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.
Health effects / Measured in people or real products
Women who had ever used underarm products had higher total paraben in the lateral breast region than never-users (median 115.7 vs 41.8 ng/g, P = 0.043), but no other region or paraben differed meaningfully.
Weak evidence.
Caveats
One of about 24 uncorrected comparisons, only 7 never-users who were also about 14 years older on average, and the authors read the overall pattern as similar levels in users and nonusers.
Source: [[src-2012-measurement-paraben-concentrations-breast-tissue-serial-locations|Barr 2012]], Journal of Applied Toxicology Link to this fact
Health effects / Other evidence
In a 1972 class experiment, 56 medical students were told to expect a sedative or stimulant but all got only inert capsules; 53 of 56 (all but 3) reported at least one change, and the most common were drowsiness (54%), sluggishness (52%) and tiredness (50%).
Weak evidence: Single small class, single-blind, no untreated control arm, self-ratings, taken during a lecture hour; shows reports follow suggestion, not the size of any placebo effect.
Health effects / Other evidence
Healthy students given only placebo also showed drug-like physical readings: pulse fell in 66% and systolic blood pressure in 71% an hour later, and 32% volunteered side effects such as headache.
Weak evidence: No untreated comparison; lecture-hour sitting and observer bias (authors flag systolic pressure and pulse as bias-prone) could explain part of the drop.
Health effects / Other evidence
In the same experiment, two placebo capsules produced stronger reported changes than one (26 of 158 changes were severity 2 vs 6 of 119), and blue capsules produced more reported drowsiness than pink (72% vs 37%).
Weak evidence: Groups of 12-16 subjects; one study; color effect may be specific to the culture, era and the sedative framing used.
Health effects / Self-reported survey
Even the unbranded inert tablet gave some relief in 73% of headaches and 40% were rated considerably or completely better, a reminder that self-reported relief includes expectation and natural resolution.
Weak evidence: No no-treatment control, so placebo response here cannot be separated from natural course or regression to the mean.
Health effects / Other evidence
Davenas et al.'s 1988 Nature paper claimed that an antibody solution diluted until no antibody molecules should remain (down to 1 in 10^120) could still make human basophils release their granules, and that this needed vigorous shaking. It is the origin of the 'memory of water' idea.
Weak evidence: The claim was not reproduced when Nature sent investigators to the lab (Maddox 1988); peaks were erratic; funding not stated; in-vitro cell assay only, no clinical outcome.
Source: Davenas et al. 1988, Nature, Nature Link to this fact
Health effects / Other evidence
A hospital pharmacist developed work-related asthma from a quaternary ammonium biocide (lauryl dimethyl benzyl ammonium chloride, a benzalkonium chloride) in a floor cleaner used when he was out of the room; a one-minute challenge with the single ingredient caused a late asthma attack (FEV1 about 4.1 to 2.3 liters), while the plain detergent and the other cleaner ingredients caused none.
Weak evidence.
Caveats
Single patient with a long history of asthma triggered by other phenols; no measured air levels, no specific IgE, blinding not described; shows that a QAC can provoke asthma in a sensitized person, not how common this is.
Health effects / Other evidence
In this case the cleaner was applied to the whole pharmacy floor when the worker was not present and he never handled it, so exposure was indirect and likely low, yet it was enough to drive his asthma until the cleaner was replaced by a simple detergent.
Weak evidence.
Caveats
The low exposure is inferred from the indirect route, not measured; one patient with pre-existing asthma and hyperreactive airways (histamine reactivity 0.9 µmol on a day away from work).
Health effects / Other evidence
A 1996 BMJ review of six trials found no consistent effect of pill color on how well a drug works: some trials showed differences between colors, but designs and outcomes differed so much that the results could not be pooled.
Weak evidence: Trials 1968-1978, mixed quality (three scored under 5 of 10), no meta-analysis; absence of a consistent trend is not proof of no effect.
