Three Things to Know
While some research suggests pet ownership may benefit human health, the findings are inconsistent. This paper describes the key differences between pet and non-pet owners and explains why researchers must control these socio-demographic distinctions in studies examining the pet-health link.
Pet owners differ significantly from non-pet owners across several socio-demographic factors. Specifically, pet owners in the study were more likely to be younger, White, female (married or single), homeowners living in rural areas, and members of fully employed households.
These socio-demographic variables are crucial because they are established determinants of health. Factors such as race, income, age, and employment strongly influence health outcomes. Therefore, researchers caution that any apparent link between pet ownership and health is likely confounded by the tendency for healthier or wealthier individuals to be more likely to own pets. Studies must account for this selection bias.
For Dog Welfare Practitioners
As dog welfare practitioners, we naturally believe in the health and mood benefits of animal companionship. While research doesn’t refute these benefits, it highlights a crucial challenge: pet owners and non-owners often differ significantly in key socio-demographic factors, complicating direct comparisons. Furthermore, knowing that some potential adopters seek dogs specifically for exercise or health goals underscores the importance of setting realistic expectations at the shelter level to reduce the risk of returns.
The Full Picture
Pet ownership is common in the U.S., and many believe it benefits human health, but research findings are inconsistent. Some studies report that pets — especially dogs — reduce anxiety and increase physical activity, while others link them to allergies, disease, or no measurable health effects. Methodological weaknesses, including small or biased samples and cross-sectional designs, limit conclusions about causality. In addition, evidence from animal-assisted therapy shows benefits in clinical settings but may not generalize to everyday pet ownership. Moreover, pet owners differ from non-owners in factors like household size, housing type, and lifestyle, which themselves influence health outcomes.
As current research suggests possible short-term psychological and physical benefits of pet ownership, but strong, population-level evidence of a causal link is still lacking. This paper, titled “Exploring the differences between pet and non-pet owners: Implications for human-animal interaction research and policy”, aims to:
- Describe how pet owners and non-pet owners differ.
- Explain why these differences must be accounted for when studying pet ownership and health.
Study Methods
This study used data from the 2003 California Health Interview Survey (CHIS), a large, population-based telephone survey designed to represent California’s diverse population. CHIS collected detailed information on health status, behaviors, access to care, and demographics through interviews conducted in multiple languages.
Study Results
The final sample included 42,044 adults with available data on individual characteristics and self-reported pet ownership. Among respondents, 26.2% owned a dog, 21.5% owned a cat, and 8.5% owned both. The sample was 49% male, with an average age of 44.4 years. About half were White, 26% Hispanic, 12% Asian, and 6% Black. Most were married (61.9%), lived in houses (66%), and owned their homes (55.9%). On average, respondents lived in households of 3.3 people, 56.6% worked full time, and 40.2% of households had all adults working full time. The average BMI was 26.6, 7.3% reported asthma, and self-rated health averaged 3.5 out of 5.
Dog Ownership
In adjusted multivariate models, dog ownership was strongly associated with being female, single, White, a homeowner, living in a house, having a higher income, a larger household, and a rural residence. Hispanic, Asian, and Black respondents were significantly less likely to own dogs than White respondents. When health variables were added, dog owners were more likely to suffer from asthma and had slightly higher BMI, though the BMI effect was minimal.
Cat Ownership
Most organisations focused on one or two species, primarily cats and Patterns were similar but not identical. Cat ownership was more common among single women, White respondents, homeowners, and those living in smaller households, houses, or rural areas. Older adults and non-White respondents were less likely to own cats. After adjusting for health factors, only asthma remained significantly associated with higher odds of cat ownership, while BMI and general health were not.(81%). Collectively, respondents cared for nearly 60,000 animals annually, highlighting the scale of companion animal homelessness and the critical role CARO play.
Dog and Cat Ownership
Individuals owning both dogs and cats shared similar characteristics with dog owners—more likely to be single women, White, homeowners, rural residents, and to live in larger households. Non-White and older respondents were less likely to own both. Asthma again showed a modest positive association, while BMI and general health were unrelated.
Overall, the results show that demographic, socioeconomic, and lifestyle factors—especially race, housing, marital status, and rural residence—pattern pet ownership. Researchers must account for these factors when studying links between pets and health.
Research and Policy Considerations
These socio-demographic differences are important because they determine health. Factors like income, race, age, and employment strongly influence health outcomes, suggesting that these variables may confound any apparent links between pet ownership and health. For instance, previous studies cut the association between pet ownership and fewer doctor visits in half once researchers controlled for income and similar predictors.
Because of this, the authors caution that studies on the health effects of pet ownership must account for selection bias—that is, the likelihood that healthier or wealthier people may be more likely to own pets. They recommend the use of quasi-experimental methods, particularly propensity score matching, to adjust for confounding factors. Propensity score modeling (especially with boosted regression techniques) helps balance groups of pet owners and non-owners on key covariates, better approximating randomized controlled conditions. Other methods such as natural experiments, instrumental variables, or regression discontinuity could also help, though each has challenges in identifying valid instruments and covariates.
Conclusion
Pet owners and non-owners differ in multiple socio-demographic ways that are themselves linked to health outcomes. Therefore, any study exploring the health benefits of pet ownership should account for selection bias to avoid misestimating effects. The authors recommend using propensity score matching with boosted regression as the most suitable method given the complex and uncertain relationships between demographic factors and pet ownership.
Miscellaneous
Data From Study:
Owned Dogs > United States > Pet Ownership
Year of Publication:
2017
External Link:
Saunders J, Parast L, Babey SH, Miles JV (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