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.
What you can do / Other evidence
A homeopathic product being licensed, registered, or offered on the NHS in the UK is a regulatory or local-funding decision, not a finding that it works: the Government said simple registration needs no proof of efficacy.
Moderate evidence: Applies to the UK 2010 regime as described; later changes are not covered.
Source: UK Department of Health, Government Response to the Science and Technology Committee report 'Evidence Check 2: Homeopathy', Cm 7914 (July 2010), Command Paper Cm 7914 (July 2010) Link to this fact
What you can do / Other evidence
For what the Australian government concluded about homeopathy, read NHMRC's Information Paper and Statement (March 2015), not this administrative report, which describes the process and reports no results.
Moderate evidence: Advice on citing, not a health claim.
Source: NHMRC 2015 Administrative Report on the homeopathy review (CAM02B) Link to this fact
What you can do / Other evidence
In a Rutgers/Columbia questionnaire study, people who always or usually consulted health apps or websites used personal care products with lower average Skin Deep hazard scores than those who rarely did.
Moderate evidence.
Caveats
Cross-sectional, one university community; the apps use the same label information the score uses, so part of the link is built in; lower score does not equal lower measured exposure.
Source: EWG 2025 (Friedman & Huang), EWG news (web page) Link to this fact
What you can do / Journalism
Skin Deep's method lists the product details EWG records (body area, leave-on vs rinse-off, spray, or powder form, intended age group), but its published scoring steps do not weight a product by concentration, amount applied, or leave-on use.
Moderate evidence.
Caveats
The page says route and demographic data are used for some ingredient restrictions; whether any of it enters the final score beyond that is not stated, so the narrower reading is of the described formula only.
Source: EWG Skin Deep methodology page (2026 capture) Link to this fact
What you can do / Other evidence
To get EWG Verified, a restricted ingredient must be backed by documentation such as a certificate of analysis or laboratory test report, and EWG reserves the right to test mark-bearing products at any time.
Moderate evidence: A reserved right, not a reported testing program; no frequency, lab or results are given.
Source: EWG Verified FAQ, verified-portal.ewg.org, accessed 2026-10-16, EWG Verified portal web page Link to this fact
What you can do / Other evidence
In its 2006 decision on benzalkonium-chloride-type disinfectants (ADBAC), EPA required the residential humidifier use to be deleted because its modeled 24-hour inhalation margins were 10 for adults and 4 for children against a target of 100.
Moderate evidence: One 2006 agency document using surrogate-based modeling; later EPA status was not checked and the use could have been reinstated or the decision updated.
Source: EPA RED 2006 Link to this fact
What you can do / Government agency
If you have eczema from cosmetics, ask for patch testing that includes the fragrance mixes, HICC, and oxidized limonene and linalool, and check ingredient labels for the named allergens; the EU committee said incomplete label information can cause missed diagnoses.
Moderate evidence: Clinical advice summarized from the committee's secondary-prevention discussion; 'fragrance' mixes still hide non-listed ingredients.
What you can do / Other evidence
FDA required this firm to submit risk assessments of US products made from un-validated blends or inadequately tested components, with particular emphasis on infants and children and on lots still within expiry, and placed the firm on Import Alert 66-40 on 3 May 2017.
Moderate evidence: The firm's response and any outcome are not in the letter.
Source: FDA warning letter 320-17-45, 2 Aug 2017 Link to this fact
What you can do / Other evidence
Suffolk County's 2018 report sorted perchloroethylene into its red (least preferable) category for both human health and environment, and liquid carbon dioxide into green for both.
Moderate evidence.
Caveats
A county health department's qualitative color code built from other agencies' reviews, not a measured or quantitative risk ranking; data volume differs a lot by solvent (decades for perc, little for newer ones). No industry funder named.
Source: SCDHS 2018 Link to this fact
What you can do / Other evidence
The county report relays Sherlach 2011: perchloroethylene residue of 10-56 nmol/cm2 in dry-cleaned clothes, falling about 50% over 7 days, and cites US EPA advice not to accept clothing back from a cleaner if it smells strongly of solvent.
