Abstract
Behavioral and psychological symptoms of dementia are prevalent; however, there are no current interventions to manage these symptoms in the outpatient clinic setting. Animal-assisted interventions (AAI) have shown to be acceptable, non-pharmacological methods for symptom management in the clinic setting in other populations but have not been researched in dementia. This mixed-methods study investigated the feasibility and acceptability of an AAI for persons with dementia in the outpatient clinic setting. Data were collected from three groups of participants: (1) persons with dementia; (2) caregivers; and (3) health care providers. Ten persons with dementia engaged in an AAI, interacting with a therapy dog for at least 10 minutes before their appointment with a health care provider in a memory clinic. Persons with dementia and their caregivers completed follow-up Likert scale surveys and semi-structured telephone interviews; health care providers also completed follow-up Likert scale surveys. Telephone interviews were transcribed, and data were analyzed using qualitative thematic analysis. Descriptive results indicated that persons with dementia felt comfortable (100%) and happy (88.9%) interacting with therapy dogs. Caregivers also felt comfortable (100%) with the persons with dementia interacting with the therapy dog and reported that the interaction made the persons with dementia happy (90%). Neither persons with dementia, caregivers, nor health care providers reported the interaction as burdensome (100%). Four themes emerged from the qualitative thematic analysis: (1) baseline behaviors and feelings, (2) feelings towards opportunity, (3) positive interaction with dogs, and (4) positive overall experience at doctor appointments. Recruitment strategies were shown to be feasible, and the intervention was deemed safe. Conclusively, this study showed an AAI to be feasible and acceptable in the outpatient clinic setting for persons with dementia. Further research is required to determine the effects of an AAI on behavioral and psychological symptoms of dementia in the outpatient clinic setting.
Keywords
Introduction
Dementia is a debilitating neurodegenerative disease characterized by a decline in cognitive functioning and behavioral abilities (National Institute on Aging, 2024). The disease is caused by deposits of plaques and tangles in the brain, which lead to neurodegeneration and brain atrophy. Clinically, dementia presents in affected individuals as memory impairment, deterioration of intellectual faculties, impaired communication, diminished reasoning and judgment, and an inability to perform activities of daily living (Mavroudis et al., 2019; Nelson et al., 2016).
Significant advancements in medical science and innovative research have contributed to increased average life expectancies; concurrently, some theoretical frameworks posit a correlation with the rising incidence of dementia (Tom et al., 2015). Alzheimer’s disease is the most common dementia and the sixth leading cause of death among Americans 65 and older (CDC, 2021). By 2060, an estimated 14 million Americans could be affected, leading to healthcare costs of approximately $1.1 trillion. It results in more deaths than breast and prostate cancer combined, and currently, there is no cure.
Behavioral and Psychological Symptoms of Dementia
Impaired brain functioning can lead to behavioral and psychological symptoms in individuals with dementia (BPSD). These symptoms affect nearly 97% of persons with dementia and can include aggression, agitation, apathy, depression, and sleep disturbances (Gerlach & Kales, 2018). BPSD present significant challenges for people with dementia, their caregivers, and health care providers (HCPs) (Gerlach & Kales, 2018; Olsen et al., 2016), as they are linked to poor health outcomes, can affect disease progression, and often lead to nursing home placement, possibly increasing stress and depression in caregivers (Choi et al., 2017; Gerlach & Kales, 2018).
Clinic Setting
The outpatient clinic setting plays a crucial role for people with dementia, serving as the primary site for ongoing management, follow-up, monitoring of comorbidities, and coordination of multidisciplinary care (Bergman et al., 2023; Prince et al., 2016). It is often where cognitive changes are first detected, treatment plans are initiated, and care is adjusted (Sideman et al., 2023). However, attending these appointments can be challenging for people with dementia, as disruption in routine and inherent stress of clinical visits may contribute to agitation and anxiety (Mario et al., 2009). These challenges are further compounded by the unfamiliarity of clinical environments, as many people with dementia value familiarity and a sense of security provided by their home setting (Forsund et al., 2018). Research highlights the importance of the physical environment in providing meaning, safety, and autonomy for people with dementia (Forsund et al., 2018; Parke et al., 2017; Woodbridge et al., 2018).
In contrast to familiar home environments, outpatient clinic settings may introduce stimuli that contribute to disorientation and adverse behaviors in people with dementia, potentially exacerbating confusion and hindering HCPs’ ability to assess and treat conditions (Mario et al., 2009). Transitions related to living spaces or environmental changes require careful planning to ensure their well-being. Despite its importance, little research has focused on interventions for BPSD in outpatient clinical settings. Comprehensive studies are needed to find effective strategies to reduce these symptoms and improve outpatient visits for people with dementia.
Animal-Assisted Interventions
Animal-assisted interventions (AAIs) are one of several non-pharmacological therapies being studied to address BPSD (Fine, 2019; Gundersen & Johannessen, 2018). AAIs represent unique approaches that engage multiple sensory modalities simultaneously, encompassing sight, sound, touch, and smell. AAIs have been studied among older adults in acute and long-term care settings (Friedmann et al., 2015; Nordgren & Engstrom, 2014; Wesenberg et al., 2019), but their application in outpatient clinics for people with dementia remains largely unexplored. While these interventions have been examined in clinics serving other groups, such as pediatric dental and adult pain clinics, they have not been specifically investigated in outpatient dementia contexts (Clark et al., 2020; Marcus et al., 2013; Nammalwar & Rangeeth, 2018). In such studies, participants interacted with therapy dogs during outpatient visits, demonstrating physiological and emotional effects (Marcus et al., 2013; Nammalwar & Rangeeth, 2018).