Source: de Craen 1996, BMJ Link to this fact
Health effects / Other evidence
The panel's own pathologists re-read the original brain slides from three Danish rat studies that reported cerebellar 'cyst-like spaces' after pulegone, peppermint oil, or menthone, and concluded the spaces were fixation artifacts, citing a failed reproduction (Mengs & Stotzem 1989) and a perfusion-fixation study (Olsen 1994) with no spaces.
Weak evidence.
Caveats
The slide re-read was done by the industry panel's own two pathologists, not an independent laboratory; the original authors' response is not given; the panel still accepts the same studies' liver findings.
Source: Adams 1996, Food and Chemical Toxicology Industry-funded Link to this fact
Health effects / Other evidence
In the NTP isophorone studies (103 weeks, 250 and 500 mg/kg/day gavage), male rats developed kidney tumors, male mice liver tumors (29/50 combined at the high dose) and fibrosarcomas (14/50 vs 6/48 controls); the FEMA panel judged all three findings irrelevant to humans eating about 6 ng/kg/day of isophorone.
Weak evidence.
Caveats
The panel's argument rests on a male-rat-specific mechanism (alpha-2u-globulin) and on high background tumor rates in the mouse strain; the NTP itself graded 'some evidence' in rats and 'equivocal evidence' in mice; the 1,000,000-fold margin is the panel's.
Source: Adams 1996, Food and Chemical Toxicology Industry-funded Link to this fact
Health effects / Other evidence
Residual PCE varied less between six dry cleaners (under fourfold) than an earlier report suggested (five orders of magnitude), so the authors judge shop-to-shop differences small, though their own pressed-sample test was significant (p = 0.0216).
Weak evidence: Dry-to-dry machines and one fabric only; n = 25 across six shops; the authors call their result tentative; the two studies differ in design. See C399.
Source: Brand 1999, Journal of the Air & Waste Management Association Link to this fact
Health effects / Other evidence
After 10 days of oxymetazoline spray containing benzalkonium chloride, patients' nasal swelling response to a histamine challenge fell by about half (4 mg/mL: 2.2 to 1.1 mm), but this did not happen in the preservative-free arm.
Weak evidence.
Caveats
The BAK arm started with a much higher histamine response than the other arm (baseline imbalance, possible regression to the mean); reduced reactivity is not shown to be harmful; manufacturer-funded; one small trial.
Source: [[src-1999-ten-days-oxymetazoline-nasal-spray-benzalkonium-chloride|Graf 1999 (Arch Otolaryngol)]], Archives of Otolaryngology–Head & Neck Surgery Industry-funded Link to this fact
Health effects / Review of other studies
In rats given BKC-containing beclomethasone or flunisolide nasal sprays for 21 days, whole-nose sections showed squamous cell metaplasia and few cilia, mainly in the front of the nose; rats given BKC-free budesonide looked like saline controls.
Weak evidence.
Caveats
Known only secondhand through a review by an author arguing the BKC-harm case; different steroids as well as BKC differ between arms, so the steroid cannot be fully excluded; rat data.
Source: Graf 2001, Respiratory Medicine Industry-funded Link to this fact
Health effects / Other evidence
Two studies reporting no BKC harm to nasal cilia (monkeys/rats and 22 rhinitis patients) have limits the review author lists: turbinate-only biopsies, no BKC-only control, structure rather than function measured, and possible mucus protection in allergic rhinitis.
Weak evidence: The critique is reasoned argument, not new data; the author led the positive trials and the review omits his own null 1999 patient trial.
Source: Graf 2001, Respiratory Medicine Industry-funded Link to this fact
Health effects / Review of other studies
For flu, a homeopathic product (Oscillococcinum) showed a small benefit over placebo in 7 trials (n = 3,459), about 0.28 days shorter illness, but the reviewers said the data were not strong enough to recommend it.
Weak evidence: Effect small and of debatable clinical relevance per the review; evidence volume small; later trials not covered.
Source: Ernst 2002, Br J Clin Pharmacol, British Journal of Clinical Pharmacology Link to this fact
Health effects / Other evidence
Even the author of this review notes that homeopathic consultations may help patients through a caring, time-intensive encounter, which is a context effect and does not show the remedy itself works.