Moderate evidence: Secondary relay; our evidence base holds the primary (Sherlach 2011, one lab, seven shops, one jacket per shop). Strong-odor advice is agency precaution, not a measured threshold.
Source: SCDHS 2018 Link to this fact
What you can do / Government agency
Under EU law, a body or spray sunscreen can contain at most 2.2% oxybenzone as a UV filter, so a higher-oxybenzone body spray is not compliant in the EU.
Moderate evidence.
Caveats
Applies to products placed on the EU market after the 2023 transition. The 2.2% cap holds unless 0.5% benzophenone-3 is also added as a formulation protectant, in which case the UV-filter level is capped at 1.7% (checked against the Official Journal print 2026-10-23).
Source: Commission Regulation (EU) 2022/1176, Annex VI entry 4 Link to this fact
What you can do / Other evidence
Do not use unapproved homeopathic eye drops to treat cataracts or other eye disease; see an eye-care professional and report reactions to FDA MedWatch.
Moderate evidence: Advice follows from FDA's unapproved-drug and route-of-administration statements; no harm data in the letter.
Source: FDA warning letter 708635 (25 Aug 2025), FDA warning letter (MARCS-CMS 708635), 25 August 2025 Link to this fact
What you can do / Other evidence
Check the ingredient list for "fragrance/parfum" first; a retailer's "clean" tier or a row of "free of" badges did not predict its absence in this audit.
Moderate evidence: Inference from a one-store label audit; says nothing about fragrance-free options elsewhere or about exposure.
Source: Madrid Larrañaga 2026, Journal of Exposure Science & Environmental Epidemiology Link to this fact
What you can do / Other evidence
Hairdressers, bakers and dental surgery assistants had the highest estimated rates of recognized hand eczema in Denmark (about 5.6, 8.4 and 4.2 cases per 1000 person-years), and 71% of hairdressers with recognized hand eczema were apprentices.
Moderate evidence: Rates cover recognized cases only and denominators were unavailable for many occupations; single country, single 58-week window.
Source: Skoet 2004, Contact Dermatitis Link to this fact
What you can do / Measured in people or real products
If you have persistent dermatitis and a negative fragrance-mix patch test, ask whether the oxidized linalool and limonene preparations and the individual EU-26 allergens were tested too.
Moderate evidence: Test availability varies by clinic; positive reactions need clinical-relevance review, and about 14-21% of oxidized-terpene readings are doubtful.
What you can do / Other evidence
Dermatitis patients who say a perfumed product has ever caused or worsened their rash had about double the odds (OR 2.16) of testing positive to a fragrance allergen, so such a history is a reason to ask for patch testing.
Moderate evidence: Self-report in a referred population; most 'yes' answers were not allergic (138 of 675).
Source: Bennike et al. 2019, Contact Dermatitis Link to this fact
What you can do / Other evidence
If eczema keeps returning, ask for extended patch testing that includes the oxidized limonene and linalool hydroperoxides, and check products against the CAMP database; basic screening can miss fragrance allergy.
Moderate evidence: Practice recommendation from a consensus committee; not every clinic uses the extended series.
What you can do / Review of other studies
A single large inhalation of an irritant gas, such as chlorine released when bleach is mixed with an acid, can cause lasting asthma within 24 hours in a person who never had it.
Moderate evidence: Review of mostly descriptive studies, mainly workplace accidents; no household incidence given.
Source: Lemiere 2022, The Journal of Allergy and Clinical Immunology: In Practice Link to this fact
What you can do / Measured in people or real products
If a rash has no obvious cause, the 1982 dermatologists' data suggest cosmetics are worth considering even when you do not suspect them, and patch testing with your own products and ingredient lists can identify the cause.
Moderate evidence: Selected referral series of contact dermatitis specialists, 1977-1980 (40 months); not general-population rates; half of cases were identified only by testing.
