Every fact on this site, with its evidence grade, its caveats, and a link to the source.
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
In healthy volunteers using an oxymetazoline nasal decongestant spray for 30 days, adding the preservative benzalkonium chloride (BKC) produced significantly greater rebound nasal mucosal swelling (P < 0.01) and evening nasal stuffiness (P < 0.05, from week 2 onward) than the same spray without BKC.
Moderate evidence.
Caveats
N=10/arm, healthy volunteers (not allergic-rhinitis patients), single research group (same authors across all 3 cited trials); the commentary itself says additional research is needed to confirm BKC's effect.
Source: Graf 1999, Clinical Therapeutics Industry-funded Link to this fact
Health effects / Measured in people or real products
In a 3-arm trial, benzalkonium chloride (BKC) alone (no decongestant) produced significantly more nasal mucosal swelling than placebo after 30 days, but did not cause subjective nasal stuffiness; oxymetazoline alone caused pronounced nasal hyperreactivity to histamine challenge and subjective stuffiness, suggesting the two agents contribute distinct, additive components to rhinitis medicamentosa.
Moderate evidence: N=10/arm, healthy volunteers, single research group.
Source: Graf 1999, Clinical Therapeutics Industry-funded Link to this fact
Health effects / Other evidence
In a standard mouse bioassay, methacrolein is primarily an upper-airway (sensory) irritant: it caused a concentration-dependent drop in breathing rate at 2-26 ppm (RD50 = 10.4 ppm) with no detectable pulmonary irritation or desensitization over 30 minutes of exposure.
Moderate evidence: Single small study (n=4 mice per concentration), one lab, acute exposure only (30 min); no chronic or repeated-exposure data.
Source: Larsen & Nielsen 2000, Toxicology Letters Industry-funded Link to this fact
Health effects / Other evidence
Methacrolein activates the sensory-irritant receptor by chemically reacting with it (like acrolein), not by simple physical adsorption, and its carbon-carbon double bond — not its aldehyde group — is the key reactive feature; it is roughly six times less potent than acrolein itself.
Moderate evidence: Mechanistic inference from a potency-ratio comparison across compounds tested in different (cited) studies, not a direct receptor-binding assay.
Source: Larsen & Nielsen 2000, Toxicology Letters Industry-funded Link to this fact
Health effects / Animal study
In a mouse inhalation screening study, all five commercial fabric-softener dryer pads tested (three scented, two labeled unscented) caused significant sensory irritation (21-58% of breaths) and airflow limitation (6-32%) compared to sham exposure, and two of the five brands also caused significant pulmonary irritation (12% and 23%); the two 'unscented'-labeled brands were just as irritating as the scented ones.
Moderate evidence.
Caveats
Single private commercial testing laboratory (Anderson Laboratories, owned by the study's two authors), unreplicated by an independent lab; raw-pad doses (190-750 ppm bulk TVOC) far exceed measured indoor air VOC levels.
Source: Anderson & Anderson 2000, Journal of Toxicology and Environmental Health, Part A Link to this fact
Health effects / Animal study
In the same study, a cotton tee-shirt that had already been dried with a fabric-softener pad — not the pad itself — still emitted enough residual chemical to cause significant sensory irritation in mice (49% of breaths, versus 6% for an unwashed new shirt), at a much lower bulk air concentration (86 ppm) than the raw pad's headspace.
Moderate evidence.
Caveats
Single lab, one brand tested, n=8 exposed mice; airflow limitation (7%, not significant except in 1 of 8 mice) and pulmonary irritation were not significantly changed from the new-shirt control.
Source: Anderson & Anderson 2000, Journal of Toxicology and Environmental Health, Part A Link to this fact
Health effects / Other evidence
Despite reporting more symptoms at higher VOC concentrations, the same subjects showed no significant change in lung function (FEV1, FVC) or in nasal-lavage and induced-sputum cell counts and inflammatory markers (neutrophils, IL-8, LTB4, albumin) at 25 or 50 mg/m3.
Moderate evidence.
Caveats
Small sample size (n=15) limited statistical power: by the authors' own calculation, only 50% power to detect a 50% change in induced-sputum cell counts and 10% power for a 50% change in nasal-lavage cell counts, so an inflammatory effect could have been missed.
Source: Pappas et al. 2000, Int J Occup Environ Health 6:1-8 Link to this fact
Health effects / Other evidence
This chamber study directly conflicts with an earlier study (Koren et al. 1992) on the same measurement: Koren found more neutrophils in nasal-lavage fluid after a similar VOC mixture at 25 mg/m3, while this study found no nasal-lavage neutrophil increase at 25 or 50 mg/m3 in the full sample.
