying”.² That is a review assertion with no measurement behind it in any paper opened here; see “What we could not verify” below. And on argan specifically, the one study that tracked water uptake reports the opposite of the folklore.

3. The one penetration experiment everybody cites, and what it was actually about

Almost every English-language article asserting that oil “penetrates and repairs” traces back to Rele and Mohile, Journal of Cosmetic Science, 2003. It reports a real result — not the one it is quoted for.

It tested three oils: mineral, sunflower and coconut. Not argan. Using a Taguchi design across hair types and pre-wash and post-wash regimes it measured protein loss, and concluded that “among three oils, coconut oil was the only oil found to reduce the protein loss remarkably for both undamaged and damaged hair”, attributing that to lauric acid's low molecular weight and straight linear chain. Sunflower oil — a triglyceride like argan, and unsaturated like argan — “does not penetrate the fiber, consequently resulting in no favorable impact on protein loss.”⁵ Two facts belong in the same breath: the endpoint is a loss prevented, not reversed; and the paper's own affiliation line places both authors in the research and development department of Marico Industries Ltd, an Indian manufacturer of coconut hair oil.

Argan-specific work exists, and deserves to be named rather than borrowed. A 2022 study in the Journal of Cosmetic Dermatology divided excised human hair tresses into three groups — oxidant alone, argan pre-treatment removed physically, argan pre-treatment removed by washing — recorded attenuated total reflectance spectra and quantified protein loss by the Lowry method. The pre-treated groups lost significantly less protein.⁶ This is in vitro (done on cut tresses in laboratory glassware, not on people), the abstract states no number of tresses, and the endpoint is again a loss prevented before the insult. A pre-treatment result is not a repair result.

4. Why “argan” is doing less work on the label than its position in the list suggests

In the EU an ingredient list is ordered by weight only down to a threshold. Article 19(1)(g) of Regulation (EC) No 1223/2009 requires ingredients in “descending order of weight of the ingredients at the time they are added to the cosmetic product”, then adds that “ingredients in concentrations of less than 1 % may be listed in any order after those in concentrations of more than 1 %.”⁷ Below that line the sequence carries no information. An oil sitting eighteenth may be present at 0.9% or at 0.001%, and the label cannot separate the two. The general case is set out in You Cannot Work Out Percentages From an Ingredient List.

The name is less specific than it looks. ARGANIA SPINOSA KERNEL OIL is entry 2221 in the glossary of common ingredient names annexed to Commission Implementing Decision (EU) 2025/1175, in force since 30 July 2026. It was 2449 until the day before that and 2155 before that — the list is renumbered wholesale each time it is reissued, which is a fair measure of what an entry in it is. It is a spelling: no grade, no purity specification, no minimum content.⁸ A pack showing an argan branch has satisfied that requirement by containing some.

What the law constrains is the inference. Commission Regulation (EU) No 655/2013 lays down six common criteria for cosmetic claims, and two of them close this door directly:⁹

Ingredient claims referring to the properties of a specific ingredient shall not imply that the finished product has the same properties when it does not. […] A claim extrapolating (explicitly or implicitly) ingredient properties to the finished product shall be supported by adequate and verifiable evidence, such as by demonstrating the presence of the ingredient at an effective concentration.

Commission Regulation (EU) No 655/2013, Annex (Common criteria), points 2(2) and 3(6). Text as published in OJ L 190, 11.7.2013, p. 31; the regulation has not been amended.

The sub-working group on claims — Member State authorities and trade bodies under the Commission's Working Group on Cosmetic Products, whose document states on its own first page that it is not a Commission document and is not legally binding — illustrates the first with a case that maps onto argan almost exactly: the claim “contains moisturising aloe vera”, or prominently picturing aloe vera, “should not be made if the product itself has no moisturising effect”. The same document cautions that “the presentation of results from in vitro or in silico studies should not suggest a result in vivo”, and that citing publications about an ingredient is acceptable only “provided that they are relevant to the cosmetic product and the claim made.”¹⁰ A coconut-oil paper hung on an argan mask fails on both counts.

