Thirty-nine adults with severe to profound intellectual disability (ID) were randomly assigned to either an experimental group (n = 21) or a control group (n = 18). Assessment was blinded and included selected items from the International Classification of Functioning, Disability and Health (ICF), the Behavioral Assessment Battery (BAB), and the Learning Accomplishment Profile (LAP). The experimental group, who attended a dog-assisted treatment intervention over a 20-week period, showed significant improvements in several cognitive domains, including attention to movement (BAB-AM), visuomotor coordination (BAB-VM), exploratory play (BAB-EP), and motor imitation (BAB-CO-MI), as well as in some social skills, as measured by LAP items. Effects were specific to the intervention and independent of age or basic level of disability.
In recent years, mental health rehabilitation professionals have been challenged by the increased demands for integrative, multifaceted treatment strategies. Specific issues of comorbidity and complex needs have raised interest in alternative medical and psychosocial interventions that may integrate with the more traditional treatments. Complementary and alternative medicines (CAM) include procedures, interventions, and therapies that are not considered conventional for a particular treatment need (Umbarger, 2007). Among these, animal-assisted therapies (AAT) have been identified to be an innovative and promising tool for mental health rehabilitation (Friedmann & Thomas, 1995; Kamioka et al., 2014).
AAT are goal-directed interventions where an animal meeting some specific criteria is an integral part of the treatment process (Kamioka et al., 2014). AAT are usually directed and delivered by health professionals with specialized expertise, and generally involve animals such as domesticated pets, farm animals, or marine mammals. Proponents have suggested they may be particularly beneficial for participants with mental and/or physical disabilities in terms of social integration (Odendaal, 2000), such as people with intellectual disability (ID). Animals are perceived to be agents of socialization and providers of social support and relaxation (Serpell, 2006). Being unconcerned with age, gender, or mental/physical performance, animals seem to naturally create positive interactions with people (Olbrich & Otterstedt, 2003; Santori, 2011). Indeed, the relationship between humans and animals gives the advantage of a direct, unmediated emotional experience (Guttman, 1984). Further, according to a holistic view, AAT are not directed to a specific body part or function; instead, they aim at a global effect on the person/environment fit by increasing motor abilities (Chandler, 2005; Guttman, 1984), communication skills (Guttman, 1984), social interaction (McNicholas & Collis, 2000), and sense of agency and self-confidence (King, Watters, & Mungre, 2011; Wilks, 1999).
Although most literature presents human-animal relationships as generally positive experiences for people with disabilities, findings indicating specific intervention benefits have been considered suspect due to methodological concerns regarding the research designs on which data were based. These concerns have raised calls for improved experimental designs when implementing research on AAT (Chur-Hansen, Stern, & Winefield, 2010). A recent review by Kamioka et al. (2014) identified 11 studies using randomized controlled trials (RCT) to investigate the effects of AAT on different target populations, including people with mental disorders, addictive behaviors, cancer, advanced heart failure, or severe ambulatory disabilities. No study in Kamioka's review was specifically directed at people with severe levels of ID, and most RCTs were of relatively low quality, having significant limitations in the method used to generate concealment.
The present study is an RCT aimed at determining the effects of a dog-assisted treatment program on the behavior, communication, and basic social skills of adults with severe to profound levels of ID. We hypothesized that AAT may have a relevant and independent effect on people's behavior and social abilities. The secondary purpose was to evaluate the effectiveness of AAT with dogs as a supportive tool for the rehabilitation of people with severe mental and/or physical disabilities.
Materials and Methods
The present study is part of a research project aimed at testing the potential rehabilitative effects of AAT with dogs on patients with severe levels of ID. The project was conducted between January, 2012, and July, 2013, within the “Pet Therapy Center” of the Local Sanitary Unit n. 7 “Pedemontana,” a center for AAT in the northern district area of Vicenza, Veneto region, Italy. It was designed according to the principles of the Declaration of Helsinki (World Medical Association, 1989) and approved by the institutional review board of the Italian National Reference Center for Animal Assisted Interventions (NRC AAI) and by the bioethics committee of the Venetian Institute for Veterinary Disease Control.