Purpose
The goal of this study was to develop an effective intervention for reducing BPSD in the outpatient clinic setting, ultimately aiming to achieve a successful visit for the patient, caregiver, and HCP. A visit is deemed successful when people living with dementia experience reduced symptoms and HCPs can perform accurate assessments, thereby ensuring optimal outcomes and the reliability of the evaluation. The present study aimed to conduct a feasibility trial of an AAI in the outpatient clinic setting for people with dementia. The rationale for this proposal is that BPSD pose detrimental consequences for people with dementia, their caregivers, and their HCPs. AAIs successfully address BPSD in acute and long-term care settings, and it is essential to know if AAIs can feasibly address BPSD in the outpatient clinic setting as well. To determine the feasibility of the intervention, the following questions were explored: (1) How feasible and suitable are the eligibility criteria, the number of eligible participants of the target population, and the recruitment rates? (2) To what extent were the study measures completed and followed per study protocol? To determine the acceptability of the intervention, the following questions were explored: (1) What is the level of safety and burdensomeness of the AAI for the person with dementia and the animal? (2) What reactions and verbal responses do people with dementia, their caregivers, and their HCPs have regarding the intervention?
Methods
Study Design, Setting, and Sample
This study used a feasibility design involving three types of participants to investigate the feasibility and acceptability of an AAI for people with dementia in the outpatient clinic setting. Researchers utilized a convergent mixed-methods approach to collect data, including quantitative Likert scale surveys and qualitative semi-structured interviews and observations. Due to the preliminary nature, a power analysis was not conducted to determine the sample size. This study intended to recruit 10 participants, congruent with other feasibility studies in similar populations (Friedmann et al., 2015; Grubbs et al., 2016; Holden et al., 2019).
Researchers used a consecutive sampling strategy to recruit participants from a memory clinic located in urban St. Louis, Missouri. Participants were recruited into three groups: (1) people with dementia, (2) caregivers of people with dementia, and (3) HCPs of people with dementia. Researchers screened for eligibility, and people with dementia were deemed eligible for participation if they met the following criteria: (1) aged 60 or older; (2) diagnosed with dementia; and (3) exhibited agitation behaviors at baseline, with a score of at least 35 on the Cohen-Mansfield Agitation Inventory (CMAI) assessment tool. Participants in the dementia group were excluded if they had an allergy, fear of, or dislike of dogs. Participants were eligible for the caregiver group if they accompanied the participant with dementia to the memory clinic; participants were eligible for the HCP group if they examined the participant with dementia in the memory clinic.
Upon confirming eligibility, and provided that the person with dementia and their caregiver expressed a willingness to participate, the PI initiated the informed consent process. This was conducted via telephone, Zoom, or during an in-person meeting prior to the clinic appointment. The objective of securing informed consent before the appointment was to reduce the additional time required and the participant burden during the clinic visit. Both the person with dementia and their caregiver were afforded ample opportunity to pose questions and to make an informed decision regarding their participation.
A registered nurse assessed the participant’s cognitive capability to determine their capacity to provide informed consent. Cognitive capacity was determined by reviewing records of previous cognitive testing, asking orientation and situational questions over the telephone, and assessing understanding of the study. When necessary, the participant’s legally authorized representative provided proxy informed consent, while the participant offered verbal assent to participate. Additionally, informed consent was obtained from caregivers and HCPs who met the eligibility criteria and agreed to participate.
Intervention and Study Procedures
Intervention
Touch Therapy Dog and Handler Certification Process
Additionally, therapy dog handlers underwent clinic orientation and project training with the PI, which emphasized the study’s objectives and highlighted the importance of answering participants’ questions only, to avoid influencing the results. During project training, each Touch team submitted a veterinary clearance form, documenting current vaccinations and acceptable health as determined by the veterinarian for the dog to participate in this study.
The safety and well-being of the therapy dog were prioritized and monitored to prevent hunger, thirst, discomfort, pain, injury, disease, fear, or distress, while allowing normal behaviors. The PI and handler ensured the dog was fed before visits and had access to water during sessions. Each dog interacted with up to three subjects daily and received bathroom breaks. The clinic was regularly cleaned, and participants sanitized their hands before and after contact to reduce the risk of zoonotic disease. Participants were screened for health conditions that could make animal-assisted interactions unsuitable, such as immunocompromised status, active infections, or open wounds. No patients were excluded due to health conditions, except for allergy concerns.
Upon arriving at the clinic, persons with dementia had their vital signs assessed and then underwent cognitive testing, part of routine care during their memory clinic visit. During this time, the caregiver met with the HCP separately to provide a history. After testing, patients usually returned alone to the waiting room while the HCP gathered information from the caregiver in another exam room. If enrolled in the study, patients were taken to a private room to interact with a therapy dog through activities like petting and talking. All but one participant did not have their caregiver present during this interaction. When ready, the HCP collected the patient from the therapy dog interaction room.
Data Collection, Management, and Analysis
Agitation eligibility was assessed using the validated and reliable CMAI, a 29-item caregiver-reported instrument that measures the frequency of agitated behaviors over the preceding 2 weeks, with higher scores indicating more severe agitation (Cohen-Mansfield, 1991; Finkel et al., 1992; Rabinowitz et al., 2005). Baseline agitation levels were included as an exploratory variable to generate hypotheses for future studies examining efficacy beyond feasibility. At the time of consent, participants in the dementia group provided basic demographic information, including age, gender, race, ethnicity, zip code, and pet ownership, along with medical history, medication use, Clinical Dementia Rating (CDR) scores, and vital signs. Study-visit source documents documented attendees during the AAI, the duration of the session, and detailed field notes describing the environment, interactions, behaviors, and observations. Overall experiences and satisfaction were assessed through post-intervention Likert-scale surveys developed by the investigators, administered to people with dementia, caregivers, and HCPs, as well as through semi-structured telephone interviews with people with dementia and their caregivers. Some survey and interview questions addressed subjective agitation reports or HCP impressions of agitation reduction. These data were not intended to support efficacy conclusions but served to inform hypotheses for future efficacy trials.
