Every cat owner in Singapore has had the conversation. A relative visits, sees the hair on the sofa, and says something about your lungs. A partner starts sneezing and blames the Maine Coon. Someone suggests you shave the cat.
So we went looking for what the measurements actually say about cat hair and human health. The short version is that the hair is close to innocent, the thing riding on it is not, and the appliance most people buy to fix the problem has a worse evidence record than its marketing suggests.
That last part includes the air purifier we sell. We will get to it.
1. The hair is the vehicle, not the cargo
The protein that makes people react to cats is called Fel d 1. It is a 36 kDa protein, and the cat manufactures it in its sebaceous, anal and salivary glands. The highest concentrations end up in the fur and the outer layer of skin, which is how it reaches the rest of your flat.
Two figures put its importance in perspective. Between 60% and 90% of all IgE reactivity to cat dander is directed against Fel d 1, and up to 95% of cat-allergic people are sensitised to it. When someone says they are allergic to cats, this single protein is usually what they mean, whether or not they know its name.
The practical consequence matters more than the biochemistry. A shed hair is a delivery mechanism. It carries Fel d 1 on its surface, along with flakes of skin. Shaving the cat does not remove the glands. Neither does a short-haired breed. The hair is what you see, so the hair is what gets blamed, but the protein is produced regardless of coat length.
2. How much, how small, how long
Airborne Fel d 1 has been measured in homes for decades, and the numbers are consistent enough to plan around.
In homes with a cat, airborne cat allergen typically runs 2 to 20 nanograms per cubic metre. In homes with no cat, it is usually below 0.2 ng/m³ — roughly one to two orders of magnitude lower, but note that it is not zero. Fel d 1 is carried into cat-free buildings on the clothing of people who own cats. It has been found in offices, schools and homes that have never had an animal in them.
Particle size is where it gets interesting. Roughly 75% of airborne Fel d 1 travels on particles 5 micrometres or larger, and about 25% on smaller particles. The large fraction settles out of the air fairly quickly. The small fraction is the one that stays suspended for long periods, and it is also the fraction small enough to reach the lower airways rather than being caught in the nose.
This is why a cat-allergic visitor can start reacting within minutes of walking into a flat, long before they have touched anything. They are not reacting to the hair on the sofa. They are breathing the quarter that never landed.
3. The vacuuming problem
Here is the measurement that changed how we think about cleaning.
In one house, during ordinary domestic cleaning, the concentration of small-particle cat allergen approached 40 ng/m³. For scale, that is in the same range as the dose a nebuliser delivers during a bronchial provocation test — the controlled challenge used in clinics to deliberately provoke an airway response under supervision.
We want to be careful about what that does and does not mean. It is one house, and a provocation test is a controlled clinical procedure, not a household event. But the direction is not in doubt: sweeping and vacuuming lift settled allergen back into the air, and for a while afterwards the room is measurably worse than it was before you started cleaning it.
If someone in the household reacts to the cat, the sequence matters. Clean when they are out. Give the room time before they come back in. Cleaning is still worth doing — it is what depletes the reservoir over time — but the hour after you vacuum is the wrong hour to sit down in that room.
4. The single biggest number in this whole article
Work by de Blay and colleagues found that carpet holds roughly 100 times the cat allergen of a wipeable floor.
Not 20% more. Around a hundredfold.
Most HDB flats are tiled or have vinyl or laminate, which means Singapore households start ahead of the British and American homes where much of this research was done. If you have added a large rug in the living room, you have added the single most effective allergen reservoir available for purchase.
The same group found something that we think every purifier buyer should read before spending money: air filtration reduced airborne allergen only when there was no carpet, and only when the floor had been cleaned first. Filtration was not an alternative to dealing with the reservoir. It only worked once the reservoir had been dealt with.
One more figure on reservoirs. When a cat is removed from a home entirely, Fel d 1 levels stay elevated for more than 20 weeks. The protein does not leave with the animal. It leaves with the soft furnishings, slowly.
5. What cat hair does not do
Because we would rather correct the myths than let them sell things for us:
Cat hair does not accumulate in your lungs. A cat hair is enormous in respiratory terms — tens of micrometres across. Particles that size do not reach the lower airway; they are caught in the nose and throat and cleared. The particles that reach deep into the lung are the sub-5 µm ones carrying protein, not the hairs themselves.
