Long-Term Psychological Benefits from Companion, Therapy and Service Dogs Questioned in Recent Scientific Papers

Research on the psychological benefits animals provide to us can be assigned to studies of (1) pets, (2) therapy animals, and (3) service animals.  Most of the studies involve dogs, though the second category in particular has some serious equine studies. 

Studies of pets usually refer to them as companion animals.  In some cases a distinction is being made or implied to suggest that the study is of animals to which the owners have an emotional connection, though I often think �companion� is preferred so that scientific papers can have a multisyllabic adjective in the title. 

Therapy animals�as I say, usually therapy dogs�may involve studies of relatively casual interactions with therapy dogs that visit patient or resident populations, often at institutions such as mental health facilities, nursing homes, hospitals, etc.  More often, however, therapy dogs have been studied when their participation in a treatment regimen involves relatively specific time frames and objectives with a specific patient or patient group, which is often called animal-assisted therapy. 

Service animals�again primarily service dogs�have been studied less than the first two groups for their psychological benefits, partly because many of the more traditional service dog categories (guide dogs, hearing dogs) are not designed to have psychological benefits.  With the rapid growth of dogs trained for patients, particularly military patients with PTSD, and with autism service dogs appealing to many parents of autistic children, this category is growing rapidly and is beginning to be the subject of intense scientific analysis. 

Some recent review articles covering these three categories of dogs sound a note of caution, even concern, in that some have found that dogs are not always beneficial psychologically or even physically, despite the fact they are trained or introduced to patients in contexts where they are supposed to be, and in some studies the dogs have even been found to work against treatment objectives.  The review articles also discuss recent research that has taken on some of the long-standing assumptions about dogs, e.g., that people with pets recover more quickly after heart attacks than people without pets.  Particularly disturbing is that much prior research is now being criticized for faulty design and overly subjective interpretation of results.

The review papers discussed below, however, are not against the use of dogs, and do not insist that there are no psychological benefits to dogs. They do suggest that to provide scientific support for certain arguments that have been made�e.g., that soldiers and veterans with PTSD need dogs to reduce psychiatric symptomatology�more rigorous research needs to be conducted. 

Negative Psychological Effects of Pets Noted in Australian Study

Studying companion animal relationships in Australia is made easy by the fact that 40% of the country�s households include a dog and 26% a cat.  A recent paper by Jasmin Peacock, Anna Chur-Hansen, and Helen Winefield (2012), though not a review paper, included a lengthy analysis of prior research on companion animals and mental health and concluded that much of it was characterized by methodological weakness:

�[F]ew controlled studies have been conducted to provide empirical support for positive physical or mental health outcomes gained from interacting with companion animals. Previous research has been largely descriptive and conducted with specific populations of convenience such as the aged.� 

These authors note that some prior studies indicate that companion animals may actually exacerbate psychological symptoms, cause higher levels of depression, and increase emotional distress and psychoticism (a personality pattern typified by aggressiveness and interpersonal hostility).  Threat of separation from a companion animal may lead to rejection of medical advice and failure to move out of inappropriate housing situations.   Both from prior studies and the part of the paper that includes original research, the authors conclude that �strong attachment bonds to a companion animal might not necessarily be beneficial and, in some circumstances, might potentially lead to poor health outcomes.� 

They also make an observation relevant to the issue of the psychological effect of service animals:

�The level of attachment between service animals (such as guide dogs for the vision impaired) has attracted little attention in the scientific literature. Given the important role service animals play, a better understanding of the working animal-human bond would be instructive.�

�Pet Effect� Called an Uncorroborated Hypothesis

Harold Herzog (2011), a professor at Western Carolina University, argues that �a generalized �pet effect� on human mental and physical health is at present not a fact but an unsubstantiated hypothesis.�  Herzog says that in modern America, �the public has come to accept as fact the idea that pets can also serve as substitutes for physicians and clinical psychologists.�  This can be called �the pet effect,� a term Karen Allen (2003) used to describe the popular idea that living with an animal improves human health, psychological well-being, and longevity. 