Weak evidence: Author's interpretation of uncontrolled outcome studies, which he says are prone to selection bias.
Source: Ernst 2002, Br J Clin Pharmacol, British Journal of Clinical Pharmacology Link to this fact
Health effects / Other evidence
The authors tested practitioners' objections (classical homeopathy, chronic conditions, longer follow-up) and found no strong evidence that any of them produced specific effects.
Weak evidence: Only 18 classical-homeopathy trials; subgroup tests have low power.
Health effects / Review of other studies
A medical historian's 2006 overview says an 1835 near-double-blind trial in Nuremberg found homeopathy ineffective, and that of the many later trials and reviews, a few showed slight benefit while a majority showed no therapeutic effect, with many trials poorly designed or too small.
Weak evidence.
Caveats
Narrative essay without systematic search; summarizes other reviews rather than pooling trials. Loudon also quotes a 2003 review saying a surprising number of high-quality studies are positive and bias in small trials cannot be disentangled.
Source: Loudon 2006, Journal of the Royal Society of Medicine Link to this fact
Health effects / Other evidence
Many illnesses people take to homeopaths are short-lived or come and go on their own, and a long, attentive consultation can itself be therapeutic, so a remission after a homeopath visit tends to be credited to the remedy.
Weak evidence: Opinion of a medical historian rather than a measured effect size; supports caution about anecdotes, not a finding that no remedy effect exists.
Source: Loudon 2006, Journal of the Royal Society of Medicine Link to this fact
Health effects / Other evidence
Migraine patients who report triggers overall also report more osmophobia (Spearman rho 0.139, p<0.001, n=818) and more premonitory symptoms (rho 0.145), though these correlations are weak.
Weak evidence.
Caveats
Weak correlations, no correction for multiple tests; consistent with either real triggering or premonitory misattribution, and the author acknowledges triggers can be hard to tell from premonitory symptoms.
Source: Kelman 2007, Cephalalgia, Cephalalgia Link to this fact
Health effects / Other evidence
In this clinic sample, perfume/odor was a trigger at similar rates in men and women (Table 1, p=0.903) even though the paper's text says women had more perfume/odor triggers; the table and text disagree.
Weak evidence: Only 143 to 145 men; the paper's own text and table conflict, so neither direction is firm.
Source: Kelman 2007, Cephalalgia, Cephalalgia Link to this fact
Health effects / Other evidence
The price effect held when only the most painful half of the shocks was analyzed (80.5% vs 56.1% reporting reduction, P = .03), and the regular-price pill produced greater pain reduction in 26 of 29 shock intensities.
Weak evidence: Same single study; high-voltage points rest on few people (as few as 4 to 8 per arm at 80 V); the per-voltage differences are shown in a figure, not tabulated.
Source: Waber 2008, JAMA Link to this fact
Health effects / Other evidence
A few patients reported breathing-related symptoms during the chamber tests, including two who rose 2-10-fold in respiratory symptom score after the control chemical geraniol as well as HICC; these symptoms did not match any measured change in the lungs.
Weak evidence: Two people drive the group-level result; the control geraniol is itself a fragrance allergen; self-reported scores.
Source: Schnuch 2010, British Journal of Dermatology Link to this fact
Health effects / Other evidence
In a small Swedish clinic sample of 60 women with migraine, 63% (38/60) named odors as a migraine trigger, and 87% of those (33/38) named perfume, followed by car exhaust (45%) and cigarette smoke (42%).
Weak evidence.
Caveats
Recruited specifically to find odor-triggered migraine and told the study was about olfaction, so prevalence is an upper bound (worst case 38/95 = 40%); self-report, no controls, no exposure measurement; does not show perfume causes migraine.
Source: Sjöstrand 2010, Curr Pain Headache Rep, Current Pain and Headache Reports Link to this fact
Health effects / Other evidence
Among those 38 women, 68% (26/38) said headache started within 20 minutes of the odor, and 5 said within seconds.
