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
Musk scent chemicals used in perfumes, deodorants, and detergents turn up in mothers' milk fat; a breastfed baby takes in an estimated ~2 micrograms of galaxolide a day.
Strong evidence.
Caveats
Intake is a model estimate from means of detects; the printed intakes do not reproduce from the stated method. Independently corroborated: Lignell 2008 (N=101, Swedish mothers, single lab) estimated median infant HHCB intake at 1650 ng/day (range 50-6950) using the same milk-intake modeling approach, closely matching this paper's printed 1830 ng/day figure in an unrelated country and cohort.
Source: Reiner 2007, Environmental Science & Technology Link to this fact
The industry / Review of other studies
In 1997 the cosmetics industry's own safety panel (CIR) formally asked for inhalation-toxicity data on benzyl alcohol, benzoic acid, and sodium benzoate, received none, and still published a 2001 "safe" conclusion that explicitly excludes any product where inhalation is a route of exposure (colognes, hairsprays, deodorants, aerosol fixatives).
Strong evidence: The gap is documented in the report's own text (Notice of Insufficient Data, no comments received); no later CIR report closing it was found in this pass.
Source: Nair 2001 (CIR Expert Panel), International Journal of Toxicology Industry-funded Link to this fact
The industry / Measured in people or real products
The independent dermatologists who ran this study concluded deodorants should contain cinnamal at less than 0.01% to protect people already sensitized, a stricter limit than the fragrance industry's later 0.05% IFRA standard.
Strong evidence: Authors' own recommendation, not a regulatory outcome; the 0.05% IFRA cap was set afterward using the induction NOEL, not this elicitation data.
Source: Bruze et al. 2003, Journal of the American Academy of Dermatology Link to this fact
The industry / Measured in people or real products
The independent dermatologists who ran this study concluded deodorants should contain isoeugenol at less than 0.0063% to protect people already sensitized, about 3x stricter than IFRA's current 0.02% ceiling for consumer products.
Strong evidence.
Caveats
Authors' own recommendation, not a regulatory outcome; the 0.02% IFRA cap (set in 1998) was not derived from this elicitation data, and 0.0063% is only the lowest concentration this study tested, not a demonstrated safe floor.
Source: Bruze et al. 2005, Contact Dermatitis Link to this fact
Skin allergy / Measured in people or real products
In people already allergic to cinnamal, a deodorant with just 0.01% of it triggered a reaction in 1 of 9, and 0.032% in 4 of 9; the industry limit is 0.05%, and it was set to prevent new allergies, not reactions in people already sensitized.
Strong evidence.
Caveats
Confirmed by the primary source, now in our evidence base: Bruze et al. 2003, an independent (non-industry, no COI declared), double-blind provocation study with a 20-patient non-allergic control group (0/20 reacted) and P<.001 statistics tying the reaction to both cinnamal-sensitization and the actively-scented axilla. The study's own authors recommend a deodorant limit below 0.01%, stricter than the 0.05% IFRA cap.
Source: RIFM Expert Panel 2005 (citing Bruze 2003), Food and Chemical Toxicology Industry-funded Link to this fact
Skin allergy / Measured in people or real products
In a double-blind study, deodorants containing as little as 0.01% cinnamic aldehyde (cinnamal) triggered allergic skin reactions within weeks in people already allergic to it, while an identical unscented deodorant on the other armpit, and the same deodorants used by 20 non-allergic people, caused no reactions.
Strong evidence.
Caveats
Small cohort (9 sensitized test patients, double-blind arm), but statistically significant (P<.001) with a concurrent negative control group and randomized/blinded contralateral-axilla design.
Source: Bruze et al. 2003, Journal of the American Academy of Dermatology 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
In a double-blind study, deodorants containing isoeugenol at concentrations from 0.0063% to 0.2% triggered allergic skin reactions within weeks in 10 of 13 people already allergic to it, while identical unscented deodorants and the same products used by 10 non-allergic people caused no reactions.
Strong evidence.
Caveats
Small cohort (13 sensitized test patients fully tabulated of 35 enrolled), but statistically significant (P<.001) with a concurrent negative control group (0/10) and a randomized, double-blind, contralateral-axilla design; unblinding confirmed the active deodorant caused every reaction.