Source: Eiermann 1982, Journal of the American Academy of Dermatology Link to this fact
What you can do / Review of other studies
If you have a proven preservative allergy, check the label for the specific name and synonyms and ask your clinic for a written list: formaldehyde releasers, MCI/MI and parabens are the common preservatives, and 17-28% of creams in three studies were labeled incorrectly.
Moderate evidence: Advice follows from patch-test and label data; it does not tell which product is safe.
What you can do / Other evidence
The authors of Li 2010 say their result may apply only to highly exposed workers; the effect of lower, environmental BPA exposure on male sexual function is unclear from this study.
Moderate evidence: This is the authors' own caveat; no general-population data in this paper.
What you can do / Other evidence
Of 65 top-rated Amazon sunscreens analyzed in December 2015, 40% (26) did not meet the American Academy of Dermatology's criteria (SPF 30 or higher, broad spectrum, water or sweat resistance), most often because they made no water-resistance claim.
Moderate evidence.
Caveats
One snapshot of one retailer's best-reviewed products; the criteria were scored from label claims, not tested. A non-water-resistant product can suit daily non-sport use, as the authors note.
What you can do / Other evidence
Sunscreen prices in that top-rated Amazon group ran from $0.68 to $23.47 per ounce (median $3.32), and 9 of the 10 cheapest products still met the AAD criteria.
Moderate evidence: Prices are a December 2015 snapshot; the cheapest products were mostly chemical-filter lotions and sprays.
What you can do / Other evidence
In 400 men, each additional type of personal care product used in the past day went with about 12% higher urinary MEP and 14 to 29% higher parabens.
Moderate evidence: Observational; crude count of products; same cohort and lab as the women's study, so not independent replication.
Source: Nassan 2017, Environmental Health Perspectives Link to this fact
What you can do / Other evidence
The industry panel warned that cetrimonium bromide and palmitoyl carnitine can enhance skin penetration of other chemicals (phenylazoaniline, benzocaine, benzoic acid), so products combining them with ingredients whose safety depends on low absorption need careful formulation.
Moderate evidence.
Caveats
Penetration enhancement shown in artificial membranes (enhanced below 0.5%, inhibited above 1.0%); no human product data. CIR is financed by the Personal Care Products Council, the cosmetics industry trade association.
Source: Becker et al. 2012 (CIR), Int J Toxicol, International Journal of Toxicology Industry-funded Link to this fact
What you can do / Other evidence
The FDA hair-dye precautions cited in the CIR hydrogen peroxide report are to follow package directions, do a 48-hour skin patch test, wear gloves, rinse the scalp well and not dye eyebrows or eyelashes.
Moderate evidence: FDA consumer guidance quoted in an industry-funded report.
Source: Becker et al. 2024, International Journal of Toxicology Industry-funded Link to this fact
What you can do / Other evidence
Because US food labels omit legally exempt incidental additives, a person who must avoid a particular substance cannot rely on the ingredient list alone and should ask the manufacturer.
Moderate evidence: The exemption is limited to insignificant levels with no functional effect; most ingredients must still be listed. The advice is our inference, not a statement in the regulation.
Source: 21 CFR 101.100 (eCFR, 2026-10-15) Link to this fact
What you can do / Other evidence
Dermatologists advise patch testing for benzyl alcohol in suspected fragrance or cosmetic contact dermatitis, treatment-resistant dermatitis, and dermatitis that worsens with topical therapy, with avoidance counseling that includes medicines and 'clean,' 'sensitive' or 'fragrance-free' products.
Moderate evidence: Expert recommendation from a narrative review; no trial of avoidance outcomes.
Source: Zheng et al. 2026, ACDS Allergen of the Year (Cutis), Cutis Link to this fact
What you can do / Measured in people or real products
If leg or foot dermatitis from venous disease is worsening or not responding to treatment, ask a dermatologist about patch testing and bring the creams, ointments and dressings you use.
Moderate evidence: Authors' recommendation from a retrospective series; no data on whether avoidance improves outcomes.