Weak evidence: Retrospective recall by a small, self-selected group; 7 of 38 gave multiple or no answers.
Source: Sjöstrand 2010, Curr Pain Headache Rep, Current Pain and Headache Reports Link to this fact
Health effects / Other evidence
Between-attack odor hypersensitivity was reported by 84% of odor-triggered migraine patients (32/38) versus 13% (3/22) of those without odor triggers in this 60-woman sample, consistent with two other studies and opposite to one.
Weak evidence: No validated hypersensitivity instrument; the paper's text is internally inconsistent about whether any non-odor-triggered women had it; small single-center sample.
Source: Sjöstrand 2010, Curr Pain Headache Rep, Current Pain and Headache Reports Link to this fact
Health effects / Other evidence
The review found larger placebo effects for physical placebos such as sham acupuncture (SMD -0.31) than for pills (SMD -0.10), and in trials that did not tell patients a placebo was possible (SMD -0.39 versus -0.19).
Weak evidence.
Caveats
Observational subgroup analyses with confounding risk; the information-to-patients factor was added post hoc and was significant only when the German acupuncture trials were included.
Source: Hróbjartsson 2010, Cochrane Database of Systematic Reviews Link to this fact
Health effects / Other evidence
In a three-week trial of 80 people with irritable bowel syndrome, patients who knowingly took pills labeled 'placebo' (after a persuasive explanation) rated their overall improvement higher than patients who got no pills (5.0 vs 3.9 on a 7-point scale; 59% vs 35% reported adequate relief).
Weak evidence.
Caveats
Single small single-center pilot, unblinded, all outcomes self-reported, 13% missing data filled by last observation carried forward; no replication in this paper; shows reporting after a placebo-plus-explanation ritual, not an effect of the capsule.
Health effects / Other evidence
The no-pill comparison group in that trial still got the same warm provider contact and also improved (46 points on a 500-point symptom scale), so some of the improvement people see in IBS studies comes from attention, time, and expectation, not only from pills.
Weak evidence: Same single pilot; the trial did not separate attention from natural fluctuation or regression to the mean.
Health effects / Other evidence
The open-label placebo benefit appeared on symptom reports but the trial gave no objective measure; quality of life improved but not significantly (11.4 vs 5.4, p = .08).
Weak evidence: Trial is underpowered for secondary outcomes; IBS has no objective marker.
Health effects / Other evidence
In a brain-imaging study of 13 migraine patients scanned within 6 hours of a spontaneous attack, a rose odor produced stronger responses in the amygdala, insula, and a pontine brainstem region than the same odor did when the same people were pain-free, even though they rated the odor equally strong.
Weak evidence: Only one unspecified rose odor at an unreported dose, given after the attack had begun; shows co-activation, not that odors cause attacks; no replication in this evidence base.
Health effects / Other evidence
Between attacks, 20 migraine patients did not differ from 20 matched healthy people in how strong they rated a rose odor (5.2 vs 5.3 on a 0-10 scale) or in brain responses to it.
Weak evidence: Small sample, one odor, patients not selected for odor sensitivity; self-reported odor hypersensitivity studies find group differences in sensitive subgroups.
Health effects / Other evidence
A 2011 study's own closet and car scenarios for dry-cleaned wool (about 50 and 126 ppm if all PCE vaporized) are theoretical ceilings that assume no ventilation, not measured exposure.
Weak evidence: The authors say definitive conclusions need skin-absorption and metabolite data; measured bedroom air in Chao 1998 was far lower; its gram conversions are about 28% low.
Source: Sherlach 2011, Environmental Toxicology and Chemistry Link to this fact
Health effects / Other evidence
Invasive melanoma was lower in the daily-sunscreen group (3 vs 11 cases, hazard ratio 0.27), but this is an exploratory finding on 14 cases that the authors say should be interpreted cautiously.
Weak evidence: Subgroup of a secondary end point; interval almost reaches 1.0; mean thickness difference (0.53 vs 1.2 mm) was not significant (P = .08).
