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.
Fragrance is everywhere / Review of other studies
Lanolin, a sheep-wool grease, is in many moisturizers, lip balms, soaps, shampoos, sunscreens, makeup, diaper-rash, wound, and nipple creams, and some topical steroids; US over-the-counter products may contain it at 12.5% to 50%.
Moderate evidence: Narrative review; the concentration range is a secondary citation (PubChem).
Source: Johnson 2023, Cutis Link to this fact
Fragrance is everywhere / Other evidence
Common household cleaners, wipes, and soaps contain quaternary ammonium disinfectants: product labels listed in one review show benzalkonium chloride at about 0.08-0.145% (wipes, all-purpose cleaner, hand soap) and didecyldimethylammonium chloride at 0.01-0.7% (including a fabric and air refresher).
Moderate evidence: Concentrations are as printed in the review with no source or date; database counts are not comparable between databases and are not market share.
Source: Okeke 2023, Clinical, Cosmetic and Investigational Dermatology Link to this fact
Fragrance is everywhere / Other evidence
For vanilla extract the panel assumed an intake of 300,000 micrograms (300 mg) a person a day, about 86 times its own survey-based figure of 3,500 micrograms, because home cooks add it directly and 2020 US vanilla-bean imports (1,470 metric tons) suggest industry survey volumes under-count it.
Moderate evidence: Method-based worst-case estimate (JECFA single-portion technique), not a measured intake; the 86-fold ratio is evidence base arithmetic on two figures in the paper.
Source: Guengerich 2026 (FEMA Expert Panel), Food and Chemical Toxicology Industry-funded Link to this fact
Fragrance is everywhere / Other evidence
BPA is nearly universal in people tested: the CDC urine data cited (Calafat 2005) found measurable BPA in 95% of US samples (median 1.3 ppb), and parent BPA was found in human fetal serum (median 2.3 ng/mL, range 0.2-9.2; Schonfelder 2002).
Moderate evidence: Cited by the authors from other studies, not measured here; presence is not harm. Urine data are spot samples from 2003-2004 era.
Source: vom Saal & Hughes 2005, Environmental Health Perspectives Link to this fact
Fragrance is everywhere / Self-reported survey
In two US national telephone surveys (2002-03 and 2005-06, about 1,057 adults each), about 3 in 10 adults (30.5% on average) said being next to someone wearing a scented product such as perfume or aftershave is irritating.
Moderate evidence.
Caveats
Self-reported irritation, not a clinical outcome; two surveys by the same authors; the question offered 'irritating or appealing' and only the irritating share is reported. Landline-era phone samples; no response rate given.
Source: Caress & Steinemann 2009, Journal of Environmental Health Link to this fact
Fragrance is everywhere / Other evidence
The 2011 gum table lists flavor chemicals that are also fragrance materials or allergens at the following maximum levels in chewing gum: trans-anethole 6,680 ppm, menthone 6,000 ppm, cinnamon 1,000 ppm, ethyl salicylate 2,000 ppm, pulegone 350 ppm, hydroxycitronellal 5 ppm.
Moderate evidence: Anticipated industry use levels, not measured content of products; the paper gives no labeling, skin or inhalation information.
Source: FEMA GRAS 25 (2011), Food Technology Industry-funded Link to this fact
Fragrance is everywhere / Other evidence
A Harvard study of hair products in eight Boston neighborhoods found communities with lower income and more residents of color had more than twice the risk of finding products with high Skin Deep hazard scores (EWG's summary).
Moderate evidence: Secondhand summary; the primary study had large missing-score rates (see its own page) and scores are label-based, not measured exposure.
Source: EWG 2025 (Friedman & Huang), EWG news (web page) Link to this fact
Fragrance is everywhere / Industry source
A 2018 draft report from the Cosmetic Ingredient Review panel concluded hydrogen peroxide is safe in cosmetics at current use levels, with consumer hair dyes up to 12.4%, hair bleaches up to 12%, dentifrices up to 4.6% and leave-on skin products up to 2.5%.
Moderate evidence.
Caveats
Draft, not the published report; CIR is funded by the cosmetics industry trade group (Personal Care Products Council) and 'safe as used' is the panel's position. Concentrations are maximum reported use levels, not typical ones.
