'Hypoallergenic' Means Whatever the Brand Decided It Means

A federal court struck down the only attempt to define it — not because the science was wrong, but because nobody could prove what shoppers thought the word meant.

25-Second Preview A quick listen to set the scene — the full story’s below.
Read the preview

What does hypoallergenic legally mean? Nothing fixed. The FDA tried to define it, and a federal court struck the rule down in 1977 — not over the science, but because nobody proved what shoppers thought the word meant. Dermatologist-tested only says testing happened, not that anything passed. So which claims do carry a real standard? The Dermington Gazette tests the claims.

The word was undefined by a court, not by neglect

The usual explanation for 'hypoallergenic' meaning nothing is regulatory indifference — the US never got around to it. That is not what happened, and the real story is much better.

In the 1970s the FDA did write a rule. Its approach was comparative rather than absolute, which was clever: to call a product hypoallergenic, a manufacturer would have to show by testing that it caused a significantly lower rate of adverse skin reactions than competing products covering a meaningful share of the market for that product type. No absolute threshold to game, just a requirement to be measurably better than the field.

Almay and Clinique challenged it. The district court sided with the agency. On appeal, the D.C. Circuit reversed in 1977, on the ground that the FDA had not demonstrated that consumers actually understood 'hypoallergenic' to mean what the regulation assumed. The rule came out of the Code of Federal Regulations, and the agency's own labelling guidance has since said plainly that ever since that decision the term has had no real meaning in the marketplace.

So the position today is that a manufacturer may print 'hypoallergenic' on anything, with no supporting evidence of any kind, and no obligation to define what it claims. Read the appellate reasoning and it is almost funny: the word was too vague to regulate, so it stayed vague. Half a century later it is one of the most trusted words on a shelf.

Tested, recommended, developed — three different claims, routinely conflated

The dermatologist family of claims is not one claim. It is three, and the differences are worth internalising because brands rely on you not making them.

'Dermatologist-tested' asserts that a dermatologist was involved in some testing of the product. It specifies nothing about the protocol, the number of subjects, the duration, the endpoints, or — and this is the part that startles people — the result. It is a statement that an event occurred. A study with a dozen participants over four days, supervised by one clinician, satisfies it. So would a much better study. You cannot tell which from the box, and crucially, a product that performed badly in testing may still carry the phrase, because the phrase does not claim the testing went well.

'Dermatologist-recommended' is usually derived from a survey — sometimes a large, methodologically stated one, often not. The number that matters is the denominator and it is almost never printed. It also captures habit rather than assessment: a product recommended for twenty years continues to be recommended partly because it is what everyone already recommends.

'Dermatologist-developed' is the strongest of the three, because it is a checkable claim about who was in the room during formulation. It still carries no standard, but it is falsifiable in a way the others are not — you can go and look at whether the named clinician exists and what they did.

None of this makes the claims worthless. It makes them weak priors. A brand that has gone to the trouble of clinical involvement is, on average, more careful than one that has not. That is a real if modest signal, and it is roughly the weight our own dermatologist-recommended moisturizer ranking gives it — the assessment underneath is built on formulation and evidence, with the claim treated as context rather than as a credential.

The rabbit's ear, and the correction that never travelled

Non-comedogenic has the most interesting provenance of the three, because it descends from a specific piece of 1970s research that was later formally challenged in the same journals — and the challenge never reached consumers.

The concept of acne cosmetica was described in the early 1970s: cosmetics could produce comedones in susceptible people. To identify the culprits, researchers used the rabbit external ear canal, an unusually comedo-prone tissue, applying raw materials and counting the follicular plugging that resulted. From this came ingredient-level comedogenicity ratings on a five-point scale, and those ratings were distributed to dermatologists as avoid-lists.

The correction arrived quickly and was ignored just as quickly. A 1982 review in the Journal of the American Academy of Dermatology asked directly whether the rabbit ear model, in its then-current state, was prophetic of acnegenicity. Its answer catalogued problems in how the test was performed, an absence of correlation with what actually happened in humans, and consequently serious limitations on any conclusion drawn from it — particularly for the practising dermatologist. That was more than four decades ago.