Researchers sought to collect follow-up surveys and conduct telephone interviews as soon as possible after the study visit, extending up to one week afterward to minimize the risk of memory lapses. Although the literature does not describe an “ideal” time frame for post-intervention interviews in the dementia population, previous studies have attempted to interview within 1–2 weeks, to minimize recall bias (Casey et al., 2024). Telephone interviews with people with dementia and caregivers were conducted using semi-structured interview guides and were recorded and transcribed by a professional transcription service. The semi-structured interview guides were developed by the PI and designed to understand the feasibility and acceptability of an AAI in people with dementia (Borda et al., 2020). An additional researcher reviewed the interview guides to ensure the appropriate inquiry into feasibility and acceptability. Questions on the interview guide and the interview protocol were carefully constructed to account for considerations necessary when interviewing people with dementia (Samsi & Manthorpe, 2020). The PI reviewed transcripts for accuracy, and all transcripts were uploaded to the qualitative analysis software NVivo 15. A qualitative thematic analysis approach, as defined by Braun and Clarke (2006), was used to analyze transcripts from the semi-structured interviews, open-ended responses on the feedback questionnaires, and the researcher’s field notes from the intervention. Within the context of this thematic analysis approach, a dyadic thematic analysis using the Framework Method (Collaco et al., 2021) was employed to help identify patterns within dementia/caregiver dyads and triads, as well as across all such pairs and groups.
All quantitative data was immediately stored in a secure electronic data capture system called REDCap© (Research Electronic Data Capture) (Harris et al., 2009, 2019). Researchers, in collaboration with a biostatistician from the University of Missouri-Columbia, analyzed quantitative data using IBM SPSS Software version 28. They employed descriptive statistics to evaluate participant demographics, visit characteristics, and post-study survey results.
Process Evaluation of Feasibility and Acceptability
Quantitative and qualitative data were compared side by side to identify confirmations or deviations in the results. The two datasets were analyzed for convergence. A joint display was created to present the quantitative and qualitative findings, along with an integrated interpretation for each research question. To establish credibility and trustworthiness, various strategies were implemented, starting with the design of the study. A convergent mixed-methods approach was adopted to facilitate a comprehensive understanding of the research questions through the integration of quantitative and qualitative data (Creswell, 2022). The researcher collaborated with experts in mixed-methods design to ensure the proper framing of research questions, the selection of appropriate data collection instruments, and the application of suitable data analysis techniques.
In-person observations provided deeper contextual insights and understanding of participants’ perceived meanings (Ahmed, 2024). Data source and method triangulation cross-verified emerging patterns, while two researchers conducted the thematic analysis independently and iteratively until consistent codes and themes emerged, enhancing reliability. Discrepancies between the researchers were resolved with the help of a third researcher.
Reflexive journaling documented the researcher’s positionality and biases, while a detailed audit trail recorded methodological choices and analytic steps, promoting transparency and dependability (Ahmed, 2024; Dodgson, 2019). For this project, the PI had prior personal experience relevant to the study population, having a previous family member who exhibited BPSD. This earlier experience may have influenced the researcher’s perspective. Potential internal bias was acknowledged and addressed through ongoing reflexivity, including self-reflection following data collection (Dodgson, 2019). This reflexive approach was employed to uphold the integrity and objectivity of data analysis. Together, these approaches strengthened the trustworthiness and validity of the study’s conclusions (Ahmed, 2024).
Results
The final data analysis included 10 people with dementia, 11 caregivers, and six survey responses from HCPs. One HCP provided care to six distinct people with dementia and completed a follow-up survey for each. The subsequent sections will provide details on the sample’s characteristics, the recruitment process, and the rates of eligibility, enrollment, and attrition. Additionally, responses from persons with dementia and caregivers will be presented to support the qualitative assessment of feasibility and acceptability. It is important to note that all statistics reported in the results section are descriptive in nature, and are not intended to support conclusions regarding efficacy.
Participant, Animal, and Visit Characteristics
Demographic Characteristics of Participants at Baseline
Visit Characteristics
Therapy Dog Characteristics
Photographs of participating therapy dogs
The PI was present at each study visit and recorded field notes based on their observations. Each study visit occurred in an exam room located in the rear hallway of the memory clinic. Upon entering the memory clinic through the front entrance, one encounters a reception area and a waiting room furnished with nine chairs arranged in a semi-circular formation, oriented towards one another. Exiting the waiting room leads into the main clinic area, which includes a checkout desk, a vital signs station, 14 exam rooms, and one physician’s workroom. Adjacent to the clinic is a rear hallway, which houses the nurse manager’s office, an undesignated office, and three additional exam rooms. Each exam room is equipped with a large window, tiled flooring, an examination table, two stationary chairs, one wheeled chair, a desk, and a computer. At the end of the back hallway, there is a staircase that descends to the lower level, leading to a grassy area. Figures 2 and 3 illustrate the floor plan of the memory clinic and exam room space where the intervention occurred. Memory clinic floor plan Exam room floor plan
During study visits, participants typically sat in stationary chairs near the window, while the dog handler occupied a stationary chair perpendicular to the participant. The researcher was seated in a wheeled chair by the desk. Handlers maintained a constant hold on a tether to the therapy dog. The researcher observed most therapy dogs sitting or lying on the floor beside the participant during the interaction.
