ROBOTIC PETS AND DEMENTIA

1

Robotic Pets: Supplemental Rehabilitation for the Dementia-Suffering Individual

Rushana M. Gill-Ramos
Edson College of Nursing and Health Innovation, Arizona State University

Author Note
Rushana M. Gill-Ramos, DNP Innovation Leadership Candidate. Edson College of
Nursing and Health Innovation, Arizona State University.
There is no conflict of interest to disclose.
Correspondence concerning this project can be emailed to rgillram@asu.edu.

ROBOTIC PETS AND DEMENTIA

2
Abstract

People with dementia (PwD) along with their caregivers currently face challenges in the disease
process because there is a major gap in available treatment and therapies, especially those that
are non-pharmacological. The literature shows an increased incidence of dementia, and currently,
there is no cure. This leaves medication as the treatment of choice. The aim of this project was to
investigate whether introducing robotic pets as a supplemental therapy to persons with mild to
moderate dementia helps in reducing their level of anxiety and agitation. Based on a review of
the literature on the benefits of robotic pet interactions with dementia-suffering individuals, a
quality improvement project was conducted in a suburban city in Arizona. Project participants
included individuals with dementia and their caregivers. They were asked to interact with the
robotic pet in their home for four, one-hour sessions, over a four-week timeframe. Analysis of
the findings revealed that individuals with dementia found a sense of joy and peace while
interacting with the pets, and they looked forward to the pet visit every week with excitement.
These results indicated that introducing robotic pets to dementia-suffering individuals was
clinically significant as it relates to their anxiety and agitation levels. Utilizing robotic pets as a
supplemental therapy can benefit dementia-suffering individuals and their caregivers. Further
evaluation is needed with a larger sample size to better understand the impact of robotic pets on
PwD.
Keywords: Therapy, Alzheimer’s, standard of care, cognitive impairment, anxiety, agitation

ROBOTIC PETS AND DEMENTIA

3

Robotic Pets: Supplemental Rehabilitation for the Dementia-Suffering Individual
One person doesn’t get dementia, the entire family does. Finding that one thing that can
help to decrease behavioral and psychological symptoms of dementia (BPSD) can make a huge
difference in the lives of people with dementia (PwD) and their caregivers. This project will
explore dementia, including its national and local significance, and will demonstrate the impact
of robotic pets on PwD.
Background and Significance
Worldwide, 50 million people are living with dementia (Alzheimer’s Association, 2020).
By 2060, the number of people aged 65 and older with dementia is projected to reach 13.8
million, barring the development of medical breakthroughs to prevent, slow, or cure the
condition (Rajan et al., 2021). The typical life expectancy after a dementia diagnosis is 4–8 years
(Centers for Disease Control, 2020). Researchers expect to spend $2.8 billion on dementia
studying everything from blood vessels to hormonal factors to solve the dementia dilemma
(Nania, 2020). In 2018, the American Association of Retired Persons (AARP) committed to
investing $60 million with Dementia Discovery Fund, a $350 million London-based venture
capital group that invests in new therapeutic projects to support the quest for a cure (Nania,
2020). The purpose of the current project is to investigate the emerging evidence that suggests
anxiety and agitation levels can be decreased by introducing robotic pets as a supplemental
therapy to people with mild to moderate dementia.
Alzheimer’s Disease and Related Dementias (ADRD)
Dementia is not a single disease; it is an overall term—like heart disease—that covers a
wide range of other dementia conditions, like Alzheimer’s disease (Alzheimer’s Association,
2020). Recent estimates indicate that Alzheimer’s may rank third, behind heart disease and

ROBOTIC PETS AND DEMENTIA

4

cancer, as a cause of death for older people (National Institute on Aging, 2019). There is a new
case of dementia in the world every 4 seconds, and every 66 seconds, someone in the United
States develops Alzheimer’s (Alzheimer’s Disease International, 2020). One in three seniors in
the United States dies from some form of dementia (Alzheimer’s Disease International, 2020).
The burden of ADRD in 2014 was 5 million people, which was 1.6% of the U.S.
population that year (Centers for Disease Control and Prevention [CDC], 2019). This burden is
projected to grow to 13.9 million, nearly 3.3% of the population, by 2060 (CDC, 2019). Thus,
the U.S. burden of ADRD will double by 2060, according to a study from the CDC (2019).
Although there is currently no cure for ADRD, health-care professionals continue to treat
individuals diagnosed with ADRD with the current standard of care, which includes medication
that may attempt to decrease the rate at which ADRD progress. Because of the increase in side
effects of ADRD medications that contribute to BPSD, researchers have been investigating the
use of robotic pet therapy to reduce or prevent BPSD. Recently, many studies have shown added
benefits to using robotic pet therapy in the treatment of PwD to decrease agitation and anxiety
levels as well as other BPSD.
Dementia
Dementia is a general term for loss of memory, language, problem-solving, and other
thinking abilities that if severe enough can interfere with quality of life (QoL) (Alzheimer’s
Association, 2020). The causes of dementia can vary, depending on the types of brain changes
that may be taking place (National Institute on Aging, 2019). Dementia is not a specific disease
but an overall term describing a group of symptoms (Alzheimer’s Association, 2020).
Alzheimer’s disease is thought to begin 20 years or more before symptoms arise, and it starts
with changes in the brain that are unnoticeable to the person affected. Abnormal brain changes

ROBOTIC PETS AND DEMENTIA

5

cause disorders grouped under the general term “dementia,” and only after years do individuals
experience noticeable symptoms such as memory loss and language problems (Alzheimer’s
Association, 2020a, 2020b). In the past, the term “Alzheimer’s disease” was often used to
describe the dementia phase of the disease, but dementia is only one part of the disease
(Alzheimer’s Association, 2020b). For example, some people have both Alzheimer’s disease and
vascular dementia (National Institute on Aging, 2019). The continuum starts with preclinical
Alzheimer’s and ends with severe Alzheimer’s dementia, and the duration for each part of the
continuum varies (Alzheimer’s Association, 2020b).
Nonpharmacological Therapy
Various nonpharmacological interventions for PwD include occupational therapies;
cognitive training, stimulation, and rehabilitation; exercise programs such as aerobic exercise
and strength training; art therapy; and music therapy. Nonpharmacological therapies are used for
PwD to help maintain or improve their cognitive function, QoL, and ability to perform activities
of daily living (ADLs). Nonpharmacological treatment in PwD has been shown to be more
effective than pharmacological therapy in reducing agitation and anxiety (Watt et al., 2019). The
difference between the current standard of care which is pharmacological therapy and
nonpharmacological therapy is that it does not interfere with the destruction of the neurons of the
brain that causes the symptoms and makes the disease fatal. There is always difficulty in
determining the effectiveness of nonpharmacological therapy because of the different types of
therapy being tested: including therapeutic aims, stages of dementia, and types of dementia the
PwD may have.

ROBOTIC PETS AND DEMENTIA

6

Robotic Pets
Dementia is an umbrella term covering diseases that affect one’s memory and alter one’s
ability to perform ADLs. Dementia has been growing fast in the absence of a cure, but major
investments are being made in research funding. A tremendous responsibility is placed on
scientists and health professionals to look deeper into every aspect of the human brain and make
sure that nothing is missed in diagnosing, slowing, and treating the disease. The standard of care
has always included medication therapy. Research now shows the benefits of robotic pet therapy
in the treatment of PwD. Introducing robotic pets can help reduce anxiety and agitation levels as
well as alleviate social isolation and feelings of loneliness, which may contribute to the
improvement in cognitive health. Robotic pet therapy can improve the physical and emotional
moods of elderly people with and without dementia. Nonpharmacological therapies such as
animal interactions, dolls, and robotic pets do not involve medication or drugs. They are used to
distract PwD from experiencing BPSD such as anxiety, agitation, wandering, aggression,
depression, and/or feelings of loneliness.
Intervention to Decrease Anxiety and Agitation
Although there are currently no FDA-approved pharmacologic treatments that prevent or
cure Alzheimer’s disease or slow its progression, several groups of researchers have estimated
the cost savings of future interventions that either slow the onset of dementia or reduce its
symptoms (Alzheimer’s Association, 2020b, 2021; Jutkowitz et al., 2017). Researchers estimated
that a treatment that slows the rate of functional decline by 10% would reduce average perperson lifetime costs by $4,504 in 2020, and a treatment that reduces the number of behavioral
and psychological symptoms by 10% would reduce average per-person lifetime costs by $789 in
2020 (Jutkowitz et al., 2017). A total of 68% of the projected increase in the global prevalence

ROBOTIC PETS AND DEMENTIA

7

and burden of dementia by 2050 will take place in low- and middle-income countries
(Alzheimer’s Association, 2020).
Animal interaction has long been known to benefit people emotionally. Robotic pet
therapy serves as a nonmedicinal intervention to enhance the QoL and reduce the negative
effects associated with dementia (Sicurella & Fitzsimmons, 2016). Robotic pets enhance
effective and social outcomes for PwD (Liang et al., 2017).
Internal Evidence
Located in a large urban area, this city was the first in Arizona to join Dementia Friendly
America’s initiative to ensure that communities in the United States are equipped to support
PwD and their caregivers. That makes this inner suburban community Arizona’s first dementiafriendly city. In 2015, at the White House Conference on Aging, this city was designated as one
of the six up-and-coming dementia communities and was tasked to help lead the national
initiative along with Dementia International. Since then, the city has partnered with a health care
organization, devoted to Alzheimer’s, to make its community a livable place for PwD by offering
training, programs, and resources to employees as well as the community. In the community,
about 1,500 people aged over 65 have dementia. Arizona will have the second-highest number of
PwD in the nation in the next 10 years, resulting in 3,000 PwD living in the community (U.S.
Census, 2020).
This inner suburban city continues to raise awareness, transform the attitudes of people in
the community, and move people to action to improve the city’s performance measures. As the
Dementia Friendly Community Designation their goal is to create an inclusive, informed, safe,
and respectful community for PwD. The city’s intentions are to raise support to improve the QoL
for people touched by the disease, including caregivers and families of PwD; working towards

ROBOTIC PETS AND DEMENTIA

8

engaging its diverse population and promoting community life, and encouraging the people who
live, work, and play in the city to become dementia friends. Introducing robotic pets to PwD to
help decrease anxiety and agitation levels will align with their City Council Priority for Quality
of Life and help to increase ratings in the city’s engagement and inclusion of PwD, their care
partners, and their families (Census.gov, 2020).
As the demographic composition of the United States changes, around 10,000 baby
boomers turn 65 each day (U.S. Census Bureau, 2020). Age is the single greatest risk of ADRD
(Alzheimer’s Association, 2020). The burden surrounding ADRD has doubled as of 2020. In
Arizona alone, the number of individuals living with Alzheimer’s has increased by 60%. ADRD
affects the entire household (about four-to-five people, if not more) (Alzheimer’s Association,
2020).
PICOT Question
Millions of older Americans are being diagnosed with ADRD. Although researchers are
striving to find a cure for families and PwD, only evidence-based practices (EBPs) to help PwD
achieve the best QoL are available. Caring for PwD may be challenging, and the responsibility
usually falls on the child, partner, or family member. Increased anxiety and agitation levels are
associated with violent/abusive episodes as well as a decline in cognitive ability, which may lead
caregivers to place PwD in nursing homes or long-term care facilities.
EBPs suggest that caring for PwD in their own homes and in familiar environments
prevents anxiety, and agitation, and slows the decline in cognitive ability. The home is
considered the preferred care environment to help PwD live the best QoL attainable. As studies
have shown, robotic pets can bridge the gap between PwD and their anxiety and agitation levels.
This finding raises the following clinically relevant PICOT question: Does the introduction of

ROBOTIC PETS AND DEMENTIA

9

robotic pets as a supplemental rehabilitation therapy help reduce anxiety and agitation levels in
mild to moderate dementia-suffering individuals?
Evidence Synthesis
Search Strategies
An extensive search of three databases, CINAHL, PubMed, and PsycINFO was
conducted to explore the current data available. These databases based were chosen to help
answer the PICOT question comparing robotic pets to the current standard of care and exploring
the possibility of reducing anxiety and agitation levels in PwD. These databases were searched
using key terms in PICOT question such as “dementia,” “Alzheimer’s,” “standard of care,”
“cognitive impairment,” “anxiety,” and “agitation.” “Robotic pet,” “robotic pet therapy,” and
“animal therapy” were not included in the key terms because they returned a significantly low
number of results. Instead, “nonpharmacological treatment and/or nonpharmacological therapy”
and “reducing anxiety and agitation” yielded more results. Filters used in the search included
articles written in the English language and publications from 2016–2021 and 2004 were
included to review the first study conducted on robotic pets and PwD.
An initial search of CINAHL with the key terms “dementia,” “Alzheimer’s,” “cognitive
impairment,” and “robotic pet” produced no results, so the search keys robot*, therap*, Agitat*,
and anxiet* along with other Mesh terms was added. These yielded 3,644 results. After adding
more filters, the search results ranged from 1–506. An initial search of PubMed using the key
terms “dementia,” “agitation,” “anxiety,” and “nonpharmacological therapy” yielded one result.
After adding Mesh words, 11,922 results were obtained. After applying filters, the results ranged
from 0–1,564. An initial search of PscyhINFO with key terms “dementia,” “Alzheimer’s,”

ROBOTIC PETS AND DEMENTIA

10

“cognitive impairment,” and “therapy” yielded 11,313 results. After applying filters and adding
Mesh terms, results more focused on my PICOT question produced a range from 0–1,190.
A total of 44 studies were deemed relevant because their titles and abstracts were closely
related to my PICOT question. A total of 10 studies were chosen for the annotated bibliography
and three for the rapid critical appraisal table. These studies consisted of a longitudinal study, a
mixed-methods trial, and a randomized control trial. Inclusion criteria included PwD, PwD living
at home, health-care professionals encountering or treating PwD, and the geriatric population.
The exclusion criteria were PwD that could not participate in care or decision-making and could
not consent.
Critical Appraisal and Synthesis
Ten studies were included in this review. Most had a high level of evidence and limited
bias. There were three systematic reviews, four randomized control trials, one mixed-methods
study, one quantitative longitudinal study, and one qualitative study (Appendix A). All studies
included PwD and older adults, and their authors examined the effects of anxiety and
agitation/loneliness and robotic pet interaction. Melnyk and Fineout-Overholt’s (2019)
(Appendix B) rapid critical appraisal process was used to determine the quality and strength of
the articles selected for this review. The articles provided limited information on demographics
and included participants aged 55 and older. Most studies were conducted internationally, with
two studies conducted within the United States and two articles within the SRs originating in the
United States. The SRs had a small sample size but rigorous inclusion criteria and measurement
tools, which increased their breadth for inclusion in this review. The implementation of
interventions involving robotic pets, real pets, and plush pets was done either in a residential
home and/or an institution or both.

