Animal-Assisted Interventions: A National Survey of Health and Safety Policies in Hospitals, Eldercare Facilities, and Therapy Animal Organizations

Abstract

Background

Animal-assisted intervention (AAI) programs are increasing in popularity, but it is unknown to what extent therapy animal organizations that provide AAI and the hospitals and eldercare facilities they work with implement effective animal health and safety policies to ensure safety of both animals and humans. Our study objective was to survey hospitals, eldercare facilities, and therapy animal organizations on their AAI policies and procedures.

Methods

A survey of United States hospitals, eldercare facilities, and therapy animal organizations was administered to assess existing health and safety policies related to AAI programs.

Results

Forty-five eldercare facilities, 45 hospitals, and 27 therapy animal organizations were surveyed. Health and safety policies varied widely and potentially compromised human and animal safety. For example, 70% of therapy animal organizations potentially put patients at risk by allowing therapy animals eating raw meat diets to visit facilities. In general, hospitals had stricter requirements than eldercare facilities.

Discussion

This information suggests that there are gaps between the policies of facilities and therapy animal organizations compared to recent guidelines for animal visitation in hospitals.

Conclusions

Facilities with AAI programs need to review their policies to address recent AAI guidelines to ensure the safety of animals and humans involved.

Introduction

The use of pets in Animal Assisted Interventions (AAI) for the benefit of human recipients of therapy has become increasingly popular.¹ According to the International Association of Human-Animal Interaction Organizations (IAHAIO), AAI is “a goal oriented intervention that intentionally includes or incorporates animals in health, education and human service (e.g., social work) for the purpose of therapeutic gains in humans.”² For the purposes of this paper, animals that are performing under the premise of the AAI activities described by IAHAIO are referred to as therapy animals.

Therapy animals should be distinguished from service animals or Emotional Support Animals (ESAs). Service animals are defined by the Americans with Disabilities Act as those trained in a specific task (e.g., guiding, signal response or alert dogs) to assist an individual with a disability and are regulated. According to the Americans with Disabilities Act, ESAs are often used “as part of a medical treatment plan as therapy animals…[but are] not considered service animals under the ADA. These support animals provide companionship, relieve loneliness, and sometimes help with depression, anxiety, and certain phobias, but do not have special training to perform tasks that assist people with disabilities.”³

Many healthcare facilities, including hospitals and eldercare facilities, have introduced programs that promote interactions between residents or patients and therapy animals in AAI. These programs can result in positive health outcomes, including reductions in blood pressure, improved mood, and delayed onset of dementia.^1,4–6 However, these programs are not without some risk if not conducted carefully, including patient allergies, fear of animals, bites, and potential for zoonotic disease transmission.⁴ Several studies have identified pathogens carried by animals, particularly those fed raw meat diets, that may pose risks to immunocompromised patients, including Campylobacter, Salmonella, and Cryptosporidium.^7–9 Studies have shown, for example, prevalence rates for contamination of commercial raw meat diets with Salmonella spp. between 21–48%,¹⁰ with high rates of resistance to antibiotics.⁸ Even in therapy animals, in which one would expect higher levels of safety, one study found that zoonotic agents could be isolated from 80% of therapy dogs, including Clostridium, Giardia, and Salmonella.¹¹ Beyond recommended guidelines from working groups,^12–14 no human or animal health regulatory agencies are currently responsible for monitoring or regulating AAI programs. The Society for Healthcare Epidemiology of America (SHEA) has produced guidelines for animals in healthcare facilities, which include important steps such as establishment of written policies, designated AAI visit liaisons, and formal training programs for both animals and handlers. However, there is no legal incentive to establish such parameters in healthcare facilities. Furthermore, therapy animal organizations are self-regulated and thus have no mandated behavior training or health requirements. Standards range from very rigorous (i.e., some national therapy animal organizations which require that volunteers be trained, undergo recurring evaluation with their animal, and attend ongoing education for professionalism, safety, and animal welfare, and have rigorous animal health and grooming requirements) to non-existent.