Moderate evidence.
Caveats
Neither study's explanation for the disagreement (lavage technique vs. repeated-lavage artifact) has been independently confirmed; treat the nasal-lavage neutrophil question as unresolved between the two studies.
Source: Pappas et al. 2000, Int J Occup Environ Health 6:1-8 Link to this fact
Health effects / Lab study (cells or chemistry)
In a lab test, four common fragrance/flavor chemicals found in essential oils – citral, geraniol, nerol, and trans-anethole – weakly activated the human estrogen receptor, and eugenol weakly blocked it, but all needed concentrations 100,000 to 1,000,000 times higher than the body's own estrogen to show any effect.
Moderate evidence: Single-lab yeast reporter assay, confirmed by direct receptor-binding but not by a second cell-based assay (Ishikawa cells, negative for all compounds).
Source: Howes et al. 2002, Journal of Pharmacy and Pharmacology Link to this fact
Health effects / Animal study
When the two most estrogen-receptor-active compounds, citral and geraniol, were applied to the skin of mice at doses about 6,000 times the effective estrogen dose, neither produced any of the classic signs of estrogen exposure (uterus growth or increased uterine blood-vessel leakiness).
Moderate evidence: N=5-6 mice per group, one species, one route (transdermal), short exposure (single dose or 3 days).
Source: Howes et al. 2002, Journal of Pharmacy and Pharmacology Link to this fact
Health effects / Lab study (cells or chemistry)
Linalool and eucalyptol (1,8-cineole) – two of the main scent chemicals in lavender and tea tree oil – showed no estrogen-receptor activity at all in this yeast test, alongside 14 other common essential-oil constituents that were tested and found inactive.
Moderate evidence.
Caveats
Single assay type (yeast ERE reporter); does not test androgen-receptor activity or steroid-synthesis enzymes, which other studies show linalool and eucalyptol can affect differently.
Source: Howes et al. 2002, Journal of Pharmacy and Pharmacology Link to this fact
Health effects / Review of other studies
Three large NEI-sponsored trials (OHTS, EMGT, BISED) each found an increased incidence of cataract or nuclear lens opacity in patients on long-term topical antiglaucoma medication compared with untreated or lower-IOP patients, but none was designed to identify a cause.
Moderate evidence.
Caveats
Association only, from observational/trial follow-up, not causation; the OHTS cataract-extraction difference (6.4% vs 4.3%) was not statistically significant (P=.06), and BISED's IOP-nuclear-opacity association was borderline (P=.07) with no dose-response relationship.
Source: Brandt 2003 (editorial), Archives of Ophthalmology Industry-funded Link to this fact
Health effects / Self-reported survey
In a 2002-03 US national telephone survey of 1,054 randomly selected adults, 11.2% self-identified as unusually sensitive to everyday chemicals such as cleaning products, perfume, and detergents, and 2.5% said they had been medically diagnosed with multiple chemical sensitivity (MCS).
Moderate evidence.
Caveats
Single national random-digit-dial telephone survey; self-report with no clinical verification of hypersensitivity or diagnosis; the earliest in a cross-country series of self-report chemical-sensitivity prevalence figures.
Source: Caress & Steinemann 2003, Archives of Environmental Health: An International Journal Link to this fact
Health effects / Self-reported survey
In the same national survey, almost 1 in 3 adults (31.1%) said being next to someone wearing a scented product was irritating, and about 1 in 6 (17.6%) said air fresheners gave them headaches, breathing difficulties, or other health problems.
Moderate evidence: Self-report, single national telephone survey; 46.3% found scented products on others appealing rather than irritating.
Source: Caress & Steinemann 2003, Archives of Environmental Health: An International Journal Link to this fact
Health effects / Self-reported survey
Among the 11.2% of US adults who self-identified as chemically hypersensitive, about 2 in 3 (66.7%) described their symptoms as severe or moderately severe, and about 1 in 7 (13.2%) said they had lost a job because of it; 72.7% had changed their household or personal-care products.
Moderate evidence: Self-report severity and lifestyle-effect items in a subgroup of n=117-122; no clinical verification.
Source: Caress & Steinemann 2003, Archives of Environmental Health: An International Journal Link to this fact
Health effects / Self-reported survey
In this national survey, women were reported as chemically hypersensitive at roughly 4 times the rate of men (about 15.9% of women vs 4.0% of men, wiki arithmetic from the paper's cross-tabulation), and prevalence rose steadily with age from about 7% under age 36 to about 15% over age 50.