Source: Bruze et al. 2005, Contact Dermatitis Link to this fact
Children and pregnancy / Measured in people or real products
In a Danish/multi-country test of 25 children's cosmetics and cosmetic-toys, all 7 hydro-alcoholic children's fragrance products (5 eau de toilette, 1 eau de parfum, 1 roll-on deodorant) contained the allergen geraniol, and a separate children's toy-cosmetic perfume-blending kit's lilac essence contained cinnamic alcohol at 3.7%, about 4.6x (the paper's abstract states "by a factor of 5") the IFRA guideline of 0.8%.
Moderate evidence.
Caveats
Now sourced from the primary study (Rastogi 1999; n=25 products, Denmark/Germany/England/Sweden, 1997-98) rather than relayed via a review. Single small study, not replicated: strength stays moderate per the grading rule. The geraniol finding is for regular children's fragrance products, not toys; only 1 of 3 toy-perfume-set essences had the elevated cinnamic alcohol — the review's original "cosmetic-toy perfumes" framing conflated the two.
Source: Pastor-Nieto 2021 (Rastogi 1999), Current Treatment Options in Allergy Link to this fact
Children and pregnancy / Measured in people or real products
In a 1999 European lab test, all 7 children's hydro-alcoholic fragrance products (eau de toilette, eau de parfum, roll-on deodorant) contained the allergen geraniol, and a children's toy perfume-blending kit's "lilac essence" contained cinnamic alcohol at 3.7%, about 4.6x (the paper's abstract states "by a factor of 5") the industry's own 0.8% safety guideline.
Moderate evidence.
Caveats
N=25 products total, n=7 hydro-alcoholic products and n=1 toy item for the two headline numbers; single small study (Denmark/Germany/England/Sweden, 1997-98), not replicated for this specific product set.
Children and pregnancy / Measured in people or real products
A deodorant vapo-spray labeled "perfume for babies" in this 1996 European sample contained 14 of 21 tested fragrance substances at levels up to 1640 ppm, including 4 fragrance-mix allergens.
Moderate evidence.
Caveats
A single product, not a category survey; the authors note sensitization is rarely seen in the first 2-3 years of life, but warn that early exposure to established contact allergens may cause problems later.
Source: Rastogi et al. 1998, Contact Dermatitis Link to this fact
Children and pregnancy / Measured in people or real products
In a study of 36 Swedish breastfeeding mothers, those who used a triclosan-containing toothpaste, soap, or deodorant had significantly higher triclosan in both their blood plasma and their breast milk than mothers who did not (median plasma 16 vs. 0.072 ng/g, milk 0.54 vs. 0.019 ng/g; p<0.001).
Moderate evidence.
Caveats
Single-cohort, single-country study; one small earlier study from an overlapping author group (Sandborgh-Englund 2006, n=10) found no group difference, likely due to being underpowered.
Source: Allmyr 2006, Science of The Total Environment Link to this fact
Fragrance is everywhere / Measured in people or real products
The more personal care products a woman used, the higher her levels of diethyl phthalate; deodorant (2.8x), nail polish (2.1x) and hair spray (1.8x) users also had higher levels.
Moderate evidence: Cross-sectional association; one urine sample.
Source: Parlett 2013, Journal of Exposure Science and Environmental Epidemiology Link to this fact
Fragrance is everywhere / Measured in people or real products
All five scented personal-care products tested, including a hand lotion and a roll-on deodorant, released terpene fragrance chemicals that react with ozone in indoor air.
Moderate evidence: 5 Swiss grocery-store products; lotion/roll-on emissions measured from a Kimwipe for 1 h, not skin.
Source: Wu 2024, Environmental Science & Technology Letters Link to this fact
Fragrance is everywhere / Measured in people or real products
A silicone ingredient in everyday deodorants, lotions, and hair products (D5) shows up in city outdoor air on a clear morning schedule tracking when people shower and dress for work, peaking around 6-7 AM and fading over the following several hours.
Moderate evidence.
Caveats
Diurnal emission-rate profile derived for one mid-sized US city (Boulder, CO, winter 2015/2017); Toronto ground-site data (2009) show a similarly timed morning peak in concentration, but the emission-rate model itself is Boulder-only.