5. “Repair”, “reconstruction”, “nutrition”: what a claim has to be able to support

Begin with what nobody does: nobody approves the sentence on the pack before it ships. The EU responsible person holds the substantiation in the product information file, and Article 20(1) states a prohibition rather than a permission — text, names, trade marks, pictures and other signs “shall not be used to imply that these products have characteristics or functions which they do not have.”¹¹ Under the common criteria the honesty rule reads “presentations of a product's performance shall not go beyond the available supporting evidence”⁹, and the sub-working group's worked example of a duration overreach sits under truthfulness instead: “The claim ‘48-hour hydration’ is not allowed if the set of evidence only supports a shorter period of hydration.”¹⁰ Applied to a mask, the claim evidence can carry is closer to “reduces protein loss during this wash”. Not “repairs”.

Brazil draws its line elsewhere, and where it draws it is instructive. RDC 907/2024 — which replaced RDC 752/2022 in September 2024 — prohibits therapeutic claims:¹²

Art. 12. A rotulagem não deve conter nome comercial, marcas, imagens, links eletrônicos ou dizeres que: […] II - representem alegações terapêuticas atribuídas ao uso do produto ou de seus ingredientes, como, por exemplo, prevenção ou tratamento de hematomas, feridas, rachaduras, dores, inflamações, câimbras, varizes […]

Resolução da Diretoria Colegiada — RDC nº 907, de 19 de setembro de 2024, art. 12, caput and item II. Consolidated text on the ANVISA legislation portal, consulted 6 August 2026; art. 51(I) of the same resolution revokes RDC nº 752/2022.

“Repair” is not on that list, because repairing a fibre is not a therapeutic claim about a body. It is a cosmetic claim, and Brazil treats it as one: item 45 of the Grau 1 product-group list in Annex I to the same resolution reads, in full, “Preparações cosméticas com enxágue para reparar, umectar, hidratar e/ou restaurar os cabelos” — rinse-off preparations to repair, moisten, hydrate and/or restore the hair; item 48 repeats the verb for leave-in products.¹³ Reparar appears in Brazilian cosmetic law as the name of a shelf, not as an effect with a definition or a threshold. Searching the consolidated text for the strings reconstru and nutri returns nothing at all.¹² The resolution does require the ingredient list in INCI codification on the secondary packaging, and complete efficacy data held by the company whenever a benefit is stated on the label.¹² Held — not filed, not reviewed, not published.

So, in both regimes: a word with no legal definition, no measurement attached to it, and a substantiation duty discharged into a file nobody outside the company reads.

6. The cronograma capilar is a schedule, not a chemistry

The cronograma capilar — a rotation of hidratação, nutrição and reconstrução across a month, one step per wash, adjusted according to how the hair is judged to be behaving — is one of the most durable ideas in Brazilian hair care, and the archive at this address is full of it. It has no traceable source. A PubMed search for the exact phrase "cronograma capilar", run on 6 August 2026, returns zero records, and neither reconstrução nor nutrição is a defined term in Brazilian cosmetic labelling law.¹²

Parts of it map onto documents. Cationic conditioning is described in the review literature.² Oils partition into the cell membrane complex and change fatigue behaviour, in directions that depend on the oil.³⁴ The protein step has the most direct evidence: a 2021 study in the International Journal of Cosmetic Science applied wool-derived hydrolysed keratins of three molecular weights — 221 Da, about 2,577 Da and about 75,440 Da — to very curly hair of African origin relaxed with a sodium-hydroxide treatment, tracking them by laser scanning micrometre and fluorescence microscopy. The low and mid-weight peptides penetrated deep into the cortex; the high-weight peptide “adsorbed onto the hair surface and possibly slightly penetrated into the outer layers of the fibre surface”. Note which one the authors' hedged conclusion is about: it is the high-molecular-weight peptide, the one that stayed on the surface, that “may repair damage on freshly relaxed textured hair”. Tensile testing ran on 50 fibres per group, the electron microscopy on 15 fibres per control group.¹⁴

What does not map onto any document is the framework: the three-way partition, the ordering, the monthly cadence, and above all the diagnostic step in which a reader decides which of the three their hair “needs”. Nothing in the literature opened for this article distinguishes a lipid deficit from a protein deficit by any test that can be performed at a basin. The chemistry under the schedule is real. The schedule is a convention.