Moderate evidence.
Caveats
Sex- and age-specific prevalence rates are evidence base arithmetic from the paper's Table 4 cross-tabulation, not stated directly by the authors; single national self-report survey.
Source: Caress & Steinemann 2003, Archives of Environmental Health: An International Journal Link to this fact
Health effects / Self-reported survey
In the ALSPAC cohort, mothers who used air fresheners daily or on most days during pregnancy more often reported depression (adjusted OR 1.19) and headache or migraine (adjusted OR 1.24) 8 months after giving birth; aerosol use was similarly associated with headache (adjusted OR 1.25) but not depression.
Moderate evidence.
Caveats
Self-report exposure and outcome; depression measured with a validated scale (Edinburgh Postnatal Depression Scale), headache was not. Single cohort, not independently replicated for this outcome.
Source: Farrow 2003, Archives of Environmental Health: An International Journal Link to this fact
Health effects / Self-reported survey
In a 1999-2000 telephone survey of 1,582 randomly selected adults in the Atlanta, Georgia metro area, 12.6% self-identified as unusually sensitive to everyday chemicals, and 3.1% reported a medical diagnosis of environmental illness or multiple chemical sensitivity (MCS).
Moderate evidence: Single regional random-digit-dial telephone survey; self-report with no clinical verification of hypersensitivity or diagnosis.
Source: Caress & Steinemann 2003 (EHP), Environmental Health Perspectives Link to this fact
Health effects / Self-reported survey
Among Atlanta-area adults who reported chemical hypersensitivity, about 1 in 7 (13.5%) said they had lost a job because of it (1.8% of the entire sample), and about 3 in 10 (29.9%) said it made shopping in stores difficult.
Moderate evidence: Self-report, single regional survey; asked only of the 199 respondents who reported hypersensitivity.
Source: Caress & Steinemann 2003 (EHP), Environmental Health Perspectives Link to this fact
Health effects / Self-reported survey
In this Atlanta sample, women were reported as chemically hypersensitive at roughly 1.75 times the rate of men (about 15.7% of women vs 8.95% of men, wiki arithmetic from the paper's cross-tabulation), a smaller sex skew than the same authors' national survey found (F:M about 4.0).
Moderate evidence: Sex-specific prevalence rates are evidence base arithmetic from the paper's Table 4 cross-tabulation, not stated directly by the authors; single regional self-report survey.
Source: Caress & Steinemann 2003 (EHP), Environmental Health Perspectives Link to this fact
Health effects / Self-reported survey
In the same follow-up, the most common symptoms reported after a chemical exposure were headache (88.4%) and burning eyes (76.8%), followed by asthma-like breathing difficulty (59.4%) and nausea or stomach distress (55.1%).
Moderate evidence.
Caveats
Small, non-random follow-up sample (n=69); self-reported symptoms with no clinical verification; the symptom-question cluster had the lowest internal-consistency score (Cronbach's alpha 0.5054) of the three question clusters tested.
Source: Caress & Steinemann 2003 (EHP), Environmental Health Perspectives Link to this fact
Health effects / Self-reported survey
In a 10-country European cohort of 3,503 adults who did the cleaning at home and were asthma-free at baseline, using household cleaning sprays at least once a week was associated with a 49% higher rate of new asthma symptoms or asthma-medication use over about 9 years of follow-up.
Moderate evidence: Single multinational cohort (not yet independently replicated by a different research group); exposure and symptom outcomes were both self-reported at one follow-up interview.
Source: Zock et al. 2007, American Journal of Respiratory and Critical Care Medicine Link to this fact
Health effects / Self-reported survey
Using household cleaning sprays at least 4 days a week was linked to roughly double the rate of new physician-diagnosed asthma (HR 2.11, 95% CI 1.15-3.89) compared with light or no spray use, with a clear dose-response across frequency categories (P for trend 0.041) and number of different sprays used (P for trend 0.022).
Moderate evidence: Physician-diagnosed asthma relied on participant report of a doctor's diagnosis, not independent verification; single cohort.
Source: Zock et al. 2007, American Journal of Respiratory and Critical Care Medicine Link to this fact
Health effects / Self-reported survey
Cleaning products that were not applied as a spray (liquid multiuse cleaners, washing powders, polishes, bleach, ammonia, decalcifiers) showed no significant association with new-onset asthma in this study, while glass-cleaning, furniture, and air-refreshing sprays did.
Moderate evidence.