Source: Coggon 2018, Environmental Science & Technology Link to this fact
Fragrance is everywhere / Measured in people or real products
In a study of 427 people smelling real body-odor samples, individuals stayed identifiable by smell even when wearing a deodorant, but any deodorant (their own or an assigned one) still made people significantly worse at telling them apart than smelling the unfragranced body odor alone.
Moderate evidence.
Caveats
Direct triangle-test design (large for this subfield); deodorants only, not perfumes; public-venue convenience sample; one subgroup (female donor odor + assigned fragrance) fell to chance level.
Source: Allen, Havlíček & Roberts 2015, Frontiers in Psychology Link to this fact
Fragrance is everywhere / Measured in people or real products
People were better at telling two individuals apart when both wore their own habitual deodorant than when both wore an unfamiliar, experimenter-assigned deodorant, consistent with fragrance choice tracking a person's own body chemistry rather than being arbitrary.
Moderate evidence.
Caveats
Own-vs-assigned effect is modest (Exp(B)=1.375, p=0.03); cannot rule out a familiarity confound (using a product you already use vs. a brand-new one) separate from body-chemistry matching.
Source: Allen, Havlíček & Roberts 2015, Frontiers in Psychology Link to this fact
Fragrance is everywhere / Other evidence
In a 36,446-subject Europe/US market-research survey, most people use only one type of deodorant rather than combining formats: 59% used roll-on only, 17% used spray only, and only about 2% used both spray and roll-on deodorant.
Moderate evidence.
Caveats
Deodorant format preference was strongly regional: in the EU roughly 30% used roll-on only and 30% used spray only, while in the US about 80% used a non-spray format and only about 5% used spray only.
Source: Comiskey 2015 (RIFM/Creme Global), Regulatory Toxicology and Pharmacology Industry-funded Link to this fact
Fragrance is everywhere / Other evidence
The most common personal-care product combination in this 36,446-subject Europe/US survey was deodorant, oral care, and a shower product used together, at about 19% of subjects; only about 1% of men but about 7% of women used products from all 7 surveyed product categories.
Moderate evidence: This reflects only the 19 personal-care/cosmetic product types surveyed; it says nothing about household, laundry, or air-care product use, which were not included in this study.
Source: Comiskey 2015 (RIFM/Creme Global), Regulatory Toxicology and Pharmacology Industry-funded Link to this fact
Fragrance is everywhere / Measured in people or real products
Besides the fragrance-mix allergens, 14 other common fragrance materials (including linalool, benzyl acetate, and hedione) were found in 40-97% of the 73 deodorants tested in 1996; for three of them (hedione, linalyl acetate, alpha-isomethyl ionone) no published human contact-allergy case existed at the time despite their widespread use.
Moderate evidence.
Caveats
Absence of published reports as of 1998 is not the same as absence of risk; testing history for these three substances specifically was not systematically reviewed by the authors beyond a literature check.
Source: Rastogi et al. 1998, Contact Dermatitis Link to this fact
Fragrance is everywhere / Measured in people or real products
Unlike the classic hazardous air pollutants in the same 482-deodorant survey (formaldehyde, benzene, methanol, ethylene oxide, trichloroethylene, chloroform), which each appeared in only 0.4-13.7% of products at generally low concentration, the two intentionally added biocides (chlorine dioxide, silver) and the fragrance allergens were found far more consistently and at far higher concentrations, consistent with the classic HAPs being incidental base-formulation contaminants rather than deliberately added ingredients.
Moderate evidence.
Caveats
An inference from detection-frequency and concentration patterns in one survey, not a source-attribution (chemical-origin) study; the paper itself does not test whether these HAPs come from the base formulation.
Source: Kim et al. 2018, Science of The Total Environment Link to this fact
Fragrance is everywhere / Measured in people or real products
Matching the ingredient lists of 7,603 Korean household chemical products (cleaners, synthetic detergents, fabric softeners, air fresheners, deodorants) against a hormone-disruption hazard database found 293 flagged chemicals, and nearly two out of three products listed more than one of them.
Moderate evidence: Counts ingredient-list matches against EPA ToxCast/CompTox hazard flags, not measured concentrations or confirmed toxicity in the products as sold; single-country database study.