ROBOTIC PETS AND DEMENTIA

11

These interventions included introducing robotic pets to dementia patients, evaluating the effects,
and comparing this group to a group that participated in the standard activity of the institution.
Sessions were held either individually or as a group, and participants were given the opportunity
to spend time with the robotic pets. Heterogeneity was significant in the tools and designs that
were used. Although there was variation in the measurement tools, three of the studies (Banks et
al., 2008; Birks et al., 2016; Libin & Cohen-Mansfield, 2004) were focused on robotic pets and
the effects of anxiety and agitation in elderly and/or elderly dementia patients. The dependent
variables in these studies also included loneliness, depression, mood, social interaction,
medication, cognitive function, and QoL. Most of the reviewed studies showed significant
improvements in BPSD. Significant changes in behavioral symptoms included positive effects on
a participant’s anxiety and agitation level, feelings of loneliness, feelings of depression, moods,
and increased social interaction with staff and residents. Studies showed a positive effect on
cognitive function and QoL. More studies need to be conducted to obtain significant data to
support the current evidence.
Theoretical Framework
The Theory of Human Caring (Watson, 2006) is applicable for this project because of the
notion that health professionals make moral, social, and scientific contributions to humankind
and by caring, which influences human development (Appendix C). Nurses are at the center of
healing. The constructs of this theory embrace the importance of helping to enhance nursing
practice, management, education, and research. Assisting with basic needs with an intentional
caring consciousness and administering “human care essentials” will potentiate the alignment of
the mind, body, and spirit and introduce wholeness into all aspects of care (Wagner, 2019).

ROBOTIC PETS AND DEMENTIA

12

This project focused on two constructs – creating a healing environment and teaching and
learning. Creating a healing environment can go both ways. For dementia sufferers, their
environment can be crucial to their QoL as the home is the recommended healing ground for
dementia. This theory is very relevant to this project because it reflects the fundamental aspects
of caring, in a holistic way. Watson formulated her theory based on Florence Nightingale’s
beliefs which can be described as the soul of nursing. Nurses go out into the environment, see
where the pain is, wrap their arms around it, and try to provide comfort. Especially, when there
is no certain solution. Providing a tool, in this case, a robotic pet that can help alleviate anxiety
and agitation, not having to worry about it causing a reaction with the medication or adding on to
the already tedious medication regime reflects caring principles. It’s a form of distraction that
can decrease these symptoms that may lead to abusive behaviors, if not treated. It instills faith
and hope knowing that your loved ones are happy, at least at that moment. PwD will be able to
stay in the home environment for longer if the BPSD that is often present due to medication can
be suppressed or controlled.
When it comes to teaching and learning, being able to utilize and explore other EBPs to
investigate the impact of dementia, and how or what has been done, helped choose the direction
for this project. Robotic pets are a readily available tool to help both dementia sufferers and
caregivers in improving their QoL as well as their mental health. The aim of the project is to
provide a caring environment that can promote healing and growth and encourage a sense of
calmness to help decrease anxiety and agitation levels in PwD.
Implementation Framework
Project implementation followed the intervention mapping framework by BartholomewEldredge et al. (2016). This framework was chosen because it encompassed the core constructs

ROBOTIC PETS AND DEMENTIA

13

of this project, focusing on knowledge from existing research to guide the implementation of
evidence-based intervention in a community setting (Appendix D). The framework further
guided this health promotion project within a dementia-populated community and explored
different responses to anxiety and agitation in PwD. The intervention mapping framework has
six steps. The first step involves the development of comprehension of the purposed issue and
formulating objectives to solve the issue. The next step involves using guidance to obtain
outcomes and objectives in an analytical manner. The other steps involve designing the project
using theories as guides, implementing the project, and evaluating the implementation method to
determine if one has achieved one’s objectives or outcomes.
A logic model was created to depict the impact of the project (Appendix E). These steps
were navigated within the context of the project by first establishing and working with a
planning group, completing a needs assessment regarding the Dementia Initiative, and
understanding other relevant evidence-based literature. Changes were made to the project design
focusing on specific needs, leadership changes, and the evolving pandemic changes. Moving
through these steps helped in creating a well-developed answer to the PICOT question. This
framework also provided an opportunity for results and data to be reexamined and corrections to
be made, while placing them in the right step within the framework. Using this framework has
made it possible for this project to achieve the best results from the data obtained during the
implementation of this project.
Methods
Recruitment Procedure and Ethical Considerations
Dementia Friendly site volunteers and the co-primary investigator (CO-PI) recruited
participants. The CO-PI solely obtained the consent of participants, and volunteers acted as

ROBOTIC PETS AND DEMENTIA

14

supplemental resources for the project. The project was introduced to the City’s Dementia
community via an email that Dementia Friendly regularly sends out to Dementia Memory Café
members and family support groups. Informational flyers (Appendix F) were handed out at the
in-person Memory Café meetings. This City designated a large community room for recruiting
and educating participants for this QI project. This area provided a safe and comfortable
environment for project participants to transition from the Memory Café. Project participants
were recruited from the attendees.
The recruitment was completed in a two-step process. Volunteers completed the first step
using a well-defined script to describe the project to participants (Appendix G). These volunteers
completed the Dementia-Friendly Information Session training to become dementia champions
to create consistency when dealing with project participants. Objectives of the session included a
description of dementia and the types of dementia, key messages about dementia, approaches to
communicating with PwD, and the commitment to serve as a “dementia friend” in their
community.
Then the CO-PI provided a detailed project presentation to the Memory Café participants,
including the site champion, host, staff, volunteers, and community members. Once primary
caregivers and PwD displayed interest and willingness to participate in the project, their names,
and numbers were recorded. The CO-PI called all interested persons (n = 30) individually to
discuss any concerns or answer any questions they had regarding the project. Five PwD and their
primary caregivers were chosen for the project. All five individuals that were chosen met the
inclusion criteria with a special focus on availability to participate in all aspects of the project.
The consent process took place in the comfort of participants’ homes on the first visit to
ensure that participants were fully informed about the QI project as well as to give participants

ROBOTIC PETS AND DEMENTIA

15

the opportunity to voice their questions and concerns. Handwritten signatures were obtained
authorizing consent on a standardized form that outlined the details of the project (Appendix H).
Obtaining the consent of participants was the sole responsibility of the CO-PI. This was to ensure
that all participants were equally educated on the project.
On the first visit, the CO-PI evaluated the decision-making capacity of the PwD using the
University of California, San Diego, Brief Assessment of Capacity to Consent (UBACC)
(Appendix I). If a score of 15 or above was obtained on UBACC, the PwD would consent, and
the primary caregiver would sign as a witness. If the PwD was cognitively unable to consent or
received a score of less than 15, then the primary caregiver, who is the legally authorized agent
for the PwD, consented on their behalf. Participants were informed of their rights and assured
that they could withdraw from the project at any time without any consequences or explanations.
Participants were informed that withdrawing from the project would not affect their relationship
with the Dementia Friendly Organization or interfere with the services the organization provided
to them. This QI project was approved by the Institutional Review Board of Arizona State
University as an Expedited Review.
Participants
Ten participants were included as the convenience sample: five being dementia-suffering
individuals and the other five their primary caregivers, who were also their spouses. Inclusion
criteria for the QI project included individuals aged 55 and over; diagnosed with mild to
moderate dementia, as disclosed by the caregiver; who understood, spoke, and read English;
lived with a consenting caregiver/guardian; individual living at home; were available for hourly
visits, and was reachable by phone. Exclusion criteria included individuals not diagnosed with
dementia; individuals unable to participate in their care; individuals with no primary caregivers

ROBOTIC PETS AND DEMENTIA

16

to give consent; individuals not living at home; individuals who could not be available for hourly
sessions; and individuals not reachable by phone. No minors, prisoners, and economically or
educationally disadvantaged individuals were included in this QI project. However, adults with
impaired decision-making capacity were included and provided consent along with their primary
caregiver.
PwD are typically diagnosed by an accredited provider as having symptoms affecting
memory, thinking, and social abilities that may interfere with their daily life. Participants and/or
primary caregivers’ recognition of PwD was based on their knowledge of healthcare
professionals’ diagnosis of dementia. Therefore, no formal written diagnosis or assessment of
dementia was required for this project. The absence of primary caregivers to give consent
excluded participants from the QI project. Recruited caregivers were the primary caregivers of
the PWD with the authority to give consent for participation.
Setting and Procedure
The quality improvement project involved the observation and management of anxiety
and agitation levels of mild to moderately diagnosed dementia-suffering individuals while they
interacted with a robotic pet in the comfort of their own homes in a large urban community in
Arizona. Primary caregivers were given a brief introduction to how the robotic pet works and an
instruction manual and care and instruction for robotic pets. Last, the pet was introduced to the
PwD. The project was thoroughly reviewed during the first visit, including visit schedule and
duration, CO-PI role, types of observations, and measurements (Appendix J).
Robotic Pet Intervention
This QI project focused on the benefits of introducing robotic pets to PwD. These robotic
pets have lifelike qualities including sound and movement and the ability to respond to

ROBOTIC PETS AND DEMENTIA

17

movement and touch. All participants preferred to interact with the golden retriever dog, instead
of the cat. This battery-operated, robotic pet was designed by the manufacturer, Joy for All, to
look and feel like a real dog. Its state-of-the-art art technology creates a life-like appearance. It
has built-in sensors that allow it to respond to presence and touch, realistically reacting to its
environment. Features of the robotic pet include a sensor of the head, cheek, mouth, back, and a
light one above the right ear. A microphone is near the upper part of its right front leg and has an
on/mute/off switch and a battery door in the tummy area. One can hear a “thump, thump” heart
sound, it “WOOFS!” and makes other realistic “pup-like” noises. Petting the pup over the
sensor areas will evoke a response and if there is no interaction after a few minutes it will go to
sleep. The pup is considered a lap pup and can’t walk or run (JoyForAll.com, 2018).
Measures
Brief Assessment of Capacity to Consent (UBACC)
UBACC is a shorter version of the MacArthur-Competence-Assessment-Tool-forClinical-Research. It is brief and screens for capacity to consent to research participation among
those at risk for decisional capacity (Seaman et al., 2015). UBACC was created as a short and
efficient way of conducting screening without extensive training of research staff. It is an
alternative rather than a comprehensive assessment for measuring capacity to consent. UBACC
contains 10 questions related to reasoning, appreciation, and understanding (see Appendix H for
a full set of questions). Individuals are scored from 0 to 2, with 0 indicating little to no
understanding of the project and 2 indicating a clear understanding of the project. A score of 15
or higher indicates that an individual can consent to the project. If the individual scores less than
15, the primary caregiver provided consent to participate in the project. The instrument
demonstrated acceptable levels of reliability with a Cronbach’s alpha of 0.77.

ROBOTIC PETS AND DEMENTIA

18

Agitated Behavior Scale (ABS)
The ABS is used to measure the extent and nature of agitation in individuals recovering
from a brain injury. This instrument may also be used to measure agitation in nursing home
residents with Alzheimer’s/progressive dementia (Bogner, 2000). The ABS contains 14
questions related to the anxiety levels of individuals (see Appendix K for a full set of questions).
Individuals are scored on the degree to which a behavior is present: 1 = Absent (not present), 2 =
Present to a slight degree (does not interfere/disrupt), 3 = Present to a moderate degree
(interferes/disrupts but can be redirected), and 4 = Present to an extreme degree
(interferes/disrupts and cannot be redirected). A pre-interaction survey was conducted before the
session started, and a post-interaction survey was conducted after the session. A score of 21 or
less is normal, a score of 22–28 is mild, a score of 29–35 is moderate, and a score of 35 or more
is severe. The instrument demonstrated good levels of reliability with a Cronbach’s alpha
between 0.801 and 0.921.
Rating Anxiety in Dementia (RAID)
The RAID scale is used to evaluate anxiety in PwD. Standardization in administration
and scoring makes RAID easy for non-clinicians to use, and it exhibits good reliability and
validity in older adults with dementia (Snow et al., 2012). The RAID scale contains 20 questions
related to anxiety (see Appendix L for a full set of questions): five questions related to worry,
four questions related to apprehension and vigilance, four questions related to motor tension, five
questions related to autonomic hypersensitivity, one question related to phobia, and one question
related to panic attacks. The RAID scale has the following scoring system: U = unable to
evaluate, 0 = absent, 1 = mild or intermittent, 2 = moderate, 3 = severe. A score of 11 or more
suggests significant clinical anxiety. Ratings in this project were based on signs and symptoms

ROBOTIC PETS AND DEMENTIA

19

occurring pre- and post-interactions with robotic pets. The instrument demonstrated good levels
of reliability with a Cronbach’s alpha of 0.81.
Caregiver Evaluation of Robotic Pet Interaction Survey
The Caregiver Evaluation of Robotic Pet Interaction Survey was newly created for the
purposes of this project. This instrument evaluated caregivers’ perceptions and opinions on the
benefits of robotic pets, user-friendliness, and the effects the pet has on the anxiety and agitation
levels of their loved ones. There are a total of seven questions with a Likert scale from 1 =
strongly disagree to 5 = strongly agree and one additional comment box is provided (see
Appendix M for a full set of questions).
Results
Outcomes
The total number of participants was 10. Of these 10 people, there were 5 male
individuals diagnosed with dementia and 5 female individuals who were their primary caregivers
as well as their spouses. The age range for participants was 74 to 82 years old and the average
age for participants was 77.4 years old. Out of the 5 male participants 2 were diagnosed with
mild dementia and the other 3 had moderate dementia. Participants received the Hasbro
companion robot of their choice – a golden retriever; none selected a cat as they were all former
dog owners. Given the changing needs during the pandemic, personal funds were used to
implement this QI project and ensure that strict infection control measures were upheld. All
participants were provided with their own robotic pet at a cost of $129.00 per dog. The total
budget for this QI project was an estimated $2,309, including personal computers and printers
and recruitment materials/stationery (Appendix N).

ROBOTIC PETS AND DEMENTIA

20

Caregivers were introduced to the robotic function and given an instruction manual for
the care and use of the robotic pet. The robotic pets were not left in the home instead the
intervention in this project consisted of one- one hour visit, one day of the week over a period of
four weeks. Pre- and post-surveys were completed; before interaction - looking at the
individual's anxiety and agitation levels over the week and post looking at anxiety and agitation
levels after interaction with the robotic pet. At the end of the 4 weeks in addition to the pre and
post survey, caregivers were asked to do a caregiver evaluation of the robotic pet interaction.
The level of anxiety and agitation was measured before interaction and after interacting with
robotic pets using the RAID and ABS scale. A Shapiro-Wilks test was done initially to test for
homogeneity, the results were not significant for RAID based on an alpha value of .05, W=0.89,
p=.349 and ABS based on an alpha value of .05, W=0.96, p=.801. The results indicated that the
data were normally distributed from the convenience sample.

Figure 1
RAID and ABS Pre and Post Totals

Note: Two-Tailed Paired Samples t-Test, pre and post test means
Statistically significant change in RAID p < .02; ABS not statistically significant p < .06

ROBOTIC PETS AND DEMENTIA

21

As shown in Figure 1, the result of the two-tailed paired sample t-test was significant for
RAID based on an alpha value of .05, t(4) = 3.53, p=.024. The result of the two-tailed paired
samples t-test was not significant for the ABS based on an alpha value of .05, t(4) = 2.50,
p=.067. It was observed that before the interaction with the robotic pets the mean total score for
RAID survey was 15.8 (SD = 11.41, range 4 – 32), which indicated that there was significant
clinical anxiety present among participants. The mean total score of ABS survey was 19.8 (SD
5.40, range 14 – 26) which falls within the normal range for agitation. At the end of interactions
between the PwD and the robotic pet, total mean score for RAID was 0.25 (SD = 0.50, range 0 –
1) and ABS was 14.60 (SD = 1.79, range 14 – 18). Both survey results indicated that they were
positive effects of interaction with robotic pets. The RAID Anxiety level mean dropped from
15.8 to 2.2; an 86% decrease. The ABS agitation level, even though it was reported in the
normal range, went from 19.8 to 14.8, a 25% decrease.
Figure 2
RAID Symptoms Pre and Post Scores

Note: Summary Statistics for pre and post scores of symptoms associated with anxiety. A notable
decrease in Trifles and Tiredness pre and post scores was observed.