Participants were recruited from seven community centers for the disabled within the district area of the “Pet Therapy Center.” Parents/legal guardians interested in the study were provided a brief summary of the project and asked for written informed consent. However, in conformity with the principles of good clinical practice (expressed by the Italian law in matter of clinical research—Dlgs N. 211/2003, art. 3 and 5), the experimenter also looked for any possible expression of consent from the person with the disability (e.g., enthusiasm, distress).
According to a standard procedure established by the regional authorities, the admission into a community center is submitted to a multidisciplinary team evaluation. The team usually includes the general practitioner, a psychologist and/or a psychiatrist, an educator, a social worker, and a family member of the person with the disability who is being considered for admission. The final decision is based on the treatment and rehabilitation needs established by the multidisciplinary team based on the SVaMDi (in Italian: Scheda di Valutazione Multidisciplinare per la Disabilità), a multidimensional evaluation form for people with disabilities. The SVaMDi is a standardized instrument based on the International Classification of Functioning, Disability, and Health (ICF; World Health Organization, 2001). The SVaMDi, which also includes important information about social, economic and household background, is used by professionals to classify people's needs for rehabilitation into 10 levels of growing intensity: residual abilities (AB), job-preparatory (PL), educational/occupational (EO-1, EO-2), educational/assistential (EA-1, EA-2, EA-3), autism (AUT 2, AUT 3), and severe brain injury (GCA).
For the present study, only people meeting criteria for the EA-2 or EA-3 levels were eligible for inclusion. People classified as EA-2 usually have relational, behavioral, cognitive, and/or motor problems that require medium intensity and time-limited daily assistance. The EA-3 level includes individuals with severe behavioral problems and/or complex needs that require intense and/or continuous daily assistance. Other inclusion criteria were: (a) an age over 17 years; (b) a regular attendance at the community center for at least 1 year; (c) no involvement into new educational, recreational, or rehabilitative programs for the duration of the study; (d) parents/legal guardian consent for participation; (e) a stable (i.e., consistent dosage) daily psychotropic drug prescription (e.g., antipsychotics, mood stabilizers, antidepressants) for at least 1 month before study entry. Type and dosage of psychotropic drugs were supposed to be stable throughout the course of the entire study. Any change in drug treatment (except for benzodiazepines or other sedative drugs) had to be registered. A review of participants' medical history was performed to screen for precautions (e.g., history of seizures) and contraindications (e.g., a history of pet allergy). Further, parents/legal guardians and community centers' professionals were interviewed regarding participants' physical, cognitive, communication, behavioral, social, and sensory processing functioning and needs.
Exclusion criteria involved: (a) a history of animal abuse or phobia to dogs; (b) a history of pet allergy; (c) a change in daily dose of psychotropic drugs less than 1 month before the start of the program; (d) a limited compliance to educational and rehabilitation activities; (e) the lack of a formal consent from the parents/legal guardians.
Sociodemographic variables including age, gender, civil status, and education level, and clinical variables including main diagnosis, comorbidities, and psychopharmacological treatment, were registered for all potential participants by means of a standardized form. Sixty potential participants responded to the invitation for the study; 21 did not meet the selection criteria (i.e., limited compliance to educational and rehabilitation activities, or no regular attendance at the community center). Thus, 39 participants (17 males, 22 females) between the ages of 20 and 58 years (mean age: 37.1±12.1) were included in the final sample. All participants were diagnosed as having a severe to profound level of ID (IQ level < 35) according to the SVaMDi. The participants' diagnoses according to ICD-10 diagnostic criteria (World Health Organization, 2005) were the following: Down syndrome (n = 7), Down syndrome with congenital hypothyroidism (n = 4), Neurofibromatosis type 1 (n = 1), Prader-Willi syndrome (n = 1), Cri du chat syndrome (n = 1), congenital encephalopathy with comorbid severe epilepsy (epileptic encephalopathy; n = 7), congenital encephalopathy with comorbid spastic tetraparesis (n = 7), congenital encephalopathy (n = 16). Seven participants (17.9%) were prescribed antipsychotics (including olanzapine, risperidone, clotiapine, levomepromazine ,and levosulpiride), nine participants (23.1%) were prescribed anticonvulsants (including carbamazepine, oxcarbazepine, valproate, lamotrigine, and levetiracetam), and 19 participants (48.7%) were prescribed other psychotropic drugs (including benzodiazepines, phenobarbital, and, in one case, citalopram).