Source: Lee & Ji 2022, Ecotoxicology and Environmental Safety Link to this fact
Fragrance is everywhere / Measured in people or real products
The study's own protocol lists 'hygiene products (e.g., deodorant sprays; aftershave lotions; perfumes; cosmetics; insect repellents)' alongside mouthwash as products that contain or may contain ethanol, and had participants avoid them as confounding ethanol sources.
Moderate evidence: This is a study-design control list, not a measurement of ethanol content or absorption from perfumes/cosmetics specifically.
Source: Reisfield 2021, Journal of Analytical Toxicology Link to this fact
Fragrance is everywhere / Measured in people or real products
Aerosol deodorant/antiperspirant delivers more product per application (mean about 1.27-1.39 g) than roll-on (about 1.0 g) or stick (about 0.6 g) formats, for both French men and women — a pattern relevant to comparing inhalation-relevant exposure across deodorant formats.
Moderate evidence.
Source: Ficheux et al. 2016 (amount data), Food and Chemical Toxicology Link to this fact
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
Most 'natural' or herbal deodorant actives described in this review (essential oils such as tea tree and usnic acid, and dozens of herbal extracts) work the same way synthetic actives like triclosan and silver compounds do: by killing or inhibiting the skin bacteria that cause odor, not through some fundamentally gentler mechanism.
Moderate evidence: Based on antibacterial potency (MIC/MBC) data only; this review does not compare skin-irritation or sensitization rates between herbal actives and synthetic anti-microbial salts.
Source: Kanlayavattanakul & Lourith 2011, International Journal of Cosmetic Science Link to this fact
Green, natural and unscented / Measured in people or real products
Not all fragrance-based odor masking works the same way: Iso E Super, a common deodorant fragrance chemical that barely blocked the malodor receptors in lab dish tests, still reduced how strong a sulfur malodor smelled to people, but less effectively than beta-ionone, suggesting this masking runs mainly through the brain blending two smells together rather than through receptor blocking.
Moderate evidence: Central-processing explanation is the authors' inference from a negative in-vitro result, not directly tested; based on the same single-lab, single-cohort study.
Source: Fukutani et al. 2023, Current Biology Industry-funded Link to this fact
Health effects / Animal study
AETT, a musk fragrance once common in soaps, deodorants, and cosmetics, caused progressive nerve-fiber demyelination and irreversible pigment build-up in the brains of rats fed the compound for 12-20 weeks; the fragrance industry voluntarily withdrew it from cosmetics in 1978 on the strength of its own data.
Moderate evidence: Small single-lab study (N=16 rats), one route (dietary), one dose pair (25/50 mg/kg/day); AETT was withdrawn from cosmetics in 1978 and is not a current fragrance ingredient.
Source: Spencer 1979, Science Industry-funded Link to this fact
Health effects / Lab study (cells or chemistry)
Among 62 individually tested deodorant ingredients, only fragrance chemicals (plus one insect-pheromone contaminant) showed any estrogenic activity; no carrier, stabilizer, emulsifier, plasticizer, antitranspirant, or other active ingredient (including triclosan and BHT) tested positive.
Moderate evidence: Only 62 of the many ingredients found by GC/MS screening were selected for individual testing, based on frequency of occurrence; ingredients not selected were not tested at all.
Health effects / Measured in people or real products
A Korean government risk assessment of 64 deodorants found that at the maximum concentrations actually measured in the products, none of the 19 fragrance and biocidal ingredients carried to detailed (tiered 2) assessment exceeded its margin-of-exposure or hazard-quotient safety threshold, for either the inhalation or the dermal route.
Moderate evidence.
Caveats
A single government risk-assessment study using modeled Korean population-average exposure factors and mostly-animal (rat) toxicity reference values with standard uncertainty factors, not a clinical safety trial; not independently replicated.
Source: Kim et al. 2018, Science of The Total Environment Link to this fact
Health effects / Other evidence
The more types of personal care product a man had used in the previous 48 hours (cologne, aftershave, hair gel, deodorant, lotion), the higher his urinary MEP, rising about 33% per additional product type.
Moderate evidence: Five product categories; very few men used four or five types (22 and 3); no health outcome measured.