Caveats
Correlation between spray use and other product use was generally weak but not zero (Spearman's r up to 0.41 for some product pairs), so residual confounding by broader cleaning intensity cannot be fully excluded.
Source: Zock et al. 2007, American Journal of Respiratory and Critical Care Medicine Link to this fact
Health effects / Self-reported survey
The household-spray-asthma association was not modified by atopy (allergic sensitization): risk was similar in atopic (RR 1.30) and non-atopic (RR 1.33) participants, supporting an irritant rather than allergic mechanism.
Moderate evidence: Single cohort; the study could not directly test cellular/immunological mechanism, only the atopy-stratified association.
Source: Zock et al. 2007, American Journal of Respiratory and Critical Care Medicine Link to this fact
Health effects / Measured in people or real products
In a PET brain-imaging study of 12 women with multiple chemical sensitivity (MCS) and 12 matched controls, MCS subjects activated the brain's core odor-processing regions (amygdala, piriform, and insular cortex) LESS than controls when smelling most odorants, not more, despite resting brain activity being normal in both groups.
Moderate evidence: N=12/group, women only, single center (Karolinska), no independent replication of this exact PET paradigm in our evidence base yet.
Source: Hillert et al. 2007, Human Brain Mapping Link to this fact
Health effects / Measured in people or real products
The same MCS subjects showed extra activation of the anterior cingulate cortex and cuneus-precuneus during odor smelling that was not seen in controls, a pattern the authors call 'highly speculative' evidence for top-down (brain-level) regulation of odor response rather than a sensitized peripheral or olfactory system.
Moderate evidence: The authors themselves label the top-down/harm-avoidance explanation highly speculative; it is one candidate account among several, not established mechanism.
Source: Hillert et al. 2007, Human Brain Mapping Link to this fact
Health effects / Measured in people or real products
MCS subjects showed real, measurable physical changes during odor exposure that controls did not: their heart rate rose (RR interval fell, P<0.0001) and their breathing pattern changed differently from controls specifically while smelling odorants (not odorless air), a difference that held up even after statistically controlling for breathing pattern itself.
Moderate evidence: N=12/group; the finding shows odor-locked autonomic distress, not that the distress is caused by a specific chemical mechanism.
Source: Hillert et al. 2007, Human Brain Mapping Link to this fact
Health effects / Measured in people or real products
Odor detection thresholds (n-butanol test) did not differ between MCS subjects and controls (P=0.8), and nasal anatomy was normal in both groups, so normal ability to smell coexisted with abnormal brain and autonomic responses to odors.
Moderate evidence: Tests only the olfactory (CN I) channel; says nothing about the trigeminal/chemesthetic channel some MCS mechanism hypotheses focus on.
Source: Hillert et al. 2007, Human Brain Mapping Link to this fact
Health effects / Measured in people or real products
In a double-blind chamber study of 130 healthy women, those who scored high on a validated chemical-odor-intolerance questionnaire did not have worse symptoms, stress-hormone (cortisol) levels, or cognitive/driving-simulation performance when exposed to an indoor VOC or VOC-plus-ozone mixture, compared to women who scored low on the same questionnaire.
Moderate evidence.
Caveats
Single lab, subsample of 130 (40 high-CI, 46 low-CI, tertile split); healthy, non-smoking women only; people with clinically significant MCS/illness behavior were excluded, and the authors could not recruit volunteers at the extremes of chemical intolerance; one 140-minute exposure, not repeated or chronic exposure.
Source: Fiedler 2008, Psychosomatic Medicine Link to this fact
Health effects / Measured in people or real products
Women prone to negative affect (anxiety, worry, distress) became more anxious after a public-speaking stressor during an indoor-chemical exposure session, regardless of what was in the air (clean air, VOCs, or VOCs plus ozone) – but they did not show higher cortisol or more physical (non-anxiety) symptoms than women low in negative affect.
Moderate evidence.
Caveats
Tertile-split subsample (45 high-NA, 48 low-NA) of 130 healthy women; excludes people with diagnosed psychiatric conditions; a public-speaking task is a milder stressor than daily-life stress.
Source: Fiedler 2008, Psychosomatic Medicine Link to this fact
Health effects / Measured in people or real products
People who scored high in chemical intolerance or negative affect rated the same chemical exposure and their own symptoms as more annoying and more distressing than other participants did, even though their measured symptom counts, cortisol, and driving/attention performance were no different.
Moderate evidence: Annoyance/distress were retrospective, subjective 5-point ratings, not the same structured symptom-severity measure used for the main null; single study, healthy women only.