We used a stratified block randomization method to assign participants to two treatment groups. The sample was divided into subgroups (strata) by community center placement. Then, a research assistant used a random number table to generate the block allocation sequence within each community center. The block length was four (2 x number of treatments). The allocation sequence and the block size were concealed until treatments were assigned (see Figure 1 for the flow chart of the study).
All the participants in the AAT group (n = 21) were enrolled for a 20-session treatment program involving a dog-assisted intervention. Each session lasted 30 minutes, was held once a week within a room at the Pet Therapy Center, and involved three participants (randomly assigned from the AAT group at the beginning of the program). The room provided was about 42 square meters and included tables, chairs, and different toys. Six specially trained dogs were chosen for the rehabilitation sessions. They were quiet and well-schooled adult dogs of different breeds and size (two Bernese Mountain Dogs, one Bergamasco Shepherd, one French Bulldog, and two medium size, mixed-breed dogs). The dogs had to meet some specific health prerequisites according to the Regional Operational Manual (in Italian, Manuale Operativo Regionale - MOR) for pet therapy interventions (Manuale Operativo Regionale, 2010). Further, to be admitted to the project, the dogs had to be: (a) reliable (i.e., show a consistent pattern of behavior in front of similar situations), (b) predictable (i.e., their behavior, in specific circumstances, can be expected); (c) controllable (i.e., their behavior can be guided or managed); (d) suitable (i.e., they are qualified because they have already experienced similar activities; and (e) reassuring (i.e., their appearance and behavior inspire a sense of security).
The rehabilitation setting included three participants, a dog, a dog handler, an expert supervisor of the intervention (e.g., a mental health professional or an educator), and at least one educator from the community center. Each dog's welfare was guaranteed by a veterinarian for the entire duration of the study. The dog/dog-handler combination was assigned randomly to each participant and was maintained throughout the program. The sessions were always preceded by an approaching meeting, aimed at preparing the person with the disability to receive the dog by recalling the animal's appearance and behavior and by showing and commenting on photos about the previous session's activities.
The activities, arranged within the 20 weekly sessions, were structured according to a pattern of growing complexity and interaction (i.e., from simple observation with no interaction toward more complex group activities). The activities are presented in Table 1 with their potential rationale for rehabilitation. The program was designed to be as inclusive as possible, and the different activities were conceived to be simple and flexible in order to adapt to each participant's needs and abilities. The team was allowed to make some simple changes so that each participant could complete all the activities (i.e., for participants having serious motor disabilities, the activities in sessions 13-15 were simplified). Further, although all participants were verbal, group activities in sessions 16-20 were adapted to each participant's ability to communicate.
Briefly, during the “observation of the dog” the team described the dog and asked questions about the animal's appearance and behavior, stimulating the participant's attention, concentration, and memory. By “calling the dog” by its name, the intent was to stimulate the participant's verbal communication abilities, and promote his or her sense of self-efficacy and self-esteem. “Taking care of the dog” (i.e., by giving the dog food) also was intended to encourage attention, concentration, and motor coordination, as well as having an effect on basic social abilities such as the respect for rules. “Grooming the dog” was intended to boost relaxation and sense of self-efficacy. “Interactive playing” and “driving the dog on leash” were thought to be potentially useful in helping participants with space orientation and motor coordination, while also reinforcing their sense of self-esteem and self-efficacy. “Sharing personal impressions” was intended to provide opportunities to improve verbal communication and basic social skills. “Group activities” generally consisted of a sequence of activities with the dog and the dog handler that was intended to help improve motor coordination and boost self-esteem and self-efficacy.
In the control group (n = 18), participants were assigned to comparable activities as the AAT program, but with no dog intervention. The setting and the participants were the same, except for the dog and the dog handler, which were generally substituted by an unanimated object. For example: “observation of the dog” was replaced by the observation and description of an object of affection, such as a stuffed animal; “taking care of the dog” was replaced by brushing and cleaning the stuffed animal; a typical throw-and-catch game with the dog was replaced by a simple game of throwing a tennis ball into a basket.
Assessment and Outcome Measures
Two experienced clinical raters (a psychologist and an educator), who were blinded to treatment assignments, administered assessment at baseline (T0), at the middle (10th week, T1), and at the end of the treatment program (20th week, T2). The raters worked together and any disagreement about rating was resolved by consensus.