Feasibility and Acceptability
Feasibility
The institution allowed researchers to access the memory clinic one day per week. During the study, nine clinic days were screened, and 80 patients were pre-screened for eligibility in the dementia group. Of these, 26 met the initial criteria of age over 60 and a dementia diagnosis. They were contacted by phone to assess interest. Seven declined due to reasons such as fear of dogs, allergies, or preferring normal conditions without stimuli. Nineteen patients were interested, and a further CMAI assessment phone screen was conducted. Three were excluded for CMAI scores below 35, and two due to appointment rescheduling to telemedicine. Ultimately, 14 people with dementia signed consent.
In the dementia group, attrition was 28.57%. Of the 26 initially eligible patients, three were excluded because their CMAI scores were below 35, leaving 23 eligible. Of these, 14 signed consent (60.87% recruitment rate), and 10 (71.43%) participated in the intervention, post-survey, and follow-up, and were included in the final analysis. Each person with dementia had at least one caregiver involved in the post-survey and follow-up, and six people with dementia had an HCP complete the follow-up survey, totaling 6 HCP survey responses. Figure 4 illustrates eligible and enrolled participants. Enrollment flow chart
The intervention was conducted per protocol with no deviations. Each session involved a participant with dementia engaging with a therapy dog for at least 10 minutes before seeing the HCP. No adverse events occurred involving participants or therapy dogs. Data collection adhered to protocol specifications, with the PI present to record observations. During follow-up, one participant with dementia was unable to recall their participation, resulting in a 90% completion rate for the Likert-scale survey; all participants completed the semi-structured interviews. All caregivers completed the Likert surveys and semi-structured interviews, achieving 100% completion. Of the HCPs, 6 out of 10 completed the Likert surveys, yielding a 60% response rate. Notably, the same clinician completed all six HCP surveys, limiting the independence of HCP survey responders. The descriptive results presented below are based on the number of actual responders: for participants with dementia, n = 9; caregivers, n = 11; and HCPs, n = 6.
Acceptability
Post-Study Survey: Participants With Dementia
Post-Study Survey: Caregivers
Post-Study Survey: Health Care Providers
Researchers conducted a qualitative thematic analysis applying the dyadic framework method to follow-up semi-structured telephone interviews and documented observations. All people with dementia and their caregivers completed the interviews; one person with dementia was accompanied by two caregivers, both of whom were interviewed. This resulted in nine dyads and one triad. Most interviews were conducted by phone, with one in person. Most dyads were interviewed together, starting with the caregiver, who then handed the phone to the person with dementia or used speakerphone; some were separate calls.
Themes, Subthemes, Exemplar Statements, and Field Notes
Baseline Behaviors and Feelings
People with dementia and their caregivers were asked about their behaviors and feelings at a baseline level, unrelated to a doctor appointment, and responses were mixed. Some caregivers described the person with dementia’s baseline behaviors as unremarkable and subdued, or insisted that certain exhibited behaviors were directly associated with situational sleep issues. However, most caregivers described behaviors such as memory issues, confusion, irritation, and agitation. One caregiver reported, “Um, yeah, and I just think she can get frustrated more easily. Most always related to trying to say something or need something or want to get something across” (Caregiver 15). Most people with dementia were reluctant to self-report any baseline behaviors unrelated to a doctor appointment. However, there were no major disagreements among the dyads and triad.
Caregivers and people with dementia both provided input regarding the person with dementia’s behaviors and feelings directly related to attending a doctor’s appointment, and again, responses were mixed. Most people with dementia reported that they do not feel any different about attending a doctor’s appointment and are generally comfortable doing so. When asked if they felt anything specific when attending the doctor appointment, one stated, “Not really. It was just another appointment to me.” (Dementia 7). Another commented, “Not unless I’m ill, like sick with something, then I’m concerned I’m gonna get a shot. Okay. But if there’s no shot, no chance of me getting a shot? No, I don’t mind going to the doctor.” (Dementia 10). Some expressed interest and curiosity in visiting the doctor to learn more about their health, and others were interested in attending an appointment at this specific memory clinic due to its excellent reputation. Conversely, some people with dementia did admit to experiencing specific feelings of anxiety, worry, and apprehension directly surrounding attending a doctor appointment. Participants described worrying about accurately completing the memory testing and having anticipatory anxiety related to cognitive deterioration. I was mostly like, you know, just curious about it, you know? So, I wasn't really afraid. The only thing that I, you know, that I was, uh, a little assistive about is I was hoping that they wouldn't find anything worse or that's how it would come out. (Dementia 18)
Caregivers in most dyads and triad agreed with the persons with dementia’s responses. Caregivers highlighted apprehension related to going to doctor appointments, with some people with dementia avoiding appointments by rescheduling or wanting to leave excessively early for the appointment to ensure they arrived on time. However, for people with dementia who did not report feeling any differently when attending a doctor appointment, some caregivers provided conflicting accounts and noted that they indeed expressed anxiousness or nervousness surrounding a doctor appointment. When asked about feelings surrounding a doctor appointment, participant with dementia 15 reported, “I was fine. Yeah, I was fine.” (Dementia 15). However, caregiver 15 provided a conflicting report. Um, well, I think she's less confident that she used to be, um, because she has trouble, um, communicating what she's trying to say. And so I think sometimes that worries or when it's someplace new, just kind of the procedures and the processes. (Caregiver 15)
Feelings Towards Opportunity
People with dementia and caregivers were asked about their thoughts and feelings regarding the opportunity to interact with a therapy dog at the memory clinic appointment. Responses were largely favorable, with people with dementia and caregivers alike describing the opportunity as a positive distraction from the appointment, a valuable addition, and a chance to relax before seeing the HCP. Multiple participants expressed, “I like animals” (Dementia 2), “I really like animals” (Dementia 10), or “I love dogs” (Dementia 5), implying their interest in participating due to the therapy dog involvement.