Source: CIR 2018 draft (Do Not Cite or Quote; PCPC-funded CIR) Industry-funded Link to this fact
Fragrance is everywhere / Other evidence
Perchloroethylene in Suffolk County groundwater: not detected in about 92% of supply wells sampled 2009-2013, but found in three times as many wells in 2013 as in 1987 with the average concentration doubled; public-supply maximum 93 ppb against a 5 ppb limit, and orders of magnitude higher at dry-cleaning sites.
Moderate evidence: Local county data summarized second-hand; trend reflects the wells sampled and is not attributed to dry cleaning alone.
Source: SCDHS 2018 Link to this fact
Fragrance is everywhere / Other evidence
Retailers diffuse a scent blend of about 20 listed materials, including limonene, coumarin, lilial and hexyl cinnamal, through a shop from a ceiling-mounted unit running from one hour before opening, with the door open so the scent also drifts outside.
Moderate evidence: Describes one store; ingredient names only, no concentrations; the paper does not say how common this is and makes no allergen or health statement.
Fragrance is everywhere / Measured in people or real products
Because most cosmetics contain fragrance, the dermatologists noted that fragrance allergy is not limited to fragrance products: fragrance ingredients were the commonest allergen group even though perfumes and colognes were only 7% of reactions.
Moderate evidence.
Caveats
Authors' inference from a selected series; 39 reactions were to unspecified fragrance and 77 to individually identified fragrance ingredients, with the 77 not reproducible from the printed table.
Source: Eiermann 1982, Journal of the American Academy of Dermatology Link to this fact
Fragrance is everywhere / Other evidence
Phthalate breakdown products were found in nearly every man tested in this 2000-2001 Massachusetts clinic sample: MEP in 100%, and MBP, MBzP and MMP in over 95%.
Moderate evidence: Clinic men, not a representative sample; consistent with national NHANES data, which showed levels 2-3 times higher.
Fragrance is everywhere / Other evidence
Parabens are the most widely used cosmetic preservatives: methylparaben was in 7,866 of 22,228 US-registered products (2005, voluntary registry) and more than 28% of 1,170 Danish products contained at least one paraben.
Moderate evidence: US registry is voluntary and 2005; counts show use, not exposure or risk.
Fragrance is everywhere / Other evidence
In markets without microbiological rules, 30% to 100% of purchased cosmetics were contaminated and pathogens were found in every study (for example Egypt and Iran), while US tester kits showed 50% bacterial contamination.
Moderate evidence: Mostly older studies from 1989 to 2005, varying methods and thresholds; shared-use testers differ from consumer products.
Fragrance is everywhere / Other evidence
Men who used hand/body lotion in the past day had about 80 to 150% higher urinary methylparaben, propylparaben and butylparaben, and sunscreen users about 66 to 156% higher.
Moderate evidence: Sunscreen use was rare (25 of about 860 samples), so sunscreen estimates are unstable; exposure markers only.
Source: Nassan 2017, Environmental Health Perspectives Link to this fact
Fragrance is everywhere / Other evidence
The links between product use and urinary MEP and parabens were generally stronger when the product was used within 6 hours of the urine sample than within 24 hours, consistent with the short half-lives (6 to 24 hours) of these chemicals.
Moderate evidence: Timing from self-report; no data on time since last void; many 6 h estimates are from small numbers of users.
Source: Nassan 2017, Environmental Health Perspectives Link to this fact
Fragrance is everywhere / Other evidence
Apart from MEP, other urinary phthalate metabolites (including DEHP, DnBP and DiBP metabolites) were not meaningfully predicted by personal care product use in men, which fits DEP being the main phthalate in these products and food being a major source of others.
Moderate evidence: Null results from a single cohort of mostly white, educated men; DEHP food source is the authors' interpretation, not measured.
Source: Nassan 2017, Environmental Health Perspectives Link to this fact
Fragrance is everywhere / Other evidence
Cigarette smoke is one of the most chemically characterized consumer mixtures: the tobacco industry's catalog lists more than 5,200 identified tobacco components and about 4,800 in mainstream smoke, and its authors say many more remain unidentified.