ROBOTIC PETS AND DEMENTIA

22

The results in Figure 2 indicated that Anxiety symptoms were significantly reduced pre
vs post RAID survey, including a notable reduction in Trifles pre 1.40 (SD = 1.34, range 0 – 3)
to post-survey 0.40 (SD = 0.89, range 0 – 2) and Tiredness pre 1.40 (SD = 0.89, range 0 – 2) to
post-survey 0 (SD = 0, range 0 – 0). Trifles are repeatedly calling for attention over trivial
matters or attention-seeking behavior. An example from one of the caregivers was, “Whenever
he’s occupied with the pet, he doesn’t shadow me around and I can get stuff done in the kitchen
while keeping an eye on him.” As for relief from tiredness, PwD was excited once they got the
pet and start cuddling, smiling, and talking to the pet. An example from a caregiver was, “He’s
much more talkative, I think he sits there imagining he has the pet and just keeps talking, I don’t
understand what he’s saying but he’s relaxed, smiling, and conversing with himself.”
Other caregiver comments included, “He is calmer and more relaxed.” “This opportunity
has proven to be so beneficial for my husband. He does not fidget with the dog in his hands. He
strokes it and smiles.” This particular caregiver had a basket filled with small gadgets/devices to
otherwise keep her husband’s hands busy. “The robotic pet; increased interest in the pet over the
phone and calmness and happiness in overall daily interaction with him.” When this individual
received a call from his friends, he would excitedly tell them about the pet and made plans to
take the pet to the community center at the end of the project.

ROBOTIC PETS AND DEMENTIA

23

Figure 3
ABS Symptoms Pre and Post Scores

Note: Summary Statistics for pre and post score of symptoms associated with agitation.
Uncooperative, demanding, and resisting care were highest and was notably less during
observation.
As indicated in Figure 3, even though the agitation level scored within normal per ABS
survey, a clinically significant drop in uncooperative, demanding, and resisting care was
observed. Before the interaction with the robotic pet, uncooperative, demanding, and resisting
care mean scores were 2.20 (SD = 0.84, range 1 – 3), the highest on the ABS scale. After the
robotic pet interaction, the average uncooperative, demanding, and resisting care values were 1
(SD = 0, range 1 – 1), a 25% decrease.

ROBOTIC PETS AND DEMENTIA

24

While not statistically significant the findings were clinically relevant as noted by an
observed reduction in specific behavior after interaction with the robotic pet. The findings
suggest that there are benefits in utilizing robotic pets as a supplemental rehabilitative therapy
and should be studied further. This gets the site for this project one step closer to achieving its
desired performance goal.
Observation of PwD
While observing the PwD interacting with the robotic pets, they were more verbal: “I
wish you could live with me,” “You should move in so the pet can stay,” “I don’t want you to
go,” “I love you,” and “I miss you.” They became very pleasantly verbal when interacting with
the robotic pets. Their mood changed. They smiled, they cuddled, and they rubbed noses with the
pets, presenting a very caring and loving demeanor in the presence of or while interacting with
the robotic pets. As the process went on, the benefit of interactions between moderately
diagnosed vs mildly diagnosed participants became obvious because these were the people who
created a real bond with their pets.
In the mildly diagnosed individuals, it was great to watch how they related to the pets and
the way they perceived themselves or their roles in interacting with the pets. The mildly
diagnosed PwD were very verbal in giving their opinion on who they thought would benefit from
having the pets. They also gave their suggestions on robots and how robots have evolved over
the years. The mildly diagnosed PwDs enjoyed interacting with the pets because this built their
curiosity about why and how the pets looked like real puppies and behaved like real puppies. As
they explored the robotic pets and listened to and felt the pets’ heartbeats, they expressed
fascination and amazement. They were in tune with the interactions and curious about what was

ROBOTIC PETS AND DEMENTIA

25

under the fur and where the batteries were located. This is important because mildly diagnosed
PwDs tend to forget they asked a question and keep asking it.
In mildly diagnosed individuals, one of the major influencing factors of their behavior or
moods is their tendency to mimic the behaviors of their caregivers. If the caregiver becomes
frustrated and agitated, the PwD perceives that as being their fault, causing their mood to change
or making them want to leave. During a home visit, when interacting with the caregiver to
complete survey questions, she became frustrated because she did not or could not understand
the questions. The PwD then stated he was going to go to his room so that he would not be in the
way. In another interaction, the caregiver was loud and aggressive, causing the PwD to respond
poorly and try to exit the room.
The clinical significance associated with the interaction of PwD and robotic pets is
evident as the changes in their behaviors and moods could be visually observed. All individuals,
whether mild or moderately diagnosed with dementia, had a more pleasant attitude and were
smiling more when the pet was in their presence. The caregiver watched the PwD from a
distance while sitting on the opposite side of the living room, cleaning up the kitchen, and
starting dinner. They were always excited/happy when anticipating the pets’ presence/visit.
Caregivers’ unanimously scored all evaluation questions as a “5”, which signaled their strong
agreement that the robotic pet was a helpful, supplemental rehabilitative tool for their loved ones.
Discussion
Summary, Conclusions, Recommendations
Although the results did not produce statistical significance, it is worth mentioning that
the interaction between these robotic pets and PwD was visually pleasing and comforting to both
the PwD and caregivers. It produced a sense of responsibility and companionship for the PwD.

ROBOTIC PETS AND DEMENTIA

26

Robotic pet therapy served as a nonmedicinal intervention to enhance PwD’s QoL and reduce the
negative effects associated with dementia (Sicurella & Fitzsimmons, 2016). Robotic pets
enhance effective and social outcomes for PwD (Liang et al., 2017). Although the standard of
care has always included medication therapy, this projects suggests the benefits of robotic pet
therapy in the treatment of PwD. Introducing robotic pets to individuals diagnosed with mild to
moderate dementia can help decrease anxiety and agitation levels, which may contribute to an
improvement in cognitive health.
Maintaining the cognitive health of PwD over a longer period is a part of the Healthy
People 2020 objective (OA-5): reduce the proportion of older adults who have moderate to
severe functional limitations, and (OA-6): increase the proportion of older adults with a reduced
physical or cognitive function who engage in moderate or vigorous leisure-time physical
activities (Healthy People 2020, 2019). Hudson et al. (2020) showed that there is an increased
benefit in using robotic pets to relieve anxiety, agitation, and loneliness in older adults,
especially in care centers. Moyle et al. (2015) stated that the use of companion robots and
investigations of the benefits of social robots compared to and in association with human
interaction were beneficial. Robotic pets enhance the effectiveness of social outcomes for PwD
(Liang et al., 2017).
According to the reviewed studies, robotic pets had beneficial effects on the emotional
and social functioning of PwD. Robotic pets played an important role in decreasing anxiety and
agitation levels, decreasing the need for mood-adjusting medications, alleviating the burden of
caregiver burnout, and enhancing autonomy while encouraging the social interaction of PwD.
Evidence shows that robotic pets can help in improving moods and fostering communication
among elderly persons with mild to moderate dementia. It also has positive effects on managing

ROBOTIC PETS AND DEMENTIA

27

BPSD. Robotic pet interaction has positive effects on agitation, anxiety, loneliness, medication
consumption, and QoL for older adults.

Strengths and Limitation
The strengths and facilitators of this project included the opportunity that was given to
work with a city that was a part of a National Initiative for Dementia. It was an honor lead this
quality improvement project with the community. The city was very supportive throughout the
entire project and community members showed significant interest in the project as 30
participants responded to the initial rollout.
Due to the impact of COVID-19, there were many challenges and limitations during this
QI project. Cost was incurred by CO-PI due to risk of infection, having to purchase additional
robotic pets since they had to be limited to one per household; Smaller sample size and
representation of PwD due to heightened precautions/transmission and fear of contact with
people outside the household; Inability to carry out the project in a memory care setting and the
small sample size, which limited the generalizability of the findings. One of the major
challenges that set this project back was the change in the site, sample size, and population. The
project went from being conducted in 10 memory care home with 10 patients each to five
individuals and their caregivers in their homes.
While there were many changes throughout the project related to the COVID-19
pandemic and concerns about the spread of COVID-19 by handling the pets, leader turn-over at
the Dementia Friendly project site was also significant. During this time, new leadership was
being onboarded, and a great deal of repetitive work needed to be done such as getting new

ROBOTIC PETS AND DEMENTIA

28

stakeholders up to date on the project. Grants and supplies that were to be used for this project
changed with the new management. Using the robotic pets that were already acquired for the
project was not an option because it was seen as a transmission risk. As a result, personal funds
were used to purchase everything needed for the project.
Sustainability, Implication and Future Recommendation
This project revealed that robotic pets can be effectively used as supplemental therapy for
PwD. Based on the findings and conclusions, the following recommendations are suggested.
Because robotic pets have been shown to decrease anxiety and agitation, providers should
incorporate or introduce the pets as a supplemental therapy for dementia-suffering individuals.
These pets will act as a tool to foster a sense of responsibility in providing care and be a sensible
distractor for symptoms associated with dementia.
Ownership of a robotic pet should be encouraged by providers and embraced by
caregivers to provide a sense of peace and calmness for PwD and to create an opportunity for
caregivers to take a break or downtime from the constant shadowing/mood changes associated
with the progression of dementia. The use of robotic pets should be encouraged as therapy for
both elderly and dementia-suffering individuals in nursing homes/long-term care facilities.
Robotic pets should be investigated further within the context of PwD, using a larger sample size
to provide significant data to help strengthen the effectiveness and benefits of robotic pet
interactions. Studies suggest that robotic pets are beneficial when used as a supplemental therapy
and they are a readily available and affordable option for improving QoL and fostering effective
experience for PwD. Recommendations for further study would include other individuals that
may also experience other BPSD. For example, the elderly population who experience
loneliness may be a good candidate for this robotic pet as a companion, along with individuals

ROBOTIC PETS AND DEMENTIA

29

who suffer mental illness and have multiple BPSD. The opportunities for robotic therapy as an
effective therapy shows great promise across many diverse popluations.

ROBOTIC PETS AND DEMENTIA

30
References

Abbott, R., Orr, N., McGill, P., Whear, R., Bethel, A., Garside, R., Stein, K., & Thompson‐
Coon, J. (2019). How do “robopets” impact the health and well‐being of residents in care
homes? A systematic review of qualitative and quantitative evidence. International
Journal of Older People Nursing, 14(3). https://doi.org/10.1111/opn.12239
Alzheimer’s Association. (2020a). Alzheimer’s and dementia. Alzheimer’s Disease and
Dementia. https://www.alz.org/alzheimer_s_dementia
Alzheimer’s Association. (2020b). Facts and figures. Alzheimer’s Disease and Dementia.
https://www.alz.org/alzheimers-dementia/facts-figures
Alzheimer’s Association. (2021). Resources. Alzheimer’s Disease and Dementia.
https://www.alz.org/help-support/resources/
Alzheimer’s Disease International. (2020). ADI – World Alzheimer reports. Alzheimer’s Disease
International. https://www.alzint.org/what-we-do/research/world-alzheimer-report/
Balouch, S., Rifaat, E., Chen, H. L., & Tabet, N. (2019). Social networks and loneliness in
people with Alzheimer’s dementia. International Journal of Geriatric Psychiatry, 34(5),
666–673. https://doi.org/10.1002/gps.5065
Banks, M. R., Willoughby, L. M., & Banks, W. A. (2008). Animal-assisted therapy and
loneliness in nursing homes: Use of robotic versus living dogs. Journal of the American
Medical Directors Association, 9(3), 173–177.
https://doi.org/10.1016/j.jamda.2007.11.007
Bartholomew, K., Markham, C. M., Ruiter, R. A. C., Fernandez, M. E., Gerjo Kok, & Parcel, G.
S. (2016). Planning health promotion programs: An intervention mapping approach.
Wiley.

ROBOTIC PETS AND DEMENTIA

31

Birks, M., Bodak, M., Barlas, J., Harwood, J., & Pether, M. (2016, November 20). Robotic seals
as therapeutic tools in an aged care facility: A qualitative study. Journal of Aging
Research. https://www.hindawi.com/journals/jar/2016/8569602/
Bogner, J. (2000). The Agitated Behavior Scale. www.tbims.org.
https://www.tbims.org/combi/abs/
Burstein, A. A., DaDalt, O., Kramer, B., D’Ambrosio, L. A., & Coughlin, J. F. (2015). Dementia
caregivers and technology acceptance: Interest outstrips awareness. Gerontechnology,
14(1). https://doi.org/10.4017/gt.2015.14.1.005.00
Casey, D., Barrett, E., Kovacic, T., Sancarlo, D., Ricciardi, F., Murphy, K., Koumpis, A.,
Santorelli, A., Gallagher, N., & Whelan, S. (2020). The perceptions of people with
dementia and key stakeholders regarding the use and impact of the social robot MARIO.
International Journal of Environmental Research and Public Health, 17(22), 8621.
https://doi.org/10.3390/ijerph17228621
Centers for Disease Control and Prevention. (2019). U.S. burden of Alzheimer’s disease, related
dementias to double by 2060. CDC. https://www.cdc.gov/media/releases/2018/p0920alzheimers-burden-double-2060.html

Chen, S.-C., Jones, C., & Moyle, W. (2018). Social robots for depression in older adults: A
systematic review. Journal of Nursing Scholarship, 50(6).
https://doi.org/10.1111/jnu.12423
Fine, A., Beck, A., & Ng, Z. (2019). The state of animal-assisted interventions: Addressing the
contemporary issues that will shape the future. International Journal of Environmental
Research and Public Health, 16(20), 39–97. https://doi.org/10.3390/ijerph16203997

ROBOTIC PETS AND DEMENTIA

32

Gitlin, L. N., Kales, H. C., Marx, K., Stanislawski, B., & Lyketsos, C. (2017). A randomized trial
of a web-based platform to help families manage dementia-related behavioral symptoms:
The WeCareAdvisorTM. Contemporary Clinical Trials, 62, 27–36.
https://doi.org/10.1016/j.cct.2017.08.001
Góngora Alonso, S., Hamrioui, S., de la Torre Díez, I., Motta Cruz, E., López-Coronado, M., &
Franco, M. (2019). Social robots for people with aging and dementia: A systematic
review of literature. Telemedicine and E-Health, 25(7), 533–540.
https://doi.org/10.1089/tmj.2018.0051
Guerreiro, R., & Bras, J. (2015). The age factor in Alzheimer’s disease. Genome Medicine, 7(1).
https://doi.org/10.1186/s13073-015-0232-5

