Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
International Journal of Nursing Studies Advances 2 (2020) 100009
Contents lists available at ScienceDirect
International Journal of Nursing Studies Advances
journal homepage: www.elsevier.com/locate/ijnsa
Acceptability of personal contact interventions to address
loneliness for people with dementia: An exploratory mixed
methods study
Hannah M. O’Rourke
a , ∗
, Souraya Sidani
b
, Nicole Jeery
b
, Judy Prestwich
a
,
Haydn McLean
a
a
Faculty of Nursing, University of Alberta, Level 3 Edmonton Clinic Health Academy, 11405 87 Avenue, Edmonton, AB T6G 1C9 Canada
b
School of Nursing, Ryerson University, 350 Victoria Street, Toronto, ON M5B 2K3 Canada
Keywords:
Dementia
Intervention development
Loneliness
Acceptability
Mixed methods
Background: Personal contact interventions involve routine visits with a person or animal to ad-
dress loneliness. Research supports the promise of these interventions to address loneliness among
cognitively intact older adults, but little is known about their use with people with dementia.
Objective: To assess the acceptability of personal contact interventions for use to address loneliness
with older people with dementia, according to formal and informal care providers.
Design: Cross-sectional, mixed methods complementarity design.
Setting: Ontario, Canada
Participants: A purposive sample of 25 family members, friends, and health care providers of
people with dementia.
Methods: Participants attended a face-to-face interview to discuss the acceptability of personal
contact interventions. Participants completed questionnaires to rate acceptability (adapted Treat-
ment Perception and Preference measure). A semi-structured interview followed to discuss the
ratings and features of personal contact (with another person or animal) in more detail. The anal-
ysis involved descriptive statistics (quantitative data) and conventional content analysis (quali-
tative data). During the interpretation of the results, the qualitative ndings were compared to
the quantitative results to provide context and understand participants’ perceptions of interven-
tion acceptability in more depth; these are presented together in the results to demonstrate their
distinct and complementary contributions to the ndings.
Results: Personal contact with a person or animal was rated as eective, logical, suitable, and low
risk to address loneliness by over 80% of participants. Participants’ willingness to engage in this
type of contact, for example as a visitor or as a facilitator of animal contact, was 72%. Participants
emphasized the benets of personal contact. The ndings highlight that individualized, exible
interventions that include appropriate facilitation are needed.
∗ Corresponding author.
E-mail addresses: hannah.orourke@ualberta.ca (H.M. O’Rourke), ssidani@ryerson.ca (S. Sidani), nicolejeery4@gmail.com (N. Jeery),
prestwic@ualberta.ca (J. Prestwich), hmclean@ualberta.ca (H. McLean).
https://doi.org/10.1016/j.ijnsa.2020.100009
Received 22 July 2020; Received in revised form 11 September 2020; Accepted 3 October 2020
Available online 8 October 2020
2666-142X/© 2020 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
( http://creativecommons.org/licenses/by-nc-nd/4.0/ )
H.M. O’Rourke, S. Sidani, N. Jeffery et al. International Journal of Nursing Studies Advances 2 (2020) 100009
Conclusions: Future studies to develop and test personal contact interventions should involve
exible delivery, assess the feasibility and acceptability of these interventions (as in a Phase 2
trial of a complex intervention), and focus on the experiences of people with dementia.
Tweetable Abstract: Tailored, routine, and facilitated contact with a person or animal shows
promise to address loneliness for people with dementia.
What is already known about this topic:
• Loneliness is emotionally painful and harms the health and quality of life of those that
experience it.
• Personal contact interventions refer to routine visits with another person or animal and have
been found eective in addressing loneliness among cognitively intact older adults.
What this paper adds:
• Friends, family members and health care providers of people with dementia view personal con-
tact interventions as logical, suitable and eective to address loneliness of older adults with de-
mentia.
• Personal contact interventions are not always easy to implement and do not automatically pro-
mote meaningful connection and prevent loneliness for people with dementia.
• Strategies to tailor and facilitate personal contact interventions are needed to promote their
eectiveness when used with people with dementia.
1. Introduction
Addressing loneliness experienced by older persons is an international priority ( Alzheimer’s Society, 2013 ). Loneliness is emo-
tionally painful and associated with depression ( Marshall, 2011 ; Van Orden et al., 2013 ) and a 26% increased likelihood of mortality
( Holt-Lunstad et al., 2015 ). Factors related to aging, such as living alone and cognitive impairment, may increase the risk of loneliness
among older persons ( O’Rourke and Sidani, 2017 ). In particular, older persons with dementia experience changes that drastically
alter their social interactions. Changes like forgetfulness progress to diculty following conversations and, nally, to severe mem-
ory impairment and an inability to formulate recognizable speech ( Alzheimer Society of Canada, 2020 ). To prevent loneliness and
promote quality of life, relationships and support systems must adapt to meet the needs of persons with dementia ( O’Rourke and
Sidani, 2017 ).