Moderate evidence: Counts are the R.J. Reynolds authors' own catalog and are internally inconsistent (4200, 5200, 5229, 8400).
Source: Rodgman & Perfetti 2013 Industry-funded Link to this fact
Fragrance is everywhere / Other evidence
In 338 Boston men, those who had used cologne in the previous 48 hours had about 2.6 times the urinary level of MEP, the breakdown product of the fragrance solvent diethyl phthalate, as men who had not (aftershave users about 1.7 times), after adjusting for age, BMI, smoking and race.
Moderate evidence: Self-reported product category only, no product DEP content measured; single clinic, mostly White men; same-group follow-up (Nassan 2017) agrees but is not independent.
Source: Duty 2005, Environmental Health Perspectives Link to this fact
Fragrance is everywhere / Other evidence
In 108 women in northern Mexico, the more personal care products a woman had used in the past two days, the higher the level of monoethyl phthalate (the breakdown product of the fragrance carrier diethyl phthalate) in her urine.
Moderate evidence.
Caveats
Small single sample, one spot urine, 48-hour recall, no brand or fragrance-content data; supports the product-count pattern seen in US cohorts but cannot separate DEP from other co-used chemicals.
Source: Romero-Franco 2011, Environment International Link to this fact
Fragrance is everywhere / Measured in people or real products
In urine from 95 Swedish mothers and 97 of their children (6-11 years), phthalate metabolites were found in essentially every sample, and methylparaben in all mothers and 86% of children.
Moderate evidence.
Caveats
One small Swedish cross-sectional sample (22% response rate); a single first-morning urine each; Sweden's paraben and BPA levels were among the lowest in the European harmonized program.
Source: Larsson 2014, Environment International Link to this fact
Fragrance is everywhere / Review of other studies
Fragrance chemicals used in sunscreens, soaps and perfumes (amyl, hexyl and benzyl salicylate and three trade-name aroma chemicals) were measured in wastewater and coastal seawater at an Antarctic research station at low ng/L levels, about 2 to 54 ng/L.
Moderate evidence: One source study and one station; no toxicity, removal or risk data for these compounds; no synthetic musks measured.
Source: Balakrishna 2023 (review), Environmental Research Link to this fact
Fragrance is everywhere / Review of other studies
UV filters were the highest personal-care signal in Antarctic station wastewater: BP-1 up to about 14,000 ng/L, 4-MBC up to about 11,700 ng/L, which the authors put at 5 to 340 times the wastewater range reported elsewhere.
Moderate evidence: Few stations and sampling events; comparison ranges come from two secondary reviews; authors' explanation (heavy sunscreen use, poor removal) is not tested.
Source: Balakrishna 2023 (review), Environmental Research Link to this fact
Fragrance is everywhere / Other evidence
Benzophenone-3 (10.8 ng/mL), methylparaben (10.4 ng/mL) and MEP (8.9 ng/mL) had the highest median levels in infant urine at 3 to 6 months; most phthalate metabolites fell between 2008 and 2019, but a metabolite of the replacement plasticizer DEHTP (MECPTP) rose significantly (Kendall tau 0.81).
Moderate evidence: One cohort in Northern California; trend tests across year of birth, not repeated measures; DEHTP is a plastics plasticizer, not a fragrance ingredient.
Source: Sotelo-Orozco 2024, International Journal of Hygiene and Environmental Health Link to this fact
Fragrance is everywhere / Other evidence
Fragrance exposure was nearly universal among surveyed UAE women: 91.5% used fragrance daily, 90.0% used fragranced personal care products weekly, and only 0.7% reported no personal weekly exposure.
Moderate evidence: Self-report from a convenience sample of women; consistent with population surveys in Sweden (98.5% weekly own use) and Germany (96.9%).
Source: AlBlooshi 2026, Frontiers in Toxicology Link to this fact
Green, natural and unscented / Review of other studies
Products labeled unscented can still contain a fragrance added to mask other smells.
Moderate evidence.
