Comedogenic ratings dataset
Every comedogenic rating on our list, 285 ingredients, as a file you can download. For 123 of them, each row also gives the grade a published rabbit-ear study printed, the name the study tested it under and how it was tested. It is free to use under CC BY 4.0.
What is in it
- 285 ingredients with the rating our list publishes, 0 to 5, and whether it is a range.
- 123 traced to Fulton 1989 or Morris and Kwan 1983, both read in full. Each grade was read twice before it went in, off the page image and again from the PDF text, or for the one table printed as a picture, from a zoomed render checked against its other columns.
- 60 untraced: we checked and could not tie the number to a paper we have read. The row says what we looked for.
- 102 not yet checked against the papers. They carry the list rating only.
- 44 of the 123 traced ingredients where the list is above the research (44 of all 285): 5 two of Fulton's bands apart, 31 one band apart, 8 within the same band. And 3 where the list is lower than every paper grade, so the comparison runs both ways.
- 6 substances the list rates twice, with numbers that do not overlap,
such as groundnut oil and peanut oil. The
same_substance_ascolumn links each pair. - All of it laid out, contradictions first, on where the list and the research disagree.
Where the list came from
This list of 285 ratings was already on the site when the current work began. It was published here on 24 February 2026 as a single table, introduced as a list "we have compiled", and it cited no sources. We have no record of who assembled it or from which source. Some of its ratings can be traced to the published rabbit-ear studies, and the missing provenance is one of the reasons we have been checking it ingredient by ingredient. We have not changed any rating.
What it is not
It is not a copy of any paper's table. Each paper grade is reported for an ingredient on our list, attributed to the paper with its year and test conditions, and only the comedogenicity grade is included. For anything beyond that, read the papers; both are open in the Society of Cosmetic Chemists library.
It is not a verdict on any product, and not medical advice. Every grade describes a raw ingredient put on a rabbit's ear, which reacts more readily than human skin. Fulton 1984 is behind a paywall and has not been read, so nothing here is attributed to it.
Columns
| Column | What it holds |
|---|---|
ingredient | The ingredient as our list names it. |
slug | A stable identifier, the last part of the page URL. |
page_url | The ingredient page on skinsafecheck.com, with the papers in full. |
list_rating | Our list’s comedogenic rating, 0 to 5, exactly as the list published it. Some entries are a range; the list does not say how a range was set. |
list_min | The low end of the list rating. |
list_max | The high end of the list rating. |
list_is_range | true when the list gives a range. |
status | traced: at least one paper we read grades it. untraced: we checked and could not tie the number to a paper we have read. not_yet_checked: not yet compared with the papers. |
fulton1989_grade | The comedogenicity grade Fulton 1989 printed, as printed. "2(1)" is crude oil 2, refined 1. Several values are separated by semicolons. |
fulton1989_tested_as | The name Fulton 1989 printed for the material tested, where it differs, with any note on it being the same substance. |
fulton1989_conditions | How Fulton 1989 tested it. |
morris1983_grade | The comedogenic grade Morris and Kwan 1983 printed. |
morris1983_tested_as | The name and dilution Morris and Kwan 1983 printed. |
morris1983_conditions | How Morris and Kwan 1983 tested it. |
other_findings | What a paper says about the ingredient without giving a 0 to 5 grade: one read at abstract level, or a human or method study read in full. |
untraced_note | For untraced rows, what we checked and did not find. |
list_above_research | true when our list’s highest figure is above every grade the papers give. |
disagreement_bands | When the list is above the research, how far apart in Fulton 1989’s bands (0 to 1 not significant, 2 to 3 borderline, 4 to 5 positive): two_bands, one_band or within_band. |
list_below_research | true when the whole list range is below every grade the papers give. |
same_substance_as | Other entries on our list that name the same substance, with their list ratings. |
list_vs_research | One sentence comparing the list with the papers. |
Licence and how to cite it
The list ratings and our provenance notes are licensed under Creative Commons Attribution 4.0. You may copy, change and build on them, commercially too, if you credit SkinSafeCheck and link here. The papers' grades are published facts: when you repeat one, credit the paper.
Cite the dataset as:
SkinSafeCheck. Comedogenic ratings, with who rated each one. Version 1.3.0, 2026-10-08. https://skinsafecheck.com/downloads/comedogenic-ratings-1.3.0.csv
Versions
Cite a versioned file. Its address names the version and its contents never change, so anyone following your citation gets exactly the figures you quoted. The unversioned file is a convenience that always points at the latest version.
- 1.3.0, 2026-10-08: Findings from the abstract of Ahn and colleagues 1985 (Korean Journal of Dermatology, rabbit ear canal). The full text is not online and the abstract gives no 0 to 5 grade, so no grade and no disagreement count changed. CSV, JSON
- acetylated-lanolin: Ahn 1985 abstract finding added to other_findings.
- butyl-stearate: Ahn 1985 abstract finding added to other_findings.
- cetyl-alcohol: Ahn 1985 abstract finding added to other_findings.
- corn-oil-zea-mays: Ahn 1985 abstract finding added to other_findings.
- isopropyl-myristate: Ahn 1985 abstract finding added to other_findings.
- myristyl-lactate: Ahn 1985 abstract finding added to other_findings.
- myristyl-myristate: Ahn 1985 abstract finding added to other_findings.
- olea-europaea-olive-oil: Ahn 1985 abstract finding added to other_findings.