Loneliness is a negative feeling resulting from a perceived decit in companionship, quantity or quality in one’s relationships with
an attachment gure or community ( El Sadr et al., 2009 ; Weiss, 1973 ). It is dierent from being socially isolated; one can be alone
but not lonely ( Holt-Lunstad et al., 2015 ). Social connectedness, the opposite of loneliness, reects feelings that one is cared about
and belongs to a group or community ( O’Rourke and Sidani, 2017 ). Older people with dementia have some needs that dier from
those without dementia and are a diverse group. Interventions to promote social connectedness need to be adapted to meet the needs
and preferences of diverse groups of older people with dementia; when an intervention meets the needs of the target population, it
is more likely to be engaged in and eective ( De Las Cuevas et al., 2018 ; Sidani, 2015 ).
Personal contact interventions have been used most frequently to address loneliness among cognitively intact older persons
( O’Rourke et al., 2018 ). Interventions are best classied according to their active ingredients which are, in this case, the mecha-
nisms hypothesized to aect loneliness ( Campbell et al., 2007 ; Sidani et al., 2020 ). A personal contact intervention has one active
ingredient: routine, close contact with others ( O’Rourke et al., 2018 ). There are several possibilities for the mode of delivery for this
contact. Contact could occur with another person (family member, friend or volunteer) or an animal (live or robotic). The contact
with the person or animal may be delivered face-to-face or may involve the telephone or other technology. It is usually delivered in
one-to-one sessions, but animal contact has also been delivered in group sessions ( O’Rourke et al., 2018 ).
Several studies have evaluated the eects of personal contact in samples comprised of people who were primarily cognitively
intact; ndings supported a small, positive impact on loneliness ( Reed, 1986 ; Robinson et al., 2013 ; Schwindenhammer, 2014 ; Tsai and
Tsai, 2011 ; Tsai et al., 2020 ; van der Heide et al., 2012 ). However, none of the personal contact interventions evaluated in these
studies was designed or evaluated specically for use with people with dementia.
Two small studies have explored the feasibility of dementia-specic interventions aimed to use contact with a person to address
loneliness for people with dementia. The rst used technology (a telepresence robot and video call) to promote personal contact
with family members ( n = 5) ( Moyle et al., 2017 ). The second assessed a software program designed to prompt caregivers to have
reminiscence-based conversations with people with dementia ( n = 3) ( Purves et al., 2015 ).
There are more studies of personal contact delivered via a robotic animal. A systematic review of robotic animals demonstrated
a small reduction in loneliness among the cognitively intact and reduction in behavioral symptoms among people with dementia
( Bemelmans et al., 2012 ). This review supported the promise of robotic animal contact interventions, but larger studies that include
comparison interventions are required to further examine their feasibility and eectiveness ( Bemelmans et al., 2012 ).
Notably absent in the dementia-specic literature are studies of low-tech personal contact interventions that involve scheduled
visits with a person face-to-face or by phone, and visits with a live animal. Also, no studies were identied that aimed to understand
H.M. O’Rourke, S. Sidani, N. Jeffery et al. International Journal of Nursing Studies Advances 2 (2020) 100009
intervention acceptability from the perspectives of key stakeholders, such as family members, friends and health care providers of
a person with dementia. Key stakeholders are those people that understand the needs of a person with dementia and can identify
whether the intervention addresses the problem in a way that makes sense, based on their experiences. Understanding acceptability
from the perspectives of key stakeholders is an important piece of intervention design; it is like an early screen to determine whether
the intervention is likely to be appropriate for use in the target population ( Sekhon et al., 2017 ). The use of mixed methods to assess
acceptability can help to identify adaptations to optimize an intervention’s eectiveness and to improve the extent to which it may
be used in practice ( Sidani, 2015 ).
The current evidence is limited in informing the design of personal contact interventions for use with people with dementia.
Assessing acceptability is a rst step to develop an intervention that will be used by people with dementia and implemented with
delity by caregivers in practice ( Sekhon et al., 2017 ). The purpose of this study was to explore formal and informal caregivers’
perceived acceptability of personal contact interventions aimed to decrease loneliness experienced by people with dementia.
2. Methods
2.1. Design
We used a cross-sectional mixed methods (complementarity) design. Participants rated the acceptability of two personal contact
interventions (i.e., with a person and with an animal) in a questionnaire. Immediately following each intervention rating, participants
discussed, in a semi-structured interview, their reasons for the ratings and suggestions for how to adapt and improve personal contact
interventions for use with people with dementia. Qualitative and quantitative data were collected concurrently from June 2016 to
April 2017, weighted equally, and compared in the analysis ( Creswell and Plano Clark, 2018 ).
2.2. Sample
We selected a purposive sample of family members, friends, and formal health care providers of a person with dementia. The
person with dementia may have lived in either a residential care setting (i.e., a long-term care home) or was community-dwelling.
Participants were ≥ 21 years old and able to read and write in English. We planned to recruit 25–30 participants. This number is
typical of qualitative interview-based studies to achieve information redundancy (usually reached at ~20 participants) and avoid
overly complex analysis ( Vasileiou et al., 2018 ). Participant availability and data saturation (i.e., our observations of information
redundancy) determined the nal sample ( Saunders et al., 2018 ).
2.3. Recruitment
Recruitment strategies included: advertisement in local newspapers and newsletters; posting yers in public spaces, on a university
campus, and in one large long-term care home in Toronto, Ontario; and word-of-mouth where participants informed their network
of the study. Individuals called the research oce for more information, to conrm their eligibility, and schedule an interview.
