Don’t take our word for it. Here’s every fact on this site, with its evidence grade, its caveats, and a link to the source, so you can check us.
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.
Children and pregnancy / Measured in people or real products
16.5% of children and adolescents patch tested in 9 population studies had a contact allergy, and in Swedish 16-year-olds fragrance (2.1%) was the second most common allergen after nickel.
Strong evidence: Any allergen, not fragrance only; some adolescent studies read too early and underestimate.
Children and pregnancy / Review of other studies
Children exposed to household chemical/cleaning products before birth or in their first year wheeze more often, now replicated in four independent prospective birth cohorts. The founding ALSPAC cohort (UK, n=7,019, Sherriff et al. 2005) found that an 11-product, frequency-weighted prenatal household chemical (TCB) score was significantly associated with persistent wheeze through 3.5 years (adjusted OR 1.06 per unit increase; top-vs-bottom-decile adjusted OR 2.3), but not with transient early or late-onset wheeze; this is the same ALSPAC cohort and composite-score methodology that both Casas 2013 and Mikeš 2019 name as their model. In the Canadian CHILD cohort (n=2,022), frequent use in early infancy was linked to higher odds of recurrent wheeze (adjusted OR 1.35) and asthma diagnosis (adjusted OR 1.37) by age 3, but not isolated skin-prick allergy (adjusted OR 1.14, ns), though wheeze combined with allergy was itself raised (adjusted OR 1.49). In the Czech ELSPAC-CZ cohort (n=3,411), each 1-SD increase in an 18-product household chemical exposure score at 6 months raised the odds of intermediate-onset transient, intermediate-onset persistent, and early-onset persistent wheeze phenotypes through age 5 (adjusted OR 1.23-1.36), and those persistent phenotypes carried 5-8x higher odds of doctor-diagnosed asthma at age 7 (pediatrician records, not self-report). In the pooled Spanish INMA cohorts (n=2,292), prenatal (third-trimester) use of cleaning sprays raised the odds of both wheezing (OR 1.37) and lower respiratory tract infection (OR 1.29) in the first year of life, with solvents raising wheezing odds (OR 1.30) and air fresheners raising LRTI odds (OR 1.29); bleach, ammonia, and non-spray cleaners showed no association.
Strong evidence.
Caveats
All four cohorts are observational and use self-reported product-use frequency, not measured indoor air chemistry. The ALSPAC/Sherriff 2005 and ELSPAC-CZ signals are specific to persistent-type wheeze phenotypes, not wheeze in general; ALSPAC found no association with transient early or late-onset wheeze. The 'not atopy' finding remains single-cohort (CHILD only); ELSPAC-CZ's atopic-rash variable is a covariate, not an outcome, and Casas 2013 measured neither atopy nor a physician asthma diagnosis, only maternal-reported LRTI and any-wheezing through 12-18 months, so it neither confirms nor refutes the atopy finding; ALSPAC/Sherriff 2005 did not assess atopy or asthma at all. Casas 2013 is the only cohort among the three post-2010 cohorts with genuinely prenatal (rather than early-postnatal) exposure measurement, but ALSPAC/Sherriff 2005 is also prenatal and its composite score behaved like CHILD's and ELSPAC-CZ's (significant), not like Casas's (null) — evidence against prenatal timing per se explaining the Casas null. Casas 2013's own summed cleaning-product-count score was NOT a significant predictor, unlike ALSPAC's, CHILD's, and ELSPAC-CZ's frequency-weighted composite scores — an open methodological tension (register entry C262) that means 'more products used' is not yet a uniformly replicated exposure metric across all cohorts, even though several specific products (sprays consistently, plus solvents and air fresheners) are.
Source: Salonen 2024, Environment International Link to this fact
Health effects / Government agency
The EU committee said in 2012 that one fragrance allergen, HICC, had caused more than 1,500 published allergy cases in a decade and should not be used in consumer products, even at 200 ppm, and that voluntary industry reductions were insufficient.
Strong evidence.
Caveats
Case counts come from published literature and are described as a severe underestimate; the 'voluntary restrictions are insufficient' judgment is the regulator's reading of available evidence; industry's counter-position is not presented in this document.
The industry / Lab study (cells or chemistry)
The lab tests that replaced animal testing for setting cosmetic and fragrance allergy safety limits explain only about 62% of how strongly a chemical causes allergy.
Strong evidence.
