Abstract
People with dementia who live in nursing homes, assisted living facilities, or other long-term care (LTC) facilities are at heightened risk for poor quality of life (QOL), given the debilitating nature of their disease coupled with their escalating environmental vulnerability. As identified by Lawton (1989; Lawton & Nahemow, 1973) and echoed by Padilla (2011b), older adults with diminishing competencies resulting from Alzheimer’s disease and related neurodegenerative dementias over time grow less able to cope with situations that exceed their capacities. Stated differently, the living environments of such persons more strongly influence how they act and what they experience over time. Consequently, there is extensive interdisciplinary consensus that the everyday environments of people with dementia profoundly influence their QOL, and there is interdisciplinary support for nonpharmacological environment-based interventions to enhance QOL (Padilla, 2011b; Sloane et al., 2005; Small et al., 1997). These interventions do not rely on antipsychotic drugs to manage the anxiety, agitation, depression, apathy, or delusions that can occur with dementia (Padilla, 2011a). Rather, on the basis of presumed likely mismatches between the environment of people with dementia and their coping capacities, these interventions use strategies that modify social and physical environmental elements. The focus of this article is animal-assisted therapies (AATs), a type of nonpharmacological environment-based interventions that health professionals have provided to institutionalized people with dementia.
Preliminary evidence has suggested that AATs are promising practices for this population. Cipriani et al. (2013) conducted an evidence-based review of research into dog-assisted therapies for LTC residents. Most participants in the reviewed studies had a diagnosis of Alzheimer’s disease, a related dementia, or a cognitive disorder. Cipriani et al. found outcomes that pertained to performance skills and areas of occupation recognized in the second edition of the Occupational Therapy Practice Framework: Domain and Process (American Occupational Therapy Association, 2008). These investigators concluded that the respective goals and strategies of dog-assisted therapy and occupational therapy were connected. Moreover, this review and two others by Bernabei et al. (2013) and Filan and Llewellyn-Jones (2006) provided preliminary evidence that various AATs held promise for enhancing the QOL of people with dementia. Specifically, AATs may elicit and improve social interactions with both animals and people; they may also help to reduce apathy, stress, and problematic behaviors.
Assessments of the quality of evidence across the aforementioned reviews indicate that intervention research concerning AATs for institutionalized people with dementia remains in early scientific development. We therefore conducted a systematic mapping review of articles on AAT for this population. Systematic mapping reviews are a method of choice when a need exists to map the literature on a specific area of inquiry that is in early development to generate preliminary guidelines for practice and research (Grant & Booth, 2009; Hammick, 2005). The Lived Environment Life Quality (LELQ; Wood, Lampe, Logan, Metcalfe, & Hoesly, 2017) Model guided data extraction in our mapping review, as well as our ultimate map of included literature.
Lived Environment Life Quality Model
The LELQ Model is a conceptual practice model designed to guide occupational therapy services for institutionalized adults with dementia (Wood et al., 2017). In the model, caregiving microsystems refer to the face-to-face settings of homes that are institutional in nature, such as nursing homes, assisted living facilities, or other residential care facilities. The model poses that transactions among people with dementia and their caregiving microsystems give rise to particular environmental presses. These environmental presses serve to shape or otherwise influence particular QOL experiences, which may range from occupationally enlivening to occupationally deadening (Figure 1). Derived from Lawton (1989), the concept of environmental press, as incorporated in the LELQ Model, refers to environmental characteristics or stimuli that ordinarily press people with dementia to act in particular ways and have particular social, emotional, and embodied experiences. Environmental presses that are more occupationally enlivening impel people toward spending time purposefully engaged and ably functioning on some level while experiencing some well-being (e.g., by attending to something of interest; by feeling safety, emotional warmth, pleasure, or gratitude; by experiencing a sense of oneself and of belonging). Environmental presses that are more occupationally deadening drive people toward passing time unengaged, not ably functioning, and experiencing feelings of ill-being such as loneliness, apathy, despair, sadness, anger, anxiety, agitation, or hopelessness.
The Lived Environment Life Quality Model.
In the LELQ Model, therefore, the lived environment domain is the focus of assessment and intervention and includes two subdomains: the caregiving microsystem and the person with dementia (Wood et al., 2017). The model prompts practitioners to gather robust occupational profiles of people with dementia and to offer environmental interventions that respect those profiles and occur where people spend considerable time (e.g., common areas, dining rooms, bedrooms or bathrooms, patio or garden areas). The QOL domain is the focus of intervention goals and outcomes and includes three subdomains of dementia-specific QOL indicators: time use, ability to function, and relative being. Altogether, the model serves as a guide to client-centered environmental interventions that aim to optimize occupational engagement, functional abilities, and well-being. Because occupation is central to the LELQ Model’s conceptualization of occupational profile, environmental press, and life quality, the model also promotes practices that are occupation focused, as defined by Fisher (2014).