2.4. Data collection
All participants reported on their age, gender, and cultural or ethnic background (in their own words); healthcare providers
also indicated their professional title and work setting. The acceptability questionnaires were adapted (to reect the interventions
examined in this study) from the Treatment Perception and Preference measure ( Sidani et al., 2018 ). The Treatment Perception
and Preference measure has demonstrated internal consistency ( 𝛼> 0.85) and there is evidence that supports construct validity
(i.e., participants’ stated preferences for an intervention correlated with their ratings for that intervention) ( Sidani et al., 2018 ). The
adapted measure was reliable for personal contact with a person or animal ( 𝛼= 0.82 and 0.93, respectively).
Participants rated and discussed personal contact with a person and then with an animal as follows. The interviewer read aloud a
lay summary of the goal, activities, benets and risks of the intervention, while the participant read a printed copy. The participant
completed ve items to rate eectiveness, logic, suitability in ameliorating loneliness, ease of use, and the study participant’s will-
ingness to engage in personal contact (e.g., as a visitor or facilitator) on 5-point scales. Higher scores indicated higher acceptability
(e.g., 0 = not at all to 4 = very much). The total score for these ve items ranges from 0 to 20. An additional item assessed risk severity,
and higher scores indicated lower acceptability (0 = not bad at all to 4 = very much bad). An interview immediately followed to ex-
plore perspectives of the intervention’s dose (i.e., length and frequency) and mode of delivery, willingness to engage in intervention
sessions, and possible modications to make the intervention highly acceptable and feasible. The lay summaries and interview guides
were based on the descriptions of personal contact interventions from the literature. Interviews were audio-recorded and transcribed.
2.5. Analysis
Quantitative data (participant characteristics and ratings) were analyzed with descriptive statistics using SPSS version 22.0. Only
the cultural background variable had missing data. Concurrently, the qualitative analytic team (JP, HM, and HMO) completed a
conventional content analysis ( Hsieh and Shannon, 2005 ). Our analysis stayed close to the participants’ intended meaning, as in
qualitative description ( Sandelowski, 2000 ). Lead analysts (JP/contact with a person, HM/contact with an animal) rst reviewed
H.M. O’Rourke, S. Sidani, N. Jeffery et al. International Journal of Nursing Studies Advances 2 (2020) 100009
Table 1
Participants’ cultural or ethnic background in their own words.
Category Participants’ words Frequency Percent
Specified a single ethnic or cultural background 16 64
African 2 8
Arab 1 4
Asian 1 4
Originally from/grew up in Guyanag 1 4
Canadian, Greek descent 1 4
Caucasian 1 4
Eastern European 2 8
Filipino 3 12
Portuguese 1 4
Third-generation Japanese Canadian 1 4
Uganda; Baganda tribe, Bantu group 1 4
White; European 1 4
Specified a mixed cultural or ethnic background 4 16
Black & Spanish 1 4
Caucasian/Scottish/French 1 4
Macedonian /Lebanese/Irish/Canadian 1 4
Mulato 1 4
Canadian Canadian 3 12
Not reported 2 8
I believe in humanity, not any particular culture 1 4
No response 1 4
transcripts and audio recordings to clean and familiarize themselves with the data. They then added codes to any phrases which
described participants’ perceptions of the interventions using the comments function in Word. JP and HM analyzed three transcripts
each, and agreed on a preliminary list of codes and code denitions for each intervention, in discussion with HMO. The team met
regularly to review the audit trail of (i) tables that linked all themes, categories, codes, and supporting data and (ii) memos of analyst
impressions of the emerging ndings. HMO lead the nal stage of the analysis to merge the data sources and generate the nal
themes. First, she wrote a description of her observations of some key similarities and dierences in intervention ratings based on
the descriptive statistics. Inferential statistics were not calculated because of the exploratory nature of the analysis (multiple tests
increase risk of type 1 error). In addition, because the sample size was based on the needs of the qualitative analysis, it was likely
underpowered to detect statistically signicant dierences (high risk of type 2 error). Second, she reviewed the complete audit trails
to compare the qualitative ndings for contact with a person to contact with an animal. This was done to identify a set of common
themes that helped to describe and explain the acceptability of both interventions. She re-organized the audit trail tables as needed.
During this stage, she referred frequently back to the description of the intervention ratings, exploring how the qualitative themes
could help to understand the observed similarities and dierences in ratings. These comparisons between contact with a person and
animal, and between the qualitative and quantitative results, generated a nal set of themes. These themes are common to both
contact with an animal and person, and provide the context to understand the ratings in more depth. The analytic team approved
the nal synthesis. To reect how qualitative and quantitative data were merged in the interpretation phase, statistics are presented
with themes in the results section. This integration and presentation gives a more complete and nuanced description of intervention
acceptability ( Onwuegbuzie and Leech, 2005 ; Younas et al., 2020 ).
2.6. Ethical approval and informed consent
Written informed consent was obtained at the research oce, before the interview. We provided $60 as an incentive and to oset
the transportation costs to all participants that arrived for an interview, regardless of enrollment. This study was approved by the
Ryerson University Research Ethics Board (REB 2016-100).
3. Results
The sample ( n = 25) included 16 family members and friends and 9 health care providers of people with dementia. The average
age was 42 years (SD 12; range 23–65). The sample was ethnically diverse, typical of the population in Toronto, ON ( Table 1 ). A
majority were women (56%, n = 14). The health care providers worked as health care aides (33%, n = 3), registered nurses (22%,
n = 2) or licensed practical nurses (44%, n = 4) in home-care (22%, n = 2) or long-term care homes (78%, n = 7).
