This is the entry on the register where the gap between what is known and what is knowable is widest. The chemistry is not in dispute. What is genuinely hard is that you cannot read most of it off a label, and in the United Kingdom that has just become true in two different ways depending on which part of the UK you are standing in.
1. Why the label cannot tell you — and the split that opened this summer
Under cosmetics law, a perfume can be declared as a single word: parfum, or fragrance. Behind that word sits anything from a handful of compounds to two hundred, and the formula is a trade secret. The only compounds that must be named separately are those on a list of recognised contact allergens.
That list just changed in the EU and did not change here.
| Named fragrance allergens | Status | |
|---|---|---|
| EU, and Northern Ireland | About 82, including several essential oils named as whole substances | Regulation (EU) 2023/1545. New products placed on the market have had to comply since 31 July 2026; existing stock may be sold through until 31 July 2028 |
| Great Britain — England, Scotland, Wales | 26, the original list | Has not adopted the expansion. The UK Cosmetics Regulation still runs the pre-2023 list, with any change a matter for OPSS. Checked August 2026 — this is a live policy area and could move |
What that means standing in a shop
- The same product can carry a different ingredients list in Belfast and in Birmingham, and the Belfast one tells you more. This is a live internal UK divergence, not a Brexit talking point — Northern Ireland follows EU cosmetics law and Great Britain does not.
- A short allergen list on a GB label does not mean a simple fragrance. It means the product was assessed against a list of 26. Fifty-odd substances that an EU label would now name can be present and unnamed.
- The threshold matters too. Even on the longer list, declaration is only required above 0.001% in leave-on products and 0.01% in rinse-off ones. Below that, nothing appears.
- And none of this covers your cleaning cupboard. Detergents, surface sprays, fabric conditioners and air fresheners are not cosmetics — they sit under detergents and chemicals law, where the labelling duty is narrower still. The cupboard with the most fragrance in it is the one with the least labelling.
The deeper problem, which is not really about labels
Everything above is about disclosure. There is a second issue underneath it, and it is the one that makes fragrance different from the rest of a product.
- A fragrance is a mixture, and it is regulated largely as a mixture. The finished product must be safe and the responsible person must hold a safety assessment — but the individual aroma chemicals inside “parfum” are not each subject to the public, substance-by-substance evaluation that a preservative or a UV filter goes through, with a published opinion and a permitted concentration.
- Much of the safety work sits with the industry's own bodies — the research institute and the association that set the standards fragrance houses follow. That work is real and technically serious. It is also self-governance, and the outputs are not equivalent to an independent regulator's published opinion.
- So a new aroma chemical can enter thousands of products without appearing anywhere a consumer, or a doctor, can see it. That is the actual gap, and it is why the labelling expansion in the EU matters more than it first appears: naming a substance is the precondition for anyone studying it, patch testing for it, or avoiding it.
- What is being discovered is mostly allergy, not toxicity — and that is worth saying clearly, because the two get conflated. Fragrance is among the commonest causes of contact allergy in Europe, the rate has risen with exposure, and the response has been regulatory: substances restricted, banned, or moved onto the naming list as the evidence accumulated. The system is slow and reactive rather than absent.
- The systemic claims are a different matter. Endocrine effects, neurological effects and cancer risk attributed to fragrance at domestic exposure rest largely on mechanism, on animal work, or on a small number of specific compounds that have themselves been restricted. Emerging for those, against Strong for sensitisation.
The reasonable conclusion is not alarm and not dismissal. It is that this is the one ingredient class where you cannot find out what you are exposed to, which is a sound reason to reduce the number of fragranced products in a house without needing to believe anything specific about any of them. That argument holds whatever the toxicology turns out to say.
2. What is actually in the air when you clean
| Class | Where from | Why it matters |
|---|---|---|
| Terpenes — limonene, pinene, linalool | Citrus and pine scent, in almost everything. Present in “natural” and plant-derived products at higher concentrations, not lower | Mild irritants themselves, and the precursor to the reaction in section 3 |
| Quaternary ammonium compounds — “quats” | Antibacterial sprays and wipes, disinfectants | Established respiratory sensitisers and among the better-evidenced causes of occupational asthma in cleaning work. Use rose sharply after 2020 and has not fallen back |
| Chlorine bleach — sodium hypochlorite | Toilet and surface cleaners | Airway irritant at ordinary use concentrations. Section 6 covers the mixtures |
| Glycol ethers, alcohols, aldehydes | Solvents in multi-surface and glass cleaners | Irritant; formaldehyde specifically is a recognised carcinogen and appears both as an ingredient and as a reaction product |
| Ammonia | Glass and oven cleaners | Strong irritant. Never with bleach |
| Isothiazolinones — MI, MCI, BIT | Preservatives in liquid detergents, wipes, paints, shampoos | A leading cause of contact allergy in the UK, and one of the commonest positive results on patch testing. Restricted in leave-on cosmetics after an epidemic of dermatitis; still widespread in household products |
3. The reaction that makes a spray worse than its ingredients
- This is the single best argument for opening a window, and for opening it afterwards rather than only while you work.