Source: Duty 2005, Environmental Health Perspectives Link to this fact
Hidden ingredients / Review of other studies
FDA's voluntary cosmetic registration listed benzyl alcohol in only 322 formulations in 1998 and 511 products in 2006, far below its real use in moisturizers, shampoos, baby products, and deodorants, which the authors call widespread under-reporting.
Moderate evidence.
Caveats
The 322-formulation count for 1998 is now confirmed against the primary source: CIR's 2001 safety report (Nair 2001) reproduces the identical FDA VCRP count in its own Table 2, with a full 34-category breakdown (largest use hair dyes and colors, 130/1572; baby shampoos 5/21; baby lotions 3/53) — this is the exact table Jacob & Barron's figure traces to, not a secondhand citation. The 511-products-in-2006 figure remains an unpublished personal communication (postdates the 2001 report, so it cannot confirm or deny it). The same 2001 report separately documents that FDA stopped collecting concentration-of-use data after 1992, an independently CIR-acknowledged registry gap that supports treating VCRP counts as an incomplete picture of real use, distinct from the authors' own 'under-reporting' framing.
Hidden ingredients / Advocacy group testing
In a 2010 survey of more than 11,000 Canadian personal-care products with ingredient lists, 'fragrance' or 'parfum' was the most common ingredient of concern, listed on 56% of products and on about two-thirds of shampoos, soaps, and deodorants.
Moderate evidence: Self-selected volunteers read labels; product entries not de-duplicated; 2010 formulations; label count is a lower bound on fragranced products.
Hidden ingredients / Other evidence
A 1998 GC-MS survey of 71 European deodorants found 226 distinct fragrance and non-fragrance chemicals – more than ten times the 21 fragrance substances the same research group had specifically targeted and quantified in a companion study on the same products.
Moderate evidence: Single study, 1998 market snapshot (product formulations have changed since); no concentrations measured for any of the 226 chemicals, only presence/absence by product type.
Source: Rastogi et al. 1998 (Contact Dermatitis 39:293), Contact Dermatitis Link to this fact
Hidden ingredients / Other evidence
Methyl anthranilate, flagged by this 1998 GC-MS/SAR survey as an aromatic amine (a structural class linked to sensitization and other toxicity concerns), was found in 22 of 71 European deodorants tested, a chemical identity a fragrance label would not disclose since it is covered by the generic term 'parfum'.
Moderate evidence: Structural classification, not a clinical allergy finding for this specific compound in this paper; presence only, not concentration.
Source: Rastogi et al. 1998 (Contact Dermatitis 39:293), Contact Dermatitis Link to this fact
Hidden ingredients / Review of other studies
Fragrance in deodorants and antiperspirants often functions as a deliberate odor-masking ingredient at percent-level concentrations (e.g., linalool combined with dihydromyrcenol at 0.2-1% w/w, or a multi-fragrance blend of musk, skatole, lemon oil, lavender oil, jasmine, vanillin, benzoin, benzyl acetate, and menthol in one patented licorice-root deodorant), separate from and in addition to any anti-microbial active in the same product.
Moderate evidence.
Caveats
Review compiling cosmetics-industry patents and older formulation studies, not new experimental data; the specific listed formulations are patent examples, not necessarily current marketed products.
Source: Kanlayavattanakul & Lourith 2011, International Journal of Cosmetic Science Link to this fact
Hidden ingredients / Lab study (cells or chemistry)
In a lab test of 25 commercial underarm deodorants, 9 (36%) showed measurable estrogen-like activity in a human breast-cancer-cell assay, concentrated almost entirely in sprays (7 of 10) versus roll-ons (1 of 9) and sticks (1 of 6) from the same product lines.
Moderate evidence.
Caveats
Single lab, single 5-day cell-proliferation assay (E-Screen/MCF-7) per product; no second confirmatory assay, no dermal absorption or human data; the assay is a mixture 'sum parameter' that cannot identify which ingredient(s) drove any one product's result.
Hidden ingredients / Lab study (cells or chemistry)
Testing 62 individually purchased fragrance-related ingredients found in the deodorants, 8 were estrogenic in the same cell assay – including the musk HHCB/galaxolide and benzyl salicylate – at potencies the study describes as comparable to parabens, a preservative class with well-documented weak estrogenic activity.