Caveats
Regression R2 against mouse LLNA, not human outcomes; industry-authored. Natsch 2023 (same author) retrains against human-anchored data: the 62% figure survives on a blended human+LLNA composite (63.4%, n=139), but against strictly human data (n=62) accuracy drops to 32-45%, and even the LLNA itself only reaches 43% predicting human potency — so '62%' should not be read as human-relevant accuracy. Api, Basketter & Lalko 2014 (RIFM's own earlier LLNA-vs-human validation, 57 fragrance chemicals) independently corroborates, by a different metric (categorical concordance, not R2) and seven years earlier, that the LLNA itself is an imperfect human proxy: 76% agreement within half an order of magnitude, but up to ~790-fold mispredicted for salicylates and ~26-42-fold for trans-2-hexenal/methyl-2-nonanoate. Roberts & Api 2018 (same fragrance dataset, direct LLNA-vs-human regression) reports a much higher R2=0.784, but only for a 35-chemical subset that already excludes those 14 known outliers a priori — demonstrating that domain-restricted R2 figures like this one can substantially overstate accuracy on unselected chemistry, a caution that applies to curated NAM training/case-study sets generally, not only to this specific figure.
Source: Natsch & Gerberick 2022, ALTEX Industry-funded Link to this fact
The industry / Lab study (cells or chemistry)
Accuracy has not improved: a 2015 model predicted allergy potency about as well as the 2022 one.
Strong evidence.
Caveats
Confirmed by reading the 2015 primary directly, not just inferred from one comparison line in the 2022 paper: Natsch 2015 Equation 7 gives R²(adj)=62.3% (Results, not the abstract's rounded 60%), matching 2022's 62% and 2023's 59-65%.
Source: Natsch & Gerberick 2022, ALTEX Industry-funded Link to this fact
The industry / Other evidence
By 1994 the fragrance industry itself had banned or restricted 82 fragrance materials, most because they cause skin allergy and a dozen because they cause skin reactions in sunlight.
Strong evidence: Counts restrictions, not harm rates; limits not printed.
Source: Ford 1994 (RIFM) Industry-funded Link to this fact
The industry / Lab study (cells or chemistry)
The lab tests that replaced animal testing for fragrance and cosmetic allergy limits have been stuck at explaining about 62% of allergy potency since 2015 — the 2015 model and the 2022 model perform the same.
Strong evidence: Regression R2(adj) against mouse LLNA potency, not human outcomes; industry-funded (Givaudan).
Source: Natsch 2015, Toxicological Sciences Industry-funded Link to this fact
The industry / Government agency
A population study widely treated in dermatology literature as independent, EU-style evidence that confirmed dermatologists' fragrance-allergy estimates over the industry's own claims was itself funded by the fragrance industry's own research body, the Research Institute for Fragrance Materials (RIFM) — a funding fact omitted by every earlier secondhand citation of this study.
Strong evidence.
Caveats
The industry-funded study still confirmed the higher, consumer-relevant estimate (1.9% clinically relevant, up from a disputed 'up to 2%') over industry's own 2000 'far lower' claim — the funding did not appear to weaken the result in this case.
Source: Diepgen 2015 (EDEN), British Journal of Dermatology Industry-funded Link to this fact
The industry / Government agency
As of 2021, the EU's cosmetics regulator states directly that it is not yet possible to use the fragrance industry's QRA2 methodology to set a concentration below which a fragrance chemical is unlikely to cause allergy in new users.
Strong evidence: A regulatory verdict on methodology validity, not a finding that current fragrance concentration limits are unsafe; QRA2 remains in industry use for setting IFRA limits regardless.
Laws and loopholes / Government agency
In 2000, EU cosmetics could declare an entire fragrance as the single word 'parfum'; the EU committee noted dermatologists' estimate that up to 2% of people have perfume allergy, against industry's claim that the true figure was 'far lower'. The EDEN population-based patch-test study (Diepgen 2015; n=3,119 patch-tested across 5 European countries) directly confirmed the dermatologists' estimate: 1.9% (95% CI 1.5-2.5) clinically relevant fragrance contact allergy, and 1.8-2.6% for fragrance mix I alone.
Strong evidence.
Caveats
EDEN is the largest general-population patch-test study of fragrance allergy to date and a large, well-controlled multi-country primary (not a single-lab or single-cohort study), which is why this moves to strong. Note EDEN was itself funded by the Research Institute for Fragrance Materials (RIFM), the fragrance industry's own body — the industry-funded study still confirmed the higher, consumer-relevant estimate over the industry's own 'far lower' claim.
Source: SCCNFP 2000 Link to this fact
Skin allergy / Review of other studies
EU scientific advisers (SCCS 2012) list 82 fragrance ingredients as established causes of skin allergy in people.
Strong evidence.