Among the handful of national organizations that register handlers and therapy animals, Pet Partners^® is the only national therapy animal organization that requires volunteer training, recurring evaluation of animal-handler teams every two years, as well as prohibiting raw meat diets.¹⁵ Other AAI organizations do not require all of these standards, and specific policies vary widely between groups. For example, while one organization prohibits animals eating raw meat diets from being registered,¹⁵ another explicitly allows it,¹⁶ and another has no policy on the issue.¹⁷ Similar variation in other practices of the national therapy animal organizations (e.g., training, re-evaluation of teams, etc.) also exists. Anecdotally, the practices of the regional and local therapy animal organizations appear to be even more variable.

Although results of a limited sample of healthcare facility policies related to animals in the United States were recently reported,¹⁴ there is a dearth of studies documenting national trends in health and safety policies for AAI in hospitals and particularly in eldercare facilities. It is essential to understand if healthcare facilities are incorporating elements of the recent SHEA guidelines into their existing AAI policies. In addition, it is important to understand the standards that are used by regional and local therapy animal organizations. Vulnerable patient populations exist not only in hospitals, but also in eldercare facilities where AAI has become popular. It is thus important to more fully understand where existing policies fall short of best-practice guidelines for protecting patients or residents and animals from harm. Therefore, the aim of this study was to investigate the AAI program policies in hospitals and eldercare facilities across the United States, as well as policies and procedures of regional and local therapy animal organizations.

Materials and Methods

Two surveys were conducted: one to hospital and eldercare facilities and one to therapy animal organizations across the United States about existing policies related to animal health and behavioral prerequisites for therapy animals and AAI programs. The study protocol was determined to be excluded from review by the University Institutional Review Board.

National Hospital and Eldercare Facility Animal Policy Survey

The first cross-sectional phone/email survey was conducted in hospital and eldercare facilities. Hospitals included public, private, and teaching hospitals. Eldercare facilities included independent living communities, assisted living facilities, and full-time nursing care facilities. These facilities were located in nine geographic regions of the United States, which were defined using the American Pet Products Association Pet Owners Survey: Pacific, Mountain, West North Central, West South Central, East North Central, East South Central, New England, Mid Atlantic, and South Atlantic.¹⁸ One state was randomly selected from each of the nine regions, numbering the states 1 through x (number of states in the region) and using a computer generated random number selector to correspond with one of the numbered states. Within the selected state, five eldercare facilities and five hospitals were selected using a web search for “eldercare facilities in [state]” and “hospitals in [state]”. Multiple search engine pages were scanned and a random sample was selected, in which facilities were listed in alphabetical order, numbered, and selected using a computerized random number generator as described above. Upon selection from the list, hospitals or eldercare facilities that had neither a listed telephone number nor an email address on their website or online directory were excluded for the purposes of this study, and a randomly selected replacement from the list of facilities was used. However, if a facility had only an email or a phone, the facility was included and only that method was used to contact them. This process was repeated until five facilities from each region were selected.

Facilities were contacted by telephone (if available) and study investigators asked to speak to the staff member responsible for AAI policies (which was often the volunteer program coordinator at the hospital or eldercare facility) or the eldercare facility manager. For those facilities with existing AAI programs, the volunteer coordinator was contacted to discuss animal and handler visitation policies. If the facility did not have a designated AAI coordinator on staff, any member of staff familiar with pet visitation policies could be contacted. If no telephone number was available for a selected facility, an email was sent to volunteer services (when an address was listed on the website) or to general inquiries.