This systematic mapping review aimed to comprehensively gather, describe, categorize and synthesize refereed papers on AAT and institutionalized adults with dementia to guide animal-assisted services and identify critical research gaps. We asked this question: In what ways does the literature about AAT for institutionalized adults with dementia map (or not map) onto the domains and subdomains of the LELQ Model?
Method
The systematic mapping review is one of 14 review types in the family of systematic review research, and it is used as an important first step in developing evidence-based practices and clarifying critical research gaps (Grant & Booth, 2009). Unlike other systematic review methods, systematic mapping reviews are broad and inclusive in scope. This type of review is a method of choice when knowledge pertaining to an area of inquiry is in early scientific development and has not been comprehensively gathered and synthesized as a guide to practice and research (Hammick, 2005). Systematic mapping reviews, therefore, may include non–research articles, and the rigor of individual studies is not assessed as a basis for inclusion. As occurs with other systematic review methods, however, the method of a systematic mapping review involves the application of three filters: search procedure, inclusion and exclusion criteria, and data extraction.
Filter 1: Search Procedure
A library scientist (the fourth author, McLure) constructed and executed comprehensive searches in nine databases to serve multiple projects related to animal-assisted interventions: AGRICOLA (EBSCO), CAB Abstracts (EBSCO), CINAHL (EBSCO), PsycINFO (EBSCO), PubMed (National Center for Biotechnology Information), Social Services Abstracts (ProQuest), Social Work Abstracts (EBSCO), SPORTDiscus (EBSCO), and Web of Science (Thomson Reuters). Search terms were selected to retrieve articles related to animal–human interactions in a therapeutic context. The searches were restricted to English-language articles published between 1980 and 2015 and, when facilitated by the given database, to retrieve peer-reviewed journal articles specifically. The database searches, which were run in February (PsycINFO and CINAHL) and April 2016 (all remaining databases), identified 4,034 articles. After 2,692 duplicates were excluded, 1,342 unique articles remained and were aggregated in one EndNote library for screening. EndNote allows for organization and management of references in review research (King, Hooper, & Wood, 2011).
Filter 2: Inclusion and Exclusion Criteria
Application of inclusion and exclusion criteria occurred in two phases. In Phase 1, we screened the 1,342 unique articles for inclusion on the basis of being peer-reviewed, primary sources, written in English, published between 1980 and 2015, and relevant to animal-assisted activities or therapies (AAATs). We excluded articles if they addressed human–animal interactions beyond AAAT, focused on therapy animals, or were intervention studies without published results. We then manually searched reference lists of included articles and located 3 more that were relevant. At the conclusion of Phase 1, 650 articles remained. In Phase 2, we reduced this set of 650 articles by applying four exclusion criteria: (1) the intervention was not described as AAT, (2) no intervention components (e.g., therapeutic activities or processes) were described, (3) institutionalized adults with dementia did not participate in the AAT, or (4) no health professional was identified as providing the AAT.
Filter 3: Data Extraction
Three reviewers were involved in data extraction: the first author and principal investigator (PI) of the study (Wood), the second author (Fields), and the third author (Rose). Each of these reviewers independently read all included articles and contributed to data extraction. Specifically, the third author first extracted the following information from each included article: (1) study aims, inclusion and exclusion criteria, participants’ ages and diagnoses, sample sizes, and research designs; (2) components of the AAT, including animals used, therapeutic activities or processes, number and duration of sessions, and involved health professionals; and (3) reported outcomes, including any established measures. She also sorted outcomes into three categories on the basis of how authors had described findings. The category of positive finding included desired or hypothesized positive outcomes of AAT that the study authors had described as statistically significant. The category of possible clinical significance included positive trends that the study authors had described as important even though they had not obtained, or did not report, statistical significance. The category of negative finding pertained to hypothesized outcomes of AAT that the study authors had reported were statistically nonsignificant or unrelated to clinical significance. Throughout, the third author consulted with the PI to ensure coding consistency and to resolve coding discrepancies. Information extracted during this primary coding process was entered into a synthesis table.