Gustafsson, C., Svanberg, C., & Müllersdorf, M. (2015). Using a robotic cat in dementia care: A
pilot study. Journal of Gerontological Nursing, 41(10), 46–56.
https://doi.org/10.3928/00989134-20150806-44
Harding, A. J. E., Morbey, H., Ahmed, F., Opdebeeck, C., Wang, Y.-Y., Williamson, P.,
Swarbrick, C., Leroi, I., Challis, D., Davies, L., Reeves, D., Holland, F., Hann, M.,
Hellström, I., Hydén, L.-C., Burns, A., Keady, J., & Reilly, S. (2018). Developing a core
outcome set for people living with dementia at home in their neighborhoods and
communities: Study protocol for use in the evaluation of non-pharmacological
community-based health and social care interventions. Trials, 19(1).
https://doi.org/10.1186/s13063-018-2584-9
Healthy People 2020. (2019, February 15). Healthy people 2020 | Healthy aging data |
Alzheimer’s disease and healthy aging | CDC.
https://www.cdc.gov/aging/agingdata/data-portal/healthypeople.html#HealthyPeopleDementias

ROBOTIC PETS AND DEMENTIA

33

JoyForAll.com. (2018). Joy for All Companion Pup : provides interactive engagement. The
Wright Stuff, Inc. | CareGiverProducts.com. https://www.caregiverproducts.com/hasbrojoy-for-all-companion-pup.html
Jutkowitz, E., Kane, R. L., Gaugler, J. E., MacLehose, R. F., Dowd, B., & Kuntz, K. M. (2017).
Societal and family lifetime cost of dementia: Implications for policy. Journal of the
American Geriatrics Society, 65(10), 2169–2175. https://doi.org/10.1111/jgs.15043
Liang, A., Piroth, I., Robinson, H., MacDonald, B., Fisher, M., Nater, U. M., Skoluda, N., &
Broadbent, E. (2017). A pilot randomized trial of a companion robot for people with
dementia living in the community. Journal of the American Medical Directors
Association, 18(10), 871–878. https://doi.org/10.1016/j.jamda.2017.05.019
Libin, A., & Cohen-Mansfield, J. (2004). Therapeutic robocat for nursing home residents with
dementia: Preliminary inquiry. American Journal of Alzheimer’s Disease & Other
Dementiasr, 19(2), 111–116. https://doi.org/10.1177/153331750401900209
Mitchell, G., McCormack, B., & McCance, T. (2016). Therapeutic use of dolls for people living
with dementia: A critical review of the literature. Dementia, 15(5), 976–1001.
https://doi.org/10.1177/1471301214548522
Nania, R. (2020, June 25). Will we ever cure dementia? AARP.
https://www.aarp.org/health/dementia/info-2020/cure-for-dementia.html

National Institute on Aging. (2019, May 22). Alzheimer’s disease fact sheet. National Institute on
Aging. https://www.nia.nih.gov/health/alzheimers-disease-fact-sheet
Petersen, S., Houston, S., Qin, H., Tague, C., & Studley, J. (2016). The utilization of robotic pets
in dementia care. Journal of Alzheimer’s Disease, 55(2), 569–574.
https://doi.org/10.3233/jad-160703

ROBOTIC PETS AND DEMENTIA

34

Pu, L., Moyle, W., & Jones, C. (2019). How people with dementia perceive a therapeutic robot
called PARO in relation to their pain and mood: A qualitative study. Journal of Clinical
Nursing, 29(3–4), 437–446. https://doi.org/10.1111/jocn.15104
Pu, L., Moyle, W., Jones, C., & Todorovic, M. (2018). The effectiveness of social robots for
older adults: A systematic review and meta-analysis of randomized controlled studies.
The Gerontologist, 59(1), e37–e51. https://doi.org/10.1093/geront/gny046
Rajan, K. B., Weuve, J., Barnes, L. L., McAninch, E. A., Wilson, R. S., & Evans, D. A. (2021).
Population estimate of people with clinical AD and mild cognitive impairment in the
United States (2020–2060). Alzheimer’s Dementia. In press.
Scerri, A., Sammut, R., & Scerri, C. (2020). Formal caregivers’ perceptions and experiences of
using pet robots for persons living with dementia in long‐term care: A meta‐ethnography.
Journal of Advanced Nursing, 77(1), 83–97. https://doi.org/10.1111/jan.14581
Scoglio, A. A., Reilly, E. D., Gorman, J. A., & Drebing, C. E. (2019). Use of social robots in
mental health and well-being research: Systematic review. Journal of Medical Internet
Research, 21(7), e13322. https://doi.org/10.2196/13322
Seaman, J. B., Terhorst, L., Gentry, A., Hunsaker, A., Parker, L. S., & Lingler, J. H. (2015).
Psychometric properties of a decisional capacity screening tool for individuals
contemplating participation in Alzheimer’s disease research. Journal of Alzheimer’s
Disease: JAD, 46(1), 1–9. https://doi.org/10.3233/JAD-142559
Snow, A. L., Huddleston, C., Robinson, C., Kunik, M. E., Bush, A. L., Wilson, N., Calleo, J.,
Paukert, A., Kraus-Schuman, C., Petersen, N. J., & Stanley, M. A. (2012). Psychometric
properties of a structured interview guide for the rating for anxiety in dementia. Aging &
Mental Health, 16(5), 592–602. https://doi.org/10.1080/13607863.2011.644518

ROBOTIC PETS AND DEMENTIA

35

Stall, N. M., Savage, R. D., & Rochon, P. A. (2019). Loneliness in older adults. Canadian
Medical Association Journal, 191(17), E476–E476. https://doi.org/10.1503/cmaj.181655
U.S. Census Bureau. (2020). Census.gov. http://www.census.gov
Victor, C. R., Rippon, I., Nelis, S. M., Martyr, A., Litherland, R., Pickett, J., Hart, N., Henley, J.,
Matthews, F., & Clare, L. (2020). Prevalence and determinants of loneliness in people
living with dementia: Findings from the IDEAL programme. International Journal of
Geriatric Psychiatry, 35(8), 851–858. https://doi.org/10.1002/gps.5305
Yu, R., Hui, E., Lee, J., Poon, D., Ng, A., Sit, K., Ip, K., Yeung, F., Wong, M., Shibata, T., &
Woo, J. (2015). Use of a therapeutic, socially assistive pet robot (PARO) in improving
mood and stimulating social interaction and communication for people with dementia:
Study protocol for a randomized controlled trial. Journal of Medical Internet Research
Protocols, 4(2), e45. https://doi.org/10.2196/resprot.4189
Zhou, Z., Mao, F., Zhang, W., Towne, S. D., Wang, P., & Fang, Y. (2019). The association
between loneliness and cognitive impairment among older men and women in China: A
nationwide longitudinal study. International Journal of Environmental Research and
Public Health, 16(16), 2877. https://doi.org/10.3390/ijerph16162877

ROBOTIC PETS AND DEMENTIA

36
APPENDIX A: Evaluation Tables

Citation

Theoretical/
Conceptual
Framework

Design/Method/
Purpose

Sample/
Setting

Variables

Measurement/
Instrumentation

Data
Analysis

Results/
Findings

(Chen et al.,
2018). Social
robots for
depression in
older adults: A
systematic
review

Inferred:
Cognitive and
behavioral
theoretical
framework

Design: SR of SRs
and Mas

N=7
N1 = 6 RCT
N2 = 1 comparison
study

IV1: Robot

Screening,
data extraction, form,
discussion,
Endnote software,
reviews

JBI-MA
StARI
PRISMA

IV1: Moderatequality evidence.
Mixed results.
6/7 studies found
significant
improvement in
dep. Active
participation and
group delivery
associated with
better results

Country:
Australia,
Denmark,
New Zealand,
Norway,
United States
Funding:
None
Bias:
Publication bias.
Implicit bias
regarding
interviews

Raw data for analysis
Blinding of treatment
allocation
Using intention to
treat
Purpose: To review
evidence from
existing SRs and Mas
of the effects of
social robot
interventions for dep.
in OA

DS: CINAHL = 21,
MEDLINE = 38,
PsychInfo = 41,
PubMed = 38, Web
of Science = 316,
Scopus = 22,
Cochrane = 43,
EMBASE = 75,
ProQuest = 51
LTC facilities: 6
studies
Home: 1 study
Inclusion criteria:
SR & Mas; included
statement of review;
adult ≥ 55; Studies
w/ young onset
dementia; social
robot intervention;
outcomes measures

DV1: Dep
DV2: OA

Level of Evidence;
Application to
Practice/
Generalization
LOE: I

Strengths:
Comprehensive search
and selection strategy
used in selecting
relevant articles.
Thorough observation of
interaction w/robot
improves health
outcomes by reducing
depressive symptoms.
Inclusion of solely SRs
increases breadth of
evidence examined
Weaknesses: No MA
d/t diversity of
intervention and low
sample size
Conclusions: Overall
evidence suggesting that
social robot
interventions have the
potential to reduce
depressive symptoms.
Evidence was not strong

Key: ABMI—Agitated Behaviors Mapping Instrument; ABS—Agitated Behavior Scale; AIBO—robotic dog; AIBOG—robotic dog group; ANCOVA—analysis of covariance; ANOVA—analysis of
variance; b/w—between; CG—control group; CI—confidence interval; CMA—Cohen-Mansfield Agitation Inventory; CMAI—Cohen–Mansfield Agitation Inventory instrument; CSDD—Cornel
Scale for Depression in Dementia; CSS—cross-sectional study; d/o—diagnosis of; Dep.—depression; DG—dog group; DM—Delphi methodology; d/t—due to; DQ—descriptive qualitative; DS—
databases searched; DSM-IV—Diagnostic and Statistical Manual of Mental Disorder; DSS—Delphi survey scale; DV—dependent variable; dz—disease; EG—experimental group; FSS—Face Scale
Score; GDS—Global Deterioration Scale; h/o—history of; hr—hour; ICC—intraclass correlation coefficient; IG—intervention group; ITT—intention to treat; IV—independent variable; JBIMAStARI—Joanna Briggs Institute Meta-Analysis of Statistics Assessment and Review Instrument; LAPS—Lexington Attachment to Pet Scale; LMBS—Lawton’s Modified Behavior Stream—LOE
—level of evidence; LS—longitudinal study; LTCF—long-term care facility; MA—meta-analyses; MANOVA—multivariate analysis of variance; MD—mean difference; min—minutes; MLAPS—
Modified Lexington Attachment to Pet Scale; MMSE—Mini-Mental State Exam; mn—months; N—number of studies; n—number of participants; NGO—nongovernmental organization; NRCT—
nonrandomized control trials; NRNCT—nonrandomized noncontrolled trial; OA—older adult; OMI—outcome measurement; OT—observation table; PI—Paro intervention; PO—potential outcomes;
PRISMA—Preferred Reporting Items for Systematic Reviews and Meta-Analyses; pt—patient; PwD—person with dementia; QD—qualitative data; QDC—qualitative data collection; QIS—
qualitative interview study; QoL—quality of life; QUALID—Quality of Life in Late-Stage Dementia scale; RAID—Rating Anxiety in Dementia; RCT—randomized control trial; RT—recreational
therapist; SCS—single case study; SD—standard deviation; SG—support groups; SPS—stated preference survey; SR—systematic review; UCLA—University of California Los Angeles; w/—with;
w/o—without; wk—weeks; wkly—weekly; y.o. —years old; y—years; ZBI—Zarit Burden Inventory; &—and; α—Cronbach’s alpha value; +—plus; ≥—greater than or equal to

ROBOTIC PETS AND DEMENTIA
Citation

Theoretical/
Conceptual
Framework

Design/Method/
Purpose

37
Sample/
Setting

Variables

Measurement/
Instrumentation

Data
Analysis

Results/
Findings

of dep.;
quantitative;
includes pre- and
posttest; English
peer-reviewed
journals

enough to formulate
recommendations on
clinical effectiveness on
dep. in OA. Rigorous
and powered studies
needed for Mas.

Exclusion criteria:
Grey literature;
non-English articles
Attrition: Not
discussed
Harding et al.
(2018).
Developing a
core outcome set
for people living
with dementia at
home in their
neighborhoods
and
communities:
Study protocol
for use in the
evaluation of
nonpharmacologi
cal communitybased health and
social care
interventions

Inferred:
Caregiver
dynamics,
middle range
theory

Four-phase mixedmethods study
Phase 1:
Identification of PO
through QDC and LR
Phase 2: DM
Phase 3: SR of OMI
Phase 4: SPS
Interviews and FG,
recruitment,
bringing together QD
and data extracted
from LR

Random 50%
sample;
N = 124
Key reviews and
qualitative studies;
N=8
Policy documents;
N = 38
DS:
MEDLINE,
CINAHL, ALOIS,
PROSPERO, and
Cochrane

Level of Evidence;
Application to
Practice/
Generalization

To establish an
agreed
standardized COS
for use when
evaluating
nonpharmacologic
al health and
social care
interventions for
PwD
Identify and
achieve a
consensus around
which outcomes
should be
measured from the
perspective of key
stakeholders

Surveys,
audiotaped
histograms,
workshops,
discussions

Logistic
analyses,
probity
regression
analyses

Identification of
potential
outcomes
through
qualitative data
collection and
literature review.
LR of existing
outcomes
reported in
studies and
review.
Bringing
together the
qualitative data
and the data
extracted from
literature.

Feasibility:
Recommended for use
in practice because of
effectiveness of
nonpharmacological
benefits for Dep. in
PwD
LOE: I
Strength:
Study design and the
Delphi method are
increasingly being
recognized as a robust
approach to forming
COSs.
Study focused on PwD.
MMS to achieve best
possible results
Weaknesses:
Data from UK only.
Sample restricted to
North-West England.