Swallowing the occasional hair is not a medical event. A hair in your food is unpleasant. It is not a health risk in the way people imagine.
Most things cats can pass to people do not travel on shed hair. Toxoplasma is acquired from faeces, not fur. Cat scratch disease comes from scratches and bites, with fleas in the chain. Neither is a reason to worry about the hair on your jumper.
The genuine exception is fungal skin infection — ringworm — which does travel on shed hair and skin scale, and which does pass between cats and people. That is a real, hair-borne risk. It is also visible, diagnosable and manageable by a vet and a doctor, which makes it a different category of problem from the invisible protein this article is mostly about.
6. The part where we argue against our own product
The obvious move here would be to explain that an air purifier captures the small airborne fraction, and then sell you one. The evidence does not support that move cleanly, and we would rather you heard it from us.
The best trial on this question is by Wood and colleagues, published in the American Journal of Respiratory and Critical Care Medicine in 1998. It is a double-blind, placebo-controlled study of 35 cat-allergic people who lived with their cats. After a one-month baseline, bedrooms received either a working HEPA air cleaner or an identical-looking dummy, for three months.
The working units did what the physics says they should: airborne allergen went down, and the difference was statistically significant (p = 0.045).
And then nothing happened. There was no improvement in nasal symptoms, no improvement in chest symptoms, no improvement in peak flow, and no reduction in medication use. Allergen in settled dust did not change either. A commentary on the trial in AAP Grand Rounds summarised it in five words: effective, but not useful.
We think the most likely explanation is the reservoir. The machine cleans the air in one room. The allergen lives in the mattress, the sofa, the curtains and the clothes, and gets re-suspended every time anyone moves. Cleaning the air without depleting the reservoir is emptying a bath with the tap running.
In fairness to the appliance, the picture is not uniformly negative. Van der Heide and colleagues (1999) found that air cleaners did reduce airway hyperresponsiveness in asthmatic children, even though allergen in floor dust was unchanged. Green and colleagues (1999) found HEPA filtration cut airborne dog allergen by 75% to 90%. So filtration measurably changes the air, and in at least one population it changed a clinical measurement.
What we cannot honestly tell you is that buying our purifier will make an allergic person in your household feel better. The one good trial that asked that exact question about cats found that it did not.
7. Reading our own label the way we would read a competitor's
Our product page for the Kopipets Smart Pet Air Purifier says two things next to each other: HEPA13, and 99.97%.
Those come from two different standards, and putting them side by side flatters the machine.
Under the European standard EN 1822, class H13 means at least 99.95% efficiency at the Most Penetrating Particle Size — the particle diameter at which the filter media performs worst, typically 0.12 to 0.25 µm. It is a deliberately unkind test, because it targets the filter's weakest point. Note also that H13 carries no local scan requirement. The next class up, H14, requires both ≥99.995% overall and ≥99.975% at every individual point, which is what catches pinhole leaks. H13 does not include that per-point check.
The 99.97% figure is the American convention: 99.97% at a fixed 0.3 µm. Because 0.3 µm is already past the most penetrating size for most media, that is the easier test. So the higher-sounding number comes from the gentler standard.
There is a second issue that applies to every purifier on the market, ours included. Both figures describe the filter media, not the finished machine. Air that slips around the edge of a filter has not been filtered at any percentage. Whole-machine performance depends on the seal, and neither number on our page is a whole-machine number.
We are going to rewrite that product page. In the meantime, this article is the correction.
8. What CADR 340 actually covers in an HDB flat
The specification sheet for our unit states CADR 340 m³/h. That number is only meaningful against a room volume, so here is the arithmetic, which anyone can check.
Air changes per hour = CADR ÷ room volume. Assume a 2.6 m ceiling.
- A 10 m² bedroom is about 26 m³. 340 ÷ 26 ≈ 13 air changes per hour.
- A 25 m² living room is about 65 m³. 340 ÷ 65 ≈ 5.2 air changes per hour.