Participant with dementia 2 demonstrated his excitement for participation by bringing dog treats to the appointment. He interacted with Patti, a 5-year-old Wolfhound-Poodle mix, who had long, wiry, black, and gray fur. When entering the exam room, participant with dementia 2 noted that he had dog treats, but they were with his caregiver. He indicated he would return after the appointment to give Patti her treats. The researcher observed the participant smiling, laughing, shaking Patti’s hand, and asking Patti repetitive questions. Patti obliged, lifting her paw when asked, stretching, and wagging her tail. Caregiver 2 described the opportunity as an excellent distraction for the participant. I thought it was an excellent opportunity. Again, I, I just feel like he, um, was able to think about something other than what was going to happen at the appointment. And, um, it, it was something for him to look forward to. I, I think it’s a great opportunity for patients. Um, you know, he loves animals, so again, I, I think it was a, a great addition to his appointment and, um, it, it’s a good addition. (Caregiver 2)
Caregiver 11 reported that the project would be a good opportunity to enhance social interaction for people with dementia. Um, well, she, as she said, her, um, so my brother had, uh, a dog also who’s very well behaved and everything, so I just thought it'd be a good, you know, just something to cheer her up a little bit and, um, something she would enjoy just putting, uh, a dog and just being in a room with a dog. I just thought it was a great social, like, just interaction type of opportunity. I just, I was just hoping she would enjoy it. (Caregiver 11)
Caregiver 18 reported that the participant with dementia 18’s initial reaction was that of apprehension and paranoia concerning participating in the study. Participant with dementia 18 suggested that the research team may pair him with a dog with rabies; however, he did not disclose these initial feelings to the researcher. He interacted with Ollie, a 5 ½-year-old Shar Pei mix. Ollie was medium-sized, with short hair and sable coloring. When participant with dementia 18 entered the exam room, the researcher noticed a hint of hesitancy, evident in a slow entrance and quiet demeanor, which faded away as soon as he sat down and began to pet Ollie. He immediately started sharing stories about his own dog, a Maltipoo named Frankie, who is 6 ½ years old.
Positive Interaction With Dog
People with dementia and caregivers overwhelmingly reported positive interactions with the dogs, with participant with dementia 1 reflecting on his interaction with Barron, a 10-year-old, smooth-coated Golden Retriever, “Oh, the dog. The dog was fabulous. I could do that all day” (Dementia 1). When he entered the exam room during the study visit, the researcher observed an immediate smile and apparent excitement wash over his face. Participant with dementia 1 happily petted and talked to Barron throughout the interaction, while Barron remained well-behaved and calm.
Participant with dementia 5 reported wanting to comfort the therapy dog. During the study visit, the researcher observed him sitting next to the dog, Moose, a 71-pound black Goldendoodle with long, curly hair. Participant with dementia 5 repeatedly stroked Moose’s back and neck while engaging in continuous conversation throughout the entire interaction. He referenced the cognitive testing he just completed, revealing to Moose, “I had to have some tests and couldn’t remember” (Dementia 5). He continued to share with Moose his experience of owning dogs for over 49 years, including his current dog, Peter, who is 14 years old. A big smile remained on his face throughout the interaction, as he talked the entire time, even stating once, “I’m talking non-stop” (Dementia 5). During the follow-up interview, participant with dementia 5 reflected on his interaction with Moose. Uh, I, I wasn’t certain the dog had been, uh, around people my age. Uh, so I had to make him feel comfortable, let him know that I, I love dogs and that I always had a dog with my wife and also growing up, and they were central part of our family. And I tried, I tried to make him feel, uh, comfortable and assured that I loved him and I enjoyed being with him and petting him and talking to him. (Dementia 5)
Others described the interaction as calming and soothing to be able to pet the dog, participant with dementia 12 noting, “So the dog to me was a pause period. Mm-Hmm. Um, so if anything, I think it, it may not have taken my apprehension away, but I felt a little calmer” (Dementia 12). She interacted with Winston, a 5-year-old Black Labrador with therapy dog experience in outpatient treatment centers and children’s hospitals. Winston lay on the floor adjacent to her feet and allowed her to pet him while she discussed her own pets, both past and present.
Although most caregivers were absent during the interactions, they commented on the persons with dementia’s behaviors post-interaction and post-appointment. I noticed even that day just when she came back in [to the physician’s exam room], you know, she just seemed more at ease and it really did seem like she was having, had an easier time communicating. And I think because one of the things she and I have talked a lot about is anticipatory anxiety, like feeling worried because it could happen, Not even that it’s happening. Right. So I’m excited about it because I think, you know, she’s got her cat, she’s had a dog before, and I think that, I think that she’s, you know, she likes animals. (Caregiver 15)
One caregiver reported that the person with dementia described the dog as their “new best friend” (Caregiver 1) and that he went on to show pictures of the dog to friends and family after the appointment.
Among all the caregivers, caregiver 11 had the opportunity to witness the interaction with the therapy dog. Participant with dementia 11 sat in a stationary chair by the window, while caregiver 11 stood silently in the opposite corner of the room, observing the interaction. Therapy dog Winston rested beside the participant, allowing her to pet and talk to him, while the participant maintained a broad smile throughout. During the follow-up interview, caregiver 11 affirmed that the experience was positive. Oh, just how she just seemed very happy and engaged and, you know, her whole focus went on the dog, and she just seemed to forget about other things. Mm-Hmm. Just kind of went in her own little, little world kind of….well, I just think it, it’s, you know, she’s smiling. Um, just the, the touching of the dog and the kind of the stimulation a little bit and just the, just the, the closeness and the contact. I just think it, it just made, it kind of brightened her day. (Caregiver 11)
Exploratory researcher observations did not identify a consistent relationship between baseline agitation levels, as measured by the CMAI, and interaction intensity with the therapy dog. Participants with varied CMAI scores displayed diverse engagement styles, ranging from sustained verbal interaction and continuous petting to quieter, positive interactions such as gentle petting or brief exchanges. Both engagement patterns occurred across different agitation levels.