- It is worse on high-ozone days — hot, still, sunny weather, which is also when people are least likely to have windows open if they are avoiding pollen or heat.
- Sprays are worse than wipes and cloths, because an aerosol puts the product into the air by design. Trigger sprays are worse than pump bottles; both are worse than a cloth.
- “Natural” and citrus-scented products are not exempt — they are often the highest terpene emitters on the shelf. Plant-derived is not the same as low-emission.
4. What the outcome evidence actually shows
The most-quoted study on this is worth reporting properly, because it is usually reported badly.
Svanes and colleagues, 2018 — the "20 cigarettes a day" study
- What it was: 6,230 adults across 22 European centres in the ECRHS cohort, lung function measured three times over more than 20 years. Published in the American Journal of Respiratory and Critical Care Medicine.
- What it found, with the numbers stated properly: in women, FEV1 declined by 22.1 mL per year in those responsible for cleaning at home and 22.4 in occupational cleaners, against 18.5 in women who did not clean. So the excess attributable to cleaning is about 3.6 and 3.9 mL per year — not 22. FVC excesses were about 4.3 and 7.1. An earlier version of this page reported the total decline as though it were the excess, which overstated the effect roughly sixfold.
- And the famous comparison is to pack-years, not to a daily habit. The authors described the effect in occupational cleaners as somewhat less than 20 pack-years of smoking — a cumulative lifetime exposure — which is a different and considerably smaller claim than "twenty a day", the form in which it is almost always repeated. It remains their comparison rather than a journalist's.
- The caveat that almost every repetition drops: the study did not find a faster decline in the FEV1/FVC ratio, and the lung function decline was not accompanied by more obstructive airway disease. So this is a measurable loss of lung volume over two decades, not a demonstration that cleaning gives you COPD. Reporting the first half without the second is how a real finding becomes an overstated one.
- No effect was seen in men — though there were few male occupational cleaners, so that is as likely to be a limit of the study as a fact about men.
- And it is observational. People who clean professionally differ from people who do not in income, housing, smoking and much else, and the authors adjusted for what they could.
Around that study, the picture is consistent and unsurprising. Occupational asthma in cleaners is well established and is one of the better-characterised causes of work-related asthma. Disinfectant use in healthcare staff has been associated with higher rates of new asthma and COPD. Strong for airway irritation and occupational asthma from regular, heavy, poorly-ventilated use. Emerging for the systemic and hormonal claims made about the same products at domestic exposure levels — those rest largely on mechanism.
5. The products whose only job is to smell
Cleaning products at least clean. This category does not, and it is where the easiest reductions are.
| Product | What it is doing |
|---|---|
| Plug-in air fresheners | The clearest thing to stop on this page. A heated, continuous terpene and fragrance source running in an occupied room for weeks — the one exposure here that is deliberately constant rather than occasional. It masks odour rather than removing it, and it has no other function |
| Aerosol and spray fresheners | The same chemistry as a burst, plus the aerosol problem. Also a common trigger for people with asthma |
| Scented candles | Combustion in a small room: fine particles and soot as well as fragrance. Paraffin and soy behave similarly — the wax is a smaller issue than the burning and the scent. An occasional candle is not the concern; several, nightly, in a closed room is |
| Fabric conditioner | Fragrance designed to persist on cloth and release slowly for days, next to skin and airway. Provides softness and static reduction and nothing else. The single highest-fragrance product in most homes |
| Scent boosters and in-wash beads | Neat perfume in bead form, added on top of an already fragranced detergent. A recent product category with no function beyond scent intensity |
| Reed diffusers, room sprays, car fresheners | Same again, in smaller spaces. A car is the smallest, least ventilated room most people occupy |
Laundry, since it touches skin all day
- Detergent residue on fabric is a genuine cause of dermatitis, and the usual culprits are fragrance and preservatives rather than the enzymes — which matters, because the standard advice is to switch to non-biological, and non-bio removes the enzymes while frequently keeping the perfume.