Source: Fiedler 2008, Psychosomatic Medicine Link to this fact
Health effects / Other evidence
In a large Canadian population survey, more frequent use of personal care products was associated with small but statistically significant reductions in objectively measured lung function.
Moderate evidence: Single cross-sectional study; exposure was self-reported, not measured; cannot establish causation.
Source: Dales 2013, Annals of Epidemiology Link to this fact
Health effects / Other evidence
In both men and women, more frequent use of hairstyle products (hairspray, gel, mousse) was associated with an approximate 2% decrease in FEV1, FVC, and the FEV1/FVC ratio.
Moderate evidence: Self-reported frequency of use, not measured exposure; no specific chemical or product identified as responsible.
Source: Dales 2013, Annals of Epidemiology Link to this fact
Health effects / Other evidence
Restricting the analysis to never-smokers did not remove the association between personal care product use and lower lung function in women.
Moderate evidence: Subgroup analysis of a single cross-sectional cohort; wide confidence intervals.
Source: Dales 2013, Annals of Epidemiology Link to this fact
Health effects / Self-reported survey
Among 543 healthcare workers in 7 French hospitals/clinics, occupational exposure to quaternary ammonium compounds (QACs, e.g. benzalkonium chloride) was linked to about 7.5x higher odds of reported physician-diagnosed asthma (adjusted OR 7.56, 95% CI 1.84-31.05) and about 3.2x higher odds of nasal symptoms at work (adjusted OR 3.21, 95% CI 1.42-7.22), after adjustment for age, gender, smoking, BMI, and atopy.
Moderate evidence.
Caveats
Single cross-sectional cohort (not replicated by an independent group); the asthma outcome relies on self-report plus doctor confirmation, not spirometric testing; the confidence interval's lower bound (1.84) shows a wide range of plausible effect sizes; healthy-worker survivor bias could have caused an underestimate, especially among cleaners.
Source: Gonzalez 2014, Clinical & Experimental Allergy Link to this fact
Health effects / Self-reported survey
In the same study, chlorine/bleach exposure (adjusted OR 1.01, 95% CI 0.47-2.18), latex glove use (adjusted OR 0.69, 95% CI 0.32-1.51), and glutaraldehyde exposure (crude OR 1.31, 95% CI 0.53-3.23) showed no significant association with reported asthma; QAC exposure was the only occupational chemical exposure that remained statistically significant after adjustment.
Moderate evidence.
Caveats
A null result in one cross-sectional study does not prove these exposures are safe; some comparisons (e.g. glutaraldehyde, chlorine/bleach) had small exposed groups and wide confidence intervals, so a true effect could have been missed for lack of statistical power.
Source: Gonzalez 2014, Clinical & Experimental Allergy Link to this fact
Health effects / Self-reported survey
The single highest-risk activity for asthma and nasal symptoms was manual dilution of concentrated disinfectant products (adjusted OR 4.01, 95% CI 1.34-12.00 for asthma; 1.90, 95% CI 1.13-3.18 for nasal symptoms), higher than general disinfection or cleaning tasks alone.
Moderate evidence.
Caveats
Task-level self-report from a single cross-sectional study; the authors interpret this as likely reflecting dilution errors that create exposure peaks to concentrated, strongly irritant product, not a tested causal mechanism.
Source: Gonzalez 2014, Clinical & Experimental Allergy Link to this fact
Health effects / Measured in people or real products
Only 6.5% of tested healthcare workers (32 of 494) had detectable specific IgE antibodies to quaternary ammonium compounds, and this was not significantly associated with reported asthma, new-onset asthma, or nasal symptoms — despite QAC exposure itself being strongly associated with those same outcomes. The study authors read this pattern as pointing to an irritant, not an IgE-mediated allergic, mechanism.
Moderate evidence.
Caveats
Absence of an IgE signal does not rule out a non-IgE immunological mechanism; specific IgE was only tested against a general quaternary ammonium radical, not against every specific QAC compound (e.g. benzalkonium chloride vs. didecyldimethylammonium chloride) separately.
Source: Gonzalez 2014, Clinical & Experimental Allergy Link to this fact
Health effects / Self-reported survey
Reported physician-diagnosed asthma prevalence varied sharply by healthcare occupation: 16.1% among nurses, 16.7% among auxiliary nurses, 7.7% among cleaners, 5.1% among other healthcare workers, and 3.4% among administrative staff (overall 11.2%, p=0.003).
Moderate evidence.
Caveats
Descriptive prevalence from a single cross-sectional cohort; differences between nurses and cleaners may partly reflect a healthy-worker survivor effect (cleaners with severe asthma more likely to have left the job) rather than only differences in true exposure or risk.
