Among homeless individuals, does owning a pet improve their mental health?

[...]

Appraisal, application and reflection

Homelessness is an increasing public health epidemic in the United States, and up to 25% of homeless individuals own pets (Rhoades et al., 2015). By understanding the potential mental health benefits that pets provide to these individuals, there is a potential to increase the amount of scientific information available to the general public, to potentially change negative perceptions associated with pet ownership in the homeless population.

While the conclusions made among the qualitative and qualitative/cross-sectional hybrid studies were near unanimous in their findings, that pets can improve the mental health status of homeless individuals (Yang et al., 2020; Labrecque & Walsh., 2011; Cleary et al., 2020; Howe & Easterbrook, 2018; Lem et al., 2013; Slatter et al., 2012; Kidd & Kidd, 1994; Rew, 2000; Brewbaker, 2012; Cleary et al., 2021; Scanlon et al., 2021; and Schmitz et al., 2021), these results should be interpreted with caution as there are multiple limitations and biases in these studies. Nine out of these 11 studies had small sample sizes ranging from 2–34 subjects, while the Kidd & Kidd (1994) and Labrecque & Walsh (2011) studies had a sample size of 105 and 51, respectively. Given the nature of qualitative studies, which generally do not report any statistical analysis, the small sample size is understandable given the degree of in-depth interviews conducted. However, the biases associated with these studies cannot be ignored; most notably the selection bias. Many of these studies focused solely on pet owners, with targeted questions regarding the effects their pet has had on them. Few studies asked similar questions to non-pet owners, so the generalisability of these results is limited. Additionally, with the exception of Kidd & Kidd (1994), all of these studies selected participants by recruiting at shelters, day clinics, free street veterinary clinics, or similar scenarios. This creates a significant selection bias towards individuals who have access to these facilities, who feel safe and welcome at them, and who knew they could bring their pet. This leaves a potentially large proportion of homeless individuals who are unreachable – whether they were not qualified to access clinics (if guests are required to have a drug test before coming, for example), or perhaps they did not think their pet would be allowed. Thus, the population of people who were already in a position to come to these clinics may already have better circumstances and thus, improved mental health compared to those who are even more marginalised. With all limitations and biases considered, there is still an effect to be noted that those who were included in the study, albeit part of a small sample size or fairly unavoidably biased population, all reported various ways their pets improved their own perception of their mental health.

Unlike the qualitative studies, the two cross-sectional studies provided more quantitative and statistically significant results with a larger sample size and a comparison group. The study performed by Lem et al. (2016) is the only study to perform a quantitative measure of effect (odds ratio) using prevalence to show the odds of depression in pet owners is three times lower compared to non-pet owners. Similarly, Rhoades et al. (2015) was able to perform a chi-square quantitative analysis showing that homeless pet owners’ scores on a depression and loneliness scale were significantly lower than non-pet owning homeless individuals. Despite the strength behind these quantitative studies compared to the qualitative studies, these studies were also limited in the 

same selection bias that results from only selecting participants who had access to shelter services. Additionally, these cross-sectional studies can only conclude there is an association between pet owners and depression, without any means of determining causality. As such, it is possible that individuals who are in a better mental health state are more likely to be a pet owner.

Based on the literature described above, it is evident there is a clear need for more quantitative and longitudinal studies in this area. However, the feasibility of performing a cohort study in this population would be very challenging. Many individuals experiencing homelessness would be likely lost to follow-up, or potentially become temporarily housed which would complicate their role in the study. Additionally, the selective bias in all of these studies indicates a need to access a proportion of the homeless population that does not have access to shelters, but attempts to reach this population may be complicated by an increased chance of encountering individuals whose ability to consent to participate may be influenced by other factors. Additionally, as Schmitz et al. (2021) pointed out, some questions may be framed to encourage a positive response regarding pet ownership, leading respondents to only report the positive mental effects associated with pet ownership.

Overall, there is a clear need to continue research in this area, as homelessness continues to increase nationwide. While the evidence does support the conclusion that there is an increase in the mental health status associated with pet ownership in this population, many individuals reported grief and stress associated with the loss of a pet. Future efforts should attempt to address these potential negative impacts to mental health as well. Thus, further research should be focused on the potential decrease in mental health associated with pet loss that many of these studies highlighted, and the long-term consequences associated with that loss.

[…]