Positive Overall Experience at Doctor Appointment
Responses from people with dementia and caregivers indicated that people with dementia had a positive overall experience at the doctor appointment. Most caregivers reported that the people with dementia seemed relaxed and in an upbeat, positive mood after interacting with the dog. Some caregivers reported that people with dementia seemed to communicate more easily with the HCPs and were more receptive to questioning from HCPs. Participant with dementia 12 described the dog as having a calming, soothing effect and helped decrease apprehension leading up to the appointment. Caregiver 2 described the excitement participant with dementia 2 had for the appointment, even though there may have been underlying anxiety. Um, he hasn’t really visited the doctor until just recently. And I, I think that of course he, he is a little bit anxious about what might happen. Uhhuh. When he goes to the doctor, but he’s, he’s okay with going Mm-Hmm. He, um, when I talked to him about, um, seeing the dog at the appointment, he, he talked about that all the way up until the day we went and, and went and bought treats for the dog. So he, he was excited about the dog being there for him. Um, and I think it helped him not worry so much about what was gonna happen at the visit. (Caregiver 2)
Some people with dementia and caregivers reported they did not feel the interaction with the dog changed their feelings or behaviors when seeing the HCP. Notably, these dyads similarly reported the people with dementia being quite comfortable attending a doctor appointment, one caregiver reporting, “I mean, she was the same way coming out as she went in, and that was comfortable and relaxed, I’d say. Okay. She’s not stressed coming there.” (Caregiver 10). However, most of these dyads agreed the interaction contributed to a positive overall experience.
One HCP in a follow-up survey said they appreciated the dogs’ presence in the clinic and would miss them after the project ended. The researcher observed most HCPs warmly welcomed the therapy dogs while accompanying the person with dementia from the study visit to the exam room. Other clinic staff not treating study participants eagerly sought out therapy dogs between patient appointments, enjoying the interaction. Additionally, clinic staff expressed joy on Fridays when the dogs visited. The first therapy dog was Ozark, a 5-year-old, 127-pound Leonberger, who did not interact with a study participant but made a big impact on staff and providers. Ozark’s long, fluffy black and gold fur quickly drew attention, and as she moved through the clinic, smiles appeared on many faces. Staff happily approached to pet her and learned she was a local celebrity from a recent Duo Dog commercial. Staff valued the chance to take short breaks with the dogs and embraced the Duo Dogs Touch Teams as part of the clinic family. Photos of the dogs were added to the staff bulletin board, and a name placard was made for the exam room.
The researcher’s observations revealed that the therapy dogs did not cause safety or infection issues among participants or clinic staff members. Therapy dog handlers demonstrated a strong interest in volunteering for this study. A majority of the handlers expressed a personal connection to dementia, which heightened their motivation to volunteer specifically for this project. All handlers indicated that their participation was a positive experience for both themselves and their therapy dogs. Additionally, most expressed a desire to learn about the outcomes of the study.
Convergence
Quantitative and qualitative findings were presented together, comparing results to assess confirmation or divergence concerning the research aims. Feasibility was evaluated through quantitative data on recruitment, attrition, retention, CMAI score completion, and logistics-related Likert survey questions. Qualitative data, including researcher observations and reports from people with dementia and caregivers about burden, were also analyzed. Both data types indicated satisfactory feasibility, with quantitative results showing acceptable recruitment and retention, and qualitative findings confirming practicality.
Joint Display
Discussion
The present study had two main objectives. The first was to determine the feasibility of incorporating an AAI for people with dementia in the outpatient clinic setting by evaluating the suitability of the eligibility criteria, recruitment metrics, and completion of study procedures as per the protocol. The second was to determine the acceptability of incorporating an AAI in the memory clinic by evaluating the level of safety and burdensomeness of the intervention, as well as participant responses.
Feasibility
Feasibility was assessed by evaluating recruitment rates, the reasons why people declined participation, and whether study procedures were completed according to protocol. Regarding recruitment, the researcher obtained permission to recruit from a memory clinic in St. Louis. The researcher was granted permission to employ the AAI exclusively on Fridays, as this day was identified as a period of lower patient volume for the memory clinic HCPs. This restricted the population for recruitment, as HCPs did not actively see patients in the memory clinic every Friday. Over the span of five months of recruiting, researchers were able to screen patients on a total of only nine Fridays.
Moreover, the participant cohort was limited due to HCPs conducting many initial evaluations of patients, which resulted in a significant portion of individuals who had not yet received a formal diagnosis of dementia. Among the 80 patients screened, only 26 fulfilled the initial diagnosis criteria. While the intent of this study was not to measure the impact of AAI on agitation, the research team wanted to measure the feasibility of the intervention specifically within the population experiencing agitation. Therefore, the CMAI tool was applied during the screening process, further narrowing the subject pool. Additionally, researchers needed to account for the subjects’ willingness to participate. Consistent with prior research, participants declined to participate for reasons such as allergies or fear or anxiety around animals (Friedmann et al., 2015; Nordgren & Engstrom, 2014). Ultimately, the recruitment rate was 60.87%. Although this rate falls short of our aspirational benchmark of 80%, it is considered moderate and consistent with previous feasibility studies within this population (Fowler et al., 2021; Kelly et al., 2021).