To address the study’s primary aim, the first and second authors reviewed the synthesis table to ensure agreement with the findings, after which they started mapping extracted information onto the LELQ Model (Figure 1). They first independently mapped extracted information from 2 studies onto the model; next, they assessed consistencies and inconsistencies in their respective mappings and came to consensus on areas of inconsistency. The second author then mapped extracted information from the remaining articles onto the model, consulting with the first author throughout this process. Last, the first and second authors created a synthesis table in which they mapped extracted information onto relevant domains and subdomains of the LELQ Model.
Results
Ten articles, all of which were research reports, met all inclusion and exclusion criteria. Supplemental Table 1 (available online at http://otjournal.net; navigate to this article, and click on “Supplemental”) presents results of the primary coding process. The 10 included studies were published in Australia, Japan, Italy, Sweden, or the United States, and the earliest study was published in 2001. Participants in 2 studies were reported to have mild dementia (Nordgren & Engström, 2014b; Travers, Perkins, Rand, Bartlett, & Morton, 2013), and participants in the remaining studies were reported to have moderate to severe dementia. Alzheimer’s disease was the most commonly reported type of dementia (Kanamori et al., 2001; Moretti et al., 2011; Nordgren & Engström, 2014a, 2014b; Richeson, 2003), followed by vascular dementia (Kanamori et al., 2001; Nordgren & Engström, 2012, 2014b). Health professionals providing AATs included occupational therapists, physical therapists, physicians, psychologists, nurses, therapeutic recreationists, and veterinarians. All 10 studies used dogs and reported as outcomes various elements of behavioral and psychological functioning.
Collectively, these 10 studies found that canine-assisted therapies generally benefited people with dementia who reside in nursing homes, assisted living facilities, and other institutional residential facilities. Accordingly, their mapping onto the LELQ Model emphasized positive statistically significant findings and possibly clinically significant findings. Table 1 summarizes this map, which is next elaborated.
AAT and Dementia Literature Mapped Onto the LELQ Model
Note. Empty cells indicate that no content mapped onto the concept of the LELQ Model. For example, no content from Berry et al. (2012) mapped onto the concepts of occupational history or ability to function. AAT = animal-assisted therapy; ADLs = activities of daily living; LELQ = Lived Environment Life Quality.
Lived Environment
Caregiving Microsystem.
The LELQ Model suggests that facets of the physical and social environment in a person’s lived environment may promote occupational opportunities (Wood et al., 2017; Figure 1). All 10 studies provided information pertaining to physical environmental elements such as location of AAT activities and animal care materials. Regarding social environmental supports, several studies explicated the importance of interdisciplinary collaboration and establishing therapeutic rapport. For example, before implementing AAT, Nordgren and Engström (2014b) encouraged collaboration with members of the health care team to help achieve positive outcomes. Richeson (2003) emphasized establishing rapport with participants by encouraging communication and using strategies such as cuing or positioning to support positive experiences. Health professionals incorporated various AAT dogs that had previous training or had otherwise been selected for their soothing or friendly temperaments.
Given these environmental supports, AATs were found to offer participants many occupational opportunities to engage in various activities and interact with both dogs and people. Some of these activities and interactions directly related to activities of daily living (ADLs; Friedmann et al., 2015) or other commonplace social activities such as greeting one another or singing songs (Sellers, 2006; Travers et al., 2013).
Person With Dementia.
The LELQ Model encourages occupational therapy practitioners to gather robust occupational profiles of people with dementia as a basis for planning interventions (Wood et al., 2017). Although providers of AAT did not gather occupational profiles to plan their interventions in any included study, they did attend to participants’ needs and preferences. Seven studies emphasized the importance of assessing whether people had a positive or negative history with dogs as a prerequisite for AAT (Berry et al., 2012; Friedmann et al., 2015; Kanamori et al., 2001; Nordgren & Engström, 2014a, 2014b; Richeson, 2003; Sellers, 2006). In 3 studies, the occupational therapists who provided AAT individually tailored intervention protocols to participants’ needs (Nordgren & Engström, 2012, 2014a, 2014b). In the Nordgren and Engström (2012) study, for instance, a participant who was often aggressive and repetitively wandered but who had enjoyed nature, walking, and animals in the past was encouraged to groom and walk a therapy dog outdoors. These activities respected the participant’s preferences and helped to mitigate problematic behaviors. In addition, Friedmann et al. (2015) spoke to the importance of a dynamically unfolding intervention in which providers helped participants master challenging activities with dogs as a way to increase physical activity and improve physical capacities and self-efficacy.
Positive Quality-of-Life Indicators
Various promising outcomes were associated with AATs (Table 1, Supplemental Table 1). Research findings reported as statistically or clinically significant also corresponded to several positive QOL indicators associated with the three QOL subdomains of the LELQ Model.