Herzog makes a crucial observation that has largely escaped other researchers, which is that there has been a strong media bias.  People like to read stories about how good their pets are for them, but �studies in which pet ownership has been found to have no impact or even negative effects on human physical or mental health rarely make headlines.�  He notes a study (Parker et al., 2010) which looked at 425 heart-attack victims and �found pet owners were more likely than non-pet owners to die or suffer remissions within a year of suffering their heart attack (22% vs. 14%).�  This throws into doubt Friedmann et al. (1980), a study on which I made much in Service and Therapy Dogs in American Society (pp. 8, 95). 

The lack of media attention may also explain part of another phenomenon, which Herzog calls the �file drawer effect.�  This is a tendency of researchers not to publish results that are not positive towards animal ownership:

�At a session at a 2009 conference on human�animal interactions, for example, one researcher reported that separation from their pets had no effect on the psychological adjustment of college students, another found that interacting with animals did not reduce depression in psychiatric nursing home residents, and a third found no differences in the loneliness of adult pet owners and non-owners. So far, none of these studies have appeared in print.�

Herzog cites studies finding that pet owners are just as lonely as non-pet owners as determined by a standardized �loneliness scale,� were no happier than non-pet owners, and might even be more depressed if they were highly attached to a dog.  One longitudinal study of nearly 12,000 American adults (Gillum and Obisesan, 2010) found that cat or dog ownership was unrelated to mortality rates. 

Herzog also discusses methodological problems with human-animal interaction studies, noting that self-reporting approaches may not be valid.  One study of people suffering from chronic fatigue syndrome (Wells, 2009) found that pet owners described numerous psychological and physical benefits to having pets, yet their scores on standardized measures �indicated that they were just as tired, depressed, worried, and stressed as chronic fatigue sufferers who did not get a pet.� 

Pet Therapy with Children Shows Positive Results, but Broad Conclusions Are Premature

A paper by Layla Esposito, James A. Griffin, and Valerie Maholmes (2011), begins with a statement on the need for more research �on the physical and psychological health benefits that can accrue to children through their interactions with pets, both in daily life and in therapeutic settings.�  They note that as far back as a 1987 National Institutes of Health conference, there has been a belief that more needs to be done, and that this is still true today. 

�To delve into the mechanisms of HAI [human-animal interaction] effects on health and determine who benefits from pet interaction and under what circumstances, larger, more focused studies that include crossover designs (in which subjects receive a sequence of different treatments) and look at environmental variables and a variety of pet species are needed. Well-controlled experiments and research that extends beyond studies of short-term physiological responses are also needed to provide more conclusive evidence of effects on health. Studies aiming to more precisely detect physiological changes that occur in people in the presence of pets could be more powerful if researchers would agree to use a common model and then conduct path analyses to show causal connections among the variables. The fact that studies have been done independently without common data elements limits the extent to which contradictions in findings can be explained.�

Need for Better Research Methods to Study Benefits of Service Animals

Melissa Winkle, Terry Crowe, and Ingrid Hendrix (2012) begin their recent paper with the observation that �few rigorous studies� exist regarding the utility of dogs as an �assistive technology option.�  The authors searched the literature for service animal studies using keywords and found 432 papers that might have been relevant.  After eliminating studies that were anecdotal, reviews, primarily qualitative, and dissertations, they were left with only 23, and 11 more were eliminated for focusing on issues the authors were not concerned with, leaving only 12. 

Unfortunately, in the opinion of the authors of this overview study, all 12 papers they looked at closely �had research design quality concerns including small participant sizes, poor description of the interventions, outcome measures with minimal psychometrics and lack of power calculations.�  The authors acknowledge that some of the weaknesses in prior studies are difficult to overcome, saying, for instance, that it �is difficult to conduct a blind investigation of the benefits of service dogs, so it is impossible to rule out the contribution of participant expectations.� 

There was also a lack of uniformity in the training of the dogs, making comparisons across studies essentially impossible.  �Without knowing the length and quality of the specialized training of the service dog, criteria for dog or person placement readiness or the content, length and quality of the training for dog or person, it is impossible to replicate these studies.�

The authors concede that service dogs were perceived by users as allowing them to decrease their reliance on other people. The studies reaching such conclusions, however, did not look at the opinions of caregivers independently of users. 