Participants were provided a brief introduction from the researcher, verbally or in the form of an email, explaining the confidential nature of the study and the goal to improve current understanding of national AAI policies. Participants were first asked whether their facility had an “animal visitation policy.” If yes, details on animal health requirements and behavioral assessment were requested. Both email and telephone responses were compiled and tallied into six policy strength categories: “no animals allowed,” “no existing policy,” “verbal confirmation that the animal was healthy,” “written confirmation of animal health,” “written confirmation of animal health and meeting animal/handler team,” or “registered animals only allowed.” Initial telephone calls were made between November-December, 2013 with one follow up call for non-respondents in January, 2014. Email surveys (to facilities without phone numbers) were distributed December, 2013, with one follow-up reminder in January, 2014. Failure to respond after a follow-up call or email (n=17 facilities of those contacted) resulted in disqualification of the facility from the study, and the randomized selection process was repeated until responses from five eldercare facilities and five hospitals from the selected state were obtained.

Therapy Animal Organization Policy Survey

The second national cross-sectional study was conducted assessing regional and local therapy animal organizations using the same nine states randomly selected for the facility survey. Three therapy animal organizations in each of the nine states were randomly selected using an internet search containing the name of the state and “Animal Therapy”. Websites generated by the search were visited and listed certification/registration credentials were noted; groups selected for this survey were those that did not state affiliation with a major national therapy animal organization (Pet Partners^®, Therapy Dogs International^®, or American Kennel Club^™ Therapy Dog Program). An introductory email including a hyperlink to the anonymous online survey was sent to each therapy animal organization’s contact email. Groups without readily available email contact information were excluded, and a re-selection process, using the random selection method described above, was used to contact a different organization to reach target participant numbers.

The online survey consisted of nine questions (Appendix A), which was derived from a previous study assessing the role of veterinary education in safety policies for animal-assisted therapy and activities in hospitals and nursing homes.¹⁹ These questions were phrased to allow a “yes” or “no” response to each. Space for elaboration on a response was provided but not required. Respondents were asked about their organization’s policies on animal health status and records, types of animals permitted in the program, and obedience testing criteria. Groups were asked to identify the state in which they were located, but no other identifying information was required. Email surveys were distributed December, 2013, with one follow-up reminder in January, 2014. Failure to respond after a follow-up email (n=5 organizations of those contacted) resulted in disqualification of the organization from the study, and the randomized selection process was repeated until responses from three therapy animal organizations from the selected state were obtained.

Data Analysis

Descriptive data were generated using commercial statistical software (Microsoft Excel v. 16.0, Excel, 2016, California, USA). “Likelihood ratio (LR) chi-square test was used for comparison of proportions between hospital policies and eldercare facility policies using commercial statistical software (SAS V. 9.4, SAS Institute, Cary, NC). The LR chi-square test was used because of its reduced sensitivity to small numbers of observations in cells compared to the standard Pearson chi-square test and is less conservative than Fisher’s Exact test.”

Results

Facilities Survey

A total of 45 eligible eldercare facilities and 45 eligible hospitals across nine selected states participated in the survey. Not all facilities allowed therapy animals, with 8/45 (18%) of hospitals and 1/45 (2%) of eldercare facilities allowing admittance to service animals only. Of those allowing animals, 100% of facilities fell short of at least one aspect of the recently published SHEA Guidelines for animals in healthcare facilities,¹⁴ based on information provided by facilities.

Surveyed facilities varied widely in the rigor of their AAI policies (Table 1). Some facilities [2/45 (4%) of hospitals and 10/45 (22%) of eldercare facilities] had no policy whatsoever for AAI. If no animals were permitted in the facility, this was considered to be a type of animal policy. Of those with a policy that did not forbid therapy animals (n=78), eldercare facilities had fewer animal health and behavior requirements than hospitals. For example, 38/45 (84%) of eldercare facilities required only a minimal written health record for the therapy animal. On the contrary, 34/45 (76%) of hospitals required at least a meeting and certification/registration from a therapy animal organization prior to participation.

Table 1.

Summary of policies for animal-assisted interventions programs in 45 hospitals and 45 eldercare facilities from nine regions of the United States.