Moderate evidence.
Caveats
Potencies were 100,000 to 100,000,000 times weaker than estradiol (EC50 3.08×10^-6 to 1.45×10^-4 M; EEF 2.34×10^-8 to 1.15×10^-6); a positive result in this cell assay does not establish that any dose reaches human tissue.
Hidden ingredients / Measured in people or real products
In a 2018 Korean government survey of 482 market deodorants, 18 of 26 screened fragrance allergens and 13 of 27 screened biocidal ingredients were actually detected in the products; some fragrance ingredients reached extreme concentrations in individual products, including benzyl benzoate at 142,344 mg/kg (about 14% of product weight) and thymol at 271,525 mg/kg (about 27% of product weight).
Moderate evidence: Single Korean-market survey (64-482 products depending on the ingredient panel); concentrations are maxima found in specific products, not typical or average levels.
Source: Kim et al. 2018, Science of The Total Environment Link to this fact
Hidden ingredients / Review of other studies
"Alum stone" deodorants are potassium alum (Al3K(SO4)2, 12H2O), which does contain aluminum, sometimes marketed under the name "potassium alum" or the misleading "double potassium sulfate," which omits the word aluminum entirely.
Moderate evidence: Single review's characterization of marketing language, not a systematic label survey.
Source: Martini 2020, Annales de Dermatologie et de Vénéréologie Link to this fact
Hidden ingredients / Measured in people or real products
Hormone-disrupting chemicals in household cleaners, synthetic detergents, fabric softeners, air fresheners, and deodorants are almost always added as fragrance or preservative, not as the product's main ingredient, so their presence rarely shows up as a headline ingredient a shopper would notice.
Moderate evidence: Based on a database of listed ingredients matched to EPA hazard flags, not chemical measurement of the actual products; a single country's product registry (South Korea, 2020).
Source: Lee & Ji 2022, Ecotoxicology and Environmental Safety Link to this fact
Hidden ingredients / Measured in people or real products
All 17 everyday product perfumes (shampoos, baby bath, lotions, deodorants) contained at least one of the fragrance-mix allergens by GC-MS, about 2.6-2.8 on average; geraniol was in 12 of 17.
Moderate evidence: Single manufacturer's lower-priced perfumes, 1990s Denmark; oak moss not analyzed; concentrations are in the perfume, not the finished product.
Source: Johansen 1997, Contact Dermatitis Link to this fact
Hidden ingredients / Other evidence
Limonene and beta-pinene scent terpenes were measurable in every perfume, deodorant and cleaning liquid tested (about 0.1-0.3% and 0.04-0.65% of product weight), even though labels only say 'fragrance' or, for the cleaners, name limonene.
Moderate evidence.
Caveats
Six products; GC-MS of three terpenes only, so other scent chemicals are unmeasured; alpha-pinene was below the detection range. The abstract's beta-pinene range (0.4-0.7%) differs from Table 1 (0.04-0.65%).
Indoor air / Measured in people or real products
Among 3,411 Czech households with an infant, the most commonly used chemical products were window cleaner (88%), deodorant (82%) and aerosol/hairspray products (74%), not air fresheners, which were the second-least-used of 18 chemical product categories (26%).
Moderate evidence: Self-reported product use frequency, one Central European cohort; product prevalence does not equal exposure dose or emission rate.
Source: Mikeš 2019, Science of the Total Environment Link to this fact
Indoor air / Measured in people or real products
A personal-care product containing the silicone D5 (such as a deodorant applied hours earlier) can raise D5 levels inside a car far above outdoor background, and this passenger-carried D5 vents out of the car and mixes with the vehicle's exhaust-associated emissions when the cabin fan runs.
Moderate evidence.
Caveats
Small controlled experiment (4 passengers, 1 vehicle, plus one deodorant-in-car test); a car cabin is a distinct, small, enclosed microenvironment, not home or building indoor air, which this paper does not measure.
Source: Coggon 2018, Environmental Science & Technology Link to this fact
Indoor air / Measured in people or real products
In a study of 60 UK homes with real air sampling (not a lab chamber), the VOC found in the highest concentration indoors was n-butane, an aerosol-spray propellant, not a fragrance chemical; it reached a wintertime maximum of 4,630 ug/m3, and aerosol antiperspirant/deodorant was the single most universally used VOC-releasing product, used in all 60 homes.