Relative Being.
Most key findings pertained to indicators of relative well-being as detailed in the LELQ Model. Five studies identified that AAT decreased problematic behaviors such as verbal or physical aggression or wandering and excessive restlessness (Kanamori et al., 2001; Nordgren & Engström, 2012, 2014a; Richeson, 2003; Sellers, 2006). Having found slight decreases in agitation over time with AAT, Friedmann et al. (2015) noted that their study did not support concerns that increased “stimulation and physical activity for individuals with dementia may increase behavioral symptoms due to overstimulation” (p. 285). Five studies associated more frequent episodes of positive affect and decreased anxiety, apathy, sadness, or depression with AAT (Berry et al., 2012; Friedmann et al., 2015; Kanamori et al., 2001; Moretti et al., 2011; Travers et al., 2013). For example, Berry et al. (2012) conducted extensive behavioral analyses of video recordings of a dog-assisted socialization group and found that play activities with dogs, especially throwing and retrieving balls, most often elicited smiling in participants. Friedmann et al. posed that reduced depression after AAT might encourage greater physical activity and improvement in physical function. Moretti et al. (2011) reported that most participants recalled pleasant past memories during AAT and would recommend it to other older adults.
Time Use.
None of the studies addressed time use as an element of QOL, but 6 studies suggested that participants occupied time in a manner that evidenced greater occupational engagement during AAT (Berry et al., 2012; Kanamori et al., 2001; Nordgren & Engström, 2012; Richeson, 2003; Sellers, 2006; Travers et al., 2013). In particular, increased participation in social interactions and conversation were reported. During one AAT session, for example, 1 participant snuggled a dog close to her, saying, “They are very cute; they are very warm; the softness of their fur is comfortable” (Kanamori et al., 2001, p. 237). As suggested by the LELQ Model, these findings are understandable given that the AAT in these studies was designed to offer opportunities to do things and interact with dogs and people. Indeed, during the AAT studied by Berry et al. (2012), participants spent significantly more time interacting with the dog than with people. AAT may thus diversify ways in which institutionalized people with dementia can spend time being occupationally engaged.
Ability to Function.
Four studies addressed the possible impact of AAT on participants’ functional capacities. Specifically, 2 studies found that participants’ ability to participate in ADLs improved after AAT (Friedmann et al., 2015; Nordgren & Engström, 2012). Two other studies suggested that AAT could help to improve participants’ cognitive function, at least for short periods after AAT (Kanamori et al., 2001; Moretti et al., 2011).
Discussion
Rigorous research on animal-assisted interventions can be quite complex, given the many associated issues (Johnson, Odendaal, & Meadows, 2002). For example, researchers need to invest substantial time and resources to gain access to clinical settings; address safety concerns about animals; select and train animal handlers; select, train, and care for animals; and design interventions and select outcomes that are acceptable and feasible within local contexts. The complexity of such research partly explains why few rigorous studies of animal-assisted interventions have been conducted (Johnson et al., 2002). Diverse approaches to addressing these methodological issues have also made explication of contextual factors in AATs—and synthesis of findings into a coherent whole—challenging. We therefore used the LELQ Model as a guide to a systematic mapping review of the literature on AAT for institutionalized people with dementia. Using this model allowed us to illustrate linkages among the lived environments of institutionalized people with dementia, their experiences related to AAT, and their QOL.
Findings from the systematic mapping review offer proof of the concept that AATs are feasible, may be occupationally enlivening, and may help to improve the QOL of institutionalized people with dementia. Guided by the LELQ Model, we found that diverse health professionals offered institutionalized people with dementia many opportunities for shared occupations with dogs and other people during AATs. Furthermore, trained therapy dogs and providers often acted as social environmental supports that helped people with dementia spend time occupationally engaged in diverse ways. Familiar physical objects such as balls, dog brushes, or leads and access to spaces such as quiet rooms, gardens, and the outdoors often functioned as physical environmental supports to engaging in interesting and pleasing activities and social interactions. Promising findings across the 10 included studies were additionally congruent with some positive indicators of QOL in the LELQ Model that pertain to relative well-being, occupational engagement, and more optimal ability to function. Numerous standardized assessments that practitioners in occupational therapy and other health professions might use to document improvements in QOL resulting from AAT were also evident (Supplemental Table 1).