Minimum requirements for animal-assisted interventions program	Hospitals, Number (%)	Eldercare facilities, Number (%)	% Facilities, Total	Hospital vs. Eldercare Chi-square (p-value)†
No requirements for documentation of animal health	2 (4%)	10 (22%)	13%	6.64 (0.01)
Verbal confirmation that animal was healthy	2 (4%)	10 (22%)	13%	6.64 (0.01)
Written confirmation that animal was healthy	7 (16%)	18 (40%)	28%	6.88 (0.009)
Meeting and written health report*	3 (7%)	3 (7%)	7%	0.00 (1.00)
Registered animals only**	23 (51%)	3 (7%)	29%	23.80 (<0.001)
Policy prohibiting all animals	8 (18%)	1 (2%)	10%	6.80 (0.009)

^*Administration required an initial meeting with animal and handler to assess appropriateness of behavior first-hand.

^**Animals were required to have an affiliation with a therapy animal organization. These were not restricted to any particular national, regional, or local organization affiliation.

^†Likelihood ratio chi-square test value and P-value for comparison of proportions between hospital policies and eldercare facility policies.

Therapy Animal Organization Survey

A total of 27 regional and local therapy animal organizations participated in the online survey (three from each of the nine geographical regions). Animal health was considered by all therapy animal organizations, though depth of policy varied among groups (Table 2). For example, the majority of surveyed therapy animal organizations (20/27, 74%) required a veterinary examination prior to permitting animals to participate in AAI; however, some therapy animal organizations (2/27, 7%) did not even have a rabies vaccination requirement. One quarter of respondents did not explicitly require a fecal test from a participating animal (26%). Almost every organization (96%) permitted animals with disabilities to participate in their program. For behavioral qualifications, 9/27 (33%) of the therapy animal organizations required only an American Kennel Club^™ Canine Good Citizen^® certificate. This is a 10-step behavioral certification course, which is not intended to demonstrate suitability for any kind of AAI. However, other respondents elaborated on additional tests including behavioral and stress testing, assessment by an animal behaviorist, and a handler-dog team evaluation by the program director in one instance. Although behavioral assessments were required for registration in many groups, only 52% of respondents (14/27) required regular retesting of animals. In direct contrast to the recent SHEA Guidelines, the majority of respondents (19/27, 70%) did not have a policy against feeding raw meat diets to therapy animals and only 19% (5/27) specifically had policies specifically prohibiting it (3 organizations chose not to answer this question, so their policy is unknown). Although 70% of respondents (19/27) stated that liability insurance was provided by their organization, two of them elaborated that handlers needed to provide their own liability insurance. Finally, almost half of all respondents (43%) permitted non-dog species in their program, including cats, rabbits, ferrets, horses and guinea pigs. The health requirements for these additional species were not investigated in this study.

Table 2.

Summary of health and safety policies in 27 regional and local therapy animal organizations

	Number of Organizations, (%)
Permits animals with disabilities	26 (96%)
Requires rabies vaccination	25 (93%)
Requires veterinary examination	20 (74%)
Requires fecal test	20 (74%)
Prohibits raw meat diet	5 (19%)1
Provides liability insurance††	19 (70%)1
Permits non-dog species*	13 (48%)
Requires only Canine Good-Citizen® certificate**	9 (33%)
Requires regular behavioral retest†	14 (52%)

^*Examples of other species included cats, guinea pigs, rabbits, and horses.

^**A 10-step behavioral certification course of the American Kennel Club^®

^†Of the 14 groups requiring retest, four reported requiring an annual retest, the others did not note frequency.

^††Two survey respondents selected “we provide liability insurance” but noted in writing that participants needed to provide their own insurance.

¹Three organizations did not answer this question.