What we could not verify

“Hygral fatigue.” The term is ubiquitous in hair-care writing. We could not verify it as a measured quantity: of the papers opened for this article, only one names it, and that is a narrative review, which glosses it in a subordinate clause — “repeated swelling and drying” — without citing a study that quantified cumulative wet-and-dry cycling damage under that name.² Swelling of the fibre in water is separately documented; the leap from one swell to accumulated damage is the part with no measurement behind it here. The argan-specific water-uptake result reported in §2 rests on a single 2024 paper.⁴ The §2 heading uses the adjacent term hygral swelling, which is what the review actually documents.

The Robbins monograph. Chemical and Physical Behavior of Human Hair is behind a paywall. What was opened for this article is the publisher's free reading sample of Chapter 2, pp. 105–180, and every quotation attributed to the book above comes from those pages (§2.2.1.6 at p. 117, §2.3 at p. 122). Chapter 6, which the aging passage points to for the detail, and Chapter 10, which carries the conditioning material, were not opened; statements made there are neither confirmed nor contradicted here.

A citation dropped. A 2013 Journal of Cosmetic Science paper on argan oil and cupuassu butter after hair dye circulates widely with the identifier PMID 23852943. That identifier resolves to an unrelated neuroanatomy paper about the cockroach antenna, and a PubMed search returns no record matching the argan paper's title. The paper may well exist in print; the identifier attached to it in circulation is wrong, so it is not cited above.

CosIng. The Commission's cosmetic ingredient database is a JavaScript application that returns no document to a plain fetch, so the declared functions listed there for ARGANIA SPINOSA KERNEL OIL could not be quoted. The glossary name and its entry number are cited from the Official Journal text instead.⁸

Concentrations in retail masks. Trade sources circulate figures for typical argan-oil inclusion levels in hair masks. None of them resolves to a document that can be opened and checked, so no such number appears in this article.

Sample sizes. The 2022 argan pre-treatment study and the 2024 three-oil study both report tress-level experiments without stating a number of tresses in the material available to us. Their findings are reported above with that limit attached.