Caveats
Masking itself is still unmeasured (a review statement). For a primary count cite Xu 2017: 18/40 (45%) best-selling US 'fragrance-free' moisturizers listed a fragrance cross-reactor or botanical, which is identity-based, not masking.
Source: Kumar 2020, Journal of Cosmetic Dermatology Link to this fact
Green, natural and unscented / Measured in people or real products
In 14 of 17 'alternative' (mostly green-marketed) products that contained fragrance chemicals, the label never mentioned fragrance.
Moderate evidence: Mostly natural terpenes at low levels; single products from one retailer.
Source: Dodson 2012, Environmental Health Perspectives Link to this fact
Green, natural and unscented / Measured in people or real products
Products labeled 'green,' 'natural' or 'non-toxic' released hazardous chemicals too, with no significant difference from ordinary ones in a test of 25 best-sellers.
Moderate evidence: 11 vs 14 products; underpowered; list-membership counts; small sample (11 vs 14 products, p>0.30): a null result, not proof of equality.
Source: Steinemann 2011, Environmental Impact Assessment Review Link to this fact
Green, natural and unscented / Measured in people or real products
Baby products labeled green, organic, or natural released no fewer chemicals than regular ones (398 vs 286 detections), and 20 of 21 in each group released at least one chemical classed as potentially hazardous.
Moderate evidence: Headspace counts, not concentrations; agrees with Lin 2020 and Steinemann 2011.
Source: Nematollahi 2018, Air Quality, Atmosphere & Health Link to this fact
Green, natural and unscented / Measured in people or real products
Natural baby products swapped one scent chemistry for another: linalool, a common fragrance allergen once it oxidizes, turned up in 16 of 21 green products vs 4 of 21 regular ones.
Moderate evidence.
Caveats
Presence only; oxidized (allergenic) forms not measured in these products. Hagvall 2008 shows linalool in an essential oil (lavender) oxidizes in air at the same rate as synthetic linalool, so the natural source gives no protection; real-product hydroperoxide levels unknown.
Source: Nematollahi 2018, Air Quality, Atmosphere & Health Link to this fact
Green, natural and unscented / Animal study
'Natural' doesn't mean allergy-free: essential oils oxidize too; in mouse tests lavender oil exposed to air for 45 weeks became about 8 times more potent as a sensitizer (EC3 36% fresh, 4.4% aged), and oxidized alpha-terpinene from tea tree oil was the most potent of all.
Moderate evidence.
Caveats
Mouse LLNA (groups of 3-4 mice; one 45-week sample) from the primary Hagvall 2008, lab open-flask exposure; aged oil reached 'moderate', not 'strong', potency. Human support: 4/4 oxidized-linalool-allergic patients reacted to oxidized lavender oil (elicitation, n=4). Earlier evidence base figure '11% -> 4.4%' was a misreading.
Green, natural and unscented / Self-reported survey
Three in four UK adults (75.0%) did not know that products called natural, green, or organic can also release hazardous air pollutants, and more than half (53.5%) said they would stop using a product if they knew.
Moderate evidence: Awareness item; the emission claim itself rests on other studies.
Source: Steinemann 2018, Air Quality, Atmosphere & Health Link to this fact
Green, natural and unscented / Measured in people or real products
Plant-derived scent chemicals drove the particle bursts, so a 'natural' citrus or pine scent is not automatically cleaner to breathe.
Moderate evidence: No unscented comparison in this study.
Green, natural and unscented / Measured in people or real products
Labels are not a guarantee: ethyl and butyl paraben actually rose after girls switched to paraben-free-labeled products, probably because some replacements contained them.
Moderate evidence: Replacement products were not tested; explanation inferred.
Source: Harley 2016, Environmental Health Perspectives Link to this fact
Green, natural and unscented / Journalism
A 2009 audit of 2,219 North American products making green claims found that more than 98% made at least one misleading or unsupported environmental claim.
Moderate evidence: TerraChoice gray literature; auditor sells eco-certification; audits substantiation not chemistry.
Source: Dahl 2010, Environmental Health Perspectives Link to this fact
Green, natural and unscented / Journalism
Green-claim auditors named 'all-natural' as a textbook example of a vague green claim and called household cleaning-product makers 'among the worst greenwash offenders'.