Discussion

One of the key findings of this study was the large policy discrepancies across hospitals and eldercare facilities that allow AAI programs. Even critical basic policies such as written documentation of up-to-date rabies vaccination status and prohibition of raw meat diets were not consistently required. Liability insurance was reported to be provided to animal therapy organization handlers in 70% of participating groups; however, the actual number is likely less considering that two organizations also elaborated that this insurance had to be provided by the handlers themselves. Furthermore, the level of coverage was not asked as a part of this study. A lack of comprehensive liability insurance puts both the organizations and healthcare facilities at risk of complicated legal consequences in the event of an animal-related injury or illness.

Standardized guidelines for safety, health, and monitoring of AAI by national hospital or eldercare facilities are limited. Recognizing the need for further guidance, basic guidelines for AAI were originally created by a working group from the American Veterinary Medical Association (AVMA) in 1999 and were revised several times since, most recently in 2015, to reflect updated knowledge of AAI.¹³ In 2008, a working group of veterinary and public health stakeholders produced additional guidelines for AAI,¹² but it was not until 2015 (after completion of the current study survey) that SHEA created comprehensive guidelines in which veterinary, medical, epidemiologic, infection control, and public health professionals collaborated on guidelines for animals in healthcare facilities.¹⁴ Within the SHEA Guidelines, a small sample of healthcare facility policies (n=18) were assessed, with 38% of facilities requiring certification of pet’s immunization status and good health, whereas our survey found that 23% of hospitals with an AAI policy required written proof of immunization status at a minimum, and an additional 51% of hospitals that left animal health policies to registered animal therapy groups.¹⁴ Compared to the current survey, a similar percentage of hospitals in the SHEA survey did not allow pets at all (18% and 17%, respectively). However, almost all hospital policies from the SHEA survey required regular evaluation of the animal by a veterinarian to confirm absence of tickborne or other bloodborne pathogens, vaccination status, and good health.¹⁴ Though our survey inquired about veterinary care, it did not evaluate requirements for laboratory analyses such as tickborne disease or bloodborne pathogen testing.

Despite the availability of AAI guidelines for facilities from the AVMA at the time of this study (SHEA guidelines were not yet available),¹³ our findings suggest that information in any of these guidelines appears not to be widely followed by hospitals and nursing/eldercare facilities. Areas of particular concern that were highlighted in the current survey findings include: lack of specific health and vaccination requirements, behavioral assessment regulations, and inquiry as to the diet of the therapy animal. Since the time of this study, SHEA guidelines have been made available and may provide additional guidance to healthcare facilities and therapy animal organizations. Comparing the two sources of guidance, SHEA guidelines are specifically written for hospitals and are not intended for use by eldercare facilities, and lack specific veterinary recommendations. On the other hand, the AVMA outlines specific criteria for a veterinary practitioner examining a therapy animal, such as parasite screening and control, rabies vaccination recommendations, behavioral health assessment, and consideration of exposure in human facilities that could be harmful to the animal. The AVMA guidelines, if followed, would help to prevent the spread of zoonotic disease (both from the therapy animal to the patients/residents and vice versa) and advise handlers on how to monitor for and minimize therapy animal stress and fatigue. Despite their differences, both sets of guidelines provide a basic standard of care in order to provide safe and effective AAI programs. While not providing specific guidance for the veterinary practitioner, the SHEA guidelines do advise formal training of both animals and handlers, proper hand hygiene and disinfection of surfaces, and health screening for handlers as well as animals. Therefore, SHEA guidelines provide a helpful baseline of recommendations for healthcare facilities implementing AAI, but AVMA guidelines more thoroughly address ways to minimize pathogen spread, behavioral incidents, and animal welfare problems.

From our study results, a lack of healthcare requirements among therapy animal organizations illustrates an important gap in knowledge and the potential for harm. Healthcare facilities must recognize this wide spectrum of policies amongst animal therapy organizations and become versed in questions for handlers prior to admittance to the facility. Though published guidelines that include all types of facilities are not currently available, the Tufts Institute for Human-Animal Interaction has developed a manual for facilities with minimum health, safety, and training requirements for handlers and therapy animals visiting all types of facilities.²⁰ Some examples include: liability insurance for the handler, adequate training of the handler, prohibition of animals being fed a raw meat diet, proper annual health screenings of animals by a veterinarian, and regular re-evaluation of the animal and handler team by the therapy animal organization.