The mean total Treatment Perception and Preference scores were similar for contact with a person (13.2, SD 3.31) and an animal
(11.92, SD 4.58) ( Table 2 ). Dierences were observed in the proportions of participants that rated the interventions as eective,
logical, suitable, and convenient (i.e., Treatment Perception and Preference item ≥ 2) ( Table 3 ). Reasons for dierences in ratings are
discussed and contextualized with the qualitative ndings derived from health care provider (HCP) and family member/friend (FF)
participants.
H.M. O’Rourke, S. Sidani, N. Jeffery et al. International Journal of Nursing Studies Advances 2 (2020) 100009
Table 2
Mean acceptability ratings for personal contact interventions ( n = 25).
Item Mean (SD)
Person contact Animal contact
Effect 2.96 (0.841) 2.48 (0.963)
Logic 2.84 (0.688) 2.56 (1.003)
Suitability 3.12 (0.726) 2.48 (1.085)
Risk severity .36 (0.700) .48 (0.586)
Ease 2.12 (1.054) 2.16 (1.106)
Willingness to engage 2.12 (1.013) 2.24 (2.24)
Scale total (without risk item) 13.2 (3.31) 11.92 (4.58)
Table 3
Frequency of acceptability ratings of ≥ 2 for personal contact interventions ( n = 25).
Item Person contact Animal contact
≥ 2% (n) 2% (n) 3% (n) 4% (n) ≥ 2% (n) 2% (n) 3% (n) 4% (n)
Effect 96 (24) 24 (6) 44 (11) 28 (7) 84 (21) 36 (9) 32 (8) 16 (4)
Logic 100 (25) 32 (8) 52 (13) 16 (4) 88 (22) 32 (8) 40 (10) 16 (4)
Suitability 100 (25) 20 (5) 28 (12) 32 (8) 80 (20) 24 (6) 40 (10) 16 (4)
Ease 68 (17) 28 (7) 32 (8) 8 (2) 72 (18) 28 (7) 36 (9) 8 (2)
Willingness to engage 72 (18) 36 (9) 28 (7) 8 (2) 72 (18) 32 (8) 28 (7) 12 (3)
Note. 2 = eective, logical, suitable, easy, willing to engage; 3 = very eective, very logical, very suitable, very
easy, very willing to engage; 4 = very much eective, very much logical, very much suitable, very much easy,
very much willing to engage.
3.1. Effective, logical, and suitable
The large majority of all of the participants rated contact with a person as eective (96%), logical (100%), and suitable (100%).
Contact with an animal was also highly rated, but the percentages of participants rating it as eective (84%), logical (88%), and
suitable (80%) appeared lower than ratings for contact with a person. The participants’ descriptions of the benets of contact with
either a person or animal supported the quantitative ndings. Participants’ perceptions of the overall benets of personal contact
existed alongside the view that contact with another person was essential to all, while contact with an animal was appropriate for
some but not all. Participants emphasized assessing t of the contact, focusing on meaningful relationships, and exible delivery of
personal contact interventions.
3.1.1. Benefits of personal contact overall
Participants supported the benets of personal contact, with another person or an animal. Potential benets of personal contact
included providing comfort, promoting happiness, decreasing loneliness or anxiety, helping the person open up, occupying their time,
and preventing responsive behaviors: “…of course it helps them to be more calm and settled and decrease anxiety and loneliness. ” (HCP2) .
Personal contact interventions may also help care providers see the person with dementia in a new light, or reassure family members
and friends that the person with dementia was cared for: “… it eases the burden even for a brief time for the family and friends. ” (FF2).
The somewhat lower ratings of contact with an animal as compared to contact with a person can be explained by the nding that
contact with a person was widely perceived as being essential for everyone. In contrast, the appropriateness of contact with an animal
depended more on the individual’s preferences and characteristics. “I think that there’s no disputing that communication with other people
like outside a clinical or outside of a caregiving role is important to people. ” (FF1) In contrast, “it really depends on which residents are you
working with but yeah, so not all enjoys animal contact in my experience. ” (HCP1)
3.1.2. Assess fit with contact
Participants saw the eectiveness of personal contact with another person or an animal as contingent upon the creation of close,
intimate relational experiences. The t of the contact with the person with dementia was therefore essential to assess. This assessment
would determine whether an intimate bond existed or was likely to develop. Assessment of the t between a person with dementia
and an animal contact depended upon a person’s history with or preferences for animals, cultural background, and allergy status:
“…if someone is an animal lover and the person tries, is not very social…that’s something that they can do …just spend time with the animal,
give them comfort. ” (FF3) For the t of the person with dementia with another person, assessment included the interests of the person
with dementia, their comfort engaging with people whose gender identities dier from their own, whether they enjoyed connecting
with people their own age or younger, and language spoken and understood by the person with dementia. Use of simple demographic
information such as age, gender or ethnicity to choose or ‘match’ a contact was not recommended, without such an individualized
assessment. “I would say it depends on …who they’d prefer to engage with… I mean everybody’s different. ” (HCP3)
H.M. O’Rourke, S. Sidani, N. Jeffery et al. International Journal of Nursing Studies Advances 2 (2020) 100009
3.1.3. Focus on meaningful relationships
Participants emphasized that the focus on intimate relationships should be maintained when considering intervention modica-
tions to t the real-world context. For contact with a person, participants discussed whether additional services like transportation,
shopping, minor housekeeping and repairs, or meal preparation should be part of the visit. In addition to having aordable help,
doing activities together was seen as one way to engage the person with dementia in a familiar, enjoyable activity, to strengthen
the relationship, and build trust. However, task completion should not be the focus and the activity should be carefully selected to
promote social connection. Ideas for such activities included looking at photos or familiar items, listening to favorite music, or writing
letters. “There are some residents who are very frustrated. They cannot really write so at least have this one-on-one friendly visit and she
knows that this lady is going to help her to do this thing …” (HCP4)
For contact with an animal, participants considered how one-to-one and group delivery may impact the development of a mean-
ingful relationship with the animal. Many supported one-to-one delivery to enhance intimacy, allow intervention personalization,
promote focus, and because people have various allergies and preferences for animals and for participation in a group: “…they just
put everybody in the group and they don’t personalize now who is this and who that –does she like animals? Are they allergic? ” (HCP5).