Source: Green 2011, Journal of Clinical Oncology Industry-funded Link to this fact
Health effects / Other evidence
A 2009 experiment by Nobel laureate Luc Montagnier that is often cited as support for homeopathy used unshielded equipment, reported no signal-to-noise figures, has no published replication, and even on its own terms found effects that vanished at higher dilution, the opposite of homeopathic doctrine.
Weak evidence: Single-author critique of one paper; the Montagnier paper itself was not examined here, and the replication statement is 'to date' as of 2012.
Source: Grimes 2012, Focus Altern Complement Ther, Focus on Alternative and Complementary Therapies Link to this fact
Health effects / Other evidence
In most workers who reacted, airway sensitivity to histamine and sputum eosinophils rose after the challenge (11 of 17), a pattern typical of allergic-type occupational asthma rather than simple irritation; in workers who did not react, neither changed.
Weak evidence.
Caveats
Sputum was available for only 13 of 17 reactors and post-challenge histamine testing for 12; small numbers, single center, no QAC-specific IgE measured. The Gonzalez 2014 healthcare cohort found little QAC IgE, so mechanism remains disputed.
Health effects / Other evidence
At this Belgian center, cleaning-agent challenges were 3.2% of all work-asthma challenges in 1992-2001 and 8.6% in 2002-2011, and 16 of the 17 positive reactions came from the later decade.
Weak evidence: Share of referrals at one center, not incidence; referral patterns, awareness and QAC use may have changed. Authors call it only 'some suggestion' of an increase.
Health effects / Other evidence
The government authors judged it 'highly unlikely' that isoeugenol in air fresheners sensitizes consumers' airways, but this rests on modeled low exposure and low potency, and they note no toxicological reference value exists for respiratory sensitization.
Weak evidence.
Caveats
Model-based with stated assumptions; printed fragrance air levels are in mg/m3 while the diisocyanate benchmark is in ug/m3, so the 'orders of magnitude lower' comparison depends on an apparent unit slip; air oxidation of limonene/linalool not modeled; covers three chemicals only.
Source: ter Burg 2014, Inhalation Toxicology Link to this fact
Health effects / Other evidence
The largest apparent benefits in animal homeopathy trials came from weaker studies: for example a 1966 pig-farrowing trial (odds ratio 151, high risk of bias, remedies donated by a homeopathic pharmacy) and a 1995 mastitis trial (odds ratio 4.9, high risk of bias).
Weak evidence.
Caveats
Across the 9 high-risk trials with data, 5 leaned toward homeopathy and 4 toward placebo; the review did not pool them or test a quality-effect trend, and reliable trials also included a significant positive result.
Source: Mathie & Clausen 2014, Vet Rec, Veterinary Record Link to this fact
Health effects / Other evidence
Expecting a side effect can produce it: reviews cited in a 2015 NEJM Perspective estimate that 4 to 26% of people randomized to placebo in trials stop it because of perceived side effects, so some side effects attributed to drugs may be anticipation.
Weak evidence: Secondhand range from unnamed reviews; the authors say 'not unlikely' rather than showing it. The paper does not address fragrance or chemical exposures.
Source: Kaptchuk & Miller 2015, New England Journal of Medicine Link to this fact
Health effects / Other evidence
In a study of men with benign prostatic hypertrophy on finasteride, those told about the drug's sexual side effects reported them at about three times the rate of those not told.
Weak evidence: Reported secondhand with no citation, sample size or confidence interval in this paper; the primary study was not checked.
Source: Kaptchuk & Miller 2015, New England Journal of Medicine Link to this fact
Health effects / Other evidence
In one migraine study, the same rizatriptan 10 mg pill gave about 50% more pain relief when labeled as rizatriptan than when labeled as placebo, and a placebo labeled rizatriptan matched a drug labeled placebo.
Weak evidence: Single small study from the authors' own group, cited secondhand; the 50% figure is the authors' summary and was not checked against the primary.