Moderate evidence: One UK town's panel-recruited cohort (Ashford, Kent); Givaudan-funded (fieldwork), analysis independent at University of York.
Source: Heeley-Hill 2021, Environmental Science: Processes & Impacts Industry-funded Link to this fact
Indoor air / Measured in people or real products
Of every ingredient/route combination in this deodorant risk assessment, the tightest safety margin was for inhaling didecyldimethylammonium chloride (DDAC), a quaternary-ammonium biocide, from spray deodorants used at home or in cars: its calculated margin of exposure was 30, only about 1.2 times the regulatory target of 25, driven by an unusually low reference concentration for DDAC rather than by high measured exposure.
Moderate evidence.
Caveats
Still above the regulatory safety threshold in this study's own framework (no risk was formally identified); a single Korean government assessment, modeled exposure rather than measured personal or biomonitoring exposure.
Source: Kim et al. 2018, Science of The Total Environment Link to this fact
Indoor air / Measured in people or real products
The plasma deodorant device itself generates ozone: about 1.5 ppm measured 3 cm from the outlet (roughly 7.5x the 15-minute occupational STEL of 0.20 ppm the authors cite), falling to about 0.16 ppm at 30 cm in a direct line and 0 ppm when the path to the sensor was blocked, which the authors say is below international occupational ozone limits at that distance.
Moderate evidence.
Caveats
Ozone measured by the device's own gas detector at fixed distances in open air with the device stationary; not measured as real personal breathing-zone exposure during actual underarm use, and occupational limits (TWA/STEL) were not designed for close-range personal-care-device exposure.
Source: Bok et al. 2024, Scientific Reports Link to this fact
Indoor air / Other evidence
For hairspray, deodorant, and eau de toilette sprayed at the head or upper body, about 85% of the sprayed product deposits on hair, skin, or other surfaces and only about 15% stays in the air available for inhalation, as a default in the absence of product-specific measurement.
Moderate evidence.
Caveats
The only cited direct measurement behind a comparable number is a 2012 deodorant/antiperspirant aerosol study (76.5-88.6% available for inhalation, i.e., 11.4-23.5% deposited) — a different, narrower product set than hairspray/eau de toilette.
Source: Steiling et al. 2025, Toxicology Letters Industry-funded Link to this fact
Indoor air / Other evidence
In the fragrance industry's 2015 aggregate exposure model, perfume, body spray, and deodorant spray were counted only as skin exposure; the paper reports no figure for how much is breathed in when a spray is used on the body.
Moderate evidence: Industry-funded paper by RIFM, Unilever, P&G, Givaudan and Firmenich staff; it shows the model did not estimate inhalation, not that inhalation is large or small.
Source: Safford et al. 2015, Regulatory Toxicology and Pharmacology Industry-funded Link to this fact
Indoor air / Other evidence
The CIR draft applies its 'most spray droplets are not respirable' reasoning to polysorbate sprays at up to 4%, while admitting some evidence that deodorant sprays release substantially larger respirable fractions and that the information is 'not sufficient'.
Moderate evidence: Inhalation data are thin (one ethoxylated sorbitan laurate rat LC50); same boilerplate challenged in the CIR 2024 respiratory draft.
Source: CIR 2015 draft Industry-funded Link to this fact
Indoor air / Other evidence
In a lab chamber with a very high ozone level (about 6.5 ppm), all six ordinary perfumes, deodorants and cleaning liquids tested formed new particles; about 0.2% to 17% of the product's evaporated mass became aerosol, with an average near 4%.
Moderate evidence.
Caveats
One lab, six products, a closed 0.68 m3 chamber at roughly 100 times typical indoor ozone, very dry air, no vapor wall-loss correction; a yield, not an indoor concentration. NSF and Sloan funded; no industry funding.
The industry / Measured in people or real products
The independent dermatologists who ran this study suggested deodorants should contain hydroxycitronellal at less than 0.032% to protect people already sensitized, well under the fragrance industry's own 1987 recommendation equating to about 1% in finished consumer products.
Moderate evidence.