Concerning fidelity, data were collected from all people with dementia and caregivers who could provide feedback. However, one person with dementia was unable to respond to the follow-up survey and provided minimal responses to the semi-structured interviews, as they did not recall participating in the AAI. For this participant, the follow-up interval was six days. Although reducing this delay could potentially improve memory retention, the chosen time frame is consistent with the timing of post-intervention interviews employed in prior studies (Casey et al., 2024). Given the nature of the disease being studied, it is anticipated that people with dementia may not retain memories of the interaction; nonetheless, obtaining their responses is crucial, allowing them to articulate their thoughts, feelings, and opinions (Goodall et al., 2021).
To address the issue of memory retention, prior qualitative research within the dementia population has employed phenomenological hermeneutic methods of data collection that rely on observational data (Swall et al., 2017). The present study aimed to address memory retention issues by combining in-person researcher observations with semi-structured interviews involving both people with dementia and caregivers. Researchers employed triangulation of observations and interviews with persons with dementia and caregivers to facilitate the identification and development of themes within the data.
Of the 10 people with dementia who participated, only six had corresponding HCPs who completed surveys. Interestingly, the same HCP was responsible for all six people with dementia, resulting in a lack of independent HCP responders. Four HCPs received reminders to complete the surveys, but did not respond. Researchers anticipated low response rates from HCPs due to their busy schedules and numerous responsibilities. Consistent with prior research, people with dementia and caregivers in this study responded to follow-up surveys more often than HCPs (Meyer et al., 2022).
The limited engagement from HCPs may reflect a broader feasibility barrier related to administrative burden in clinical settings. HCPs often operate within tightly scheduled workflows, where additional research tasks, such as completing surveys, may be perceived as non-essential or difficult to prioritize. The lack of response, despite reminders, suggests that even low-intensity data collection efforts can exceed what is feasible in routine clinical practice. As such, these findings not only highlight a limitation in HCP representation, but also provide important insight into the practical challenges of integrating research activities into clinical environments. Future studies should consider strategies to reduce participation burden, such as embedding data collection within existing clinical documentation systems.
Acceptability
Acceptability was evaluated by assessing the safety and burden of the AAI, as well as participant responses gathered through follow-up surveys and semi-structured interviews. The AAI was deemed safe, as no adverse events occurred amongst the people with dementia, caregivers, dogs, dog handlers, research team members, or ancillary clinic staff. The Duo Dogs Touch teams agreed to adhere to the clinic’s infection prevention policy, which included submitting medical clearance from a certified veterinarian and committing to strict hygiene routines. The absence of adverse events and infection can be attributed to the extensive training and experience of dog handlers and therapy dogs affiliated with the Duo Dogs organization.
As anticipated, people with dementia, caregivers, and HCPs generally reported a low level of burden. One caregiver mentioned that participating in the study contributed to her overall list of responsibilities and duties as a caregiver. However, this was understood to refer to the specific tasks related to the study, such as giving consent and completing follow-up surveys, rather than the AAI itself. Notably, this caregiver did not indicate any burden on the follow-up survey.
Interestingly, caregivers had mixed responses regarding participants’ baseline agitated behavior. Surprisingly, some caregivers denied that participants exhibited any behaviors; however, when caregivers were directly asked about specific behaviors while completing the CMAI, they conceded that participants did exhibit behaviors. A potential explanation for this phenomenon is caregiver denial. Some caregivers might struggle to accept their family members’ diagnosis, particularly when it is relatively recent. As a result, they may persist in denying certain behaviors associated with the condition (Boots et al., 2015).
Additionally, some people with dementia reported they were generally comfortable attending doctor appointments and did not report any signs of “white-coat” anxiety typically associated with such visits (Barber, 2020; Sindi et al., 2013). This contradicts research suggesting that healthcare environments may increase stress and produce unfavorable effects. A possible explanation is that some people with dementia may be reluctant to disclose specific vulnerabilities, or perhaps lack awareness of the expression of behavioral symptoms (Klosinska & Leszko, 2024). For example, one caregiver noted that they observed the participant appearing nervous and anxious, but the participant did not express these feelings verbally and instead withdrew. However, consistent with prior research, some people with dementia reported feeling anxious or concerned prior to attending a doctor appointment, a sentiment that their caregivers corroborated (Barber, 2020; Sindi et al., 2013). Incidentally, the average systolic blood pressure for people with dementia was elevated upon arrival to the clinic, which may be hypothesized due to stress, among other contributing factors.
People with dementia and their caregivers reported feeling comfortable interacting with therapy dogs, a finding consistent with prior research (Friedman & Krause-Parello, 2018). Most people with dementia and caregivers expressed anticipation regarding their participation in the intervention, as it provided a welcome diversion from their healthcare appointments. One person with dementia, reflecting behaviors commonly associated with dementia, expressed concerns characterized by paranoia, specifically fearing that the research team might pair him with a dog that was infected with rabies. Despite this apprehension, people with dementia reported experiencing positive emotions during interactions with the therapy dog.
The lack of a clear correlation between initial agitation severity and engagement with the therapy dog underscores the variability in participants’ responses. Engagement was observed across a broad range of BPSD levels, suggesting that responsiveness may not be solely determined by agitation severity. Factors such as personal affinity for animals, prior pet ownership experience, and comfort within the clinical environment could potentially shape interaction patterns. This finding supports the notion that AAIs may be acceptable for individuals with a range of BPSD. Nevertheless, because this relationship was not formally measured, future research should incorporate standardized assessments to more accurately determine whether baseline agitation affects the intensity of the therapy dog interaction. Due to the limited sample size and the feasibility study design, these findings are preliminary and serve as a basis for generating hypotheses.