- “Non-bio” is not the same as “non-fragranced”. If someone in the house has eczema or contact dermatitis, the product to look for says fragrance-free and dye-free, whether or not it is biological.
- Enzymes are a poor suspect anyway. They are protein-degrading, deactivate in the wash, and the evidence that residual enzyme causes skin problems in ordinary domestic use is weak. The reputation is largely historical.
- Overdosing detergent is common and leaves more residue. Dose to water hardness, and an extra rinse does more for sensitive skin than switching brands.
- Fabric conditioner is the first thing to drop in any household with eczema, asthma or unexplained skin irritation. It is the most fragranced, the most persistent, and the least necessary.
Drying washing indoors
It also releases the fragrance the detergent and conditioner were designed to leave behind, slowly, into a room people are sitting in — which is the same objection as the plug-in, arrived at differently.
- Dry outside whenever the weather allows, even partially.
- If drying indoors, do it in one room with the door shut and a window open or an extractor running, rather than spread through the house. A closed bathroom with the fan on is better than a living room.
- A vented or heat-pump tumble dryer or a dehumidifier removes the water rather than moving it; a condenser dryer that is not vented externally must be emptied and still leaks some.
- Never dry washing on radiators in a closed room, which is the highest-moisture, lowest-ventilation version of this.
- If you rent and there is mould, this is not simply a lifestyle problem to be blamed on. Damp and mould in rented housing carry legal duties on landlords, tightened after a child's death from mould exposure. Report it in writing.
New clothes, and where the flame retardants actually are
Washing new clothes before wearing them is sensible, and the reason is usually not the one given. Textiles arrive with residues from dyeing and finishing — disperse dyes, which are a recognised cause of contact dermatitis, and formaldehyde-based resins used for crease resistance — plus whatever they picked up in transit and in the shop. A wash removes much of it, and it matters most for anything worn against skin and for children.
But the flame retardants are in your sofa, not your shirt — and Britain has more of them than almost anywhere
- Ordinary UK clothing is not flame-retarded. The exception is nightwear, which has its own standard. So "wash new clothes to remove fire retardants" is aimed at the wrong product.
- Upholstered furniture is a different story entirely. The Furniture and Furnishings (Fire) (Safety) Regulations 1988 require an open-flame ignition test that manufacturers have largely met by adding chemical flame retardants — and a group of scientists convened by UK Research and Innovation has observed that the UK has some of the highest usage of flame retardants in the world, largely because of that test.
- The exposure route is household dust, not the fabric touching you. Additive flame retardants migrate out of foam and textile into dust, which settles, is disturbed, and is inhaled or ingested — which is why small children crawling on floors are the most exposed group in the house.
- And the regulator has now said so, for one group. The Amendment Regulations 2025, in force from 30 October 2025, removed a list of baby and children's products from the scope of the 1988 regulations — cot mattresses, pram sacks, car seat inserts and similar — on the explicit basis that for those products the chemical exposure risk is greater than the fire risk. Further reform, including a move to a smoulder test and a requirement to justify using chemical flame retardants at all, is under consultation.
- What to do about it is dull and effective: damp-dust and vacuum with a HEPA filter, wash children's hands before eating, and ventilate. You cannot buy your way out of this one — but dust is the route, and dust is manageable.
“It all gets absorbed through your skin” — what is actually true
This deserves a straight answer, because it is used to sell things in both directions.
- Skin is a good barrier, not a sponge. The stratum corneum is specifically evolved to keep things out, and the widely repeated claim that most of what you apply enters the bloodstream is not true.
- But it is not impermeable, and what crosses is predictable. Small, fat-soluble molecules cross best; large or water-soluble ones largely do not. That is why some drugs are given as patches and most are not.
- Absorption rises where the barrier is thinner or broken — eyelids, genitals, armpits after shaving, eczematous or inflamed skin, and anywhere occluded under a dressing or tight clothing. Infant skin is more permeable than adult skin.
- The demonstrated example is sunscreen filters, which the FDA showed reach the bloodstream above its screening threshold — and, as that chapter sets out, absorption is not harm. Crossing the skin is the beginning of a toxicological question, not the answer to one.
- The practical conclusion is narrow: for leave-on products applied daily over a large area, or applied to broken skin, or applied to a baby, prefer short ingredient lists and no fragrance. For a rinse-off product used once a day, this matters much less than the marketing suggests.