Key: ABMI—Agitated Behaviors Mapping Instrument; ABS—Agitated Behavior Scale; AIBO—robotic dog; AIBOG—robotic dog group; ANCOVA—analysis of covariance; ANOVA—analysis of
variance; b/w—between; CG—control group; CI—confidence interval; CMA—Cohen-Mansfield Agitation Inventory; CMAI—Cohen–Mansfield Agitation Inventory instrument; CSDD—Cornel
Scale for Depression in Dementia; CSS—cross-sectional study; d/o—diagnosis of; Dep.—depression; DG—dog group; DM—Delphi methodology; d/t—due to; DQ—descriptive qualitative; DS—
databases searched; DSM-IV—Diagnostic and Statistical Manual of Mental Disorder; DSS—Delphi survey scale; DV—dependent variable; dz—disease; EG—experimental group; FSS—Face Scale
Score; GDS—Global Deterioration Scale; h/o—history of; hr—hour; ICC—intraclass correlation coefficient; IG—intervention group; ITT—intention to treat; IV—independent variable; JBIMAStARI—Joanna Briggs Institute Meta-Analysis of Statistics Assessment and Review Instrument; LAPS—Lexington Attachment to Pet Scale; LMBS—Lawton’s Modified Behavior Stream—LOE
—level of evidence; LS—longitudinal study; LTCF—long-term care facility; MA—meta-analyses; MANOVA—multivariate analysis of variance; MD—mean difference; min—minutes; MLAPS—
Modified Lexington Attachment to Pet Scale; MMSE—Mini-Mental State Exam; mn—months; N—number of studies; n—number of participants; NGO—nongovernmental organization; NRCT—
nonrandomized control trials; NRNCT—nonrandomized noncontrolled trial; OA—older adult; OMI—outcome measurement; OT—observation table; PI—Paro intervention; PO—potential outcomes;
PRISMA—Preferred Reporting Items for Systematic Reviews and Meta-Analyses; pt—patient; PwD—person with dementia; QD—qualitative data; QDC—qualitative data collection; QIS—
qualitative interview study; QoL—quality of life; QUALID—Quality of Life in Late-Stage Dementia scale; RAID—Rating Anxiety in Dementia; RCT—randomized control trial; RT—recreational
therapist; SCS—single case study; SD—standard deviation; SG—support groups; SPS—stated preference survey; SR—systematic review; UCLA—University of California Los Angeles; w/—with;
w/o—without; wk—weeks; wkly—weekly; y.o. —years old; y—years; ZBI—Zarit Burden Inventory; &—and; α—Cronbach’s alpha value; +—plus; ≥—greater than or equal to

ROBOTIC PETS AND DEMENTIA
Citation

Theoretical/
Conceptual
Framework

Country:
United Kingdom
Funding:
ESRC and NIHR
Bias:
None stated by
authors

38

Design/Method/
Purpose

Sample/
Setting

Variables

Purpose:
To establish an
agreed COS for use
when evaluating
nonpharmacological
health and social care
interventions for
PwD

Demographics:
20–30 PwD
and care partners

(people living at
home with
dementia, care
partners, health
and social care
professionals,
researchers, and
policymakers).
Review and
recommend how
outcomes should
be measured

Setting:
Homes across
north-west of
England
Inclusion criteria:
d/o dementia,
living in home in
North-West
England,
capacity to
understand and
consent to
participate in study

Diversity/sample
bias

Measurement/
Instrumentation

Data
Analysis

Results/
Findings
Delphi
methodology

Level of Evidence;
Application to
Practice/
Generalization
Use of online survey
tool will limit
participation.
Study limited to mild to
moderate PwD
Feasibility:
Recommended for use
in practice owing to
effectiveness of
nonpharmacological
benefits in caring for
and treating PwD

Exclusion Criteria:
Inability to consent,
PwD
institutionalized,
non-English
speaking,
too unwell to
participate
Attrition: Not
discussed
(Pu et al., 2018).
The effectiveness

Inferred:

Design: MMSR of
RCT’s

DS:2202 (A)
N: 13 (A)

IV: Robotic Pet

QUALID
CMAI

Interrater
reliability

(α- = 0.86)

LOE: I

Key: ABMI—Agitated Behaviors Mapping Instrument; ABS—Agitated Behavior Scale; AIBO—robotic dog; AIBOG—robotic dog group; ANCOVA—analysis of covariance; ANOVA—analysis of
variance; b/w—between; CG—control group; CI—confidence interval; CMA—Cohen-Mansfield Agitation Inventory; CMAI—Cohen–Mansfield Agitation Inventory instrument; CSDD—Cornel
Scale for Depression in Dementia; CSS—cross-sectional study; d/o—diagnosis of; Dep.—depression; DG—dog group; DM—Delphi methodology; d/t—due to; DQ—descriptive qualitative; DS—
databases searched; DSM-IV—Diagnostic and Statistical Manual of Mental Disorder; DSS—Delphi survey scale; DV—dependent variable; dz—disease; EG—experimental group; FSS—Face Scale
Score; GDS—Global Deterioration Scale; h/o—history of; hr—hour; ICC—intraclass correlation coefficient; IG—intervention group; ITT—intention to treat; IV—independent variable; JBIMAStARI—Joanna Briggs Institute Meta-Analysis of Statistics Assessment and Review Instrument; LAPS—Lexington Attachment to Pet Scale; LMBS—Lawton’s Modified Behavior Stream—LOE
—level of evidence; LS—longitudinal study; LTCF—long-term care facility; MA—meta-analyses; MANOVA—multivariate analysis of variance; MD—mean difference; min—minutes; MLAPS—
Modified Lexington Attachment to Pet Scale; MMSE—Mini-Mental State Exam; mn—months; N—number of studies; n—number of participants; NGO—nongovernmental organization; NRCT—
nonrandomized control trials; NRNCT—nonrandomized noncontrolled trial; OA—older adult; OMI—outcome measurement; OT—observation table; PI—Paro intervention; PO—potential outcomes;
PRISMA—Preferred Reporting Items for Systematic Reviews and Meta-Analyses; pt—patient; PwD—person with dementia; QD—qualitative data; QDC—qualitative data collection; QIS—
qualitative interview study; QoL—quality of life; QUALID—Quality of Life in Late-Stage Dementia scale; RAID—Rating Anxiety in Dementia; RCT—randomized control trial; RT—recreational
therapist; SCS—single case study; SD—standard deviation; SG—support groups; SPS—stated preference survey; SR—systematic review; UCLA—University of California Los Angeles; w/—with;
w/o—without; wk—weeks; wkly—weekly; y.o. —years old; y—years; ZBI—Zarit Burden Inventory; &—and; α—Cronbach’s alpha value; +—plus; ≥—greater than or equal to

ROBOTIC PETS AND DEMENTIA
Citation

of social robots
for older adults:
A systematic
review and metaanalysis of
randomized
controlled
studies.
Country:
Denmark
Norway
New Zealand
United States
Australia
Japan
Spain
Funding:
PhD Scholarship
from Griffith
University & the
Chinese Council
Bias:
None reported by
author.
High in
allocation
concealment,
random sequence
and blinding;
Selection bias

Theoretical/
Conceptual
Framework
Health belief
model

Design/Method/
Purpose

4 trials:
25–45 min sessions
12 wks
FU- 12–6–5 wks
after intervention
Purpose:
Summarize the
effectiveness of
social robots on
outcomes
(psychological,
physiological, QoL,
or medication) of OA
from RCT’s.

39
Sample/
Setting

Variables

Measurement/
Instrumentation

Data
Analysis

Results/
Findings

n: 11 (RCT)
n: 9 (MA)
N=1042
DS: (n=2202)
Scopus (n=748)
ProQuest (n=84)
PubMed (n=724)
EBSCO (n=152)
PsycINFO (n=41)
Science Direct
(n=143)
Web of Science
(n=289)
Cochrane (n=21)

DV1: Effects on
PwD
DV2: Effects on
professional
caregiver
DV3: Effects on
relative

Introduction and
training
coding of patterns
using qualitative
description.
Interviews

High test–
retest
reliability,
internal
consistency
Cronbach’s
alpha

Range = 11 to 59
lower QoL.
Interrater
reliability (0.69),
high test–retest
reliability (0.86)
(α = 0.74 and
0.77)
CMAI- 50–83.
Baseline (mean
= 55.9).
7-wks
intervention,
(mean = 62.0).
Follow-up period
(mean = 69.7).
Throughout all
phases:
range = 17–24.
Mean values of
19.1 (A—
baseline), 19.3
(B—
intervention),
and 19.0 (A—
follow-up).
Visual
inspection.
CMAI: 28–61.
Overall CMAI
scores low, mean
range 34.5–40.7

Setting:
Hospital
LTC Facilities
Daycare Centers
1 A -Home setting
Inclusion criteria:
OA; 55y +
RCT’s using social
robots w/o
restriction of type
English articles
Exclusion criteria:
Subject were
children.
Younger adults
Reviews

Definition:
Interaction—
Impact/Implicatio
ns of robotic cat
Communication
—Using robotic
cat
Usable—Qualities
of robotic cat

Visual inspection
increased.
Good internal
consistency

Level of Evidence;
Application to
Practice/
Generalization
Strengths:
Thorough discussion of
numerous components
in study to improve
health outcomes.
Inclusion of SRs
increases breadth of
evidence examined.

Weakness:
Sampling limited to
participants at dementia
care home. Limited
discussion of sample
demographics
Conclusions: Results
indicate improvement in
QoL. Scores were
higher after intervention.
Further research with a
larger sample size is
necessary to determine
broader impact
Feasibility/
Applicability to pt.
population: Feasible
intervention. Could be
applicable but may
witness more
advancement at time of
intervention, so results

Key: ABMI—Agitated Behaviors Mapping Instrument; ABS—Agitated Behavior Scale; AIBO—robotic dog; AIBOG—robotic dog group; ANCOVA—analysis of covariance; ANOVA—analysis of
variance; b/w—between; CG—control group; CI—confidence interval; CMA—Cohen-Mansfield Agitation Inventory; CMAI—Cohen–Mansfield Agitation Inventory instrument; CSDD—Cornel
Scale for Depression in Dementia; CSS—cross-sectional study; d/o—diagnosis of; Dep.—depression; DG—dog group; DM—Delphi methodology; d/t—due to; DQ—descriptive qualitative; DS—
databases searched; DSM-IV—Diagnostic and Statistical Manual of Mental Disorder; DSS—Delphi survey scale; DV—dependent variable; dz—disease; EG—experimental group; FSS—Face Scale
Score; GDS—Global Deterioration Scale; h/o—history of; hr—hour; ICC—intraclass correlation coefficient; IG—intervention group; ITT—intention to treat; IV—independent variable; JBIMAStARI—Joanna Briggs Institute Meta-Analysis of Statistics Assessment and Review Instrument; LAPS—Lexington Attachment to Pet Scale; LMBS—Lawton’s Modified Behavior Stream—LOE
—level of evidence; LS—longitudinal study; LTCF—long-term care facility; MA—meta-analyses; MANOVA—multivariate analysis of variance; MD—mean difference; min—minutes; MLAPS—
Modified Lexington Attachment to Pet Scale; MMSE—Mini-Mental State Exam; mn—months; N—number of studies; n—number of participants; NGO—nongovernmental organization; NRCT—
nonrandomized control trials; NRNCT—nonrandomized noncontrolled trial; OA—older adult; OMI—outcome measurement; OT—observation table; PI—Paro intervention; PO—potential outcomes;
PRISMA—Preferred Reporting Items for Systematic Reviews and Meta-Analyses; pt—patient; PwD—person with dementia; QD—qualitative data; QDC—qualitative data collection; QIS—
qualitative interview study; QoL—quality of life; QUALID—Quality of Life in Late-Stage Dementia scale; RAID—Rating Anxiety in Dementia; RCT—randomized control trial; RT—recreational
therapist; SCS—single case study; SD—standard deviation; SG—support groups; SPS—stated preference survey; SR—systematic review; UCLA—University of California Los Angeles; w/—with;
w/o—without; wk—weeks; wkly—weekly; y.o. —years old; y—years; ZBI—Zarit Burden Inventory; &—and; α—Cronbach’s alpha value; +—plus; ≥—greater than or equal to

ROBOTIC PETS AND DEMENTIA
Citation

Theoretical/
Conceptual
Framework

Design/Method/
Purpose

Performance bias
Detection bias
Attrition bias

40
Sample/
Setting

Variables

Measurement/
Instrumentation

Data
Analysis

Nonrandomized
studies
Study protocols
Case
Studies
Observational
studies
Pre/post studies w/o
CG
Conference
abstracts w/o full
text

(A—baseline:
mean = 34.9)
B—intervention:
mean = 34.5;
follow-up: mean
= 40.7).
QUALID = 11–
22
all phases, mean
values of 15.3
(A—baseline),
13.8 (B—
intervention),
14.0 (A—
follow-up

Attrition: Not
discussed
(Banks et al.,
2008). AnimalAssisted Therapy
and Loneliness in
Nursing Homes:
Use of Robotic
versus Living
Dogs.
Country:
United States
Funding:
None stated

Inferred to the
SocialEmotional
Selectivity
Theory

Quantitative Study
Design: RCT (pre- &
posttest, 8 - wks
follow up test)
Wkly 30 mins visits
Lonely elderly
LTCF pt: 3 groups
w/ wkly 30-min visit
w/ AIBO/dog over
8wks. Reviewed
initial pretest &
posttest 7wks to
compare effects of
loneliness.

N: 38
n1: 13 (initial group
w/ dog)
n2: 12 (initial group
w/ AIBO)
n3: 13 (CG)
Setting: All in
LTCF. In pt’s room.
Sample
Demographics: No
significant
differences between
DG, AIBOG, or
CG.

Results/
Findings

IV: Lonely elderly
LTCF pt.

UCLA loneliness
scale

DV1: CG
DV2: DG
DV3: AIBOG

Delta Loneliness
Score

ANOVA
Student 2
tailed t-test

MLAPS & LAPS

Pearson
Correlation

Newman-Keuls range
test

Subscale
analysis

Prism 4.0 statistical
package

1 tailed ttest

Survey/
Questionnaire

There were no
statistical
differences
among pretest
UCLA loneliness
scale for 3
groups.
CG, DG,
AIBOG; mean
loneliness score=
45.9±1.16
(n=38)
(F(2,35)=37.3,
p<.01)
CG; (N=13)
AIBO (p<.05,
n=12)

Level of Evidence;
Application to
Practice/
Generalization
may not be
generalizable.
Readers recognize
descriptions or
interpretations as
comparable to their own
experiences.
Transferred to
comparable situations
and contexts

LOE: II
Strengths: RCT design,
relatively high attrition
rate for short duration
study (over 8 wks),
adequately powered.
Weaknesses: Initial
sampling limited to
participants in LTCF.
No intervention for
those who did not get
any visits. Details not
provided regarding the
health status of pt.’s.

Key: ABMI—Agitated Behaviors Mapping Instrument; ABS—Agitated Behavior Scale; AIBO—robotic dog; AIBOG—robotic dog group; ANCOVA—analysis of covariance; ANOVA—analysis of
variance; b/w—between; CG—control group; CI—confidence interval; CMA—Cohen-Mansfield Agitation Inventory; CMAI—Cohen–Mansfield Agitation Inventory instrument; CSDD—Cornel
Scale for Depression in Dementia; CSS—cross-sectional study; d/o—diagnosis of; Dep.—depression; DG—dog group; DM—Delphi methodology; d/t—due to; DQ—descriptive qualitative; DS—
databases searched; DSM-IV—Diagnostic and Statistical Manual of Mental Disorder; DSS—Delphi survey scale; DV—dependent variable; dz—disease; EG—experimental group; FSS—Face Scale
Score; GDS—Global Deterioration Scale; h/o—history of; hr—hour; ICC—intraclass correlation coefficient; IG—intervention group; ITT—intention to treat; IV—independent variable; JBIMAStARI—Joanna Briggs Institute Meta-Analysis of Statistics Assessment and Review Instrument; LAPS—Lexington Attachment to Pet Scale; LMBS—Lawton’s Modified Behavior Stream—LOE
—level of evidence; LS—longitudinal study; LTCF—long-term care facility; MA—meta-analyses; MANOVA—multivariate analysis of variance; MD—mean difference; min—minutes; MLAPS—
Modified Lexington Attachment to Pet Scale; MMSE—Mini-Mental State Exam; mn—months; N—number of studies; n—number of participants; NGO—nongovernmental organization; NRCT—
nonrandomized control trials; NRNCT—nonrandomized noncontrolled trial; OA—older adult; OMI—outcome measurement; OT—observation table; PI—Paro intervention; PO—potential outcomes;
PRISMA—Preferred Reporting Items for Systematic Reviews and Meta-Analyses; pt—patient; PwD—person with dementia; QD—qualitative data; QDC—qualitative data collection; QIS—
qualitative interview study; QoL—quality of life; QUALID—Quality of Life in Late-Stage Dementia scale; RAID—Rating Anxiety in Dementia; RCT—randomized control trial; RT—recreational
therapist; SCS—single case study; SD—standard deviation; SG—support groups; SPS—stated preference survey; SR—systematic review; UCLA—University of California Los Angeles; w/—with;
w/o—without; wk—weeks; wkly—weekly; y.o. —years old; y—years; ZBI—Zarit Burden Inventory; &—and; α—Cronbach’s alpha value; +—plus; ≥—greater than or equal to

ROBOTIC PETS AND DEMENTIA
Citation

Bias:
None stated by
author.
Sampling and
Informational
Bias was noted.