In addition, participants suggested that using the same animal each visit may promote a feeling of familiarity and connection. “It’s
better with the same animal because they build a relationship…” (FF5)
However, small group sessions (e.g., of four) people with dementia and several animals were also considered potentially appro-
priate to meet the needs of people with dementia. Participants thought these sessions would still be small enough to ensure a quiet
environment and to allow the session facilitator to feasibly manage multiple animals. Some participants thought that personal contact
may also occur between group members, amounting to a more potent intervention that would target development of relationships
with the animal and with other people: “…so they can share the animal together and talk and play with it. And just having the animal
might encourage conversation and more interaction. ” (FF4)
3.1.4. Flexible program delivery
Opinions of the appropriate length and frequency of personal contact interventions varied. Consistently, the ndings demonstrated
a need for exibility in intervention dose to meet individual’s needs. The appropriate length of visits could depend on factors including
how much other social contact the person receives, and the person’s current mood and interest in having more social contact. In some
situations, 30-minute visits were seen as appropriate given the limited attention span or energy of the person with dementia. “Maybe
about half an hour. Because you cannot really get the attention of one resident focusing on one animal. ” (HCP6) Others suggested longer
visits to develop rapport and for bonding, if the person is ‘chatty’, and to help the person with dementia remember the contact. “I’m
thinking an hour would be good, a good start… You need some kind of bonding time. ” (FF6)
Participants thought routine visits may give the person with dementia ‘something to look forward to’. Some thought weekly visits
were sucient, while others thought visits should occur more frequently. More frequent visits could help the person with dementia
remember the contact, and would give them someone to talk to or connect with more often. “Even an hour a week I think is too short.
Every day someone should pop by and say hello, how are you, good morning…” (FF7)
3.2. Ease of delivery and willingness to engage
From the descriptive statistics, it appeared that ease of delivery and willingness to engage in personal contact interventions were
rated lower than eectiveness, logic and suitability for both contact with an animal and with a person. Participants considered contact
with an animal as being about as easy to deliver as contact with a person (68% and 72%, respectively). They appeared equally willing
to engage in contact with a person (e.g., as the visitor) or contact with an animal (e.g., as the facilitator) (both 72%). The qualitative
ndings helped to explain the quantitative results related to ease of delivery and willingness to engage. Participants identied several
key issues that should be considered to select and prepare the contact. The participants also described how facilitation of the sessions
was essential, and described barriers to participation.
3.2.1. Preparing for contact with a person
Participants described that the contact should be patient, calm, empathetic, understanding, and easy going. Experience with people
with dementia may help to prepare the contact. Three main strategies were discussed to prepare the person to feel comfortable, listen
actively, and to promote engagement of the person with dementia. The rst strategy was for the community, clinical or research
group organizing the intervention to provide education related to the stages of dementia, how to recognize and respond to behaviors,
and appropriate verbal and non-verbal communication strategies. “…the person helping them has to be trained in like …coping strategies
and just what it is, and how do deal with those moments, for example, when the person loses their memory. ”(FF8)
Second, participants identied that some family members and friends may believe that visits no longer mattered because the
person with dementia would not remember. Telling family members and friends about potential benets of one-to-one visits may
increase their willingness to engage, and to consent for the person with dementia to participate (in cases where proxy consent is
required). “So if there’s some type of like educational program or authority outside of the caregiver that can promote the benefits of this to
loved ones and maybe it can encourage more participation. ” (FF4)
Third, participants emphasized having knowledge of the person with dementia including their culture, religion or beliefs, interests,
previous hobbies or jobs, and topics to avoid. A volunteer contact could provide family members and friends ‘a break’ and augment
existing visits, but development of knowledge about the person with dementia was critical to forming a meaningful relationship. For
some, contact may be more eective with a family member or friend visitor, to capitalize on an existing bond. “If they are trying to
H.M. O’Rourke, S. Sidani, N. Jeffery et al. International Journal of Nursing Studies Advances 2 (2020) 100009
reduce loneliness…you have to learn the person, what do they like? …What are they more interested in? Music? Going out for activities? They
like taking walks? ” (HCP5)
3.2.2. Preparing for contact with an animal
Participants supported that animals should be temperament tested, free of disease, and vaccinated. To select someone to facilitate
contact with an animal, participants discussed assessing allergies, interest and skills with the animal, and ability to support the person
with dementia during the session. Participants supported that healthcare providers who met these criteria could be the facilitator,
and some considered it their duty. “If you are in the field of care it’s like you are a –you’ve sacrificed to do that, it’s your duty. ” (HCP7)
However, compensation and workload mattered. “…I have 12 residents… are you going to pay me for that? ” (HCP8) Many participants
identied that family members could bring an animal for a visit, but this would be limited by other demands on their time and energy.