The current study has several limitations that should be considered in interpreting the results. The study was limited in sample size given the large number of hospitals, eldercare facilities, and therapy animal organizations in the United States, though it was designed to be a representative sample. The study also excluded facilities that had neither telephone nor email contact information available on their websites, which may have resulted in the exclusion of facilities or organizations that are smaller and potentially less likely to have stringent policies in place. A further consideration is that there is no guarantee that policies reported by participants either in facilities or therapy animal organizations are being adhered to at all times. It is possible actual adherence and practice of policies or standards may result in practices that are worse than reported. Furthermore, facilities were not surveyed on their knowledge of appropriate health and safety measures in AAI programs (e.g., AVMA or the more recent SHEA guidelines which were not available at the time of the survey). Future studies should ask facilities and therapy animal organizations about their knowledge of existing guidelines to further understand how these guidelines should best be disseminated.

One issue that not addressed in the survey was animal welfare considerations. Visitation can be a tiring and stressful ordeal for therapy animals, therefore animal temperament in such situations should be tested (and retested at regular intervals), handlers should be trained to monitor for and minimize stress, and policies should be in place to remove animals from situations in which they display behavioral signs of stress, exhaustion, fear, or aggression. Animal welfare considerations should be addressed in future studies, given the availability of behavioral information in AVMA guidelines.

Conclusion

The findings of this study should serve as a call to action for hospital and eldercare facilities and therapy animal organizations to use evidence-based policies to strengthen the safety measures of their AAI programs. Our results demonstrate that both therapy animal organizations and facilities engaging in AAI programs may benefit from educational interventions on basic standards that improve health and safety for both humans and animals, so that the benefits of AAI may continue to outweigh the risks.

Highlights.

• Eldercare facilities, hospitals, and therapy animal organizations were surveyed.

• Health and safety policies for animal-assisted interventions (AAI) varied widely.

• Some AAI policies potentially compromise human and animal safety.

• In general, hospitals had stricter AAI requirements than eldercare facilities.

• Programs should institute recent AAI guidelines to ensure human and animal safety.

Abbreviations

AAI

Animal-assisted interventions

AVMA

American Veterinary Medical Association

ESA

Emotional Support Animal

IAHAIO

International Association of Human-Animal Interaction Organizations

SHEA

Society for Healthcare Epidemiology of America

Appendix A. Health and safety policy survey administered to 27 therapy animal organizations

1. In what state are you located?

2. Do you require animals to be examined by a vet prior to participating in visits?

   • Yes

   • No

   • Depends, please specify:

3. Do require an up-to-date rabies vaccination and negative fecal test for participating animals?

   • Yes, both.

   • Only rabies.

   • Only negative fecal.

   • Neither.

   • Additional comments (optional):

4. Do you allow animals other than dogs to participate in your program?

   • No

   • Yes, please specify:

5. Do you allow animals with disabilities (eg. blind, leg amputation) to participate in your program?

   • Yes

   • No

   • Depends, please specify:

6. Do you allow animals to participate in visitation if they can pass a test of basic obedience skills or have a Canine Good Citizen certificate?

   • Yes, this is sufficient for participation.

   • No, this is not sufficient for participation.

   • Additional comments (optional):

7. Do you require retesting of the animal to continue visitation?

   • No.

   • Yes. If yes, how often do you require a retest?

8. Do you prohibit raw meat diets from being fed to your therapy animals?

   • Yes, raw diets are forbidden.

   • No, we allow raw diets.

9. Do you provide liability insurance to your participants?

   • Yes.

   • No.

   • Additional comments (optional):