For the present study, outcome measures were represented by a selection of items from three rating scales: the International Classification of Functioning, Disability and Health (ICF), the Behavioral Assessment Battery (BAB), and the Learning Accomplishment Profile (LAP). The tests were chosen for being simple, reliable, and based on a direct observation of behavior. Assessment was generally performed within the community center of each participant, on a date different from the rehabilitation sessions.
The ICF (World Health Organization, 2001) is based on the biopsychosocial model of functioning, disability, and health, and was approved by the World Health Organization in 2001 as the replacement for the International Classification of Impairments, Disabilities, and Handicaps (ICIDH-2). The framework for the ICF includes three major sections and four subsections, where each subsection or construct is labeled by a letter-number sequence. These include: (1) Body, function (b1-7) and structure (s1-8); (2) Activities and Participation (d1-9); and (3) Environmental Factors (e1-5). For the present study, we considered only the second section, Activities and Participation, where focus is placed on the completion of a task by an individual. Tasks can vary within specific domains (e.g., learning and applying knowledge includes “d140 = learning to read” as well as “d145 = learning to write”; general tasks and demands includes “d210 = undertaking a single task” and “d220 = undertaking multiple tasks”). Furthermore, in the case of the Activity and Participation section, two important qualifiers are provided. Together, these qualifiers enable the user to code essential information about disability and health. The capacity qualifier describes an individual's ability to accomplish a task or an action. This construct indicates the highest probable level of functioning a person can achieve in a given domain at a given moment. The performance qualifier describes what an individual actually does in his or her current environment (accounting for any facilitator or barrier).
The categories of the ICF have already been used as components of outcome measures with a good interrater and intra-rater reliability (Kohler et al., 2013). For the present study, a short version (48 items) of the Activities and Participation section was used (http://www.who.int/classifications/icf/icfchecklist.pdf?ua=1). Each item was assigned a score according to the ICF qualifier by using a four-point Likert scale (from 0 = no impairment, to 4 = complete impairment). A final total score was calculated for each subsection by summing up single items' scores. Total scores for both the ICF Capacity (ICF-C) and the ICF Performance (ICF-P) were used as a measure for overall outcome.
The BAB (Kiernan & Jones, 1977) is a multidimensional test specifically designed for the behavioral assessment of people (both children and adults) with severe ID. It consists of 13 sections concerned with different areas of functioning: (1) reinforcement (R) and usual activities; (2) visual tracking (T); (3) visual attention to movement (AM); (4) visuomotor coordination (VM); (5) auditory-responsiveness (AU); (6) postural control (PC); (7) exploratory play (EP); (8) constructive play (CP); (9) search strategies (SS); (10) perceptual problems solving (PS); (11) communication skills (CO); (12) self-help skills (SH); and (13) social behavior (S). The titles of these sections reflect the aim of the BAB in tapping basic components of behavior (e.g., visual tracking), as these are essential prerequisites to most of the behavioral landmarks identified in more conventional tests. Many BAB items are considered formal tests in so far as the participant is confronted by objects or situations to which a response is required (World Health Organization, 1992). Because our purpose was to determine the effects of AAT on the behavior and communication abilities of adults with severe levels of ID, we only considered the sections of visual attention to movement (AM), visuomotor coordination (VM), exploratory play (EP), and communication (CO). For communication, we addressed in particular individual tests for motor imitation (MI), expressive language (EL), and receptive language (RL). Each item received a score on the basis of a 4-point Likert scale ranging from 0 = no response to 3 = correct response (1 = inappropriate response and 2 = approximate response). A total score was calculated for each section and individual test by summing up single items' scores.
The LAP (Sanford & Zelman, 1981) focuses on gross and fine motor skills, prewriting, cognitive, language, and social/emotional aspects of early childhood development. It was designed to assess the development of children by providing tasks or situations typical of young children's development that would be interesting and stimulate an observable response. The LAP was designed for developmental age 36–72 months of age, but is also considered appropriate for older participants with delays (Sanford & Zelman, 1981). It includes nearly 400 items, and each item addresses a specific learning accomplishment. It is specifically designed to be a flexible instrument for the assessment of individual educational needs. For the present study, 23 items from the Personal/Social area section were selected in order to address specific issues related to social behavior skills, such as the ability to play in groups, or cooperation skills (selected items are not reported in this article but are available on request). Each item received a score on the basis of a 5-point Likert scale ranging from 0 (never) to 5 (always). A total score was calculated by summing up single items' scores at baseline (T0) and at the end of the treatment program (T2). Items were selected due to their alignment with the purpose of the study (i.e., to investigate specific behavioral and social skills of the participants) and the extent to which they were appropriate given the heterogeneity of the participants due to different ages, IQ scores, adaptive skills, and verbal/motor abilities.