“If they have kids that they really need to look after, like pick up from school, feed, take them to bed, and it’s taking a whole lot of their time.
And they don’t have time for the pet and the dementia folk. ” (FF9)
3.2.3. Facilitation is key
For contact with a person, many highlighted videoconference (e.g., Skype) as an excellent option to allow for the person to see
their contact, promoting focus and engagement. “… like it’s more therapeutic when you see the person, the reaction on the face. You believe
okay, this person is still there, they really love me. ” (HCP9) However, use of a phone or videoconference technology required support by
a facilitator because the equipment was potentially unfamiliar “…just the act of holding a phone and knowing that you’re supposed to be
talking into the receiver or that someone’s talking to you on the phone, she wouldn’t be able to comprehend I think... (FF4) The facilitator’s
assessment would determine equipment accessibility, how to introduce the person to the equipment, the response to equipment use,
and whether continued support was required following equipment set-up.
For contact with an animal, participants discussed how the facilitator should promote both safety and engagement in sessions.
The facilitator would observe the responses of the person with dementia throughout the session, promote positive interactions, and
consider the needs of the animal. “It should be like a room like this with a facilitator and to see there’s no violent interaction. Just to make
sure nobody’s getting hurt. ” (FF5) Participants discussed that sessions that were delivered one-to-one or in a small group (e.g., < 5)
with multiple animals were ideal to promote one-to-one contact with the animal. Larger groups would prevent the facilitator from
giving individualized attention, and undermine feasibility and eectiveness of the session. “…say with five patients, each patient with
a different animal, so each patient will have different activity with that different animal, so you will not be able …to give them the right time
and if it’s one hour and you have five people then practically you are doing nothing. ” (HCP7)
3.3. Risk severity
Personal contact with either a person or animal was seen as low risk. The majority rated the risk severity as 0 = ‘not bad at all’
(72% for contact with a person and 56% for contact with an animal), or 1 = ‘somewhat bad’ (24% for contact with a person and 40%
for contact with an animal). Participants identied risks to address in order for personal contact interventions with a person or animal
to have a positive eect.
3.3.1. Risks to address
For contact with a person, participants believed that people with dementia may be cautious of new people and situations, under-
mining eectiveness. “So my mom would go into the room and say “I’m not sitting on the couch with her. I don’t even know her. Why would
I be there? ”…she just didn’t like the idea of having a stranger in her home. ” (FF10) The criteria provided for including animals in the
intervention (i.e., temperament tested, free of disease, vaccinated) led many participants to rate contact with an animal as low risk.
However, participants considered animals unpredictable. “I don’t care how tempered the animal is, animals like to bite and scratch out of
love too. ” (FF6)
Participants supported that a six-week period was sucient to see benet from a personal contact intervention, but worried that
withdrawing the intervention would lead to feelings of abandonment. Overwhelmingly, participants wished for personal contact to
be made continuous, and were concerned that withdrawing the intervention at 6 or 12 weeks would negatively impact the person
with dementia who had bonded with the contact. “The patient might really bond with this person and then three months later you’re done
with the person and it’s almost like a sense of abandonment at that point if the person with dementia can process it. ” (FF2)
4. Discussion
Our ndings support that formal and informal caregivers value personal contact with a person or animal, and view them as
eective, logical, suitable, and low risk interventions to address loneliness for people with dementia. Past research has highlighted
that there are several variables, including previous contact patterns, dementia status, and ethnic background, that inuence how
much current contact a person with dementia is likely to receive as a long-term care home resident ( Port et al., 2001 ). In other
words, people with dementia have diverse needs and opportunities to benet from personal contact interventions, and this was
also reected in our ndings. One way to meet diverse needs is by developing a targeted intervention. A targeted (as opposed to
tailored) intervention would specify dierent standardized versions of the intervention for use with dierent subgroups. The sub-
groups are typically determined from their demographic features (e.g., age, cultural background) ( Tancredi et al., 2013 ). For example,
participants with dementia may be matched with a contact of a similar age, cultural background and gender. Or, the researcher would
H.M. O’Rourke, S. Sidani, N. Jeffery et al. International Journal of Nursing Studies Advances 2 (2020) 100009
select the length and frequency of contact in advance based on their assessment of the amount of baseline contact of the person with
dementia. However, most participants in our study did not support using demographic proles to select a standardized version of
personal contact, and did not see this as an acceptable way to meet diverse needs and preferences.
In contrast to targeting, tailoring personal contact would involve a more individualized assessment of each person with dementia
to understand their needs and preferences. We found that personal contact should be individually tailored, and exible to meet
the diverse preferences and needs of the person with dementia. Participants in our study held varied perceptions of what was most
appropriate for people with dementia, in particular in relation to the length and frequency of visits. A scoping review supports that
the one active ingredient of a personal contact intervention is routine, close contact with a person or animal ( O’Rourke et al., 2018 ).