Moderate evidence: Opinion of TerraChoice staff; '3 of 397 cleaners' figure not text-verified.
Source: Dahl 2010, Environmental Health Perspectives Link to this fact
Green, natural and unscented / Measured in people or real products
In a test of 79 US feminine hygiene products, labels such as organic, natural, or for sensitive skin did not predict lower levels of volatile chemicals.
Moderate evidence: Product content, not exposure; small subgroups.
Source: Lin 2020, Environment International Link to this fact
Green, natural and unscented / Journalism
Perfumers themselves say every perfume blends synthetic and natural materials: "As all perfumers, I create perfume formulas with both synthetic and natural perfumery raw materials" (a Charabot perfumer). A "natural" scent is rarely all natural.
Moderate evidence: One perfumer's essay; no proportions given; industry author.
Source: Ellena 2008, Chemistry & Biodiversity Industry-funded Link to this fact
Green, natural and unscented / Self-reported survey
'Natural' essential oils have not been shown safe for hormone-sensitive children: the only survey claiming so had about a 5% chance of detecting even a tripling of risk.
Moderate evidence: Power figure is evidence base arithmetic from the paper's group sizes and observed baseline; absence of proof of safety is not proof of harm.
Source: Hawkins 2021, International Journal of Pediatrics and Adolescent Medicine Industry-funded Link to this fact
Green, natural and unscented / Review of other studies
Many products sold as "fragrance free" in natural-product stores are free only of synthetic fragrance and still contain essential oils, which make up 80 of the 162 fragrance ingredients documented to cause skin allergy.
Moderate evidence.
Caveats
80/162 is a count of documented allergens (de Groot), not a risk comparison; the store claim is unquantified. Klaschka 2016 supports the mechanism (supplier SDSs: orange oil ~95% limonene, lavender ~40% linalool) but counted no 'fragrance-free'-labeled products, and in its EU sample 58/100 products did declare natural-origin allergens, so on EU labels the ingredient list often shows them.
Green, natural and unscented / Lab study (cells or chemistry)
'Natural' scent is not biologically inert: essential oils such as lavender and tea tree are chemical mixtures, and in lab tests they acted on estrogen and androgen receptors.
Moderate evidence: Shows biological activity, not harm at use levels.
Source: Henley 2007, New England Journal of Medicine Link to this fact
Green, natural and unscented / Review of other studies
Even a review that found little evidence of harm from lavender and tea tree oil in children warned that 'a lack of evidence of harm cannot be interpreted as evidence of safety' for essential oils in children's products.
Moderate evidence: Direct quote; the review's own conclusion is that the link is unsupported.
Source: Hawkins 2020, Complementary Therapies in Medicine Industry-funded Link to this fact
Green, natural and unscented / Lab study (cells or chemistry)
Natural does not mean hormonally inert: the hormone-active chemicals tested here are ordinary constituents of lavender and tea tree essential oils, not synthetic fragrance chemicals.
Moderate evidence: Cell assays only, at concentrations far above plausible human exposure; whole-animal guideline tests of the two main lavender constituents were negative.
Source: Ramsey 2019, The Journal of Clinical Endocrinology & Metabolism Link to this fact
Green, natural and unscented / Lab study (cells or chemistry)
Terpenes do not behave alike: alpha-terpineol and terpinen-4-ol were the most hormone-active of the eight essential-oil chemicals tested, while eucalyptol was inactive on both the estrogen and the androgen receptor.
Moderate evidence: Receptor assays only; eucalyptol is separately reported to inhibit the steroid-synthesis enzyme CYP17A1, which this study did not test.
Source: Ramsey 2019, The Journal of Clinical Endocrinology & Metabolism Link to this fact
Green, natural and unscented / Measured in people or real products
A fragrance-free body lotion still gave the highest VOC emissions measured in the study, mostly from a preservative (likely phenoxyethanol), and its plant oils reacted with ozone to form aldehydes such as nonanal and nonenal.
Moderate evidence: One lotion, 4 men, chamber; phenoxyethanol identity inferred from PTR-MS ions and the ingredient list.