Source: Gonzalez 2014, Clinical & Experimental Allergy Link to this fact
Health effects / Review of other studies
Benzalkonium chloride (BAK), the preservative in most glaucoma eye drops, accumulates in the trabecular meshwork and other deep ocular tissues after chronic topical use and has a tissue half-life of about 20 hours in the corneal/conjunctival epithelium.
Moderate evidence: Accumulation is well documented; a causal link between that accumulation and disease progression is not established by these detection studies alone.
Source: Rasmussen et al. 2014, Journal of Ocular Pharmacology and Therapeutics Industry-funded Link to this fact
Health effects / Other evidence
BAK-exposed trabecular meshwork tissue and cells share the same qualitative cellular and molecular damage signature — cell loss, apoptosis, increased reactive oxygen species, extracellular matrix accumulation, decreased outflow, fibronectin upregulation, and cellular senescence — seen in the trabecular meshwork of primary open-angle glaucoma patients.
Moderate evidence.
Caveats
This shared signature is correlational. The review states explicitly that the causal hypothesis (BAK worsens glaucoma, rather than merely resembling it) is unproven and was, at the time of writing, only beginning to be tested in a nonhuman-primate model. See C324.
Source: Rasmussen et al. 2014, Journal of Ocular Pharmacology and Therapeutics Industry-funded Link to this fact
Health effects / Measured in people or real products
Burning candles indoors measurably shifted a marker of heart-rhythm regulation (heart rate variability) toward a more 'rest and digest' pattern within about 30-90 minutes, in a chamber study using candle-smoke levels similar to real dinner-table use.
Moderate evidence: Single small (n=22) study, one lab; heart rate itself did not change; the effect (HF-HRV +22.3%, p=0.01 vs. clean air) is not yet independently replicated.
Source: Hagerman et al. 2014, Atmospheric Environment Link to this fact
Health effects / Other evidence
In a controlled bronchoprovocation study of 28 adults with mild, stable asthma, doubling doses of inhaled benzalkonium chloride (BAC) produced bronchospasm (at least 20% FEV1 fall) in 17 of 28 subjects (61%, mean FEV1 decrease 28%), with a cumulative threshold dose of about 300 micrograms (1.00 micromole); pretreatment with the mast-cell stabilizer cromolyn, but not ipratropium, blocked the response.
Moderate evidence: Single study, small sample (28 subjects), adults with mild/stable (not severe) asthma; not a systematic review or meta-analysis.
Source: Prabhakaran 2017 Link to this fact
Health effects / Other evidence
In a double-blind randomized crossover study of 18 young adults with stable asthma, repeated 600-microgram doses of benzalkonium chloride (up to 4 doses, 20 minutes apart) produced a mean 17% fall in FEV1, versus no significant difference between EDTA and placebo; 6 of 18 subjects had an FEV1 fall of more than 20%.
Moderate evidence: Small sample (18 subjects), single study; EDTA and BAC are both nebulizer-solution additives but were tested for different purposes.
Source: Prabhakaran 2017 Link to this fact
Health effects / Other evidence
A randomized double-blind study of a single, clinically realistic dose of albuterol with benzalkonium chloride (100 micrograms BAC plus 2.5 mg albuterol) found no decrease in bronchodilator response compared with preservative-free albuterol in 22 adults with mild stable asthma.
Moderate evidence.
Caveats
Single low dose, well below the approximately 300-microgram cumulative bronchospasm threshold, and mild (not severe) asthma; does not generalize to continuous, hours-long nebulization at higher cumulative doses.
Source: Prabhakaran 2017 Link to this fact
Health effects / Self-reported survey
In a cohort of 4,102 US nurses already diagnosed with asthma (Nurses' Health Study II), weekly use of disinfectants to clean medical instruments was associated with poorly controlled asthma (OR 1.37, 95% CI 1.05-1.79) and very poorly controlled asthma (OR 1.88, 95% CI 1.38-2.56) versus no weekly instrument disinfection, after adjusting for age, smoking, BMI, race and ethnicity.
Moderate evidence.
Caveats
Cross-sectional, self-reported exposure and outcome in a single cohort; weekly surface-only disinfection (as opposed to instruments) showed no association, so the effect is task-specific, not a generic 'disinfectant use' signal.
Source: Dumas 2017, European Respiratory Journal Link to this fact
Health effects / Self-reported survey
Weekly use of disinfectant/cleaning sprays at work was associated with worse asthma control in US nurses (OR 1.50, 95% CI 1.16-1.94 for poorly controlled; OR 1.38, 95% CI 1.00-1.92 for very poorly controlled), and the association held even among nurses who used sprays less than once a week.
