Every fact on this site, with its evidence grade, its caveats and a link to the source.
We start you on the strong ones: findings that have been replicated or come from one large, well-controlled study. Moderate and weak facts are here too, labeled for what they are. Use “Link to this fact” to share a single card.
Hidden ingredients / Measured in people or real products
In lab tests of 213 US household and personal-care products, not one label listed a phthalate, yet the phthalate DEP, a common fragrance solvent, turned up in half of the pooled conventional product samples.
Strong evidence: Products bought 2007-08; composited samples; content not dose.
Source: Dodson 2012, Environmental Health Perspectives Link to this fact
Hidden ingredients / Review of other studies
US law does not require the individual chemicals in a fragrance to be listed, so a shopper cannot tell from the label whether a scented product contains diethyl phthalate.
Strong evidence: Regulatory background statement, as of 2013.
Source: Parlett 2013, Journal of Exposure Science and Environmental Epidemiology Link to this fact
Hidden ingredients / Review of other studies
None of the products used by the four affected children was ever chemically analyzed; the researchers could only assume that a product labeled as containing lavender oil actually met the international composition standard.
Strong evidence.
Caveats
Stated by the authors as a limitation; an industry letter later claimed one named cologne contained diethyl phthalate and no lavender oil, which has also never been independently verified.
Source: Ramsey 2019, The Journal of Clinical Endocrinology & Metabolism Link to this fact
Hidden ingredients / Review of other studies
The industry's own safety panel wrote that a single fragrance can contain 50-300 ingredients, that formulas are 'carefully guarded', and that the industry 'is often perceived as rather less than forthcoming by consumers and their physicians'.
Strong evidence: Direct quotation from an industry-authored 2003 paper; 50-300 is a range, not a measured count.
Source: Bickers 2003 (RIFM Expert Panel), Regulatory Toxicology and Pharmacology Industry-funded Link to this fact
Hidden ingredients / Advocacy group testing
Ingredient labels cannot show what is inside 'parfum': a 2010 Canadian cosmetics survey had to leave out diethyl phthalate, a common fragrance solvent, because 'short of a laboratory analysis, there's no way to know'.
Strong evidence: Illustrative quote; for measured prevalence cite Dodson 2012 (DEP in 21/42 conventional composites, no label listed a phthalate).
Hidden ingredients / Government agency
US cosmetic labels can still say just "fragrance". The 2022 law requires companies to name fragrance allergens on the label, but only the ones the FDA lists in a regulation, which the law told the FDA to propose by June 2024.
Strong evidence: Statute §609(b); allergens, number and thresholds left to FDA; check whether the final rule is in force before saying labels carry them.
Source: MoCRA 2022 Link to this fact
Hidden ingredients / Government agency
Even the FDA can only ask what is in a cosmetic's fragrance after a serious adverse event it has reason to blame on the fragrance; the company may answer with "categories of ingredients", and the answer is exempt from public-records requests.
Strong evidence: Statute §605(f); 30-day deadline; records access separately excludes formulas (§610).
Source: MoCRA 2022 Link to this fact
Hidden ingredients / Review of other studies
About 1 in 10 (370 of 3,820) topical medicines in a Belgian survey were labeled as containing a fragrance ingredient, and there is no EU or Belgian legislation restricting fragrance in topical medicines.
Strong evidence.
Caveats
Now sourced directly from the primary (Nardelli 2009), not second-hand via Pastor-Nieto 2021; exact figures confirmed (370/3,820 = 9.7%, rounds to the paper's stated 10%). This is a label survey (fragrance listed as an ingredient), not a reaction-rate or chemical-analysis figure; 127 of the clinic's 3,378 iatrogenic-ACD patients (3.8%) had a confirmed reaction to one of the fragranced products. Nardelli 2009's own abstract misprints the product-count denominator as 3,280 — use 3,820 from its Methods/body.
Source: Pastor-Nieto 2021 (Nardelli 2009), Current Treatment Options in Allergy Link to this fact
Hidden ingredients / Review of other studies
Sanitary pads, panty liners and tampons can contain undisclosed fragrance allergens — chiefly alpha-isomethyl ionone (in every scented tampon tested), hexyl cinnamal and benzyl salicylate, plus trace limonene, linalool and geraniol — and in the EU they do not have to declare any of them.
Strong evidence.
Caveats
Primary source (Desmedt 2020, validated HS-GC-MS/FID, 10 scented Belgian AHPs) now in our evidence base. The compounds actually found above the EU 10 ug/g labeling threshold were alpha-isomethyl ionone (all 4 tampons, 12.6-19.1 ug/g), hexyl cinnamal and benzyl salicylate (one panty liner); limonene, linalool and geraniol were all trace (<10 ug/g) — the original claim named the wrong allergens as the notable ones.
Source: Pastor-Nieto 2021 (Desmedt 2020), Current Treatment Options in Allergy Link to this fact
Hidden ingredients / Measured in people or real products
Unlisted PFAS impurities (perfluorocarboxylic acids, including chain lengths that overlap PFOA) were found in nearly every PFAS-containing cosmetic tested in a 2024 study, and these unlisted impurities transferred into artificial sweat far more readily (43-76%) than the PFAS ingredient actually printed on the label (which showed negligible transfer).
Strong evidence.
Caveats
Bioaccessibility (partitioning into artificial sweat) is a precursor to, not proof of, skin absorption; sebum was not included in the assay and may raise this further; small sample; no PTFE-containing products were tested.
Source: Namazkar 2024, Environmental Science: Processes & Impacts Link to this fact
Hidden ingredients / Measured in people or real products
In Belgium, about 1 in 10 topical (skin-applied) pharmaceutical products — wound-healing ointments, NSAID gels, antiseptics, antihemorrhoidals, even corticosteroid-antibiotic creams — listed a fragrance ingredient, sometimes disclosed only as the single word "Perfume".
Strong evidence: 370/3,820 products in a Belgian label survey (2009); a label/composition survey, not chemical analysis; historical product list, not a current market snapshot.
Hidden ingredients / Measured in people or real products
A single undisclosed "Perfume" in one antibiotic-corticosteroid cream (Mycolog/Triadcortyl) caused allergic reactions in 34 patients — more than any other product in a 30-year clinic series — and its 28-ingredient composition was only obtained by directly asking the manufacturer, not from the label.
Strong evidence: Single-clinic case series; one product's undisclosed formula, not a general claim about all perfume-labeled products.
Hidden ingredients / Measured in people or real products
Health Canada's own testing of 252 Canadian cosmetic and personal care products found the fragrance solvent DEP in 70% of fragrance products, at levels up to 2.6% by weight, yet every DEP-positive product was labeled only with the generic term "parfum" — never DEP or any phthalate.
Strong evidence.
Caveats
N=252 individually tested products, Canadian market, 2007-08; detection frequency is lower than two smaller earlier surveys (Hubinger & Havery 2006, EWG 2002) found for fragrances specifically.
Source: Koniecki 2011, Environmental Research Link to this fact
Information, not medical advice. See also: myths we won’t tell you.