Caveats
Authors' own recommendation, not a regulatory outcome; 0.032% is only the lowest concentration this single-center study tested, not a demonstrated safe floor, and the industry cap was set from human sensitization (induction) testing, not this elicitation data.
Source: Svedman et al. 2003, Contact Dermatitis Link to this fact
The industry / Government agency
A 2024 CIR draft stated that industry-submitted laser-diffraction survey data show an average of 26.6% (+/- 13.4% SD) of particles released from propellant deodorant/antiperspirant sprays are small enough to reach the deep lung (<10 micron), versus under 1% for pump sprays.
Moderate evidence.
Caveats
Document is stamped 'Distributed for Comment Only — Do Not Cite or Quote' and was still under Panel discussion as of the Dec 2024 meeting; the underlying particle-size data are unpublished industry submissions (self-reported, small n=3 companies), not an independently published study.
Source: CIR draft (Zhu) 2024 Industry-funded Link to this fact
The industry / Government agency
A 2024 CIR draft's meeting transcript recorded that an earlier version of its propellant-deodorant exposure calculation used a 5% respirable-particle-fraction assumption, which two Expert Panel members identified as very likely wrong given the same document's own data showing a median particle diameter of about 10 micron for these sprays (implying roughly half the particles should be below the respirable threshold); the assumption was later revised upward (33%, then 26.6-40%).
Moderate evidence.
Caveats
This is a self-corrected internal drafting error surfaced in transcript discussion, not a peer-reviewed finding; the transcript indicates the error was first raised by an outside commenter (Women's Voices for the Earth) rather than caught internally.
Source: CIR draft (Zhu) 2024 Industry-funded Link to this fact
The industry / Other evidence
As of 2025, the cosmetics industry's own safety scientists say the 'airborne fraction' values used to calculate inhalation exposure to hairspray, deodorant, and perfume sprays are a heuristic convenience carried over from a 2009 model, not physically grounded measurements, and a dedicated industry project (ICCS) to generate real product-specific values is still ongoing.
Moderate evidence: This is the industry's own admission of a data gap in its own exposure model, not an outside critique.
Source: Steiling et al. 2025, Toxicology Letters Industry-funded Link to this fact
The industry / Other evidence
The daily-use amounts the EU's official cosmetic-ingredient safety assessments (SCCS Notes of Guidance Table 3A) use for body lotion, lipstick, and deodorant non-spray trace directly to a Colipa (the EU cosmetics trade association)-commissioned industry study of 44,100 households and 18,057 individual consumers in five countries, and check out against that study's own published P90 data.
Moderate evidence.
Caveats
Industry-funded and industry-run (Colipa, with authors from L'Oreal, P&G, Unilever, Henkel, Beiersdorf and the commercial modeling contractor CREMe Software); the paper's own Conflict of Interest statement says "there are no conflicts of interest" despite only one of eight authors being academic.
Source: Hall 2007, Food and Chemical Toxicology Industry-funded Link to this fact
The industry / Other evidence
For body lotion, facial moisturizer, toothpaste and shampoo, the more often a person uses the product the less they use per application (an inverse correlation between frequency and amount-per-use); no such correlation exists for lipstick or deodorant/antiperspirant, so multiplying the maximum daily frequency by the maximum amount-per-event overstates exposure to the first four products — for body lotion, that naive worst-case calculation (47 g/day) is well above the study's own calculated 99.9th percentile of 17 g/day.
Moderate evidence.
Caveats
Industry-funded (Colipa); the correlation and its practical consequence are the study's own stated headline finding, not independently replicated elsewhere in this evidence base yet.
Source: Hall 2007, Food and Chemical Toxicology Industry-funded Link to this fact
The industry / Other evidence
This 2007 industry study found Europeans' deodorant/antiperspirant non-spray use had roughly tripled versus the ~15-year-old figure it replaced — from 0.5 g/day to 1.5 g/day (90th percentile) — which the authors attribute to changing personal-hygiene habits rather than a change in measurement method.
Moderate evidence: Industry-funded (Colipa); the attribution to genuine habit change versus improved/different measurement methodology is the authors' own interpretation, not separately tested.
Source: Hall 2007, Food and Chemical Toxicology Industry-funded Link to this fact
Information, not medical advice. See also: myths we won’t tell you.