Methodology
Where comedogenic ratings come from, what's wrong with them, and why this checker still uses them — with the caveats attached to every number.
What the numbers are
Every 0–5 rating on this site descends from one experimental lineage: the rabbit-ear assay. A test material — usually a single raw ingredient at 10% in propylene glycol — is applied daily to the inner ear of a rabbit for two weeks, and the follicles are then measured for keratin plugging. Fulton's 1989 paper in the Journal of the Society of Cosmetic Chemists is the largest single table in that lineage (about 150 ingredients) and the primary source for this site; his 1984 paper in JAAD and Kligman and Mills's 1972 "acne cosmetica" work are the roots.
The scale is: 0 no plugging, 1–2 minimal to moderate, 3–5 significant to severe, with isopropyl myristate as the grade-5 positive control. Nearly every "pore-clogging ingredients" list circulating online is a reproduction — often several copies removed — of those tables.
What's wrong with them
They test ingredients, not products. A raw material at 10% on a rabbit's ear says little about a finished formula at 0.5% on a human face. Draelos and DiNardo (JAAD 2006) put finished products containing "comedogenic" ingredients on humans using the Mills–Kligman follicular-biopsy method and found the products did not produce comedones. Fulton's own vehicle studies (his Tables II and III) showed the same pigment grading 5 in mineral oil and 0 in PEG-400.
The rabbit ear over-predicts. It is far more sensitive than human skin. Fulton wrote that negative results are more reliable than positive ones; the 1982 JAAD paper "Is the rabbit ear model, in its present state, prophetic of acnegenicity?" asked the question in its title. The 2025 JAAD Reviews survey of the field reports inconsistent translation to human skin and no regulatory standard behind the word "non-comedogenic" on a label.
The copies drift. Reproduced lists disagree with each other and with the original table. This site's Contested tier exists because of that: about one row in six carries a disclosed range rather than a single number.
The popular-science fact-checks — Michelle Wong's Lab Muffin piece on how to use comedogenicity ratings is the clearest — reach the same conclusion: the ratings are being used exactly the way their own source papers say not to use them.
Why we still flag at 3 and up
Because a flag is a prompt to look closer, not a verdict on a product. The 1989 American Academy of Dermatology Invitational Symposium on Comedogenicity concluded that rabbit-ear scores of 0–2 are unlikely to clog human pores, while higher scores warrant more caution. That is the threshold this checker uses: 3/5 and up is flagged, 2/5 is shown as "worth knowing about", 0–1 is not raised at all.
What a flag means here: a published test graded the raw ingredient in the range where caution was recommended. What it does not mean: that the product will break you out. Concentration, position on the label, the rest of the formula, and your own skin decide that — and no ingredient list can.
How to read our confidence tiers
Every rating on this site carries its source and a tier, shown on the verdict card and the ingredient page:
- Established — the number is the primary-source value (Fulton 1989) and no published disagreement is on record.
- Single-source — only the reproduced compilation lists carry the number; it has not been verified against the original table. About half the database sits here. We say so rather than pretend otherwise.
- Contested — published values disagree, or the grade depends on refinement or vehicle. The note on the row says which value is shown and why.
Precedence when values disagree: a named primary source outranks later unattributed lists, so its value is shown and the others are disclosed. Where only compilations disagree, the higher-caution value is shown. Where no nameable publication carries a number at all, no number is shown — the ingredient appears as "not in our database". That is not a pass: nothing we hold flagged it, and nothing cleared it. (An unmatched name in the checker can also be a spelling or alias we don't know yet — the database is finite.)
What this site will not do
- Invent a rating, or round a range to whichever number looks tidier.
- Rate finished products by name. Formulas change without notice; a confident wrong verdict is worse than none.
- Diagnose, promise results, or advise treatment. This is an ingredient reference, not medical advice.
Bibliography
Primary sources (the ratings)
- Fulton JE. Comedogenicity and irritancy of commonly used ingredients in skin care products. J Soc Cosmet Chem. 1989;40:321–333.
- Fulton JE, Pay SR, Fulton JE III. Comedogenicity of current therapeutic products, cosmetics, and ingredients in the rabbit ear. J Am Acad Dermatol. 1984;10(1):96–105.
- Kligman AM, Mills OH. Acne cosmetica. Arch Dermatol. 1972;106(6):843–850.
- Mills OH, Kligman AM. Comedogenicity of sunscreens. Arch Dermatol. 1982;118(6):417–419.
- American Academy of Dermatology Invitational Symposium on Comedogenicity. J Am Acad Dermatol. 1989;20(2):272–277. (The 3+ flag threshold.)
Critiques and re-evaluations (why the numbers are hedged)
- Draelos ZD, DiNardo JC. A re-evaluation of the comedogenicity concept. J Am Acad Dermatol. 2006;54(3):507–512.
- Is the rabbit ear model, in its present state, prophetic of acnegenicity? J Am Acad Dermatol. 1982.
- Comedogenicity in cosmeceuticals: a review of clinical relevance, regulatory gaps, and future directions. JAAD Reviews. 2025.
- Safety assessment on comedogenicity of dermatological products containing d-alpha tocopheryl acetate in Asian subjects: a double-blind randomized controlled trial. 2021. (Human testing of finished products.)
- Wong M. Fact-check: how to use comedogenicity ratings. Lab Muffin Beauty Science.