May 29, 2025 Identifying the Health Benefits of Companion Animals: What About George?
Dr. Boris Levinson (1907-1984), a psychologist, reportedly coined the term “pet therapy” in 1964. However, Levinson’s initiative was criticized by his colleagues. Thirteen years later, the Delta Society, founded by a group of veterinarians and a psychiatrist in 1977 (now doing business as Pet Partners), took up the challenge of demonstrating through scientific research the beneficial impact of companion and other animals on human health.
This demonstration of the human health benefits associated with companion animals has proved more challenging than expected. In a review of the available scientific evidence, Dr. Harold Herzog argues that there are many conflicting research reports and that “three decades of research on the pet effect have produced a muddle of mixed results. Some investigators have reported that pet owners are better off. Others, however, have found that pet owners have more psychological and health problems than non-owners.”
In one blog, for example, Herzog cites a 2017 study[i] published in PLoS ONE using a large dataset of California residents available from the UCLA Center for Health Policy Research. The study concludes that “pet owners and non-pet owners differ across many socio-demographic variables” and that these variables are also associated with health outcomes. The authors warn that, when trying to draw a causal relationship between pet ownership and health, the potential for selection bias should be acknowledged because such bias could lead to an inappropriate estimation of pet ownership’s actual health effects. The authors reported that pet owners were more likely to have asthma, while dog owners were more likely to have a higher Body Mass Index (BMI). Otherwise, there were no general health or BMI differences between pet owners and non-owners.
Methodological challenges have also been encountered in small-scale projects. For example, consider a study of the potential benefits of companion animals to prisoners in Virginia’s high-security Lorton State Penitentiary. [Lorton housed inmates from the District of Columbia from 1910 to 2001.] The results of this research project were reported in Anthrozoos.
The pet therapy project at Lorton State Penitentiary was initiated by a Washington, D.C. organization called People, Animals, Love (PAL). It was designed and overseen by two research psychiatrists and a public health expert who, between them, had extensive experience researching human-animal interactions. The project involved an experimental group of prisoners who received pets compared to a control group of prisoners on a waiting list to receive pets. The evaluation measures included physiological factors, such as blood pressure, and behavioral assessments related to acts of aggression and prison disciplinary incidents.
During a Delta Society conference in Washington, D.C., in the 1980s, the Lorton project was the focus of a panel discussion including the Warden of Lorton, a psychiatrist involved in the project, and the woman responsible for the daily management of the project and overseeing the daily care of the animals. The psychiatrist presented the findings of the project and concluded that there were no statistical differences between the experimental group (which had interactions with animals) and the control group (which did not). In other words, he concluded that the project yielded no beneficial results. This conclusion prompted an anguished outcry of “What about George?” from the woman managing the project. She insisted that one prisoner, George, had definitely benefited from the project. This then led to a heated exchange among the panelists, with the psychiatrist arguing that both the experimental and control groups exhibited the same number of disciplinary infractions for aggressive behavior, and there was no difference between the control and experimental groups.
The Warden then resolved the dispute by pointing out that measuring aggressive behaviors and disciplinary infractions might not have been the best indicators for the project in the high-security facility, where aggression was not only common but also a primary means for prisoners to maintain their status and safety during incarceration. The Warden further noted that he had observed significant positive differences between the experimental and control groups of prisoners, George’s rehabilitation being among those positive changes, and expressed his intention to continue the PAL project.
Ever since, that one impassioned cry—”What About George?”—has influenced my approach to research and data collection. Data is an essential part of academic investigations, but it comes in many forms, including qualitative and quantitative results. Both research approaches are necessary to understand the human-animal bond field. Statistical significance is vital, but so is a powerful anecdote. The Human-animal bond field needs both.
[i]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