Responses from people with dementia, caregivers, and HCPs regarding the effectiveness of the AAI in alleviating anxious or agitated feelings were varied. As the HCPs interacted infrequently with the specific person with dementia they were assessing, they often compared current behaviors to those from earlier assessments from previous visits to judge whether the AAI had any effect. This approach proved challenging, particularly given the time interval between appointments. It is worth noting that, in certain instances, a different HCP had evaluated the patient during prior visits, indicating that this may have been the first opportunity for the current HCP to assess the patient with dementia. Consequently, certain HCPs were compelled to depend on their existing, albeit limited, interactions with the person with dementia to formulate their conclusions. An additional limiting factor was the inability to blind HCPs to the intervention, which may have influenced their perceptions and evaluations. Awareness that a patient had participated in the AAI could have introduced expectancy effects, whereby HCPs consciously or unconsciously anticipated improvement, potentially shaping their interpretation of patient behaviors. Conversely, some HCPs may have been more skeptical of the AAI, which could also bias their responses.
Regarding logistical challenges within the clinic, the descriptive results showed that 100% of HCPs who completed the surveys strongly agreed that the introduction of the AAI did not hinder clinic workflow or cause disruptions. However, participation in the follow-up surveys was not universal among HCPs. Additionally, researchers made concerted efforts to coordinate the availability of clinic rooms for interventions in collaboration with clinic supervisors. Effective communication with clinic supervisors, clinic staff, and HCPs was essential to ensure the seamless operation of study procedures and prevent interruption of standard clinic flow.
Several limitations must be acknowledged when interpreting the results of this study. As this investigation was a feasibility study, the sample size was relatively small and predominantly composed of Caucasian and non-Hispanic individuals. This homogeneity restricts the generalizability of the findings to more diverse populations. In particular, cultural perceptions of animals, especially dogs, vary widely across racial, ethnic, and religious groups and may influence the acceptability of AAIs. While dogs are often viewed positively in many Western contexts, they may be perceived as unfamiliar, undesirable, or inappropriate in others, potentially affecting willingness to engage and comfort during interactions (Sinclair et al., 2022). As such, the high level of acceptability observed in this study may not translate to more culturally diverse populations without adaptation or consideration of these differing perspectives.
Due to clinic logistics and efforts to minimize workflow disruptions, a consistent duration of AAI was not uniformly applied among participants. Consequently, some participants were able to interact with therapy dogs for longer periods than others, which may have influenced their responses. Future studies involving a larger and more diverse sample of participants could investigate both the impact of interaction duration and the role of cultural factors in shaping engagement and outcomes.
Follow-up surveys and interviews were intended to be collected as soon as possible, but within seven days of the clinic appointment. However, scheduling constraints for people with dementia and caregivers made this time frame challenging. This represents a notable limitation of this study, particularly given that memory retention challenges are a common characteristic among people with dementia. Delays in data collection may have contributed to recall bias. Additionally, most interviews were conducted via telephone, primarily due to participant convenience and the researcher’s limited ability to conduct in-person interviews. Telephone interviews restrict the researcher’s ability to control the interview environment, minimize distractions, and confirm the participant is adequately hearing and understanding the questions being asked (Samsi & Manthorpe, 2020).
The study’s goals were limited to evaluating the feasibility and acceptability of the intervention; therefore, definitive causal relationships among variables cannot be established. The data collection methods, such as Likert scale follow-up surveys, introduced the possibility of subjectivity and biased responses. Future research should emphasize the use of objective data measurements and the application of rigorous statistical methods to evaluate the causal relationship between AAIs and agitation in people with dementia.
In research related to AAI, there exists a potential for bias introduced by the dog handlers. All handlers received explicit instructions to minimize engagement with participants, intervening only when directly solicited for questions or in situations where essential guidance was required. Notably, no references were made by the persons with dementia regarding the dog handlers during the follow-up surveys or telephone interviews.
Finally, the potential for researcher bias must be acknowledged. Although the PI did not have a family member living with dementia at the time of data collection, they previously had a close relative experiencing BPSD. This prior experience may have influenced the PI’s perspective and interpretation of participant experiences. However, efforts were made to minimize internal bias through self-reflection and reflexivity during and following data collection.
Conclusion
This study aimed to provide insight into the feasibility and acceptability of an AAI in people with dementia in the outpatient clinic setting. Feedback was collected from three groups of individuals impacted by the AAI: participants with dementia, caregivers, and HCPs. Results showed that people with dementia and caregivers were excited and felt comfortable about the opportunity to participate, had positive interactions with the dog, and felt the intervention contributed to an overall positive experience at the doctor appointment. HCPs agreed that therapy dogs did not cause burdens or logistical issues with patients or the clinic workflow. The results of this study established the feasibility and acceptability of AAI in people with dementia in an outpatient clinic setting; however, further research is warranted to address the measurable effects of AAI on agitation in people with dementia within this setting.
Footnotes
Acknowledgements
We would like to acknowledge the Duo Dogs organization and their volunteer Touch Teams for their generous dedication of time and effort to this project. Their partnership is profoundly appreciated. We also extend our gratitude to the staff of the Washington University School of Medicine Department of Neurology clinic. Their collaboration was essential to the success of this project.
Ethical Considerations
This study was approved by Washington University’s Institutional Review Board (identification number: 202306151) on August 8, 2023. The study methodology was conducted following the principles of the Declaration of Helsinki.
Consent to Participate
Informed consent was obtained from the subjects or their legally authorized representatives before study participation.
Consent for Publication
The corresponding author holds written consent for all participating individuals and for all published images of therapy dogs.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
The datasets generated during and/or analyzed during the current study are available from the corresponding author on request.