Student's t-test and Fisher's exact test were used to compare the two treatment groups (i.e., AAT vs. control) for continuous and categorical variables, respectively. Any association between continuous variables, including age, years of education, and baseline scores of outcome measures (ICF-C, ICF-P, BAB, and LAP) was determined by Pearson correlation. Two-way repeated-measures analysis of variance (ANOVA) was used to compare the course of outcome measures between the two groups. Finally, in order to control for potential confounders, a series of mixed effect models was performed with the individual outcome measure at T1 and T2 (i.e., ICF-C total score) as the dependent variable. The treatment group (AAT vs. control) was considered the fixed effect variable, and community center placement was the random effect variable. Age and baseline score of the individual outcome variable were introduced as covariates. All tests were two-tailed and significance was conservatively set with an alpha value of 0.01. Statistical analyses were performed by using the STATISTICA 6.0 software package.
Features of Study Sample
Age was significantly and negatively correlated with ICF-P total scores at baseline (Table 3). Conversely, age was also positively correlated with baseline scores of BAB subscales measuring exploratory play (BAB-EP), motor imitation (BAB-CO-MI), expressive language (BAB-CO-EL) and receptive language (BAB-CO-RL; Table 3). Also, we found a positive correlation between age and LAP total score at baseline (Table 3). Level of education had no significant correlation with any variable. LAP scores at baseline showed a significant and negative correlation with ICF-P scores, and were positively correlated with BAB subscales for communication skills (i.e., motor imitation [BAB-CO-MI], expressive language [BAB-CO-EL], and receptive language [BAB-CO- RL]).
Treatment and Efficacy
As shown in Table 4, no significant effect was determined by the group of treatment on the course of ICF total score (both Capacity and Performance) over the treatment period. Otherwise, a significant improvement was detected in the AAT group as regards to attention to movement (BAB-AM), visuomotor coordination (BAB-VM), exploratory play (BAB-EP), and motor imitation (BAB-CO-MI). Otherwise, no significant difference was detected between groups in terms of expressive (BAB-CO-EL) or receptive (BAB-CO-RL) language performance.
A significant improvement in social behavior skills, as measured by LAP items, was detected at the end of treatment in the AAT group (Table 3). A series of mixed effect models was performed with the individual outcome measure at T1 and T2 (i.e., ICF-C total score) as the dependent variable. Our models showed no significant effect of the treatment group on the course of ICF-C and ICF-P score. Overall, the effect of AAT on attention to movement (BAB-AM) was confirmed at both T1 (F = 46.12; p = 0.0008) and T2 (F= 95.56; p < 0.0001). The same trend was observed for visuomotor coordination (BB-VM) at T1 (F = 19.15; p = 0.0050), and T2 (F = 56.50; p = 0.0018), and for exploratory play (BAB-EP) at T1 (F = 20-08; p = 0.0027) and T2 (F = 32.50; p = 0.0028). The effect of AAT on motor imitation abilities (BAB-CO-MI) at T2 was at the limit of significance (F = 17.56; p = 0.0130). No treatment group effect was detected for expressive language (BAB-CO-EL) and receptive language (BAB-CO-RL) functions. For the LAP score, an effect of AAT on basic social skills was confirmed at the end of the treatment program (F = 5.72; p = 0.0019). All effects were independent of age, baseline score, and community center placement.
To our knowledge, this is the first study addressing the effects of an AAT intervention on the behavior, communication, and basic social skills of adult participants with severe to profound levels of ID. Overall, our results seem to point at a specific effect of our dog-assisted intervention on visuomotor abilities and basic social skills of participants. The presence of the dog was reported to have a positive impact on some basic cognitive functions, such as attention to movement or visuomotor coordination. The corresponding BAB sections (AM, VM, and EP) have been specifically designed to describe fine changes in the participant's visuomotor abilities and exploratory behaviors (for example, the AM section ranges from “follows an object with eyes” to “anticipates a movement”). Most important, these effects were specific to AAT and independent of age, basic level of disability, or community center placement.