The figure usually cited as a target for allergen removal is around 5 air changes per hour. So if the stated CADR is accurate, this unit is sized for roughly a 25 m² room at that target — the living room of a four-room flat, not the whole flat. One unit will not service an entire HDB unit, and any brand implying otherwise is not doing the division.
9. What we have not tested
Consistent with everything else we publish, here is what we do not know about our own product:
- We have not had this unit tested by an accredited laboratory. The CADR, the filtration class and the noise figures are the manufacturer's, not ours.
- We do not know which standard the CADR 340 m³/h was measured under. AHAM AC-1 and GB/T 18801 do not produce interchangeable numbers, and the specification sheet does not say. We have asked.
- We have not measured whole-machine efficiency, only quoted the media class. See section 7.
- We have not measured the noise level in night mode, and we are not going to describe it as quiet until we have a decibel figure to publish.
- The rated power is 36 W. At 36 W, a genuine CADR of 340 m³/h would be at the efficient end of what we would expect. We are not saying it is wrong. We are saying we have not verified it, and that combination is the one we would want a customer to ask a supplier about.
When we have answers, they will appear on the page where we publish our numbers, including any that turn out to be worse than the label.
10. What actually works, in order
Ranked by the strength of the evidence behind it, not by what we sell:
First, see a doctor. If someone in the household reacts to the cat, medication and allergen immunotherapy have an evidence base that no household appliance comes close to. This is the step that gets skipped because it is less fun than shopping, and it is the one that works.
Second, hard floors over carpet. The hundredfold difference is the largest single number in this article. If you are choosing between a rug and an air purifier, the floor decision matters more.
Third, keep the cat out of the bedroom. You spend roughly a third of your life with your face in that mattress. It is the highest-value room to protect and the cheapest intervention on this list.
Fourth, wash the textiles. Bedding, throws, cushion covers, curtains. That is where the reservoir lives, and depleting it is what makes everything else work.
Fifth, clean when the allergic person is out, and give the room time afterwards. See section 3.
Sixth, then consider filtration. It reliably lowers airborne allergen. Whether it lowers symptoms in cat-allergic people is not established, and the best trial says it did not. Buy it as one layer on top of the five above, not as a substitute for them.
One Singapore-specific note before you spend anything. Research on allergic airway disease in tropical urban Southeast Asia has repeatedly found that sensitisation here is dominated by house dust mite, not cat. If someone in your household is reacting and nobody has tested what they are reacting to, there is a real chance the cat is being blamed for the mattress.
11. Where our purifier honestly fits
The Kopipets Smart Pet Air Purifier is S$160, with replacement filters at S$30. It has a HEPA13 media stage, an electrostatic pre-filter for hair, activated carbon for odour, and app control. It also has a scratching surface on the outside, which is a genuinely sensible piece of design given where cats choose to sit.
What it is good at, on the physics: capturing floating hair before it lands, and reducing airborne particulate in the room it is standing in. Odour control from the carbon stage is real and is probably the benefit most owners will actually notice day to day.
What we will not claim: that it will stop anyone's allergy symptoms. We have shown you the trial. If your reason for buying is a person who reacts to your cat, please read section 10 first and treat the machine as the sixth item on that list rather than the first.
It is currently out of stock. Restock is on the way. We would rather publish this now and sell it later than time the article to the inventory.
If you want to see how we handle numbers on the other machine in the same category, the automatic litter box article takes the same approach, including the specification we admitted we could not obtain.
Sources: Fel d 1 production sites, molecular weight and sensitisation rates from the Phadia/Thermo Fisher allergen encyclopedia. Particle size distribution, airborne concentrations, the 40 ng/m³ cleaning measurement and the carpet comparison from published measurements of cat allergen in homes, including de Blay and colleagues (1991) and subsequent aerodynamic characterisation work. HEPA air cleaner trial: Wood, Johnson, Van Natta and colleagues, American Journal of Respiratory and Critical Care Medicine, 1998, 158:115–120. Supporting trials: van der Heide and colleagues (1999); Green and colleagues (1999). Filter classes from EN 1822. This article is general information about measurements, not medical advice; if someone in your household has allergy or asthma symptoms, please see a doctor.