Source: Wang 2024, Environmental Science & Technology Link to this fact
Green, natural and unscented / Lab study (cells or chemistry)
There is no known 'safe terpene'. Eucalyptol is inactive on hormone receptors but acts weakly on a hormone-making enzyme; alpha-terpineol is the most active on receptors and only middling on the enzyme; limonene is near-inert on receptors and a comparable enzyme inhibitor. The two rankings disagree in both directions.
Moderate evidence: Both rankings are in vitro and at concentrations far above real exposure. The point is that one screen does not clear a molecule, not that any of these molecules is proven harmful.
Source: Sharma 2024 + Ramsey 2019, Biomolecules Link to this fact
Green, natural and unscented / Review of other studies
Natural exfoliants such as salt, sugar, ground seeds or shells, and cellulose beads can replace plastic microbeads.
Moderate evidence: Environmental substitution only; skin safety of the alternatives was not assessed.
Green, natural and unscented / Review of other studies
Essential oils oxidize in air into the same allergens as synthetic fragrance, and 'natural' cosmetics routinely carry fragrance allergens: 91% of 'natural' perfumes on the Danish market contained at least one target fragrance sensitizer (chemical analysis, 1996) — but only 35% of the other 'natural' cosmetics (shampoos, creams, lotions) in the same study did, at far lower concentrations — and 82 of 100 European natural-brand products checked in 2013-15 listed at least one allergen on the label.
Moderate evidence.
Caveats
91%/35% is Rastogi 1996's own perfume-vs-other-cosmetics split, now confirmed by reading the primary directly (a review had dropped the perfume qualifier and stated 91% as if it covered 'natural cosmetics' generally; corrected 2026-09-27). Rastogi 1996 also found synthetic-only fragrance chemicals (hydroxycitronellal, alpha-hexylcinnamic aldehyde) in some 'natural' products, at concentrations that in one case (6.04% hydroxycitronellal) exceeded the level shown to sensitize 36% of volunteers in patch testing. 82/100 is Klaschka 2016's independent label analysis (10 products from each of 10 brands, not representative, declarations only). The oxidation half rests on Karlberg/Bråred Christensson primaries, not on these surveys. Single small studies (n=42 and n=100), not replicated across labs, so strength stays moderate per the grading rule.
Source: Pastor-Nieto 2021 (Rastogi 1996; Klaschka 2016), Current Treatment Options in Allergy Link to this fact
Green, natural and unscented / Measured in people or real products
In a check of 100 'natural' personal-care products from 10 European natural-cosmetics brands, 82 listed at least one of the EU's labeled fragrance allergens (up to 11 in one product); limonene was on 71 labels and linalool on 70.
Moderate evidence: Label declarations only, no chemical analysis; 10 products from each of 10 anonymous brands, not representative; 2013-15, when the EU list had 26 allergens.
Source: Klaschka 2016, Environmental Sciences Europe Link to this fact
Green, natural and unscented / Measured in people or real products
'For sensitive skin' is not a promise of no fragrance allergens: all 15 natural products named 'sensitive', 'soothing' or 'soft' listed between 1 and 7 of the EU's labeled fragrance allergens.
Moderate evidence: N=15 products, labels only, European natural brands.
Source: Klaschka 2016, Environmental Sciences Europe Link to this fact
Green, natural and unscented / Measured in people or real products
In a 2016 survey of 174 best-selling US body moisturizers, 18 of the 40 labeled "fragrance free" (45%) still listed at least one ingredient that dermatologists' allergen database (ACDS CAMP) flags for fragrance-allergic patients.
Moderate evidence.
Caveats
Ingredient-label survey, one product class and snapshot; CAMP list is broad (plant extracts, multifunctional chemicals such as benzyl alcohol); flagged ingredients not itemized; not added scent or masking.
Green, natural and unscented / Measured in people or real products
Among the same best-selling US moisturizers, 15 of 18 labeled "hypoallergenic" (83%) and 95% of those labeled "dermatologist recommended" contained at least one common skin allergen; only 12% of all products contained none.
Moderate evidence: Listed allergens (NACDG series), not reactions; label data, 2016 formulations.
