What’s Different About Dog Owners?
Study links dogs and health but highlights need to quantify ownership decisions.
Posted Nov 22, 2017
“Want to live longer? Get a dog.” This was the headline of an article recently published on CNN about a large-scale survey of the health of people in Sweden. Such phrases are eye-catching and commonplace in the reporting of links between pet ownership and health. But they’re wrong: So far, there is no direct evidence that someone who gets a dog automatically accrues several more years on this planet.
The Swedish study, although the most recent such research to be misrepresented in the press, is at least based on a very large sample — more than three million Swedes between ages 40 and 80. Moreover, in the paper itself, the authors are careful to point out that the associations they found are no proof of causation. Unfortunately, the press release issued by Uppsala University included the phrase, “Dog ownership was especially prominent as a protective factor [against cardiovascular disease] in persons living alone,” which is easy to interpret as “Dogs protect their owners against disease.”
Dogs love to protect their masters, right? Moreover, the very first sentence in the abstract of the paper describes two mechanisms whereby dog ownership might improve cardiovascular health (providing social support and motivating physical activity), neither of which were addressed directly by the data presented. So perhaps it’s understandable that this study has been widely misinterpreted, despite lead researcher Tove Fall explicitly stating, “These kind of epidemiological studies...do not provide answers on whether and how dogs could protect from cardiovascular disease.”
It’s so widely believed that pets are good for their owners’ cardiovascular health that almost every press report on this study that I’ve seen has ignored Tove’s remark and transformed correlation into causation. The Guardian wrote, “Dog ownership had a dramatic effect on people who live alone, cutting the risk of death from cardiovascular disease by 36 percent”; The Independent claimed, “Owning a dog could lengthen your life”; the BBC’s article ran under the heading, “Dog ownership lowers early death risk, study finds." (My italics, in each case.) The (London) Times headline ran, “Want to live longer? Give a dog a home," which could encourage an ill-prepared leap into a 15-year, $20,000 commitment.
The inherent problem in these types of surveys is that all the owners have, at some time in the past, chosen to get a dog, and all the non-owners have decided that a dog is not for them. It’s in this sense that this way of investigating what factors promote good health differs fundamentally from, say, a trial of a new drug, for which it’s usual to compare the drug with similar drugs or a placebo. There is no equivalent of a placebo for getting a dog, although comparisons might fruitfully be made with other types of pet, such as cats. Unfortunately, when these studies are reported in the press, such fundamental distinctions become blurred.
So what might be the explanation for the link between cardiovascular health and owning a dog, if it isn’t the exercise or the company? The answer may be found in another study, published earlier this year but not mentioned in the Swedish report, which showed that dog owners in California differ from people who don’t own pets in many ways, including gender, age, race/ethnicity, living arrangements, and income. All of these factors are known to impact health and lifespan.
It is almost impossible to tell from epidemiological studies what is cause and what is effect when a link between, say, dog ownership and health is found. Although it’s almost always assumed that dogs somehow improve health, it could equally well be the other way around — that health influences dog ownership. For many such studies, a plausible though rarely mentioned interpretation is that people who aren’t feeling very well tend not to get a new dog (though not so much this one, which tracked cardiovascular health only in people who had been symptomless for 12 years before the data was gathered). However, what is also possible is that both dog ownership and cardiovascular health are affected by some complex interplay between other lifestyle and psychological factors.
Such differences may explain some otherwise puzzling differences between owners of different types of dogs that emerged in the Swedish study: Owners of retrievers and pointers had the lowest levels of heart disease, while owners of mixed breeds had the highest — worse than people with no dog. Is it logical that a purebred Labrador should be so much more effective than a Labrador cross in combating loneliness or motivating physical exercise? Surely it’s much more likely that other lifestyle factors are affecting both what kind of dog people choose to own, and how health-promoting other parts of their life are. For example, might those who choose gun-dogs have taken more exercise throughout their lives?
All of this points to a significant gap in our knowledge of pet ownership, which is the reason behind why some people decide to get a dog and others a cat, while still others don’t want an animal in their house under any circumstances. If we knew enough about what motivates these decisions, the most important factors could be plugged into the analyses of surveys, and we could obtain a much more accurate picture of the links between pet ownership and health, and how these differ from one culture to another. The major players in the pet industry may already have such information, but if they do, they’re keeping most of it to themselves.
Facebook image: Soloviova Liudmyla/Shutterstock
Mubanga, M. et al. (2017) Dog ownership and the risk of cardiovascular disease and death – a nationwide cohort study. Scientific Reports 7, Article number: 15821 (2017) doi:10.1038/s41598-017-16118-6
Saunders, J., Parast, L., Babey, S.H., Miles, J.V. (2017) Exploring the differences between pet and non-pet owners: Implications for human-animal interaction research and policy. PLoS ONE 12(6): e0179494. https://doi.org/10.1371/journal. pone.0179494