6. Cosmetics and personal care
Applied deliberately, daily, to a large surface area, often left on. That is a different exposure pattern from a spray used once a week, and it deserves a separate look — and a calmer one, because this is where alarm outruns evidence most often.
| Concern | Where it stands |
|---|---|
| Fragrance in leave-on products | The best-evidenced concern, and the least discussed. Contact allergy to fragrance is common, frequently unrecognised, and diagnosed by patch testing. If you have unexplained facial or eyelid dermatitis, fragrance in a leave-on product is a leading candidate — and the GB label names fewer of them than an EU one |
| Isothiazolinone preservatives | A genuine epidemic of contact allergy led to restriction in leave-on cosmetics. A real, regulated, resolved problem — and a good example of labelling working |
| Parabens | The most over-stated item in this table. Weakly oestrogenic in laboratory systems; the longer-chain ones are restricted or banned in the EU and UK, and the ones remaining in use are permitted at concentrations regulators have repeatedly reassessed. No demonstrated human harm at cosmetic exposures, and paraben-free products use other preservatives which are not automatically safer |
| Phthalates | Largely removed from EU and UK cosmetics; the ones of toxicological concern are restricted. Diethyl phthalate can still appear inside “parfum”, which returns you to the labelling problem rather than a toxicology one |
| Formaldehyde releasers | Preservatives that liberate small amounts of formaldehyde — DMDM hydantoin, imidazolidinyl urea and relatives. Relevant mainly as a contact allergen, and required to carry a warning above a threshold |
| Talc | Reclassified by IARC in 2024 as probably carcinogenic to humans (Group 2A), on limited human evidence for ovarian cancer and stronger mechanistic and animal data. Historic contamination with asbestos confuses the picture badly. Cornflour-based alternatives exist and cost nothing extra, which makes this an easy substitution regardless of where the evidence settles |
The pattern worth taking from that table. The items with the strongest evidence — fragrance allergy, preservative allergy — are unglamorous, treatable and diagnosable, and they get a fraction of the attention paid to parabens. “Free from” marketing tracks public anxiety rather than evidence, and a product free of the thing you read about is not free of the thing that is actually causing your rash.
7. The mixtures that cause real injury
- Bleach + ammonia (including many glass and multi-surface cleaners) → chloramine gas.
- Bleach + acid (limescale removers, some toilet cleaners, vinegar) → chlorine gas.
- Bleach + hydrogen peroxide, and bleach with almost anything else → assume the worst.
8. What to actually do
In order of how much they achieve relative to the effort involved.
Ventilate, and keep ventilating after you finish
The single highest-value action on this page, and free. Open a window while cleaning and for twenty minutes afterwards, because the reaction products in section 3 form after the spraying stops. Extractor fans in kitchen and bathroom, actually switched on.
Stop the products that only add scent
Plug-ins first, then scent boosters, then fabric conditioner. No cleaning capacity is lost, because they never had any. This is the largest single reduction available in most homes.
Use fewer, simpler products — and a cloth rather than a spray
Microfibre and warm water handles most routine cleaning. Where a product is needed, one multi-surface cleaner and one bathroom cleaner covers nearly everything; the cupboard full of specialists is a marketing achievement rather than a cleaning one.
Skip antibacterial for everyday cleaning
Detergent and physical removal do the work at home. Quats are respiratory sensitisers and there is little domestic case for routine use. Disinfect where there is a reason — raw meat, illness in the house — not by default.
Choose fragrance-free where something touches you all day
Laundry detergent, leave-on skincare, anything used on a child. Fragrance-free, not “unscented” — unscented can mean a masking fragrance was added to cover a base odour.
Never mix, and open the window when you use bleach
Section 7. One product, one surface, rinse between.
If something is irritating you, get patch tested rather than guessing
Persistent hand, face or eyelid dermatitis has a diagnostic pathway. Patch testing identifies the specific allergen, which is far more useful than eliminating products one at a time for a year. Ask your GP about dermatology referral.
And the proportion to keep. This entry is graded on the register well below ultra-processed food, air pollution and radon, and that ordering is deliberate. The realistic gain here is avoiding a preventable dermatitis or a preventable asthma exacerbation, and reducing a lifetime of low-level airway irritation — worth having, cheap to get, and not in the same category as the things above it. A person who ventilates, drops the plug-ins and stops mixing bleach has captured most of what is available.