The deeper problem is one of unit. A 2006 study applied a modified human assay to finished cosmetic products rather than raw materials: six volunteers with prominent follicular orifices and a demonstrated ability to form comedones, patches applied to the upper back three times weekly for four weeks, then cyanoacrylate biopsies to count follicles and microcomedones per square inch. Its conclusion was direct — finished products using comedogenic ingredients are not necessarily comedogenic.

That single sentence invalidates the entire practice of reading a formula by scanning it for high-rated ingredients. Concentration matters, the vehicle matters, what else is in the formula matters, and the interaction of all three is not predictable from a list. Yet the ingredient charts persist, screenshotted and reshared, and they are the reason people avoid perfectly reasonable products. Mineral oil and petrolatum both carry the scars of that era; both are among the least reactive materials in common use, and both spent decades on avoid-lists generated in a rabbit.

So what is 'non-comedogenic' worth?

Not nothing, and this is where reflexive debunking overshoots.

There is no legal definition and no required test, so the claim's evidentiary floor is zero. Some brands genuinely run a human comedogenicity study on the finished product, which is the right test and a meaningful one. Others screen their ingredient list against the old rating tables, which we have just established is testing the wrong unit. Most sit somewhere between, and you cannot tell from the package which one you are holding.

What the claim reliably signals is intent. A formulator who put 'non-comedogenic' on a moisturiser was thinking about acne-prone users, and that concern usually shows up elsewhere in the formula — a lighter emollient system, fewer heavy occlusives, a texture built for oilier skin. Treat it as a tiebreaker between two otherwise similar products, not as a guarantee about your face.

And notice how narrow the claim is even when true. Comedogenicity is about follicular plugging. It says nothing about whether a product will cause inflammatory acne through a different route, nothing about irritation, and nothing about Malassezia-driven folliculitis, which responds to a completely different set of ingredients. A product can be genuinely non-comedogenic and still break you out.

If you are trying to work out which step in a routine is responsible, our pore-clog checker will map the likely candidates across your products, and our answer on what non-comedogenic means handles the label question directly. For the specific case that generates the most questions, the answer on sunscreens that don't break you out deals with texture and filter systems rather than with badges.

The test that works is the one you run on yourself

All three claims are attempts to answer a question that is irreducibly individual: will this product suit me. No population-level label can answer it, which is why the honest replacement is a personal protocol rather than a better badge.

For allergy and irritation, patch test. One product at a time, a coin-sized area on the inner forearm or behind the ear, twice daily for five to seven days, read at forty-eight and ninety-six hours rather than immediately — allergic reactions are delayed, and a product that feels fine on application can produce dermatitis three days later. Our patch test tool sets out the timing. If you already have an established reaction and want to know what a formal clinic workup involves, the contact dermatitis pathway covers it.

For comedogenicity, the timeline is longer and the discipline is harder. Comedones take weeks to become visible papules, so a fortnight tells you nothing. Introduce one new product at a time, hold everything else constant for four to six weeks, and if you want a genuinely controlled result, apply it to one side of the face only — split-face is the design researchers use precisely because it removes every confounder except the product.

A bland baseline makes this far easier, because it minimises the number of variables you are holding constant. Vanicream Daily Facial Moisturizer and Cetaphil Moisturizing Cream are both deliberately unexciting in ways that suit an elimination protocol; CeraVe Daily Moisturizing Lotion and Neutrogena Hydro Boost Gel-Cream are lighter options for oilier skin running the same experiment. Our fragrance-free moisturizer ranking and oil-free ranking are organised around the two variables that actually predict outcomes, rather than around what the front of the box says.

The claims that do mean something

It would be a poor article that dismantled three labels and left you with nothing to trust. Several claims on a package are rigorously defined, and the pattern is worth noticing: they are the ones nobody bothers to market, because they are boring.