Theoretical/
Conceptual
Framework

41

Design/Method/
Purpose

Sample/
Setting

Purpose:
Compare the ability
of a living dog (Dog)
and a robotic dog
(AIBO) to treat
loneliness in elderly
patients living in 3
LTCF in St. Louis,
MO.

Exclusion Criteria:
Score <24 on
MMSE
Allergic to dogs or
cats
Score < 30 on
UCLA Loneliness
scale
Known h/o
Psychiatric dz or
Alzheimer’s dz.
Inclusion Criteria:
LTCF pt.
Attrition: Not
discussed

Variables

Measurement/
Instrumentation

Data
Analysis

Results/
Findings
DG; (p<.05,
n=13)
m=0.437
i= (-18.9)
r= 0.459
n=38
p<.005
MLAPS &
LAPS
Pearson r=
0.9937
r2=0.9874
slope 1.01;
intercept (-3.04)
AIBO
(47.2±4.03,
n=12)
DG (55.0±3.73,
n=13)

Level of Evidence;
Application to
Practice/
Generalization
Limited discuss of
sample demographics.

Conclusions: Results
demonstrate a
significant decrease in
the feeling of loneliness
after visits compared to
those who didn’t get
visits. Improvement in
feelings of loneliness
noted when using either
dog or AIBO with no
significant difference in
type of pet use in visit.
Feasibility/
Applicability to pt.
population: A visit
session similar to that
described in the study
could be feasible with
caregivers. The
applicability is
relevantly easy since
caregivers provide care
and can help in
reporting effects of
intervention preintervention &
compared to post
intervention.

Key: ABMI—Agitated Behaviors Mapping Instrument; ABS—Agitated Behavior Scale; AIBO—robotic dog; AIBOG—robotic dog group; ANCOVA—analysis of covariance; ANOVA—analysis of
variance; b/w—between; CG—control group; CI—confidence interval; CMA—Cohen-Mansfield Agitation Inventory; CMAI—Cohen–Mansfield Agitation Inventory instrument; CSDD—Cornel
Scale for Depression in Dementia; CSS—cross-sectional study; d/o—diagnosis of; Dep.—depression; DG—dog group; DM—Delphi methodology; d/t—due to; DQ—descriptive qualitative; DS—
databases searched; DSM-IV—Diagnostic and Statistical Manual of Mental Disorder; DSS—Delphi survey scale; DV—dependent variable; dz—disease; EG—experimental group; FSS—Face Scale
Score; GDS—Global Deterioration Scale; h/o—history of; hr—hour; ICC—intraclass correlation coefficient; IG—intervention group; ITT—intention to treat; IV—independent variable; JBIMAStARI—Joanna Briggs Institute Meta-Analysis of Statistics Assessment and Review Instrument; LAPS—Lexington Attachment to Pet Scale; LMBS—Lawton’s Modified Behavior Stream—LOE
—level of evidence; LS—longitudinal study; LTCF—long-term care facility; MA—meta-analyses; MANOVA—multivariate analysis of variance; MD—mean difference; min—minutes; MLAPS—
Modified Lexington Attachment to Pet Scale; MMSE—Mini-Mental State Exam; mn—months; N—number of studies; n—number of participants; NGO—nongovernmental organization; NRCT—
nonrandomized control trials; NRNCT—nonrandomized noncontrolled trial; OA—older adult; OMI—outcome measurement; OT—observation table; PI—Paro intervention; PO—potential outcomes;
PRISMA—Preferred Reporting Items for Systematic Reviews and Meta-Analyses; pt—patient; PwD—person with dementia; QD—qualitative data; QDC—qualitative data collection; QIS—
qualitative interview study; QoL—quality of life; QUALID—Quality of Life in Late-Stage Dementia scale; RAID—Rating Anxiety in Dementia; RCT—randomized control trial; RT—recreational
therapist; SCS—single case study; SD—standard deviation; SG—support groups; SPS—stated preference survey; SR—systematic review; UCLA—University of California Los Angeles; w/—with;
w/o—without; wk—weeks; wkly—weekly; y.o. —years old; y—years; ZBI—Zarit Burden Inventory; &—and; α—Cronbach’s alpha value; +—plus; ≥—greater than or equal to

ROBOTIC PETS AND DEMENTIA

42

Citation

Theoretical/
Conceptual
Framework

Design/Method/
Purpose

Sample/
Setting

Variables

Measurement/
Instrumentation

Data
Analysis

Results/
Findings

(Liang et al.,
2017) A Pilot
Randomized
Trial of a
Companion
Robot for People
with Dementia
Living in the
Community.

Psychodynamic
Theory
Rogers'
phenomenologi
cal theory

RCT – Pilot Block

Weiner’s
ThreeDimensional
Model

IV – Companion
Robot (Paro)
DV1 – Effects in
home setting
DV2 – Effects in
day care center
Definition:
Companion robot
– an animal with
whom one spends
a lot of time with
Home setting –
residential area
where action is
taken place.
Day Care center –
Facility where
action is taken
place

TSM
Responses recorded
Physiological – B/P,
Salivary Cortisol,
hair cortisol
CMAI-SF
NPI-Q
ACE- New Zealand
Version
CSDD

FET
MD
MMA
ANOVA
BA
IStT
MWUT
SD

Country: New
Zealand

N = 74
Excluded N = 44
Not meeting
inclusion criteria:
N=9
Non-English
speaking: N = 2
Moved away: N = 5
No dementia
diagnosis: N = 1
Care recipient
passed away: N = 1

Paro had
beneficial effects
on emotional &
social
functioning in
people with
dementia.
Based on:
Effects in day
care center n =
13
Observation:
Facial expression
SD = 17.5% PG
33.8% CG
r=0.49
z=-2.22
Social
interaction
SD = 29.2% PG
25.1% CG
r=0.11
z=-0.56
Agitated
behavior
SD = 11.4% PG
21.6% CG
r=0.29
z=-1.14

Funding:
University of
Auckland
Faculty Research
Development
Fund to E.B.
(Grant 360813)
Sasakawa Young
Leaders
Fellowship
Funds.
Bias:
The authors
declare no
conflict of
interest.

Kelley’s
Covariation
Model

Purpose – To
investigate the
effective, social,
behavioral, and
psychological effects
of the companion
robot Paro for people
with dementia in both
day care center and
home setting.

Declined
Participation: N =
35
Time constraints N
= 23
Caregiver health
concerns N = 2
Care recipient
health concern N =
4
Setting:
2 Dementia day
care centers
Residential homes
of participants
Formal diagnosis of
Dementia

Dementia –
complex, brain
deteriorating
disorder that
results in
significant
cognitive and
functional decline.
ace

Effects in home
setting n= 14
Cognitive score

Level of Evidence;
Application to
Practice/
Generalization
LOE: II

Strengths:
Examined individual
characteristic to
determine who
benefitted most.
Examined range of self
report, observational, &
physiological variable
providing
comprehensive
assessment.
First study providing
home context for robot.
Weakness:
Sample size was small.
Maintaining participants
was difficult.
Challenging group to do
research many drop out
d/o deterioration in
health /inability to
complete assessment.
Conclusion:
EG showed
significantly more
positive facial
expressions, talked more
to staff & researchers

Key: ABMI—Agitated Behaviors Mapping Instrument; ABS—Agitated Behavior Scale; AIBO—robotic dog; AIBOG—robotic dog group; ANCOVA—analysis of covariance; ANOVA—analysis of
variance; b/w—between; CG—control group; CI—confidence interval; CMA—Cohen-Mansfield Agitation Inventory; CMAI—Cohen–Mansfield Agitation Inventory instrument; CSDD—Cornel
Scale for Depression in Dementia; CSS—cross-sectional study; d/o—diagnosis of; Dep.—depression; DG—dog group; DM—Delphi methodology; d/t—due to; DQ—descriptive qualitative; DS—
databases searched; DSM-IV—Diagnostic and Statistical Manual of Mental Disorder; DSS—Delphi survey scale; DV—dependent variable; dz—disease; EG—experimental group; FSS—Face Scale
Score; GDS—Global Deterioration Scale; h/o—history of; hr—hour; ICC—intraclass correlation coefficient; IG—intervention group; ITT—intention to treat; IV—independent variable; JBIMAStARI—Joanna Briggs Institute Meta-Analysis of Statistics Assessment and Review Instrument; LAPS—Lexington Attachment to Pet Scale; LMBS—Lawton’s Modified Behavior Stream—LOE
—level of evidence; LS—longitudinal study; LTCF—long-term care facility; MA—meta-analyses; MANOVA—multivariate analysis of variance; MD—mean difference; min—minutes; MLAPS—
Modified Lexington Attachment to Pet Scale; MMSE—Mini-Mental State Exam; mn—months; N—number of studies; n—number of participants; NGO—nongovernmental organization; NRCT—
nonrandomized control trials; NRNCT—nonrandomized noncontrolled trial; OA—older adult; OMI—outcome measurement; OT—observation table; PI—Paro intervention; PO—potential outcomes;
PRISMA—Preferred Reporting Items for Systematic Reviews and Meta-Analyses; pt—patient; PwD—person with dementia; QD—qualitative data; QDC—qualitative data collection; QIS—
qualitative interview study; QoL—quality of life; QUALID—Quality of Life in Late-Stage Dementia scale; RAID—Rating Anxiety in Dementia; RCT—randomized control trial; RT—recreational
therapist; SCS—single case study; SD—standard deviation; SG—support groups; SPS—stated preference survey; SR—systematic review; UCLA—University of California Los Angeles; w/—with;
w/o—without; wk—weeks; wkly—weekly; y.o. —years old; y—years; ZBI—Zarit Burden Inventory; &—and; α—Cronbach’s alpha value; +—plus; ≥—greater than or equal to

ROBOTIC PETS AND DEMENTIA
Citation

Theoretical/
Conceptual
Framework

Design/Method/
Purpose

43
Sample/
Setting

Variables

Measurement/
Instrumentation

Data
Analysis

Attrition: Not
discussed

(Libin & CohenMansfield,
2004).
Therapeutic
robocat for
nursing home
residents with
dementia:
Preliminary
inquiry.
Country:
United States
Funding:
None Stated
Bias:
Gender
homogeneity
Small sample
size
Short term
sessions

Inferred Theory
of Planned
Behavior

RCT (pre- &
posttest)
Comparison
Condition
Experimental
Designs
Analysis of
Agitation/ Affect/
Age/ Level of
cognitive impairment
2-10 min sessions;
1-5min pre & 5min
post w/ cat
Examine what kind
of effects occur when
a cognitively
impaired person
interacts w/ a robotic
pet; does a robotic cat
trigger more positive
experience than a
plush toy; and to
what degree does a
nursing home

SD = 19.2% PG
20.9% CG
CMAI-SF Score
SD = 10.4% PG
7.16% CG
NPI-Q Score
SD = 6.83% PG
2.43% CG

N: 10; candidates
n= 9; consented
n=6 first session w/
robotic pet
n=3 first session w/
plush toy

IV- Time
(baseline vs
intervention)
DV1- Robotic cat
DV2- Plush toy
cat

Setting: Large
suburban not-forprofit nursing home.

Time: 2, 10-min
interactive session
w/ resident
1 session per day
conducted by
research assistance
w/ 1 year+
experience.

Sample
Demographics:
PwD. Age range 83
– 98y. Mean age of
90y. Female gender.
Inclusion criteria:
d/o Dementia
nursing home
resident
Exclusion criteria:

Results/
Findings

GDS
ABMI
LMBS
3 constructs on 5point scale
Interactive Sessions
Protocol
Observation

Pearson
correlations
Paired
sample ttest
Separate ttest

GDS= 5.4 (range
4-7)
r=.76, p<.008
/duration
r=.63, p<.018
/attitude
r=.67, p<.012
/attention
r=.66, p<.014
/intensity
(t(8) =2.0,
p<0.036 and t(8)=
2.4, p<0.046)
(t(8)= 2.0, p<
0.078)/overall
agitation
(t(8)= 3.6,
p<0.007)
/pleasure
(t(8)= 2.7,
p<0.028)
/interest
22% held robotic
cat

Level of Evidence;
Application to
Practice/
Generalization

compared to those in the
CG.
Feasibility
Recommended for use
in practice due to
effectiveness of
companion robot in
decreasing loneliness.
LOE: II
Strengths: RCT design,
relatively. Relatively
low attrition rate.
Weaknesses: Sample
limited to female
participants diagnosed
w/ dementia at a
suburban nursing home.
Small sample size.
Conclusions: Results
demonstrate a
significantly increase in
the among of pleasure
residents had from their
interest in pets.
Agitation behavior also
decreased.
Conclusion:

Key: ABMI—Agitated Behaviors Mapping Instrument; ABS—Agitated Behavior Scale; AIBO—robotic dog; AIBOG—robotic dog group; ANCOVA—analysis of covariance; ANOVA—analysis of
variance; b/w—between; CG—control group; CI—confidence interval; CMA—Cohen-Mansfield Agitation Inventory; CMAI—Cohen–Mansfield Agitation Inventory instrument; CSDD—Cornel
Scale for Depression in Dementia; CSS—cross-sectional study; d/o—diagnosis of; Dep.—depression; DG—dog group; DM—Delphi methodology; d/t—due to; DQ—descriptive qualitative; DS—
databases searched; DSM-IV—Diagnostic and Statistical Manual of Mental Disorder; DSS—Delphi survey scale; DV—dependent variable; dz—disease; EG—experimental group; FSS—Face Scale
Score; GDS—Global Deterioration Scale; h/o—history of; hr—hour; ICC—intraclass correlation coefficient; IG—intervention group; ITT—intention to treat; IV—independent variable; JBIMAStARI—Joanna Briggs Institute Meta-Analysis of Statistics Assessment and Review Instrument; LAPS—Lexington Attachment to Pet Scale; LMBS—Lawton’s Modified Behavior Stream—LOE
—level of evidence; LS—longitudinal study; LTCF—long-term care facility; MA—meta-analyses; MANOVA—multivariate analysis of variance; MD—mean difference; min—minutes; MLAPS—
Modified Lexington Attachment to Pet Scale; MMSE—Mini-Mental State Exam; mn—months; N—number of studies; n—number of participants; NGO—nongovernmental organization; NRCT—
nonrandomized control trials; NRNCT—nonrandomized noncontrolled trial; OA—older adult; OMI—outcome measurement; OT—observation table; PI—Paro intervention; PO—potential outcomes;
PRISMA—Preferred Reporting Items for Systematic Reviews and Meta-Analyses; pt—patient; PwD—person with dementia; QD—qualitative data; QDC—qualitative data collection; QIS—
qualitative interview study; QoL—quality of life; QUALID—Quality of Life in Late-Stage Dementia scale; RAID—Rating Anxiety in Dementia; RCT—randomized control trial; RT—recreational
therapist; SCS—single case study; SD—standard deviation; SG—support groups; SPS—stated preference survey; SR—systematic review; UCLA—University of California Los Angeles; w/—with;
w/o—without; wk—weeks; wkly—weekly; y.o. —years old; y—years; ZBI—Zarit Burden Inventory; &—and; α—Cronbach’s alpha value; +—plus; ≥—greater than or equal to

ROBOTIC PETS AND DEMENTIA
Citation

Theoretical/
Conceptual
Framework

44

Design/Method/
Purpose

Sample/
Setting

resident w/ dementia
& a robotic cat
depends on resident’s
level of cognitive
impairment.