As for psychological benefits, studies reported �significant increases in self-esteem, internal locus of control, well-being, and positive affect.�  Persons with progressive conditions such as multiple sclerosis and Parkinson�s disease demonstrated significantly higher positive affect scores, with service dogs moderating the effects of depression.  Psychosocial characteristics did not differ significantly between those partnered with service dogs and those without.

The authors conclude that in order for occupational therapists to make recommendations for the use of service dogs, �the evidence to support such decisions must be strengthened.�  They note that this is particularly critical because of the limited availability of dogs:

�Given the extreme shortage of trained dogs and the potential cost of the dog/person partnership training and care, predicting positive outcomes based on person and dog characteristics are vital. Finally, if we are to recommend service dogs (or any kind of assistance dog) as an assistive technology option, we must study the dog-person evaluation and matching process, training and placement procedures and content, and outcomes for both the person and the dog, across assistance dog training organizations.�

Conclusions

The authors of these overviews do not say that there are no long-term psychological benefits to dogs, but they do say that claims of such benefits need more scientific support than they have received so far.  It is to be hoped that future research, such as that being conducted by NEADS, a service dog training organization, to determine the benefits of service dogs for veterans with PTSD, will have the sort of scientific objectivity that will withstand the criticisms of the authors of the papers discussed here. 

These studies will not stay in ivory towers.  Lawyers will scrutinize them for arguments about whether a tenant who wants to keep his assistance animal can really establish that the animal is providing him with psychological benefits such that a landlord must make a reasonable accommodation to a no-pets policy.  People wanting to take their service dogs to work will have to be more careful about the benefits they claim to receive from the dog, because if the matter goes to trial the lawyer for the employer is likely to bring up such studies in cross-examining the service dog user�s psychologist or doctor. 

The research should also be reviewed closely by mental health and medical professionals who are asked to sign letters for patients who want to live with service animals or take such them onto airplanes.  Professionals should be particularly careful about adopting draft language for such letters when the language is reverse engineered from regulations or legal decisions in order to convince an airline or a landlord that a service animal is providing concrete psychological benefits.  A letter that says the patient is getting better by having the service animal is more than an airline or a landlord will need and may not be supportable by the research.  Given that airlines are now being given the ability to contact a professional�s licensing authority regarding a service dog support letter, medical and psychological professionals must be increasingly cognizant of what they sign. 

Sources:

1. Allen, K. (2003).  Are Pets a Healthy Pleasure: The Influence of Pets on Blood Pressure. Current Directions in Psychological Science, 12, 236-9.
2. Esposito, L., McCune, S., Griffin, J.A., and Maholmes, V. (2011). Directions in Human-Animal Interaction Research: Child Development, Health, and Therapeutic Interventions.  Child Development Perspectives, 5(3), 205-211.
3. Friedmann, E., Katcher, A. H., Lynch, J. J., and Thomas, S. A. (1980). Animal Companions and One-Year Survival of Patients after Discharge from a Coronary Care Unit. Public Health Reports, 95(4), 307.
4. Gillum, R.F., and Obisesan, T.O. (2010). Living with Companion Animals, Physical Activity and Mortality in a US National Cohort.  International Journal of Environmental Research and Public Health, 7, 2452-9.
5. Herzog, H. (2011). The Impact of Pets on Human Health and Psychological Well-Being: Fact, Fiction, or Hypothesis?  Current Directions in Psychological Science, 20(4), 236-239.
6. Parker, G., Gayed, A., Owen, C., Hyett, M., Hilton, T., and Heruc, G. (2010). Survival Following an Acute Coronary Syndrome: A Pet Theory Put to the Test. Acta Psychiatrica Scandinavica, 121, 65�70.
7. Peacock, J., Chur-Hansen, A., and Winefield, H. (2012). Mental Health Implications of Human Attachment to Companion Animals. Journal of Clinical Psychology, 68(3), 292-303 (March 2012).
8. Wells, D.L. (2009). Associations between Pet Ownership and Self-Reported Health Status in People Suffering from Chronic Fatigue Syndrome. Journal of Alternative and Complementary Medicine, 15, 407-413.
9. Winkle, M., Crowe, T.K., and Hendrix, I. (2012). Service Dogs and People with Physical Disabilities Partnerships: A Systematic Review.  Occupational Therapy International, 19(1), 54-66.