Source: Kaptchuk & Miller 2015, New England Journal of Medicine Link to this fact
Health effects / Other evidence
The authors argue that placebo effects are real biological and psychological responses to the care encounter, but modest, and call for more trials of open-label placebos; the argument rests on small pilot trials, many from the authors' own group.
Weak evidence: Position paper by leading placebo-research advocates; no data of its own. Contested by the Cochrane view that general placebo effects are not clinically important.
Source: Kaptchuk & Miller 2015, New England Journal of Medicine Link to this fact
Health effects / Self-reported survey
Studies of whole worker groups have not shown how much occupational asthma is due specifically to quats: none of the epidemiologic studies reviewed used an incident-asthma design, most relied on recalled exposure, and only two measured exposure (air levels all below detection, or IgE).
Weak evidence.
Caveats
Conclusion comes from an HCPA (cleaning-products trade association) funded, non-systematic review; it shows the evidence is insufficient, not that quats are safe. Other evidence base cohorts (e.g., French healthcare workers) do report associations.
Source: LaKind 2019, Allergy, Asthma & Clinical Immunology Industry-funded Link to this fact
Health effects / Other evidence
A 2020 narrative review of the human literature reports that prenatal studies of whether endocrine-disrupting chemicals lower birth weight or shorten pregnancy give mixed results; for BPA, one meta-analysis found no significant link and another review saw differences of only about 100-200 g, possibly not clinically relevant.
Weak evidence: Narrative review, not systematic; no quality appraisal; authors' selection of studies; BPA only is quantified.
Source: Street 2020, International Journal of Molecular Sciences Link to this fact
Health effects / Other evidence
A 2021 review of human studies found phthalate exposure linked, in many but not all studies, to reproductive problems in women and men, including early puberty in girls, preterm birth, lower IVF success, and poorer semen quality.
Weak evidence.
Caveats
Narrative review with no study grading; the evidence is observational association (mostly urinary metabolites), and the paper itself says 'there is no consensus' and lists null and opposite-direction studies.
Source: Mesquita 2021, Molecular Reproduction and Development Link to this fact
Health effects / Other evidence
In one review's own table of 19 sperm-concentration results, 13 reported lower sperm concentration with phthalate exposure, 4 no change, and 2 higher.
Weak evidence: The count is our tally of the review's Table 2 rows (different phthalates, cohorts and metabolites); it is not a pooled analysis and says nothing about size of effect or cause.
Source: Mesquita 2021, Molecular Reproduction and Development Link to this fact
Health effects / Review of other studies
There is no clear evidence that routine aluminum exposure from ingestion or consumer products causes breast cancer or Alzheimer's disease, according to this review.
Weak evidence: One-sentence relay of other reviews; not a systematic assessment and not tested here. The register entry C341 notes some authors argue otherwise.
Source: Novack 2022, Cutis Link to this fact
Health effects / Review of other studies
Dermatologists judge that the allergy seen in a small proportion of children given aluminum-containing vaccines does not outweigh the benefit of vaccination for almost all children.
Weak evidence: Authors' opinion citing FDA and CDC positions; the paper presents no benefit-risk analysis of its own.
Source: Novack 2022, Cutis Link to this fact
Health effects / Measured in people or real products
Cardiovascular death was significantly linked to only one DEHP breakdown product, MEOHP (continuous HR 1.18, 1.04-1.35; top third vs bottom third 1.74, 1.05-2.88); the overall DEHP group was only borderline for cardiovascular death.
Weak evidence: Few cardiovascular deaths per tertile give wide intervals; one metabolite out of several tested, with no correction for multiple comparisons; one cohort.
Source: Trasande, Liu & Bao 2022, Environmental Pollution Link to this fact
Health effects / Other evidence
NYU researchers extrapolated that phthalate exposure might account for about 91,000 to 107,000 deaths a year and $40-47 billion in lost productivity among Americans aged 55-64, assuming the cohort associations are causal.
Weak evidence.
Caveats
A modeled extrapolation from a single observational study, assuming causation; the published incremental-rate figure does not reproduce from the stated inputs; the authors themselves say corroboration is needed.