Inability to partake
in study.

Variables

Measurement/
Instrumentation

Data
Analysis

Results/
Findings
78% held plush
cat

Attrition: Not
discussed

Level of Evidence;
Application to
Practice/
Generalization

Robots provided a
response to nursing
home residents’ need for
social contacts. Level of
overall agitation
decreased significantly
when residents were
interacting with pet.
Feasibility/
Applicability to pt.
population:
Introduction of pets
similar to that described
in the study could be
feasible to PwD. The
applicability is
Beneficial due to the
effects pets have on
resident in a short period
of time. Also trained
staff at nursing home is
readily available to carry
out intervention

(Yu et al., 2015).
Use of a
therapeutic,
socially assistive
pet robot
(PARO) in
improving mood
and stimulating

Standardized
framework

RCT—single-blind,
methodologically
rigorous RCT
treatment outcomes
assessed at baseline
(pre- during, and
posttreatment).
2 groups,

N=30
n=15 (CG)
n=15 (IG)
Demographics:
Communitydwelling older
Chinese

IV1—PARO (TG)
IV2—
Psychosocial
activities (CG)
DV1—Mood

FSS
OT
MMSE
CSDD
ZBI
Subjective
impression,
questionnaire,

ANOVA
ANCOVA
Independen
t t tests
Wilcoxon
signed
ranks test

FSS (0.63, SD
0.74)
t test (alpha=.05
& power=.9)
PARO TG 78.6
(range 72–87,
SD 5.3)

LOE: II
Strengths:
First RCT to use PARO
in improving mood and
stimulating social
interaction and
communication.

Key: ABMI—Agitated Behaviors Mapping Instrument; ABS—Agitated Behavior Scale; AIBO—robotic dog; AIBOG—robotic dog group; ANCOVA—analysis of covariance; ANOVA—analysis of
variance; b/w—between; CG—control group; CI—confidence interval; CMA—Cohen-Mansfield Agitation Inventory; CMAI—Cohen–Mansfield Agitation Inventory instrument; CSDD—Cornel
Scale for Depression in Dementia; CSS—cross-sectional study; d/o—diagnosis of; Dep.—depression; DG—dog group; DM—Delphi methodology; d/t—due to; DQ—descriptive qualitative; DS—
databases searched; DSM-IV—Diagnostic and Statistical Manual of Mental Disorder; DSS—Delphi survey scale; DV—dependent variable; dz—disease; EG—experimental group; FSS—Face Scale
Score; GDS—Global Deterioration Scale; h/o—history of; hr—hour; ICC—intraclass correlation coefficient; IG—intervention group; ITT—intention to treat; IV—independent variable; JBIMAStARI—Joanna Briggs Institute Meta-Analysis of Statistics Assessment and Review Instrument; LAPS—Lexington Attachment to Pet Scale; LMBS—Lawton’s Modified Behavior Stream—LOE
—level of evidence; LS—longitudinal study; LTCF—long-term care facility; MA—meta-analyses; MANOVA—multivariate analysis of variance; MD—mean difference; min—minutes; MLAPS—
Modified Lexington Attachment to Pet Scale; MMSE—Mini-Mental State Exam; mn—months; N—number of studies; n—number of participants; NGO—nongovernmental organization; NRCT—
nonrandomized control trials; NRNCT—nonrandomized noncontrolled trial; OA—older adult; OMI—outcome measurement; OT—observation table; PI—Paro intervention; PO—potential outcomes;
PRISMA—Preferred Reporting Items for Systematic Reviews and Meta-Analyses; pt—patient; PwD—person with dementia; QD—qualitative data; QDC—qualitative data collection; QIS—
qualitative interview study; QoL—quality of life; QUALID—Quality of Life in Late-Stage Dementia scale; RAID—Rating Anxiety in Dementia; RCT—randomized control trial; RT—recreational
therapist; SCS—single case study; SD—standard deviation; SG—support groups; SPS—stated preference survey; SR—systematic review; UCLA—University of California Los Angeles; w/—with;
w/o—without; wk—weeks; wkly—weekly; y.o. —years old; y—years; ZBI—Zarit Burden Inventory; &—and; α—Cronbach’s alpha value; +—plus; ≥—greater than or equal to

ROBOTIC PETS AND DEMENTIA
Citation

social interaction
and
communication
for people with
dementia: Study
protocol for a
randomized
controlled trial
Country:
China
Funding: None
Stated
Bias:
Takanori Shibata
is the developer
of PARO

Theoretical/
Conceptual
Framework

45

Design/Method/
Purpose

Sample/
Setting

Variables

Measurement/
Instrumentation

30-min wkly session;
6-wk period.
Examine whether
robot-assisted
intervention using
PARO in older
Chinese adults with
mild to moderate
dementia improves
mood and stimulates
social interaction and
communication
compared to
psychosocial
activities

≥ 60 years.
Mild to moderate
dementia referrals
from community
dementia day care
centers, geriatric
outpatient clinics,
nurse-led memory
clinics, day
hospitals

DV2-—Social
Interaction

qualitative
comments,
piloting,
video analysis

Setting:
Shatin Hospital,
geriatric day
hospital in Shatin,
new territories,
Hong Kong
Inclusion Criteria:
Communitydwelling older
Chinese adults aged
≥ 60 years MMSE
score b/w 10 and 24
d/o dementia
Exclusion criteria:
Individuals w/
CMAI scoring ≥ 2.
Severe medical
conditions limiting
ability to
participate.

DV3—
Communication
DV4- Cognitive
Function
DV5- Depressive
Symptoms
DV6- Caregiver
Burden
Pet Robot (PARO)
Both treatments
consist of six 30minute weekly
sessions, which
will be conducted
in a geriatric day
hospital. Subjects
in both groups will
be assessed by a
trained research
assistant at
baseline (pre-,
during, and
posttreatment)

Data
Analysis

Results/
Findings
(Cronbach’s
alpha = .86),
test-retest
reliability (alpha
= .78), and good
inter-rater
reliability (ICC =
.99),
CSDD
(Cronbach’s
alpha = .84),
inter-rater
reliability (kappa
= .43–.89)
ZBI- (ICC=.99,
split half
correlation
coefficient=.81)

Level of Evidence;
Application to
Practice/
Generalization

Videotaped observations
of facial expressions and
reactions that standard
questionnaires or proxy
interviews may miss.
Modified OT assesses
the degree to which
PwD will respond w/ the
treatment. Advantage in
ratings developed in
context of social
isolation and
communication. Interrater and intra-rater
reliabilities of OT have
been developed.
Measurement of various
variables reflects mood
and social behaviors
from another
perspective
Weaknesses:
Confine study
population to mild to
moderate dementia.
Limiting
generalizability of
results to a wider
population by excluding
behavioral and
psychological symptoms
of dementia.

Key: ABMI—Agitated Behaviors Mapping Instrument; ABS—Agitated Behavior Scale; AIBO—robotic dog; AIBOG—robotic dog group; ANCOVA—analysis of covariance; ANOVA—analysis of
variance; b/w—between; CG—control group; CI—confidence interval; CMA—Cohen-Mansfield Agitation Inventory; CMAI—Cohen–Mansfield Agitation Inventory instrument; CSDD—Cornel
Scale for Depression in Dementia; CSS—cross-sectional study; d/o—diagnosis of; Dep.—depression; DG—dog group; DM—Delphi methodology; d/t—due to; DQ—descriptive qualitative; DS—
databases searched; DSM-IV—Diagnostic and Statistical Manual of Mental Disorder; DSS—Delphi survey scale; DV—dependent variable; dz—disease; EG—experimental group; FSS—Face Scale
Score; GDS—Global Deterioration Scale; h/o—history of; hr—hour; ICC—intraclass correlation coefficient; IG—intervention group; ITT—intention to treat; IV—independent variable; JBIMAStARI—Joanna Briggs Institute Meta-Analysis of Statistics Assessment and Review Instrument; LAPS—Lexington Attachment to Pet Scale; LMBS—Lawton’s Modified Behavior Stream—LOE
—level of evidence; LS—longitudinal study; LTCF—long-term care facility; MA—meta-analyses; MANOVA—multivariate analysis of variance; MD—mean difference; min—minutes; MLAPS—
Modified Lexington Attachment to Pet Scale; MMSE—Mini-Mental State Exam; mn—months; N—number of studies; n—number of participants; NGO—nongovernmental organization; NRCT—
nonrandomized control trials; NRNCT—nonrandomized noncontrolled trial; OA—older adult; OMI—outcome measurement; OT—observation table; PI—Paro intervention; PO—potential outcomes;
PRISMA—Preferred Reporting Items for Systematic Reviews and Meta-Analyses; pt—patient; PwD—person with dementia; QD—qualitative data; QDC—qualitative data collection; QIS—
qualitative interview study; QoL—quality of life; QUALID—Quality of Life in Late-Stage Dementia scale; RAID—Rating Anxiety in Dementia; RCT—randomized control trial; RT—recreational
therapist; SCS—single case study; SD—standard deviation; SG—support groups; SPS—stated preference survey; SR—systematic review; UCLA—University of California Los Angeles; w/—with;
w/o—without; wk—weeks; wkly—weekly; y.o. —years old; y—years; ZBI—Zarit Burden Inventory; &—and; α—Cronbach’s alpha value; +—plus; ≥—greater than or equal to

ROBOTIC PETS AND DEMENTIA
Citation

Theoretical/
Conceptual
Framework

Design/Method/
Purpose

46
Sample/
Setting

Variables

Measurement/
Instrumentation

Data
Analysis

Results/
Findings

Participating in
other studies

SFS, CSDD, ZBI, and
the subject impression
questionnaire are selfreported.
Misreporting and
nonreporting may occur.

Attrition:
43% (12/30).
Causes included
death and inability
to participate

(Zhou et. al.,
2019)
The association
between
loneliness and
cognitive

Inferred:
Health belief
model

Quantitative LS
Purpose:
Close the gap in the
existing literature by
evaluating the

N =6 898
65 y/o +
Sociodemographic
factors:

Level of Evidence;
Application to
Practice/
Generalization

Conclusions: Strong
evidence suggested that
PARO improved mood
and stimulated social
interaction and
communication in older
PwD, as well as
provided an evidence
base for the use of such
social robots.

IV1—Men
IV2—Women
DV1—
Cognitive
impairment
DV2—

Self/proxy report
MMSE
CES-D
UCLA Loneliness
Scale

Logistic
regression
analysis

OR = 95%
CI = 95%
Cognitive
impairment:
30%
Loneliness w/

Feasibility/
Applicability to pt.
population:
The intervention
described is feasible at
the nursing home.
Applicability slightly
limited by this study
occurring in Hong Kong
LOE: II
Strengths:
Added from previous
studies indicating that
loneliness had a

Key: ABMI—Agitated Behaviors Mapping Instrument; ABS—Agitated Behavior Scale; AIBO—robotic dog; AIBOG—robotic dog group; ANCOVA—analysis of covariance; ANOVA—analysis of
variance; b/w—between; CG—control group; CI—confidence interval; CMA—Cohen-Mansfield Agitation Inventory; CMAI—Cohen–Mansfield Agitation Inventory instrument; CSDD—Cornel
Scale for Depression in Dementia; CSS—cross-sectional study; d/o—diagnosis of; Dep.—depression; DG—dog group; DM—Delphi methodology; d/t—due to; DQ—descriptive qualitative; DS—
databases searched; DSM-IV—Diagnostic and Statistical Manual of Mental Disorder; DSS—Delphi survey scale; DV—dependent variable; dz—disease; EG—experimental group; FSS—Face Scale
Score; GDS—Global Deterioration Scale; h/o—history of; hr—hour; ICC—intraclass correlation coefficient; IG—intervention group; ITT—intention to treat; IV—independent variable; JBIMAStARI—Joanna Briggs Institute Meta-Analysis of Statistics Assessment and Review Instrument; LAPS—Lexington Attachment to Pet Scale; LMBS—Lawton’s Modified Behavior Stream—LOE
—level of evidence; LS—longitudinal study; LTCF—long-term care facility; MA—meta-analyses; MANOVA—multivariate analysis of variance; MD—mean difference; min—minutes; MLAPS—
Modified Lexington Attachment to Pet Scale; MMSE—Mini-Mental State Exam; mn—months; N—number of studies; n—number of participants; NGO—nongovernmental organization; NRCT—
nonrandomized control trials; NRNCT—nonrandomized noncontrolled trial; OA—older adult; OMI—outcome measurement; OT—observation table; PI—Paro intervention; PO—potential outcomes;
PRISMA—Preferred Reporting Items for Systematic Reviews and Meta-Analyses; pt—patient; PwD—person with dementia; QD—qualitative data; QDC—qualitative data collection; QIS—
qualitative interview study; QoL—quality of life; QUALID—Quality of Life in Late-Stage Dementia scale; RAID—Rating Anxiety in Dementia; RCT—randomized control trial; RT—recreational
therapist; SCS—single case study; SD—standard deviation; SG—support groups; SPS—stated preference survey; SR—systematic review; UCLA—University of California Los Angeles; w/—with;
w/o—without; wk—weeks; wkly—weekly; y.o. —years old; y—years; ZBI—Zarit Burden Inventory; &—and; α—Cronbach’s alpha value; +—plus; ≥—greater than or equal to

ROBOTIC PETS AND DEMENTIA
Citation

impairment
among older men
and women in
China: A
nationwide
longitudinal
study
Country:
China
Funding:
US NIA,
National Natural
Science
Foundation of
China & Natural
Science
Foundation of
Fujian Province
Bias:
Female
prominence
noted in oldest
group.
Most of the
women were
illiterate

Theoretical/
Conceptual
Framework

47

Design/Method/
Purpose

Sample/
Setting

Variables

Measurement/
Instrumentation

association between
loneliness and
cognitive impairment
among older men and
women in China
using a nationally
representative
longitudinal sample.