Source: Silverberg et al. 2021, Arch Dermatol Res, Archives of Dermatological Research Link to this fact
What you can do / Government agency
In EFSA's exposure model, fine bakery wares were the main source of polysorbate intake in every age group (up to 91-98% in some surveys), followed by soups, desserts, edible ices, emulsified sauces and, for children through the elderly, food supplements (up to 70-88%).
Moderate evidence: Percent contributions from modeled mean exposure per survey; supplements were assessed without brand or form, and supplement use levels ranged to 425,000 mg per kg.
Source: EFSA ANS Panel (2015), full opinion, EFSA Journal Link to this fact
What you can do / Other evidence
CTV reported that the Environmental Health Association of Québec (ASEQ-EHAQ), after the woman's death, launched a national campaign and fundraiser to get housing for people with MCS across Canada, and its president advocates a national program to build chemical-free homes.
Moderate evidence: Advocacy group's own account as relayed by CTV; the group is not independent of the cause. Article subject is a death by MAiD; add 988 line in published use.
Source: CTV News, April 2022 (via Smoke-Free Housing BC; redacted copy), CTV News (news article) Link to this fact
What you can do / Other evidence
As displayed when captured on 2026-10-21, the GoFundMe page for the 2022 campaign showed "$10,220 raised of $10K CAD" from 82 donations.
Moderate evidence.
Caveats
Snapshot as of 2026-10-21, more than four years after the campaign date (2022-01-28); it may include donations made after her death, and it is not a total as of any 2022 date. Partial capture ending before the campaign description; the figure is a 2026-10-21 snapshot.
Source: GoFundMe campaign page, "Save Sophia's Life!" (2022; captured 2026-10-21; redacted, partial) Link to this fact
What you can do / Other evidence
The campaign page is dated January 28th, 2022 and filed under "Emergencies"; its opening appeal states that she would die by lethal injection on February 22, 2022 unless she found healthy housing.
Moderate evidence: This is the organizer's wording in an advocacy appeal, not an independent record. Partial capture ending before the campaign description; the figure is a 2026-10-21 snapshot.
Source: GoFundMe campaign page, "Save Sophia's Life!" (2022; captured 2026-10-21; redacted, partial) Link to this fact
What you can do / Other evidence
The committee's recommendations ask Health Canada to consider guidance on how to approach MAiD requests when suffering is predominately derived from an unmet social need (e.g., housing), and ask the Ontario Ministry of Health and Ontario Health to consider a provincially coordinated MAiD care system with interprofessional assessment and enhanced care coordination for people with unmet social needs such as housing.
Moderate evidence: All recommendations are worded "to consider"; the report does not say whether any recipient responded.
Source: Ontario MDRC Report 2024-3 (OCC, 2024; non-official copy) Link to this fact
What you can do / Government agency
Under CHRC guidance, a person whose environmental sensitivity is triggered at work or at a federal service must tell the employer or service provider, which the CHRC calls the duty to inform; the organization then has a duty to accommodate in good faith.
Moderate evidence: Guidance, federal jurisdiction only.
Source: CHRC 2026 web Link to this fact
What you can do / Government agency
The CHRC says anyone who experiences disability discrimination or does not receive appropriate accommodation from a federally regulated employer or service provider can file a complaint with the Commission.
Moderate evidence: Federal jurisdiction only; provincial and territorial bodies handle other employers, which the page does not discuss.
Source: CHRC 2026 web Link to this fact
What you can do / Other evidence
OSHA tells employers to build air quality into purchasing decisions and avoid products that can cause IAQ problems, and tells workers not to bring in products that release harmful or bothersome odors or contaminants.
Moderate evidence: Advisory wording; no product category (such as fragranced products) is named.
Source: OSHA 3430-04, 2011 Link to this fact
What you can do / Other evidence
OSHA recommends scheduling cleaning or maintenance work when other occupants are absent and limiting the amount and location of chemicals used near workers.
Moderate evidence: Advisory administrative controls; no evidence of effectiveness given.
