ResearchSocial connection
Do pets improve wellbeing?
Last reviewed 17 August 2026
The verdict
Across 24 observational studies, pet ownership showed no consistent association with loneliness before COVID-19, with one study finding owners lonelier than non-owners. After the pandemic, about half found reduced loneliness. Social isolation is the outcome with the clearer signal. A therapy dog trial did reduce loneliness in hospitalised older adults, which is a different question from owning one.
Strength of evidence
- traditional
- emerging
- informed
- established
Early findings that look promising and need better studies before anyone can be sure. How we grade
What does owning a pet do?
Less than the culture assumes, and it depends when you ask. Kretzler, König and Hajek reviewed 24 observational studies and found the picture too heterogeneous to pool into a single number.
Among adults before COVID-19, most studies showed no significant connection between owning a pet and loneliness. One found pet ownership associated with higher loneliness.
Among studies conducted after the pandemic began, roughly half found pet ownership associated with lower loneliness. The authors suggest dog walking may have supplied social contact when other routes closed.
Why would the answer change?
Because loneliness is relative to what else is available. When ordinary social life is intact, a dog appears to add little to it. When it is suspended, the same dog becomes a larger share of your contact with the world.
That reading is speculative and it fits the data better than the alternative, which would require pets to have changed.
It also carries a warning about the wider literature. Findings collected during an unusual period may describe that period rather than the thing being studied.
What about social isolation?
Different outcome, clearer signal. In the smaller number of adult studies examining social isolation rather than loneliness, owning a pet was associated with lower levels of it.
The distinction is worth holding. Isolation counts your contacts. Loneliness measures the gap between the connection you have and the connection you want. A pet plausibly changes the first without touching the second.
Most studies found no differences between dogs, cats and other pets, which sits oddly with the dog walking explanation and is worth noting rather than smoothing over.
Does a dog visit help if you are already lonely?
In one small trial, yes. Gee and colleagues randomised 42 hospitalised older adults to twenty minute daily sessions over three days: dog visits, conversation with the handler alone, or usual care.
Loneliness improved significantly in the dog condition compared with usual care, and the conversation-only condition did not differ from usual care. On the analogue loneliness measure the effect was clearer again.
Forty-two people across three arms is a pilot. The authors list a small convenience sample, recruitment difficulties limiting power, and a heterogeneous group across several hospital sites.
What is worth doing?
Get a pet because you want one. The evidence does not support acquiring one as a treatment for loneliness, and the pre-pandemic studies mostly found no effect.
Notice which problem you have. If your contact with people is thin, a dog that gets you outside plausibly helps. If you have people and still feel unseen, that is the outcome pets moved least.
Treat the isolation finding as the usable one, and put the direct work elsewhere. The evidence on building friendships as an adult and the research on connection and lifespan both address the gap a pet appears to leave open.
Questions people ask
- Do pet owners feel less lonely?
- Before COVID-19, mostly no, and one study found owners lonelier. After, about half of studies found lower loneliness.
- Are dogs better than cats?
- Most of the 24 studies found no differences between dogs, cats and other pets.
- Why is this evidence weak?
- All 24 studies were observational, heterogeneous enough that they could not be pooled, and drawn entirely from high-income countries.
- Do therapy animals work?
- A pilot trial found loneliness improved with dog visits against usual care, while conversation with a handler alone did not. Forty-two people, three days.
Sources
Every linked source below was opened and checked on the review date. Where a record could not be opened, it is named in the text with no link.
Kretzler B, König HH, Hajek A (2022). Pet ownership, loneliness, and social isolation: a systematic review. Social Psychiatry and Psychiatric Epidemiology.
Twenty-four observational studies, too heterogeneous to meta-analyse. Mostly no association with loneliness among adults before COVID-19 and one finding higher loneliness, with about half of later studies finding lower loneliness. Lower social isolation among owners in the few studies examining it, and no consistent differences between pet types. All studies from high-income countries, half rated good quality and half fair.
Gee NR, Townsend L, Friedmann E, Barker SB, Mueller MK (2024). A Pilot Randomized Controlled Trial to Examine the Impact of a Therapy Dog Intervention on Loneliness in Hospitalized Older Adults. Innovation in Aging.
Forty-two hospitalised adults aged 59 and over across three arms, twenty minute sessions for three days. Loneliness improved in the dog condition against usual care while handler conversation alone did not. Small convenience sample, limited power, heterogeneous across hospital sites.