Age, sex, education
level, employment
status, BMI

Feelings of
loneliness

3-model analysis,
sociodemographic
factors,
lifestyle factors,
health status

Setting:
22 provinces in
China
Inclusion criteria:
OA
80y+
Exclusion criteria:
< 65y/o
Attrition: Not
discussed

Data
Analysis

Results/
Findings
OR = 1.30;95%,
CI 1.01–1.69
Men: p < 0.05, p
< 0.01,
P < 0.001
Women:
p < 0.05, p <
0.01,
P < 0.001
Age w/ OR =
1.85; 95%, CI
0.68–0.98
More education;
OR = 0.78; 95%,
CI 0.68–0.98
Employed;
OR = 3.34; 95%,
CI 0.19–0.61

Level of Evidence;
Application to
Practice/
Generalization
significant impact.
Multiple imputations
were applied to address
missing data

Weakness:
Use of validated scale
rather than single
question to accurately
measure loneliness.
50% of participants lost
to follow-up or died
Conclusion:
Impact of loneliness on
cognitive impairment
was significant among
elderly men but not
elderly women
Feasibility:
Applicable to PwD
population and w/
favorable effects on
loneliness. Can be
applied in PwD and
readily available grants
to purchase
nonpharmacological
therapy for maintaining
QoL in PwD

Key: ABMI—Agitated Behaviors Mapping Instrument; ABS—Agitated Behavior Scale; AIBO—robotic dog; AIBOG—robotic dog group; ANCOVA—analysis of covariance; ANOVA—analysis of
variance; b/w—between; CG—control group; CI—confidence interval; CMA—Cohen-Mansfield Agitation Inventory; CMAI—Cohen–Mansfield Agitation Inventory instrument; CSDD—Cornel
Scale for Depression in Dementia; CSS—cross-sectional study; d/o—diagnosis of; Dep.—depression; DG—dog group; DM—Delphi methodology; d/t—due to; DQ—descriptive qualitative; DS—
databases searched; DSM-IV—Diagnostic and Statistical Manual of Mental Disorder; DSS—Delphi survey scale; DV—dependent variable; dz—disease; EG—experimental group; FSS—Face Scale
Score; GDS—Global Deterioration Scale; h/o—history of; hr—hour; ICC—intraclass correlation coefficient; IG—intervention group; ITT—intention to treat; IV—independent variable; JBIMAStARI—Joanna Briggs Institute Meta-Analysis of Statistics Assessment and Review Instrument; LAPS—Lexington Attachment to Pet Scale; LMBS—Lawton’s Modified Behavior Stream—LOE
—level of evidence; LS—longitudinal study; LTCF—long-term care facility; MA—meta-analyses; MANOVA—multivariate analysis of variance; MD—mean difference; min—minutes; MLAPS—
Modified Lexington Attachment to Pet Scale; MMSE—Mini-Mental State Exam; mn—months; N—number of studies; n—number of participants; NGO—nongovernmental organization; NRCT—
nonrandomized control trials; NRNCT—nonrandomized noncontrolled trial; OA—older adult; OMI—outcome measurement; OT—observation table; PI—Paro intervention; PO—potential outcomes;
PRISMA—Preferred Reporting Items for Systematic Reviews and Meta-Analyses; pt—patient; PwD—person with dementia; QD—qualitative data; QDC—qualitative data collection; QIS—
qualitative interview study; QoL—quality of life; QUALID—Quality of Life in Late-Stage Dementia scale; RAID—Rating Anxiety in Dementia; RCT—randomized control trial; RT—recreational
therapist; SCS—single case study; SD—standard deviation; SG—support groups; SPS—stated preference survey; SR—systematic review; UCLA—University of California Los Angeles; w/—with;
w/o—without; wk—weeks; wkly—weekly; y.o. —years old; y—years; ZBI—Zarit Burden Inventory; &—and; α—Cronbach’s alpha value; +—plus; ≥—greater than or equal to

ROBOTIC PETS AND DEMENTIA

48

Citation

Theoretical/
Conceptual
Framework

Design/Method/
Purpose

Sample/
Setting

Variables

Measurement/
Instrumentation

Data
Analysis

Results/
Findings

(Birks et al.,
2016). Robotic
seals as
therapeutic tools
in an aged care
facility: A
qualitative study

Inferred
physiological
model

Qualitative
descriptive,
exploratory

N = 3 RT
n2 = Selected aged
care residents
Setting:
Aged care facility in
a regional
Australian city

Animal-assisted
therapy reduces
depressive
symptoms and
agitation and
results in an
overall positive
experience for
residents

Journal/logs (notes),
interview,
1:1 observation,
group activity,
diversional therapy

Inductive
thematic
analysis

Findings support
the use of
therapeutic tool
revealing
improvement in
emotional state,
reduction of
challenging
behaviors, and
improvement in
social
interactions of
residents

Country:
Australia
Funding:
None
Bias:
Author declared
none.
Implicit bias
regarding
interviews

Daily 1-hr semistructured interview
over 4 mths.
Diversional therapy
Purpose:
Aims to contribute to
the existing body of
knowledge by
exploring the
experiences of
therapists using Paro
as a therapeutic tool
with a more diverse
group of residents in
an aged care facility
in regional Australia

127-bed facility
operated by an
NGO
125–130 staff
Inclusion criteria:
Elderly PwD
aged care facility
resident
Exclusion criteria:
Unable to
participate in study
Attrition: Not
discussed

A therapeutic
tool that’s not for
everybody
Every interaction
is powerful
Keeping the
momentum

Level of Evidence;
Application to
Practice/
Generalization
LOE: I

Strengths:
Strong design,
dependable, transferable
Weaknesses: Small n;
limited generalizability;
Focus on one facility;
uses perspective of RT.
Limited information on
sample size and
demographics.
Funding not disclosed
Conclusions: Evidence
suggesting improvement
in emotional,
behavioral, and social
well-being following PI.
Feasibility/
Applicability to pt.
population:
Feasible intervention.
Could be applicable to
PwD. Evidence of
improvement and easyto-use process for both
care providers and
caregivers.

Key: ABMI—Agitated Behaviors Mapping Instrument; ABS—Agitated Behavior Scale; AIBO—robotic dog; AIBOG—robotic dog group; ANCOVA—analysis of covariance; ANOVA—analysis of
variance; b/w—between; CG—control group; CI—confidence interval; CMA—Cohen-Mansfield Agitation Inventory; CMAI—Cohen–Mansfield Agitation Inventory instrument; CSDD—Cornel
Scale for Depression in Dementia; CSS—cross-sectional study; d/o—diagnosis of; Dep.—depression; DG—dog group; DM—Delphi methodology; d/t—due to; DQ—descriptive qualitative; DS—
databases searched; DSM-IV—Diagnostic and Statistical Manual of Mental Disorder; DSS—Delphi survey scale; DV—dependent variable; dz—disease; EG—experimental group; FSS—Face Scale
Score; GDS—Global Deterioration Scale; h/o—history of; hr—hour; ICC—intraclass correlation coefficient; IG—intervention group; ITT—intention to treat; IV—independent variable; JBIMAStARI—Joanna Briggs Institute Meta-Analysis of Statistics Assessment and Review Instrument; LAPS—Lexington Attachment to Pet Scale; LMBS—Lawton’s Modified Behavior Stream—LOE
—level of evidence; LS—longitudinal study; LTCF—long-term care facility; MA—meta-analyses; MANOVA—multivariate analysis of variance; MD—mean difference; min—minutes; MLAPS—
Modified Lexington Attachment to Pet Scale; MMSE—Mini-Mental State Exam; mn—months; N—number of studies; n—number of participants; NGO—nongovernmental organization; NRCT—
nonrandomized control trials; NRNCT—nonrandomized noncontrolled trial; OA—older adult; OMI—outcome measurement; OT—observation table; PI—Paro intervention; PO—potential outcomes;
PRISMA—Preferred Reporting Items for Systematic Reviews and Meta-Analyses; pt—patient; PwD—person with dementia; QD—qualitative data; QDC—qualitative data collection; QIS—
qualitative interview study; QoL—quality of life; QUALID—Quality of Life in Late-Stage Dementia scale; RAID—Rating Anxiety in Dementia; RCT—randomized control trial; RT—recreational
therapist; SCS—single case study; SD—standard deviation; SG—support groups; SPS—stated preference survey; SR—systematic review; UCLA—University of California Los Angeles; w/—with;
w/o—without; wk—weeks; wkly—weekly; y.o. —years old; y—years; ZBI—Zarit Burden Inventory; &—and; α—Cronbach’s alpha value; +—plus; ≥—greater than or equal to

ROBOTIC PETS AND DEMENTIA

49
APPENDIX B: Synthesis Table

Study
Author
Year
Design/LOE
Measurement
Tools

# of Studies/
Participants
Age (y/o)
Country

Residential
Institution
Robotic Pet
Real Pet
Plush Pet
Standard
Activity
No Activity
Group
Individual

Banks
2008
RCT/II
UCLA-LS
D-LS
MLAPS
and LAPS
NewmanKeuls
range test

Birks
2016
QS/VI
Interviews

Chen
2018
SR/I
GDS
QoL-AD
CSDD
BARS
RAID
UCLA-LS
MMSE

Gustafsson
2015
MMS/III
QUALID
CMAI

38

Not stated

7

18

≥ 55
United
States

Not stated
Australia

≥ 55
Australia
Denmark
New
Zealand
Norway
United
States

≥65
Sweden















Harding
2018
SR/I
JBIMAStARI
COSMIN
COMET
Delphi =
Surveys,
interviews

Liaag
2017
RCT/II
CSDD
CMAI-SF
NPI-Q
TSM
ACE-New
Zealand
version

Libin
2004
RCT/II
ABMI
LMBS

Yu
2015
RCT/II
MMSE
CSDD
ZBI

Zhou
2019
QLS/II
CLHLS
MMSE
CES-D

Sample
7

Pu
2018
SR/I
MMSE
GDS
ACE
BARS
CSDD
NPI-Q
NPI
RAID
CMAI-SF

30

10

9

30

6898

Not stated
United
Kingdom

≥ 55
New
Zealand

83–98
United
States

≥ 55
Denmark
Norway
New
Zealand
United
States
Australia
Japan
Spain

≥ 60
China

≥ 80
China

Setting

















IV—Intervention













































ROBOTIC PETS AND DEMENTIA
Length Per
Session
Session Per
Week
Length of
Intervention
Loneliness
Anxiety
Agitation
Depression
Mood
Social
Interaction
Cognitive
Function
QoL
Medication

50

30 mins

60 mins

Not stated

Unlimited

Not stated

30 mins

1

7

Not stated

Unlimited

Not stated

8 weeks

4 months

Not stated

7 weeks

Not stated










Not stated

30 mins

Not stated

3

2; 5
mins/day
1

Not stated

1

Not stated

12 weeks

1 time

Not stated

6 weeks

Not stated

DV—Outcomes Identified


















































ROBOTIC PETS AND DEMENTIA
APPENDIX C: Watson’s Theory of Human Caring
Figure 1
Theory of Human Caring

Parker, 2001)

51

ROBOTIC PETS AND DEMENTIA

APPENDIX D: Framework
Figure 2
Innovation Mapping Framework

(Bartholomew-Eldredge et al., 2016)

52

ROBOTIC PETS AND DEMENTIA
APPENDIX E: Logic Model

53

ROBOTIC PETS AND DEMENTIA

APPENDIX F: Project Informational Flyer

54

ROBOTIC PETS AND DEMENTIA
APPENDIX G: Recruitment Script

55

ROBOTIC PETS AND DEMENTIA

56

APPENDIX H: Consent Form
PARTICIPANT CONSENT FORM
To Participants:
I am a graduate student under the direction of Dr. Jacqueline Medland at the Edson College of
Nursing and Health Innovation at Arizona State University. I am inviting you and your family
member/care recipient to participate in this quality improvement project. My goal is to determine
whether robotic pets affect the anxiety and agitation levels of people with dementia.
The City of Tempe volunteers will serve as supplemental resources for this robotic pet project. In
your role as primary caregivers, we will pair you with your loved ones during this project. You
will help by introducing robotic pets to your loved ones and encouraging their interaction.
During the interaction, I will complete a 60-minute observation. Then, I will conduct two preand post surveys. The first two surveys will be completed in the first 2–5 minutes of presenting
caregivers with the robotic pets. The post surveys will be completed 2–5 minutes after each 60minute observation. Caregivers will be required to fill out a post study survey after the 4-week
sessions. This survey will assess caregiver views on how robotic pets affected their loved ones
and should take 2–5 minutes to complete.
The project will take place in the comfort of your homes once per week for 4 weeks with a
duration of approximately 90 minutes per session. There is little to no risk to the participants
because the interactions will be conducted within their regular routine.
Every effort will be made to ensure privacy and confidentiality throughout the project. All notes
collected will be kept on a safe computer and password protected. Hard copies of data received
will be kept in a locked filing box. The data may be used in reports, presentations, or
publications and reported in aggregate. Individual data will not be disclosed.
You will pay nothing to participate in this study. The same robotic pet will be provided to
participants each time to prevent contact/sharing. As a token of appreciation and keeping in mind
the current pandemic, participants will be able to keep their robotic pets at the end of the study
period. Participation is completely voluntary. Participants can refuse to participate at any time
during this project without consequence.
If you have any questions concerning the project, please feel free to contact Rushana Gill-Ramos
or Dr. Jacqueline Medland at any time at their phone numbers or email addresses given below. If
at any time you have questions about your rights as a subject/participant, or if you feel you are at

ROBOTIC PETS AND DEMENTIA

57

risk, contact the Chair of the Human Subjects Institutional Review Board, through the ASU
Office of Research Integrity and Assurance, at (480) 965-6788 or research.integrity@asu.edu.
By signing below, you are agreeing to be a participant as well as to extend your right as guardian
over your loved one with dementia to participate in the robotic pet project.
Client Name: _______________________________
Signature: _______________
Date: _______
Caregiver Name: _______________________________
Signature: _______________
Date: _______
Please see attached UCSD Brief Assessment of Capacity to Consent
Thank you for your consideration and participation,
Rushana Gill-Ramos, RN-BSN-MHA
ASU – DNP Innovation Leadership Student
Co-Primary Investigator
602-315-0645
rgillram@asu.edu
Dr. Jacqueline Medland
Faculty Mentor
ASU-Edson College of Nursing and Health Innovation
Primary Investigator
jmedland@asu.edu

NOTE: A COPY OF THE SIGNED AND DATED CONSENT FORM WILL BE KEPT BY
THE CO- INVESTIGATOR AND PROVIDED TO THE PARTICIPANT AND CAREGIVER.

ROBOTIC PETS AND DEMENTIA
APPENDIX I: University of California, San Diego, Brief Assessment of Capacity to
Consent (UBACC)

58

ROBOTIC PETS AND DEMENTIA

59

ROBOTIC PETS AND DEMENTIA
APPENDIX J: Project Outline

60

ROBOTIC PETS AND DEMENTIA
APPENDIX K: Agitated Behavior Scale (ABS)

61

ROBOTIC PETS AND DEMENTIA
APPENDIX L: Rating in Anxiety Scale (RAID)

62

ROBOTIC PETS AND DEMENTIA

APPENDIX M: Caregiver Evaluation of Robotic Pet Interaction

63

ROBOTIC PETS AND DEMENTIA

64

APPENDIX N: DNP Project Budget

Phase
Preparation

Delivery

Evaluation

Activities
Design and print educational
materials for staff and potential
audiences
Provide video of use and function
of robotic pets (provided by
manufacturer Joy for All)
Design, develop, and print
evaluation tools (surveys), ability
to consent forms, and contracts for
participants of project. Purchase
ink and paper
Send out letters reaching out to
memory care homes for
participation in project (includes
multiple attempts to reach
memory care homes and
resending and follow-up of letters)
Travel expenses (gas/food) to help
network with the city/memory
care homes to ensure everyone is
on board with project
Provide robotic pets to five
individuals from the memory care
community
10@ $130 (paid for by CO-PI
related to the pandemic and fear
of transmission of COVID-19)
Reach out to memory care
staff/family members and PwD
for postsurvey (gas/food/phone
calls/time)
Review and analyze results
(30days@$99/day)

Cost
$400

Subtotal
$400

$0

$0

$240

$240

$120

$120

$680

$680

$650

$650

$120

$120

$99

$99

Total

$2,309