Source: Trasande, Liu & Bao 2022, Environmental Pollution Link to this fact
Health effects / Other evidence
Healthcare workers' hand-hygiene and surface-disinfection routines are a plausible source of quaternary ammonium skin reactions, and the review says the pandemic increased this exposure, but it presents no pandemic-era incidence data.
Weak evidence: Inference from exposure scale and older case reports; no measured rise in dermatitis is reported.
Source: Okeke 2023, Clinical, Cosmetic and Investigational Dermatology Link to this fact
Health effects / Other evidence
A 2005 commentary by a University of Missouri biologist counted 115 published animal studies of low-dose BPA; 94 reported significant effects, 31 of them at or below the then-accepted safe daily dose of 50 µg/kg/day. These are mostly rodent studies, tallied by the authors themselves, not human evidence.
Weak evidence: Animal studies; vote count without study-quality weighting; the first author is an interested researcher; later regulatory-grade work not covered.
Source: vom Saal & Hughes 2005, Environmental Health Perspectives Link to this fact
Health effects / Other evidence
EWG's 2006 report said its biomonitoring program ('toxome') found 455 industrial chemicals in the blood, urine, or breast milk of more than 100 individuals, including babies in the womb.
Weak evidence: Advocacy-group summary with no sample description, methods or per-chemical data in this document; not population-representative as presented, and detection is not harm.
Source: EWG 2006 Annual Report, EWG annual report (PDF) Link to this fact
Health effects / Other evidence
Maltol and furaneol, common sweet flavor chemicals, damage DNA in laboratory dishes by binding iron or copper and producing hydroxyl radicals; the flavor panel says they did not cause cancer in rodents at doses at least 10,000 times food intake.
Weak evidence: In vitro mechanism plus industry-summarized rodent studies; no human data; says nothing about inhaled or skin exposure to the same chemicals in perfumes.
Source: Smith et al. 2009 (FEMA Expert Panel), GRAS Flavoring Substances 24, Food Technology 63(6), Food Technology Industry-funded Link to this fact
Health effects / Review of other studies
EWG identified 12 chemicals in the tested fragrances with published lab evidence of hormone activity (11 of 12 estrogenic), an average of four per product.
Weak evidence.
Caveats
Evidence is mostly cell-culture and animal studies; the report itself says human data at current exposures are scarce; no dose or exposure assessment; EWG is an advocacy organization.
Source: Sarantis et al. 2010 (Not So Sexy, Canadian edition) Link to this fact
Health effects / Other evidence
EASAC cites an FDA investigation that found severe adverse events, including infant deaths, reported for homeopathic teething products, with the toxicity associated with varying levels of belladonna in the product, and says dilution cannot be assumed to make homeopathic products safe in practice.
Weak evidence.
Caveats
Secondhand via a news article; the Statement gives no numbers, and an FDA report of adverse events does not by itself prove causation. See the FDA recall letter page for the primary document.
Source: EASAC 2017, EASAC Statement (September 2017) Link to this fact
Health effects / Other evidence
A serious consumer complaint of a seizure was on file for Infants' Teething Tablets lot 41116, and FDA found the firm's investigation inadequate.
Weak evidence: A single complaint; the letter does not establish that the tablets caused the seizure.
Source: FDA warning letter 17-NWJ-09, 20 Jun 2017 Link to this fact
Health effects / Other evidence
In a small German EEG study of 28 women tested between attacks, those with migraine with aura (n = 13) showed larger brain responses than those without aura (n = 15) to a nasal irritant (CO2) and to chocolate odor in the left nostril (P3 amplitude P = 0.013 and 0.023), but not the right.
Weak evidence.
Caveats
N = 13 vs 15, no healthy controls, effects on one side only, fixed (not counterbalanced) stimulus order, Novartis-funded, single center; CO2 and chocolate are not fragrance products.
Source: Mignot et al. 2023, J Headache Pain, The Journal of Headache and Pain Industry-funded Link to this fact
Information, not medical advice. See also: myths we won’t tell you.