Caveats
Primary now ingested (SCCS/1459/11): 82 = 54 single chemicals + 28 natural extracts on clinical evidence alone, including the 26 labeled in 1999; the count measures reported allergy evidence, not safety.
Source: Kumar 2020 (SCCS 2012), Journal of Cosmetic Dermatology Link to this fact
Skin allergy / Measured in people or real products
The fragrance chemicals most often found in household products were mostly not in the standard fragrance-allergy patch test used at the time: 6 of its 8 chemicals were in 12% of products or fewer.
Strong evidence.
Caveats
Frequency in products is not the same as clinical sensitization rate; FM II and hydroperoxide testing later added. Clinically confirmed in Heydorn et al. 2003 (n=658 hand-eczema patients, 3 centers): only 3 of the 14 chemicals selected from this paper's household-product survey were FM I constituents, and FM I caught only 45.5% of the fragrance-allergic patients (5.6% of all patients tested missed). The exposure-frequency leader, limonene (78% of products), caused only 0.8-0.9% oxidized-limonene reactivity clinically — product frequency of a chemical still does not predict its clinical allergy rate.
Source: Rastogi 2001, Contact Dermatitis Industry-funded Link to this fact
Skin allergy / Measured in people or real products
Common scent chemicals such as linalool hardly cause allergy when fresh, but once exposed to air they form new chemicals that do: in 1,511 dermatitis patients at six European clinics, 1.3% were allergic to air-oxidized linalool (later studies with a stronger test found about 7%).
Strong evidence.
Caveats
Dermatitis patients, not general population; 2005 test concentration (2.0% pet.) was later shown too low. Clinical relevance no longer just inferred: Andersch Björkman 2014 (ROAT) shows patients identified this way go on to develop dermatitis from twice-daily use of a cream/perfume containing oxidized linalool at concentrations as low as 0.3% (0.056% linalool hydroperoxides, ~27 µg/cm²) — inside the 0.1–2% range measured for individual fragrance ingredients in real products.
Skin allergy / Measured in people or real products
Fresh limonene and linalool rarely cause allergy; the allergens form when these chemicals react with air, which also happens in essential oils and perfumes during storage.
Strong evidence.
Caveats
A '0.1-0.3% of patients' figure was removed because it does not trace to a source in our evidence base. Parent-compound sensitization rates are low but vary by dataset; see the Dittmar 2019 entry for oxidized-terpene positive rates.
Source: Bråred Christensson 2016, Contact Dermatitis Industry-funded Link to this fact
Skin allergy / Measured in people or real products
60.5% of patients allergic to oxidized linalool would never have learned they had a fragrance allergy from the standard fragrance patch tests.
Strong evidence: Dermatitis patients.
Source: Christensson 2012, Contact Dermatitis Link to this fact
Skin allergy / Review of other studies
Standard allergy tests with fresh limonene and linalool miss most of these allergies, so European experts now recommend adding the oxidized forms to routine patch testing.
Strong evidence.
Skin allergy / Review of other studies
Scent chemicals such as limonene and linalool rarely cause skin allergy when fresh, but once exposed to air they form much stronger allergens; products and oils kept open or stored a long time can contain more of them.
Strong evidence.
Caveats
Review of experimental and clinical work; no new data. Hydroperoxide content depends on storage, light, air exposure and antioxidants; no study here shows that a product's age alone predicts content, so do not claim 'older means worse'.
Source: Karlberg & Lepoittevin 2021, Contact Dermatitis Link to this fact
Skin allergy / Measured in people or real products
After nickel, the standard fragrance allergy test (Fragrance Mix I) was the most common positive in the general population: about 3.5% reacted, across 19 studies and 19,440 people.
Strong evidence: Pooled 1966-2017 (CI 2.1-5.4%); the largest modern study found 1.8-2.6% for the same test.
Skin allergy / Review of other studies
The standard fragrance patch test covers only 8 of the 82 fragrance chemicals known to cause allergy in people, so any single test undercounts fragrance allergy.
Strong evidence: Authors' discussion; oxidized limonene/linalool never tested in a general-population sample.
Skin allergy / Measured in people or real products
Among about 4,100 North American patients referred for allergy patch testing in 2019–2020, 12.8% reacted to the standard fragrance mix and 11.1% to oxidized linalool, both among the ten most common skin allergens and both up from 2017–2018.
Strong evidence: Referral population, not the general public; primary is DeKoven 2023 (NACDG), cited by the review.