Therefore, there are many other features that could be co-designed with the participant or their designated decision-maker without
aecting this active ingredient. Components that may be tailored include who the contact is (which person or animal), length of visit
(30 min or more), frequency of visit (weekly or divided over multiple visits per week), and mode of delivery (face-to-face or using
technology).
Our ndings support that tailoring to meet individual expectations will be more acceptable than assigning participants a standard-
ized intervention. Higher acceptability can improve enrollment in trials, treatment engagement, and outcome achievement. Taken
together, this leads to increased power to detect signicant intervention eects ( Sidani, 2015 ). Evidence supports that tailored in-
terventions can achieve large eect sizes. For example, an activity program tailored to the capabilities of the person with dementia
found a large, and statistically signicant, improvement in dementia-related behaviors and a reduction in required hours of infor-
mal care ( Gitlin et al., 2010 ). Similarly, a review of reviews and guidelines indicated that the eects of psychosocial interventions
were maximized when interventions were tailored, embedded in daily care, and considered the individual’s life context ( Vernooij-
Dassen et al., 2010 ). Furthermore, tailored interventions better reect practice, as treatment follows clinical assessment ( Kovach et al.,
2006 ). However, to mitigate the risk of variability in delivery, tailored interventions should be well designed, implemented using an
algorithm, and monitored for delity of implementation ( Sidani, 2015 ). The association between treatment variability and outcome
achievement can then be formally analyzed ( Czaja et al., 2003 ).
Knowledge of the person with dementia was identied by participants as a key to promoting a meaningful connection. Family
members and friends may be especially well suited to serve as contacts or facilitators for this reason. It is especially important to
support family members’ involvement in interventions with people with dementia who live in long-term care home settings ( Vernooij-
Dassen et al., 2010 ), but there are barriers. We identied that transportation, the nature of the relationship with the person with
dementia, and time constraints aect family members’ and friends’ willingness to engage in personal contact interventions. These
are similar to the barriers that prevent family members and friends from visiting people with dementia in long-term care homes
( Miller, 2019 ). In short, we cannot rely alone on family members and friends; strategies to prepare volunteers or health care providers
to have close and meaningful contact with people with dementia are needed to increase the feasibility and reach of personal contact
interventions.
More research to identify how best to facilitate involvement of family members and friends in personal contact interventions is
needed. The personal contact intervention, while aimed to promote feelings of social connection for the person with dementia, may
also have a secondary impact on visiting patterns of the family member or friend. In a systematic review, psychological barriers had a
major impact on whether family members or friends visited a person with dementia in a long-term care home. Psychological barriers
included feelings of guilt, depression, being overwhelmed, and discomfort with dementia or long-term care homes ( Miller, 2019 ). A
study of a video-conferencing intervention with cognitively intact residents of a long-term care home found that allowing the family
member to see parts of the resident’s life in the care home was helpful to promote feelings of connection ( Schuster and Hunter, 2019 ).
This suggests that using a personal contact intervention to involve the family members or friends in the current life of the person
with dementia may help to overcome these psychological barriers and could, in theory, promote more visiting after the study period
ends. This is an area for future research.
Living with dementia is a risk factor for decreased social contact (in person and over the phone) following diagnosis ( Hackett et al.,
2019 ). This issue persists and may worsen over time. People who are admitted to care home environments experience a reduction
in weekly visits and number of contacts by about 50%, phone calls decrease by 56% upon admission, and people with dementia
have even less contact than residents that are cognitively intact ( Port et al., 2001 ). Similarly, our results emphasize that facilitation
of personal contact by a third party is very important in many situations. Such facilitation may be needed to promote engagement
by family and friends or to ensure that technology is used appropriately, for example. We should not expect personal contact to
‘just happen’. Even among cognitively intact residents, use of videoconference equipment has required facilitation by care sta
( Schuster and Hunter, 2019 ).
Individualizing (or tailoring) interventions, preparing contacts following assessment, and ensuring appropriate facilitation may
mitigate risks related to unpredictable animal behaviors or responses to ‘strange’ visitors. However, participants’ desires for an ongoing
intervention is challenging to accomplish in a research setting, where funding and activities end. Researchers need to identify ways
to support regular personal contact following the study period. For example, the research assistant could discuss with the person with
dementia or their decision maker how to facilitate visits from family members, friends or from a voluntary visitor or animal therapy
organization after the study period ends.
The current study ndings can prompt continuing care providers to consider critically how visits are promoted within their
practice settings. Our ndings support that health care providers should reect upon several issues. Relevant questions include: (1)
What are my assumptions related to the importance of personal contact for people with dementia? (2) What does my organization do
to assess the contact received by people with dementia, and how individualized is this assessment? (3) What could be done within my
organization to facilitate people with dementia to have regular contact with others? and (4) What are the opportunities to assess and
H.M. O’Rourke, S. Sidani, N. Jeffery et al. International Journal of Nursing Studies Advances 2 (2020) 100009
promote engagement by the person with dementia during personal contact? Continuing care providers can also use the ndings from
future intervention studies to inform the implementation of programs aimed to promote regular meaningful visits, prevent loneliness,
and promote quality of life of people with dementia.