Green, natural and unscented / Measured in people or real products
All 16 UK day face moisturizers tested gave off fragrance terpenes (measured together as 'limonene'), and the 8 'green'-marketed ones released the same amount as the 8 regular ones (0.25 vs 0.26 µg per second per gram), as well as the same ethanol.
Moderate evidence.
Caveats
Lab headspace (SIFT-MS) of 16 products from UK shops; 'limonene' is the sum of all monoterpenes; no statistical test; green products did have less synthetic-only solvents (t-butyl alcohol, 1,3-butanediol).
Green, natural and unscented / Review of other studies
In 1999 a dermatologist listed ten US "fragrance-free" skin products that named fragrance chemicals in their own ingredient lists, including a "fragrance-free" cleansing bar containing rose, rosewood, cedarwood, and almond oils; one manufacturer admitted the benzaldehyde in its fragrance-free product was there to mask an unpleasant smell.
Moderate evidence.
Caveats
Ten named examples from 1999 labels, not a systematic survey; formulations have likely changed since. Masking is one manufacturer's admission and remains unquantified across products.
Source: Scheinman 1999, Journal of the American Academy of Dermatology Link to this fact
Green, natural and unscented / Measured in people or real products
In a 1996 Danish GC-MS analysis, 91% (20 of 22) of perfumes marketed as based on natural ingredients contained 1 to 7 of 11 well-known fragrance sensitizers at concentrations up to 7.7%, including one perfume with 6.04% hydroxycitronellal — above the concentration shown to sensitize 36% of healthy volunteers in patch testing.
Moderate evidence.
Caveats
42 products, one country (Denmark), 1996 market snapshot; parent-compound GC-MS only, not label declarations; the 91% figure applies to the natural-ingredient PERFUME subset only, not natural cosmetics generally.
Green, natural and unscented / Measured in people or real products
The same 1996 study found far fewer fragrance sensitizers in other 'natural' cosmetics (shampoos, creams, lotions, deodorants, massage oils): only 35% (7 of 20) contained 1-3 of the target substances, at concentrations (0.0003-0.082%) roughly 100x lower than the natural perfumes.
Moderate evidence: Small non-representative sample (n=20); shows fragrance-sensitizer exposure from 'natural' products is concentrated in perfume-type products, not uniform across the category.
Green, natural and unscented / Review of other studies
Essential oils are complex mixtures of hundreds of components whose composition varies by producer, plant origin, climate, and extraction method, so a product labeled as containing 'lavender oil' or 'tea tree oil' does not guarantee a fixed chemical composition or a fixed level of biological activity.
Moderate evidence: Stated as general essential-oil chemistry background in this review; not measured for any specific consumer product here.
Source: Ramsey 2020 (YJBM), Yale Journal of Biology and Medicine Link to this fact
Green, natural and unscented / Self-reported survey
More than half of Germans (55.6%) wrongly believe products with natural fragrance ingredients are healthier than products with synthetic ones, and the false belief is highest (67-69%) among people who read ingredient labels or prefer fragrance-free products.
Moderate evidence: Belief survey, not a chemical measurement; the 'wrong' judgment is the author's own hazard-classification argument, not tested in this paper.
Source: Klaschka 2020a, Environmental Sciences Europe Link to this fact
Green, natural and unscented / Measured in people or real products
A headspace test of the fragranced and fragrance-free versions of the same detergent brand found 34 volatile chemicals in the fragranced version but only 7 in the fragrance-free version, and no fragrance terpenes (such as D-limonene or linalool) at all in the fragrance-free version.
Moderate evidence: Single product pair tested (one Australian brand); presence/absence only, not a full quantitative comparison.
Source: Goodman 2019, Air Quality, Atmosphere & Health Link to this fact
Green, natural and unscented / Government agency
Allergy UK's Product Endorsement Scheme is run by Allergy Research Ltd, a private limited company that is a legally separate trading subsidiary of the charity British Allergy Foundation (which operates publicly as Allergy UK); ARL donates its profits to the charity.
Moderate evidence: Organizational/legal fact stated plainly on the page itself; not independently verified against Companies House or charity filings.
Source: Allergy UK 2026 Link to this fact
Information, not medical advice. See also: myths we won’t tell you.