Two things are true at once about pets and health, and most writing on the subject only manages one of them. Animals matter enormously to the people who live with them. And the specific health claims made for pet ownership are considerably larger than the evidence supports.
Holding both is more useful than either alone, because a reader deciding whether to take on an animal deserves the real picture rather than the brochure.
The claim in circulation is that dog owners live substantially longer — a meta-analysis published in 2019 put it at around a quarter lower mortality, and that number went round the world.
It did not hold. A reanalysis that adjusted for how owners differ from non-owners found no significant association. A separate study of some twenty-two thousand owners and nineteen thousand non-owners found that owners were younger, wealthier, better educated and in better health to begin with, and that the apparent physical health advantage disappeared once those differences were accounted for.
The pattern across the wider literature is that the finding does not replicate outside Scandinavia:
Even within Sweden, where the large registry study reporting about twenty per cent lower mortality was conducted, the effect vanished in a subset of twins once body weight, smoking, long-term conditions, employment and exercise were taken into account — which is close to a definition of confounding, since twins share the background that ordinary studies cannot adjust away.
This is the ordinary difficulty with observational research on a voluntary behaviour. The people who choose to take on a dog are not a random sample of the population, and no amount of statistical adjustment fully fixes that.
The robust finding is narrower and more useful. Dog owners walk more. That one replicates, it has an obvious mechanism, and it matters, because moving from inactivity to regular moderate activity is among the larger health changes a person can make.
The honest framing is therefore that a dog is a reliable commitment device for daily walking rather than a health intervention in itself. The benefit belongs to the walking, and it accrues to the owner who actually does it in February rain. It does not accrue to a dog that is let into the garden.
Which also means the benefit is available without a dog, to anyone who will walk anyway — and that a person who would not walk may well get it from one. Both of those are worth knowing before taking an animal on.
The more interesting findings are not about owners in general but about particular circumstances. A Danish study found no mortality difference among married people but around fourteen per cent lower mortality among unmarried dog owners. A Swedish study of people who had had a heart attack or stroke found the association strongest among those living alone.
If there is a real effect in this literature, that is where it appears to sit: not in pets being good for everyone, but in an animal mattering most to someone who would otherwise have very little daily company or structure. That is a plausible and much more modest claim than the headline version.
"Cat owners suffer fewer strokes." This traces to a single conference abstract and has never been reliably replicated. It is repeated as established fact across a great deal of writing on pets.
Stress, cortisol and weight. The chain usually runs: pets lower stress, therefore lower cortisol, therefore prevent weight gain. Each link is weak and the conclusion has not been demonstrated. Stroking an animal is pleasant and can be calming in the moment; that is a fair thing to say and the endocrinology is not.
The mental health picture is likewise mixed rather than settled. Companionship clearly helps some people a great deal, particularly through bereavement or isolation, and the trial evidence that acquiring a pet improves depression is thin. A pet is not a treatment for depression and should not be presented as an alternative to one.
Any honest account has to include the other side of the ledger, because these are the things that actually go wrong.
Visiting animals in care homes and hospitals are genuinely popular with residents and staff, and the short-term lift in mood and engagement is reasonably well described. The stronger claim — that they reverse depression in residents — goes further than the evidence, and the studies are generally small and hard to blind.
These visits also have requirements that get glossed over: the animal has to be temperamentally suited to it, infection control applies, and the risk of a fall or a scratch among frail residents is real. Organised schemes exist precisely because they handle those things, and they are the route rather than an informal arrangement.
Older writing on this subject usually closes by advising the reader to go and get a pet, which is the one recommendation the evidence does not support.
An animal taken on as a health measure is the wrong reason and tends to end badly for the animal, which is the party with no say in the decision. The right question is whether a household wants an animal in it, can afford one, can commit for its whole life, and has the time and space the particular species needs. Where the answer to all of those is yes, the companionship is likely to be worth a great deal and the walking, if it is a dog, is a genuine bonus. Where any of them is no, nothing in the health literature is strong enough to tip the balance.