Source: OSHA 3430-04, 2011 Link to this fact
What you can do / Self-reported survey
Swedish mothers who used body lotion every second day or more had about 7 times the urinary methylparaben (46 vs 6.6 micrograms per gram creatinine) and propylparaben (17.6 vs 2.3) of those who used it weekly or less; make-up showed the same pattern. Parabens are preservatives, not fragrance ingredients.
Moderate evidence.
Caveats
Cross-sectional, 95 mothers (76 with paraben data), frequency-of-use questionnaire, no multiple-testing correction; group sizes were uneven (16 low-lotion users); it shows exposure, not harm.
Source: Larsson 2014, Environment International Link to this fact
What you can do / Other evidence
The authors point out that phasing out lead in petrol delivered an estimated US$200 billion economic benefit per annual US birth cohort and that every US$1 spent reducing lead hazards is estimated to return US$17-220.
Moderate evidence: Model-based estimates; a case for regulating a single well-understood chemical, not for any particular consumer product.
Source: Grandjean & Landrigan 2014, Lancet Neurology, The Lancet Neurology Link to this fact
What you can do / Animal study
Triclosan is an antimicrobial used in toothpaste, soaps and cleansers, not a fragrance ingredient, so this rat neurotoxicity study gives no reason to link fragranced products to ADHD-like behavior.
Moderate evidence: Scope statement: the study tested only triclosan; it does not address fragrance chemicals either way.
Source: Cui 2024, Ecotoxicology and Environmental Safety Link to this fact
What you can do / Measured in people or real products
In a study of 110 Baltimore children with asthma (86% Black, mostly low-income), children from homes that had used an air freshener or room deodorizer in the past day or two had urine propyl paraben about 3 times higher (ratio 3.03, 95% CI 1.64-5.60) and BPA about 1.9 times higher (1.92, 1.36-2.71) than other children.
Moderate evidence.
Caveats
One city, one urine sample on one day, cross-sectional, no multiple-comparison correction, no product or air content measured, and the BPA and BPS lower confidence limits sit at about 1.0-1.4; the same effect was not tested in a replication cohort.
Source: Fandiño-Del-Rio 2024, Journal of Exposure Science & Environmental Epidemiology Link to this fact
What you can do / Measured in people or real products
Children who ate food from metal cans had about 1.7 times higher urine BPA (1.71, 95% CI 1.20-2.43) and 2.05 times higher BPS (1.18-3.56), and those who drank from plastic bottles had BPA about 1.65 times higher, in 110 Baltimore children with asthma.
Moderate evidence: Not fragrance or personal care; single spot urine and a 1-2 day recall window; consistent with prior NHANES work cited by the authors but not independently replicated here.
Source: Fandiño-Del-Rio 2024, Journal of Exposure Science & Environmental Epidemiology Link to this fact
What you can do / Measured in people or real products
In the Baltimore asthma-children study, lip balm was the only product linked to urine benzophenone-3 (a UV filter), with about 2.4 times higher levels in users (2.39, 95% CI 1.37-4.16); no children reported using makeup and only 2 reported sunscreen.
Moderate evidence: Single cohort; lip balm formulations not identified (sun-protective or not); the study could not check labels.
Source: Fandiño-Del-Rio 2024, Journal of Exposure Science & Environmental Epidemiology Link to this fact
What you can do / Measured in people or real products
Parabens are cosmetic and personal-care preservatives, not fragrance ingredients, so this child-behavior study provides no evidence about perfume or fragrance.
Moderate evidence: Paper measured only methyl-, ethyl- and propylparaben; no fragrance compounds, phthalates or MEP.
What you can do / Other evidence
In 2022, 30.1% of US children with current ADHD received neither ADHD medication (53.6% did) nor behavioral treatment (44.4% did); the share with no treatment was higher in 2022 than in 2016 (23.0%).
Moderate evidence: Cross-sectional: cannot separate new diagnoses, never-treated and stopped treatment; treatment items are broad; 2022 medication shortages may play a part.
Source: Danielson 2024 (NSCH 2022), Journal of Clinical Child & Adolescent Psychology Link to this fact
Information, not medical advice. See also: myths we won’t tell you.