Sources

Robbins CR. Chemical and Physical Behavior of Human Hair, 5th edn. Springer Berlin Heidelberg, 2012. ISBN 978-3-642-25610-3. Quotations from §2.2.1.6 (Cosmetically Altered Hair — Bleached Hair, p. 117) and §2.3 (Aging Influences on Hair, p. 122). The book is paywalled; the quoted pages were read in the publisher's free reading sample of Chapter 2, which is the openable document and is linked second here. 10.1007/978-3-642-25611-0 · 9783642256103_excerpt_001.pdf ↩
Gavazzoni Dias MFR. “Hair cosmetics: an overview.” International Journal of Trichology 2015;7(1):2–15. A narrative review; the conditioner and protein-hydrolysate statements quoted above are review statements, not primary measurements. 10.4103/0974-7753.153450 ↩
Marsh JM, Whitaker S, Felts T, et al. “Penetration of oils into hair.” International Journal of Cosmetic Science 2024;46(6):905–917. LC–MS composition, differential extraction, NanoSIMS of 13C-labelled triolein, single-fibre fatigue testing. Coconut, Camellia oleifera and safflower seed oils; argan was not among them. 10.1111/ics.12978 ↩
Lourenço CB, Gasparin RM, Thomaz FM, da Silva GC, Martin AA, Paiva-Santos AC, Mazzola PG. “Impact of hair damage on the penetration profile of coconut, avocado, and argan oils into Caucasian hair fibers.” Cosmetics 2024;11(2):64. Tensile and fatigue testing plus Raman spectroscopy on virgin and bleached tresses; oils applied neat for 24 h at 25 °C. Author affiliations as printed include the Symrise Global Hair Care Center, Cotia, and the Faculty of Pharmaceutical Sciences, University of Campinas. 10.3390/cosmetics11020064 ↩
Rele AS, Mohile RB. “Effect of mineral oil, sunflower oil, and coconut oil on prevention of hair damage.” Journal of Cosmetic Science 2003;54(2):175–192. Taguchi design; protein-loss endpoint. Author affiliation as printed: Research and Development Department, Nature Care Division, Marico Industries Ltd, Mumbai. PMID 12715094 ↩
Sharifi N, Hamedeyazdan S, Shokri J, Monajjemzadeh F. “Argan oil as a pretreatment of human hair before exposure to oxidative damage: attenuated total reflectance and protein loss studies.” Journal of Cosmetic Dermatology 2022;21(10):5010–5017. In vitro; excised human hair tresses in three groups. 10.1111/jocd.14885 ↩
Regulation (EC) No 1223/2009 of the European Parliament and of the Council of 30 November 2009 on cosmetic products, Article 19(1)(g) — list of ingredients, descending order of weight and the 1 % rule. Consolidated text of 1 May 2026. CELEX 02009R1223-20260501 ↩
Commission Implementing Decision (EU) 2025/1175 of 16 June 2025 laying down rules for the application of Regulation (EC) No 1223/2009 as regards the glossary of common ingredient names for use in the labelling of cosmetic products. Annex, Entry 2221, ARGANIA SPINOSA KERNEL OIL. OJ L, 2025/1175, 10.7.2025; applies from 30 July 2026 and repeals Commission Implementing Decision (EU) 2022/677 (same name at Entry 2449), which had itself repealed Commission Decision (EU) 2019/701 (Entry 2155). Most secondary sources still quote one of the two superseded numbers. CELEX 32025D1175 ↩
Commission Regulation (EU) No 655/2013 of 10 July 2013 laying down common criteria for the justification of claims used in relation to cosmetic products. Annex, points 2(2), 3(6) and 4(1). OJ L 190, 11.7.2013, p. 31. CELEX 32013R0655 ↩
Technical document on cosmetic claims, agreed by the Sub-Working Group on Claims and endorsed by the Working Group on Cosmetic Products, version of 3 July 2017. Annex I (common criteria, with the aloe vera and 48-hour hydration examples and the in vitro / in vivo caution) and Annex II (evidential support, including the condition on citing publications about ingredients). The document states on its first page that it is not a European Commission document and is not legally binding. ec.europa.eu/docsroom/documents/24847 ↩
Regulation (EC) No 1223/2009, Article 20(1) — product claims. Consolidated text of 1 May 2026. CELEX 02009R1223-20260501 ↩
ANVISA, Resolução da Diretoria Colegiada — RDC nº 907, de 19 de setembro de 2024, arts. 8(IX), 9(II), 12, 13(II)(i) and 51(I). Consolidated text consulted 6 August 2026; the strings reconstru and nutri do not occur anywhere in it. RDC 907/2024 ↩
ANVISA, RDC nº 907, de 19 de setembro de 2024, Anexo I (Classificação de produtos de higiene pessoal, cosméticos e perfumes), Lista de grupos de produtos de Grau 1, items 45 and 48. Annex I in the redaction given by RDC nº 949 of 12 December 2024. The annex states that the lists are indicative, exemplary and non-exhaustive. RDC 907/2024, Anexo I ↩
Malinauskyte E, Shrestha R, Cornwell PA, Gourion-Arsiquaud S, Hindley M. “Penetration of different molecular weight hydrolysed keratins into hair fibres and their effects on the physical properties of textured hair.” International Journal of Cosmetic Science 2021;43(1):26–37, published online 15 October 2020. Methods as printed: 150 fibres per group for cross-sectional area, 50 per group for tensile testing, 15 fibres from each control group for scanning electron microscopy. 10.1111/ics.12663 ↩

Corrections to this article

None yet. Errors of fact are fixed in public and logged with the date, what we said and what it says now. The corrections log · Report an error.