Moderate evidence.
Caveats
Self-reported spray use; the paper attributes the effect to higher inhalation potential from spray application rather than to a specific chemical, since sprays covered multiple product types (surface, patient care, air-refreshing).
Source: Dumas 2017, European Respiratory Journal Link to this fact
Health effects / Self-reported survey
Using an expert job-task-exposure matrix, high occupational exposure to formaldehyde (OR 1.33), glutaraldehyde (OR 1.18), hypochlorite bleach (OR 1.18), hydrogen peroxide (OR 1.19) and enzymatic cleaners (OR 1.33, the only one significant at P<0.001) was each associated with poorer asthma control in US nurses, versus low exposure to the same chemical.
Moderate evidence.
Caveats
Exposure levels come from a job-task-exposure matrix built from a separate reference sample's questionnaire responses, not direct measurement; 95% confidence intervals for these specific odds ratios were shown only as graphical whiskers in the source figure, not printed, so are not reported as exact numbers here.
Source: Dumas 2017, European Respiratory Journal Link to this fact
Health effects / Self-reported survey
US nurses exposed to a combination of aldehydes (formaldehyde or glutaraldehyde) plus hypochlorite bleach/hydrogen peroxide plus enzymatic cleaners had the strongest asthma-control signal in the study: odds ratio 1.91 (95% CI 1.31-2.79) for very poorly controlled asthma versus nurses with low exposure to all seven disinfectant categories studied (P-trend=0.001).
Moderate evidence.
Caveats
Almost no nurses were exposed to only a single product, so this combination effect could reflect any one chemical in the mix, an interaction between them, or a marker for a more hazardous job/task profile overall; the paper cannot isolate independent single-chemical causation.
Source: Dumas 2017, European Respiratory Journal Link to this fact
Health effects / Self-reported survey
Occupational exposure to quaternary ammonium compounds ('quats') and alcohol-based disinfectants showed no significant association with asthma control in a cohort of 4,102 US nurses (quats: OR 1.10, n=1,548 exposed, P-trend=0.14; alcohol: OR 1.11, n=1,591 exposed, P-trend=0.12), and the null for quats held in two further models adjusting for exposure to the other six disinfectant products.
Moderate evidence.
Caveats
The study's own authors note quaternary ammonium compound exposure is particularly hard to evaluate by questionnaire and may be under-estimated even with brand-name/safety-data-sheet review; a French healthcare-worker cohort (Gonzalez 2014) found a strong association between QAC exposure and physician-diagnosed asthma, so this null does not generalize to all QAC-asthma questions.
Source: Dumas 2017, European Respiratory Journal Link to this fact
Health effects / Review of other studies
A review of controlled chamber studies (carbon dioxide, ammonia, formaldehyde, sulfur dioxide, glycol/glycol-ether mixtures) found no consistent evidence that people with allergic rhinitis or mild asthma are more sensitive to sensory-irritant chemicals than non-allergic people.
Moderate evidence.
Caveats
The underlying chamber studies excluded moderate and severe asthma/allergy by design and tested mostly non-fragrance irritants; one exception was found (allergic subjects more sensitive to n-propanol specifically), and the review is narrative, not a systematic review with a defined search protocol.
Source: Nielsen & Wolkoff 2017, Regulatory Toxicology and Pharmacology Industry-funded Link to this fact
Health effects / Review of other studies
The same authors' own worked example for ammonia — comparing a mouse near-zero-effect threshold (42 ppm) to the human no-effect level, then adding a further factor to account for ordinary variation between people — lands on a combined safety margin of about 4 to 5, not the smaller factor of about 2 that a later 2024 review attributes to this paper for protecting sensitive groups.
Moderate evidence.
Caveats
Single worked example for one chemical (ammonia); the 'about 2' figure the authors separately cite is explicitly for age/sex/lifestyle/disease combined in a general population, not derived fresh in this paper, and not tested against moderate or severe asthma/allergy.
Source: Nielsen & Wolkoff 2017, Regulatory Toxicology and Pharmacology Industry-funded Link to this fact
Health effects / Self-reported survey
The study found no association between childhood or adult hair product or hair oil use and mammographic breast density, a separate, strong breast-cancer risk factor, suggesting that if hair-product chemicals affect breast cancer risk, it may be through pubertal timing rather than breast density.
Moderate evidence: Small sample (N=248) limits power to detect a true but modest effect on breast density; one adult product subgroup (hair dye) did show a small positive association with density.