SPF is the strongest. It is not an adjective but the output of a specified experiment — product applied at 2 milligrams per square centimetre, a solar simulator, minimal erythemal doses read on protected and unprotected skin, a minimum of ten valid results. 'Broad spectrum' requires passing a separate spectral test with a mean critical wavelength of 370 nanometres or greater. Water resistance may only be claimed as forty or eighty minutes, on the basis of the corresponding immersion protocol. And the terms 'sunblock', 'waterproof' and 'sweatproof' are prohibited outright — which is why you no longer see them, and a useful demonstration that the framework can bite when a claim is defined.

OTC drug actives are the second solid category. When a package says two-and-a-half percent benzoyl peroxide, or salicylic acid at an acne-monograph strength, or zinc oxide as a skin protectant, those are regulated actives at regulated concentrations, subject to specified labelling. The active ingredients panel on a Drug Facts box is a fundamentally different kind of statement from a marketing line on the front.

Third are voluntary seals with published criteria and an application process behind them — the National Eczema Association's Seal of Acceptance being the most familiar in this category. It is not a regulation, but it is an actual standard somebody has to meet, which puts it in a different class from a self-declared adjective.

And fourth, quietly, is 'fragrance-free'. It has no legal definition either — but unlike the other three it is verifiable from the panel, because you can read the ingredient list and look for parfum, fragrance and the declarable allergens. That is the real dividing line running through this whole article. Trust the claims you can check against the ingredient list, and treat the ones that can only be checked by trusting the brand as what they are: a tone of voice. Run the panel through our ingredient decoder and the box stops mattering.

References

  1. Is the rabbit ear model, in its present state, prophetic of acnegenicity? — Journal of the American Academy of Dermatology, 1982
  2. A re-evaluation of the comedogenicity concept — Journal of the American Academy of Dermatology, 2006
  3. Acne cosmetica — Archives of Dermatology, 1972
  4. Comedogenicity of current therapeutic products, cosmetics, and ingredients in the rabbit ear — Journal of the American Academy of Dermatology, 1984
  5. Comedogenicity of sunscreens. Experimental observations in rabbits — Archives of Dermatology, 1982

Every cited link is verified to resolve to the paper named. Where a source could not be independently confirmed, it is listed without a link rather than pointing you somewhere wrong.

Frequently asked

Why does 'hypoallergenic' have no legal definition in the US?

Because the definition was struck down rather than never written. FDA's rule would have required comparative testing showing a lower adverse reaction rate than competing products, and the D.C. Circuit vacated it in 1977 on the ground that the agency had not demonstrated consumers understood the term that way. The regulation was removed and never replaced.

Does 'dermatologist-tested' mean the product passed a test?

No. It asserts only that testing involving a dermatologist occurred — no required protocol, sample size, duration, endpoint or outcome threshold. A product that performed poorly can still carry the phrase, because the phrase makes no claim about the result.

Are comedogenic ingredient ratings reliable?

They were derived from a rabbit external ear canal assay, and a 1982 review in JAAD reported an absence of correlation with human experience and serious limitations on the conclusions available from it. More decisively, a human study on finished products concluded that products containing comedogenic ingredients are not necessarily comedogenic — the ratings assess the wrong unit.

Should I ignore 'non-comedogenic' entirely?

Treat it as a weak signal of intent rather than a guarantee. A formulator who printed it was thinking about acne-prone skin, which usually shows in the emollient system and texture. But it has no required test, and it says nothing about irritation or about Malassezia-driven folliculitis, which follows a different mechanism.

Which label claims actually carry a defined standard?

SPF and broad spectrum, which are outputs of specified test protocols including a critical wavelength threshold of 370 nanometres; water resistance at forty or eighty minutes; OTC drug actives at monograph concentrations on a Drug Facts panel; and voluntary seals with published criteria such as the National Eczema Association's. 'Fragrance-free' is undefined but verifiable from the ingredient list, which is the practical dividing line.

Products named in this piece

Independently reviewed, with a verdict per concern — the real products behind this piece.

Listen next

All episodes
BRAND PORTRAIT106Bioderma

Brand Portrait Ep. 1063:47

Bioderma

0:00 -0:00
Was this piece useful?
What missed? Optional

Not sure where to go next?