Descriptions of the AATs in the 10 included studies accordingly conveyed characteristics of complex interventions, as defined by Craig et al. (2008). Namely, we identified many interacting AAT components, especially interactions among social environmental elements (e.g., therapy animal, animal handler, health care provider, and person with dementia) and physical environmental elements (e.g., selective use of objects and therapeutic spaces). Interactions among these elements gave rise to diverse occupational opportunities and supported occupational engagement (Table 1). Providers also flexibly delivered or tailored interventions to recipients’ needs (e.g., Friedmann et al., 2015; Nordgren & Engström, 2012, 2014a, 2014b; Sellers, 2006), and reported outcomes were numerous and varied (Table 1, Supplemental Table 1). Although no strict boundary exists between simple and complex interventions, these characteristics typify complex interventions (Craig et al., 2008).
Our findings further suggest that research on AATs as complex interventions is in early theoretical development, hence early scientific development. Early steps in establishing the scientific bases of complex interventions involve the development of theoretical frameworks. Theoretical frameworks promote the examination of processes that lead to desired effects, thereby avoiding simplistic outcomes-focused approaches to intervention research (Craig et al., 2008; Fleury & Sidani, 2012). Only the study by Friedmann et al. (2015) used an established theory (self-efficacy theory) to guide the design of AAT. Researchers and practitioners are thus encouraged to formulate theoretical frameworks underlying diverse AATs. This review has identified that one can draw on established theories such as self-efficacy theory, or on profession-specific conceptual practice models such as the LELQ Model, to help integrate animals into comprehensive approaches to unique professional services. A clear theoretical framework for an AAT could clarify how disciplinary perspectives influenced the design of its interacting components and selection of outcomes; it could also clarify mechanisms, or active ingredients, of the AAT, including whether and how it minimizes or prevents excess disability over time or disability beyond that directly attributable to dementing disease processes.
Last, although all participants in the included studies lived in a nursing home or assisted living facility, participants in 1 study received AAT in an adult day care program (Kanamori et al., 2001). The idea that institutionalized people with dementia may travel off site for AAT is intriguing and leads to possibilities for engagement with other animals in therapeutic contexts. Indeed, Dabelko-Schoeny et al. (2014) studied equine-assisted therapy for adults with dementia who attended an adult day care center. Similar to components of canine-assisted therapies found in this review, participants in equine-assisted therapy groomed, led, watched, painted, photographed, and freely interacted with horses. Findings suggested that equine-assisted therapy was feasible and possibly beneficial for adults with dementia. Although all studies in this mapping review incorporated dogs with gentle and sociable temperaments, other species such as horses may possess qualities that lend themselves to AAT with this population.
Limitations
Consistent with the intended scope of systematic mapping reviews, we did not assess the methodological rigor of studies. Although findings pertaining to QOL and AAT are promising, they remain preliminary. With respect to the executed searches, we searched broadly to retrieve articles of relevance to human–animal interactions in therapeutic contexts. Because no search terms were specific to dementia, however, we may have missed sources relevant to AAT and dementia. We may also have missed relevant sources that were not indexed, or not yet indexed, in the databases searched.
Implications for Occupational Therapy Practice
In this systematic mapping review, we comprehensively gathered, described, categorized, and synthesized the literature on AAT and institutionalized adults with dementia to help guide animal-assisted services for this population and to identify critical research gaps. The resulting mapping of key components of interventions and promising findings onto the LELQ Model suggests the following implications for occupational therapy practice:
The LELQ Model may help occupational therapy practitioners incorporate animals into comprehensive client-centered, occupation-focused, and ecologically valid services for institutionalized adults with dementia.
Qualities of dogs related to their temperament, training, and resulting willingness to interact with humans in safe, friendly, and nonthreatening ways constitute social environmental characteristics that can press people at early, middle, and late stages of dementia to have positive QOL experiences.
Practitioners who provide canine-assisted occupational therapy are encouraged to collaborate with formal and family care providers to ensure compatibility of prospective participants with dogs and to deliver services where participants spend considerable time each day in their institutional homes.
Supplemental Material
Supplementary material for Animal-Assisted Therapies and Dementia: A Systematic Mapping Review Using the Lived Environment Life Quality (LELQ) Model
Supplementary material, sj-pdf-1-aot-10.5014_ajot.2017.027219.pdf for Animal-Assisted Therapies and Dementia: A Systematic Mapping Review Using the Lived Environment Life Quality (LELQ) Model by Wendy Wood, Beth Fields, Michelle Rose and Merinda McLure in The American Journal of Occupational Therapy
Footnotes
*Indicates studies that were included in the systematic mapping review.
Acknowledgments
The Carl and Caroline Swanson Foundation supported this study. Michelle Rose’s thesis project contributed to the findings reported in this article.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