Source: McDonald 2018, Environmental Health Link to this fact
Health effects / Animal study
In rats exposed by nose-only inhalation to concentrated benzalkonium chloride (BKC) aerosols (37.6-52.8 mg/m3 for 4-6 hours over 3 days), animals developed increased relative lung weight and elevated total protein, LDH, and IgE in bronchoalveolar lavage fluid, indicating severe lung damage; the highest dose killed 2 of 5 animals.
Moderate evidence.
Caveats
Estimated daily doses (~10.6-11.3 mg/kg/day) were roughly 10,000-fold above the estimated human clinical inhaled dose from approved pharmaceutical inhalers; not evidence about typical disinfectant-spray exposure levels, which were not measured or modeled in this review.
Source: Johnson 2018, Journal of Aerosol Medicine and Pulmonary Drug Delivery Link to this fact
Health effects / Animal study
In mice, 30-minute inhalation exposure to benzalkonium chloride aerosol up to 19 mg/m3 produced a concentration-dependent decrease in tidal volume, an increase in respiratory rate, and neutrophilic lung inflammation; even the lowest tested dose (0.049 mg/m3) significantly decreased tidal volume.
Moderate evidence.
Caveats
No no-effect level was identified below 0.049 mg/m3 for the tidal-volume endpoint in this review; single study, dose range not directly compared to occupational or consumer spray exposure levels.
Source: Johnson 2018, Journal of Aerosol Medicine and Pulmonary Drug Delivery Link to this fact
Health effects / Other evidence
Nebulized benzalkonium chloride has caused bronchoconstriction in sensitive severe asthmatics after prolonged (15-30 minute), repeated exposure to cumulative doses of about 300 micrograms; the affected multi-dose nebulizer formulations have since been replaced with BKC-free unit-dose vials.
Moderate evidence.
Caveats
Reported as a narrative-review synthesis of prior clinical reports, not this paper's own new data; doses (~300 micrograms cumulative) are roughly 200-fold higher than the Respimat inhaler's per-actuation BKC dose (1.1 micrograms), which has not shown this effect.
Source: Johnson 2018, Journal of Aerosol Medicine and Pulmonary Drug Delivery Link to this fact
Health effects / Measured in people or real products
In a population-representative sample of about 2,900 Canadians (Canadian Health Measures Survey, 2012-2013), a 100% increase in measured residential indoor-air limonene was associated with a 17-18% relative increase in the adjusted odds of both wheezing (OR 1.17, 95% CI 1.16-1.18) and doctor-diagnosed asthma (OR 1.18, 95% CI 1.17-1.19), plus a 1.55% (95% CI 1.28-1.89) increase in exhaled nitric oxide (FeNO), an objective marker of airway inflammation.
Moderate evidence.
Caveats
Single cross-sectional study; cannot establish temporal sequence or fully rule out reverse causation, though the authors argue reverse causation would plausibly run the opposite way (airway-hypersensitive people avoiding scented products). Lung function (FEV1, FVC) showed no significant association with limonene.
Source: Dales & Cakmak 2019, Environmental Pollution Industry-funded Link to this fact
Health effects / Lab study (cells or chemistry)
In a laboratory study, benzalkonium chloride (BAK), the most common eye-drop preservative, killed 50% of human corneal epithelial cells within about 1 to 5 minutes of exposure at concentrations (10-30 ug/mL) that fall within BAK's own reported clinical eye-drop concentration range (10-200 ug/mL).
Moderate evidence.
Caveats
Single-lab, 2D immortalized human corneal epithelial cell line (HCE-T), short static exposure (1-30 min) with no tear-fluid dilution or blink clearance modeled; CDT50 values are exact printed table values, not chart-read.
Source: Kabashima 2020, Journal of Ocular Pharmacology and Therapeutics Industry-funded Link to this fact
Health effects / Lab study (cells or chemistry)
BAK exposure disrupted and reduced the tight-junction protein occludin and delayed recovery of transepithelial electrical resistance (a barrier-function measure) in a dose-dependent manner in human corneal epithelial cells, while a preservative-free trometamol/boric-acid/EDTA mixture (TBE) did not.
Moderate evidence: Single-lab, in vitro monolayer model; occludin densitometry and TEER recovery magnitude are partly chart-read (approx., read from Fig. 3 and Fig. 4F); no independent replication.
Source: Kabashima 2020, Journal of Ocular Pharmacology and Therapeutics Industry-funded Link to this fact
Information, not medical advice. See also: myths we won’t tell you.