The original study, led by Putri and colleagues, is unusually rigorous for a question this messy. The team used validated diagnostic algorithms to confirm asthma and allergy diagnoses, cross-referenced a national register of cats to determine which households actually kept one, and measured five distinct dimensions of asthma outcomes. They even ran a probabilistic bias analysis to account for the possibility that some cat-owning households were missed by the register. Their conclusion was blunt: blanket advice to avoid cats may not be uniformly beneficial for children who already have asthma and airway allergy.

But the Zurich commentators — Milad Ameri, Pål Johansen, and Marie-Charlotte Brüggen of the University of Zurich and the Christine Kühne Center for Allergy Research and Education — argue that the study’s central exposure variable deserves closer scrutiny. Cat ownership, they write, is a decision, not a dose. It reflects which families choose to keep a cat, not how much allergen a child actually breathes in. Once you take that distinction seriously, the behavior of families stops being background noise and becomes part of what the study is actually measuring.

The commentators point to a striking pattern in the Swedish data: severe allergic rhinitis at baseline was less common among children living with cats (11.5 percent) than among those without one (14.2 percent), even though initial asthma severity and parental atopy were comparable across groups. This hints at what epidemiologists call confounding by indication — parents may avoid getting a cat precisely because their child’s allergies are worse. The direction of this selection is not random, and it could bias the observed association toward showing no effect, even if a real effect exists.

How strong is this selection effect? A 1997 survey cited in the commentary offers a clue. Among 341 adults with diagnosed cat or dog allergy who had been explicitly advised by doctors to stop living with their animals, only 21.4 percent actually removed the pet or excluded it from the home. Among a subset of 122 whose animal had since died, 70.5 percent went on to acquire another. Families who relinquish a pet appear to be a minority — and they may represent a subgroup in whom allergic symptoms or medical advice have become severe enough to override the emotional bond. Cat-owning households, in other words, may be a self-selected population in which the presence of a cat has already proven tolerable.

The commentators also flag a subtler problem: the absence of a cat at home is not the same as the absence of cat exposure. Children classified as living without cats may still encounter substantial amounts of the major cat allergen, Fel d 1, through schools, childcare centers, relatives, and friends. This was demonstrated in Sweden itself, in a study co-authored by the senior author of the current commentary. Among more than 400 asthmatic children sensitized to cats but with no cat at home, those attending classrooms where more than 18 percent of classmates owned cats showed decreased peak expiratory flow, more symptom days, and greater medication use once the school term began. Passive allergen transport is real, measurable, and clinically meaningful.

There is also a question about how cat ownership was ascertained in the first place. Sweden’s National Cat Register was only established in 2023, and while the study authors appropriately modeled incomplete registration in their bias analysis, the commentators note that the cat-exposed group was markedly enriched for Nordic-born parents — 92.2 percent of mothers and 89.9 percent of fathers, compared with 78.9 percent and 77.7 percent in the non-exposed group. Cat ownership, in other words, is socially patterned. If the completeness of registration or record linkage varies with sociodemographic characteristics that are themselves related to asthma outcomes, the resulting misclassification could be differential — a form of bias that can push estimates in either direction, and one that a bias analysis assuming non-differential ascertainment cannot address.

None of this, the commentators stress, is meant to dispute the null finding, which was consistent across all five outcome measures. Rather, they argue that the behavioral constructs lurking behind ownership decisions are not impenetrable black boxes — they are measurable. Family attachment to the animal, for instance, can be quantified using validated instruments such as the Lexington Attachment to Pets Scale, a 23-item questionnaire, or its briefer seven-item version, which has been tested in more than 2,000 pet owners across three countries and shown to be measurement-invariant across four. Cat-keeping practices — whether the cat is allowed in the bedroom, kept indoors or outdoors, washed regularly, or fed an allergen-reducing diet — can be captured through structured exposure inventories and environmental Fel d 1 measurements.

Each of these constructs occupies a different position in the causal chain, and that position determines how it should be handled analytically. Attachment to the animal is a determinant of ownership and a confounder only if it also affects asthma independently of cat exposure; otherwise, it simply characterizes which families select into ownership. Cat-keeping and avoidance practices sit between ownership and actual allergen dose, so adjusting for them statistically would overadjust and mask real effects — but recording them would allow researchers to grade exposure by dose rather than by binary ownership status. Symptom attribution, the tendency of families to interpret a child’s breathing difficulties as cat-related, may itself be influenced by the outcome being studied and resists simple covariate adjustment.

The clinical implications could be significant. If blanket avoidance advice is not uniformly warranted, what replaces it is a conversation — one that depends on knowing how attached the family is, what practices they already follow, and what changes they are willing to make. And that conversation is increasingly backed by options that do not require giving up the pet at all. Allergen-reducing cat diets and experimental immunization of cats against Fel d 1, designed to induce neutralizing antibodies against the major feline allergen, have made what researchers call keep the cat, change the care pathway a testable clinical strategy rather than a slogan. The Zurich commentators propose that the field stop treating the human–animal bond as an unmeasured limitation and start recording it systematically — because the real question is not simply whether cat-owning families differ in asthma outcomes, but whether any difference reflects allergen dose or the selection of families into ownership in the first place.

Subject of Research: The relationship between household cat exposure and asthma outcomes in children with asthma and allergy

Article Title: Commentary: Cat exposure and asthma outcomes in a cohort of children with asthma and allergy

Article References: Ameri, M., Johansen, P., & Brüggen, M.-C. (2026). Commentary: Cat exposure and asthma outcomes in a cohort of children with asthma and allergy. Frontiers in Allergy, 7, Article 1997917. https://doi.org/10.3389/falgy.2026.1997917

Image Credits: AI Generated

DOI: 10.3389/falgy.2026.1997917

Keywords: asthma, cat allergy, Fel d 1, allergen exposure, pet ownership, confounding by indication, exposure misclassification, human-animal bond, pediatric allergy, register-based epidemiology, allergen avoidance, Lexington Attachment to Pets Scale