Visuomotor coordination and fine motor skills are important for the performance of many activities (e.g., self-care and social activities) and are necessary not only for motor development but also for emotional and social development (Piek, Barrett, Smith, Rigoli, & Gasson, 2010; Whittingham, Fahey, Rawicki, & Boyd, 2010). People engage with their environment and learn about their bodies, the space around them, and each other, mainly through the use of motor skills (Bushnell & Boudreau, 1993). Thus, the effect of AAT with dogs on these basic cognitive functions would be particularly relevant to efforts to promote the development of communication skills and positive social behaviors in participants with severe levels of ID.
Positive social behaviors are verbal or nonverbal social and communicative behaviors that indicate social interest or provide appropriate social initiations or responses in specific situations (e.g., showing an interest in others, smiling or looking at other people, playing simple interactive games; Walton & Ingersoll, 2013). Our results showed no specific effect of AAT with dogs on primary communication skills, such as receptive language (RL) or expressive language (EL). However, we found AAT had a significant effect on basic social skills, as represented by LAP items, such as participation in simple group activities and cooperation. No previous study has investigated the role of an animal in the development of social abilities in adults with severe levels of ID. However, other studies have reported a positive effect of AAT on social skills in different populations of participants. For example, Bass, Duchowny, & Llabre (2009) found improvements in social functioning for an experimental group (19 children diagnosed with autism) after a 12-week program of therapeutic horseback riding (THR). Gabriels et al. (2012) reported a similar effect of THR in self-regulation behaviors, expressive language, and motor skills for a group of 26 children with autism spectrum disorders. The use of dogs with persons suffering from Alzheimer's disease resulted in positive effects on communicative variables such as smiles, laughs, touches, and verbalizations (Kongable, Buckwalter, & Stolley, 1989). A similar effect of the presence of the dog was reported by Fick (1993) in regard to social interactions among male nursing home residents. Pets can decrease anxiety and sympathetic nervous system arousal (Handlin et al., 2011; Kanamori et al., 2001; Miller et al., 2009; Odendaal & Meintjes, 2003) by providing a pleasant external focus for attention, promoting feelings of safety, and providing a source of contact comfort. Findings from this study supported the hypothesis that humans witnessing “cheerful” behavior in a friendly dog (such as attentiveness to the participant, mouth slightly open and “smiling,” and tail wagging) might result in empathic imitation of cheerful behavior.
Although some relevant and specific effects of AAT were found in specific cognitive and social skills domains, no difference was found between groups on the ICF scores of the Activities and Participation section. This finding could be an artifact of either the heterogeneous nature of the sample or on the low specificity of the selected items. The need for additional outcome measures that are sensitive to changes in cognitive and social skills displayed by people with severe levels of ID is apparent. Assessing participants with severe disabilities is a challenge, due to severe physical problems, serious attentive deficits, noncooperation, and low motivation in test situations. Our outcome measures, such as BAB and LAP, were not specifically designed for adults with severe to profound levels of ID, but we felt they were the best options available.
The major limitation of our study is the limited number of participants and the heterogeneity of the sample. A larger and more diverse sample would have allowed us to make more robust knowledge claims. Nevertheless, data from this pilot study suggests that the “AAT with dogs” intervention that was described in this study may be a promising practice for future researchers to investigate and build upon.
To our knowledge, there are no specific rehabilitation standards for intervention in regard to the skills and abilities we selected as the outcome of interest in this category of participants. Therefore, although our data showed that the AAT treatment was more effective than the control treatment (which did not involve an animal), there is no basis to suggest that the AAT would be better (or worse) than alternative interventions. Our AAT with dogs may not be applied extensively because of logistical and financial reasons, as it usually requires health professionals with specialized expertise and a specifically designed environment of treatment.
Additional limitations include the lack of a standardized AAT intervention protocol and the use of a single blind rater, which may have limited the generalizability of our results.
Our study was a pilot research project that focused on an innovative rehabilitation approach for people with serious mental and physical disabilities. To our knowledge, it is the first to quantify and systematically examine the effects of an AAT intervention on specific behavioral, communication, and basic social skills of participants with severe to profound levels of ID. Future researchers can build on this work by including a standardized AAT intervention protocol and a clearly defined and empirically supported AAT curriculum manual, allowing for the comparison of standardized and comparable therapeutic interventions.