Skin allergy / Measured in people or real products
The industry's safety panel cleared linalool, a lavender-scented chemical in most fragranced products, as raising 'no safety concerns' after testing the fresh chemical; independent clinics later found that air-exposed (oxidized) linalool causes allergic reactions in about 7% of patients patch-tested for skin allergy (2,900 patients in 9 centers; 5,773 in Sweden, with rates rising).
Strong evidence: Clinic rates are in dermatitis patients, not the general population; test preparations debated; fresh linalool is genuinely a weak sensitizer.
Source: Bickers 2003 (RIFM Expert Panel), Regulatory Toxicology and Pharmacology Industry-funded Link to this fact
Skin allergy / Measured in people or real products
In a patch test of 3,119 people randomly sampled from the general population of five European countries, 2.6% tested positive to fragrance mix I and 1.9% to fragrance mix II, and 1.9% had a fragrance allergy judged clinically relevant (skin symptoms plus product avoidance).
Strong evidence.
Caveats
The largest, best-controlled general-population fragrance patch-test study to date (n=3,119 of 12,377 interviewed, harmonized protocol across 5 countries); the raw marker-positive rate is higher (4.1%) before the clinical-relevance filter, and about half of marker-positive people do not meet the stricter definition.
Source: Diepgen 2015 (EDEN), British Journal of Dermatology Industry-funded Link to this fact
Skin allergy / Measured in people or real products
In a 5-year Danish patch-test survey of 17,716 eczema patients, 10.1% had fragrance allergy, and cosmetic products caused 42.1% of it; deodorants (25.3% of specific products listed) and scented lotions (24.4%) were the two leading culprits, together accounting for roughly half of all cases.
Strong evidence.
Caveats
Clinic-referred eczema patients (Danish Contact Dermatitis Group members' consecutive patients), not a general-population sample; retrospective database analysis, not a prospective cohort.
Source: Heisterberg 2011, Contact Dermatitis Link to this fact
Skin allergy / Measured in people or real products
Deodorants were significantly more likely to be the reported cause of fragrance allergy in men than women (37.9% vs 21.1% of listed products, odds ratio 2.3), while scented lotion (odds ratio 2.6) and fine fragrances (odds ratio 4.2) were significantly more often the reported cause in women; shampoo showed no sex difference.
Strong evidence: Reflects differential product use by sex, not necessarily differential susceptibility to sensitization.
Source: Heisterberg 2011, Contact Dermatitis Link to this fact
Skin allergy / Measured in people or real products
Standard fragrance allergy screening tests (the fragrance mixes, HICC, Myroxylon pereirae, colophonium) missed about 70% of patients allergic to oxidized limonene or linalool; only 30-33% of hydroperoxide-allergic patients also reacted to any standard baseline marker.
Strong evidence: None.
Source: Deza 2016, Contact Dermatitis Industry-funded Link to this fact
Skin allergy / Measured in people or real products
Women's overall contact allergy rate (35.5%) was roughly double men's (17.1%) in this general-population sample.
Strong evidence: OR 2.6 (95% CI 2.1-3.1) adjusted for age and skin complaints; nickel allergy (much more common in women, 22.2% vs 5.2%) drives most of the gap.
Source: Diepgen 2016 (EDEN), British Journal of Dermatology Industry-funded Link to this fact
Skin allergy / Review of other studies
Fragrance ingredients rank alongside nickel as a leading cause of contact allergy in dermatology-clinic eczema patients, diagnosed in roughly 10-12% of those tested.
Strong evidence: Clinic (patch-tested, already-eczematous) populations, not the general population; figures are for FM I specifically.
Source: Johansen 2003, Am J Clin Dermatol 4(11):789-798, American Journal of Clinical Dermatology Link to this fact
Skin allergy / Measured in people or real products
In a 2018 meta-analysis pooling 28 general-population patch-test studies (20,107 people), the pooled prevalence of contact allergy to Myroxylon pereirae (balsam of Peru) was 1.8% (12 studies, 8,002 people, 95% CI 1.0-2.7%).
Strong evidence.
Caveats
Meta-analysis with substantial heterogeneity across the 12 pooled studies (I2 85.3%); general-population samples, not clinic-referred dermatitis patients, so this is a lower-bound estimate compared to referred populations.
Skin allergy / Government agency
In 2012 the EU's scientific committee on consumer safety (SCCS) classified 82 fragrance ingredients (54 chemicals and 28 natural extracts, including all 26 then labeled) as established causes of contact allergy in people, on clinical evidence alone.
Strong evidence.
Caveats
Counts reported patch-test and case-report evidence from the literature to October 2010; many counts rest on small or single-center series and publication bias; the classification measures allergy evidence, not a safe-use verdict.
Information, not medical advice. See also: myths we won’t tell you.