- sesame-oil-sesame-seed-oil: Ahn 1985 abstract finding added to other_findings.
- sodium-lauryl-sulfate-sls: Ahn 1985 abstract finding added to other_findings.
- 1.2.0, 2026-10-08: Findings from two more papers, read in full: Arquette and colleagues 1998 (jojoba oil, human back skin) and Kligman 1996 (petrolatum, acne patients and the revised rabbit ear). Neither gives a 0 to 5 grade, so no grade and no disagreement count changed. CSV, JSON
- jojoba-oil-butter: Arquette 1998 microcomedone counts for refined and pure jojoba oil against a blank patch.
- acetulan: Arquette 1998 count for Acetulan as the positive control.
- acetylated-lanolin: Kligman 1996 found it comedogenic undiluted in the revised rabbit-ear assay, not below 15 percent.
- isopropyl-myristate: Kligman 1996 on undiluted against below 15 percent.
- mineral-oil: Kligman 1996 on the 1972 rabbit-ear result being a false positive.
- 1.1.1, 2026-10-08: Documentation only. No row changed. CSV, JSON
- The description of list_rating no longer says a range means published ratings disagree. We do not know how the list set its ranges.
- Added where_the_list_came_from to the JSON: the list was already on this site when the current work began, and no record says who assembled it or from which source.
- No value in any row changed; the rows are identical to 1.1.0.
- 1.1.0, 2026-10-08: Three columns added. No rating and no paper grade changed. CSV, JSON
- Added disagreement_bands: how far the list sits above the papers, in Fulton 1989's bands. Of the 44 rows above the research, 5 are two bands apart, 31 one band, 8 within a band.
- Added list_below_research: the 3 rows where the whole list range is below every paper grade.
- Added same_substance_as: 19 substances the list names more than once. 6 have ratings that do not overlap, 6 differ but overlap, 7 agree.
- No value in any 1.0.0 column changed.
- 1.0.0, 2026-10-08: First release. 285 ingredients; 123 traced to Fulton 1989 or Morris and Kwan 1983, every grade read twice; 60 checked and untraced; 44 where the list is above every grade the papers give. CSV, JSON
- First release.
Common questions
Can I use this data on my own site or in my own checker?
Yes. The list ratings and our notes are under CC BY 4.0: use them, change them, sell what you build, as long as you credit SkinSafeCheck and link to this page. The papers’ grades are facts from published research; credit the papers when you repeat them.
Why do some ratings disagree with the papers?
The list is a compilation, and some of its numbers are higher than any grade in the two rabbit-ear papers we have read in full. 44 rows are flagged that way. We publish the list as it stands and show the papers beside it, so you can see both. The comparison is laid out on its own page.
Is this medical advice?
No. Every grade is for a raw ingredient tested on a rabbit’s ear, which reacts more readily than human skin. None of it predicts how a finished product will behave on a person.
Sources in the dataset
- Fulton JE Jr. “Comedogenicity and irritancy of commonly used ingredients in skin care products.” Journal of the Society of Cosmetic Chemists, November/December 1989;40(6):321–333. (opens in a new tab) Rabbit ear. Each ingredient at 10 percent in propylene glycol unless stated, applied five days a week for two weeks, and graded 0 to 5 from the width of the follicles, measured by micrometer and under the microscope. The paper counts 0 to 1 as not significant, 2 to 3 as borderline and 4 to 5 as positive. Some grades depended on the supplier of the raw material, and some oils were also tested refined.
- Morris WE, Kwan SC. “Use of the rabbit ear model in evaluating the comedogenic potential of cosmetic ingredients.” Journal of the Society of Cosmetic Chemists, August 1983;34(5):215–225. (opens in a new tab) Rabbit ear canal, at the Revlon Research Center. Each material applied undiluted unless stated, once a day five days a week for 14 applications, and graded 0 to 5 by microscope only. The authors set out to recheck earlier results because investigators disagreed.
- Fulton JE Jr, Bradley S, Aqundez A, Black T. “Non-comedogenic cosmetics.” Cutis, February 1976;17(2):344–351. (opens in a new tab) Rabbit ear. Read at abstract level: it names butyl stearate, isopropyl myristate and sodium lauryl sulfate among the emulsifiers found comedogenic.
- Nguyen SH, Dang TP, Maibach HI. “Comedogenicity in rabbit: some cosmetic ingredients/vehicles.” Cutaneous and Ocular Toxicology, 2007;26(4):287–292. (opens in a new tab) Rabbit ear. Read at abstract level: of fourteen materials, the first nine named (isopropyl palmitate, isopropyl myristate, butyl stearate, isopropyl isostearate, decyl oleate, isostearyl neopentanoate, isocetyl stearate, myristyl myristate and cocoa butter) were deemed positive; cetyl alcohol, paraffin, stearyl alcohol, sodium lauryl sulfate and petrolatum were not among them.
- Kligman AM, Kwong T. “An improved rabbit ear model for assessing comedogenic substances.” British Journal of Dermatology, June 1979;100(6):699–702. (opens in a new tab) The rabbit-ear assay behind published 0 to 5 comedogenicity grades.
- Draelos ZD, DiNardo JC. “A re-evaluation of the comedogenicity concept.” Journal of the American Academy of Dermatology, March 2006;54(3):507–512. doi:10.1016/j.jaad.2005.11.1058 (opens in a new tab) Found that animal-model results did not reliably predict what happened on human skin, and that a rating for a single ingredient does not tell you how a finished product will behave.