4.1. Strengths and limitations
This mixed methods study used quantitative ratings and qualitative semi-structured interviews to gain a nuanced understanding
of the acceptability of personal contact interventions. A larger sample size would have provided a more stable quantitative estimate
of intervention acceptability. However, our sample of 25 met our exploratory aims, allowed for in-depth analysis of the qualitative
data with the resources available for the project, and we achieved data saturation ( Saunders et al., 2018 ). Data saturation refers to
our observation of similar comments from many participants, and redundant information leading to the generation of very few new
codes after about 15–20 interviews. In addition, the nal themes have analytic generalizability, in that they can be used to think
about the problem beyond the study sample ( Polit and Beck, 2010 ).
The views of older persons with dementia are essential to consider in order to fully understand intervention acceptability. It is a
limitation that people with dementia were not included in this study. They were not included because their views may dier from
those of their family members, friends, and health care providers and because a study to understand perspectives of people with
dementia must employ methods dierent from those used here. While people with mild dementia may be able to make sense of
and recall written descriptions in order to provide ratings and answer questions about personal contact, these approaches are not
likely appropriate for use with people with moderate to severe dementia. To understand acceptability and how people with dementia
experience and respond to personal contact interventions, data should be collected during future feasibility studies: the researcher
can both observe their engagement during visits (for those with mild, moderate or severe dementia) and ask several questions to
understand their experiences immediately following each visit (for those with mild to moderate dementia who can respond verbally).
5. Conclusions
Personal contact with a person or an animal was assessed by family members, friends and health care providers of people with
dementia. People with dementia are a group at risk of loneliness and subsequent poor quality of life. The ndings support that personal
contact is potentially eective, logical, and suitable, to address loneliness of people with dementia. A personal contact intervention
must involve routine, close contact with others, but it has many other elements that can and should be tailored to meet the individual
needs of the person with dementia. Our ndings highlight that simply putting someone in contact with another person or animal
will not necessarily be eective to prevent loneliness. Personal contact visits should be tailored to meet the individual needs of the
person with dementia, have facilitation approaches in place that aim to promote intimacy and meaningful connection, and be exible
enough to meet people’s diverse and changing needs. While our ndings are important and highlight specic adaptations for use
with people with dementia, they are perhaps unsurprising. People with dementia are human beings with an innate need for social
connection; their needs for personal contact will resonate with many of us who are (for the moment) counted among the ‘cognitively
intact’.
Funding sources
This work was supported by the Canadian Institutes of Health Research Post-Doctoral Fellowship program, and a Faculty of
Nursing, University of Alberta Professorship in Dementia Care Interventions (both held by the rst author). Funders were not involved
in any aspect of the research, in the writing of the report, or in the decision to submit the article for publication.
Declaration of Competing Interest
None.
References
Alzheimer’s Society, 2013. Dementia 2013: The Hidden Voice of Loneliness [WWW Document]. URL https://www.basw.co.uk/system/les/resources/basw_105857-6_0.pdf
(accessed 7.8.20).
Alzheimer Society of Canada, 2020. Stages of Alzheimer’s Disease [WWW Document]. URL https://alzheimer.ca/en/Home/About-dementia/Alzheimer-s-disease/Stages-
of-Alzheimer-s-disease (accessed 5.25.20).
Bemelmans, R., Gelderblom, G.J., Jonker, P., de Witte, L., 2012. Socially assistive robots in elderly care: a systematic review into eects and eectiveness. J. Am. Med.
Dir. Assoc. 13, 114–120.e1. doi: 10.1016/j.jamda.2010.10.002 .
Campbell, N.C., Murray, E., Darbyshire, J., Emery, J., Farmer, A., Griths, F., Guthrie, B., Lester, H., Wilson, P., Kinmonth, A.L., 2007. Designing and evaluating
complex interventions to improve health care. BMJ 334, 455–459. doi: 10.1136/bmj.39108.379965.BE .
Creswell, J.W. , Plano Clark, V.L. , 2018. Designing and Conducting Mixed Methods Research, 3rd ed. Sage Publications, Thousand Oaks, CA .
Czaja, S.J., Lee, C.C., Schulz, R., Belle, S.H., 2003. A methodology for describing and decomposing complex psychosocial and behavioral interventions. Psychol. Aging
18, 385–395. doi: 10.1037/0882-7974.18.3.385 .
De Las Cuevas, C., Motuca, M., Baptista, T., de Leon, J., 2018. Skepticism and pharmacophobia toward medication may negatively impact adherence to psychiatric med-
ications: a comparison among outpatient samples recruited in Spain, Argentina, and Venezuela. Patient Prefer. Adherence 12, 301–310. doi: 10.2147/PPA.S158443 .
El Sadr, C.B., Noureddine, S., Kelley, J., 2009.
Concept analysis of loneliness with implications for nursing diagnosis. Int. J. Nurs. Terminol. Classif. 20, 25–33.
doi: 10.1111/j.1744-618X.2008.01110.x .
H.M. O’Rourke, S. Sidani, N. Jeffery et al. International Journal of Nursing Studies Advances 2 (2020) 100009
Gitlin, L.N., Hodgson, N., Jutkowitz, E., Pizzi, L., 2010. The cost-eectiveness of a nonpharmacologic intervention for individuals with dementia and family member