This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Research Protocols, is properly cited. The complete bibliographic information, a link to the original publication on http://www.researchprotocols.org, as well as this copyright and license information must be included.
Socially assistive robotics is a growing area for geriatric research.
This single-blind, randomized controlled trial (RCT) aims to investigate the use of PARO, a therapeutic, socially assistive pet robot, in improving mood, and stimulating social interaction and communication for people with dementia in the community.
For the study, 40 community-dwelling older Chinese adults (≥60 years) with mild to moderate dementia will be recruited and randomly assigned to the PARO therapy group or the psychosocial activities control group. Both treatments consist of six, 30-minute 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 post-treatment. Mood (assessed with a simplified face scale), social interaction, and communication (ie, facial expressions and reactions towards each treatment, assessed with an observation table) will be the primary outcome measures. Secondary outcome measures will include assessments on cognitive function (Mini-Mental State Examination) and depressive symptoms (Cornell Scale for Depression in Dementia), as well as caregiver burden (Zarit Burden Inventory). Subjective impression towards each treatment and qualitative comments from the caregivers, facilitator, and therapists will also be obtained.
Recruitment to the pilot study began in 2014 and the last subject is expected to complete their post-treatment assessment in 2015.
This will be the first RCT using PARO to improve mood, and stimulate social interaction and communication in the care of older people with dementia, as well as provide an evidence basis for the use of PARO in dementia care in Hong Kong.
The Australian New Zealand Clinical Trials Registry (ANZCTR): ACTRN12614000037606; https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?ACTRN=12614000037606 (Archived by WebCite at http://www.webcitation.org/6Xi7uXdu9).
With a rapidly ageing population, dementia has become an important public health issue worldwide [
Activities, especially in group settings such as social support groups, may be of psychological and social benefits to people with dementia by reducing depression and improving quality of life [
In recent years, socially assistive robots have been developed for elderly care, particularly companion robots [
Amongst the recent literature on the use of socially assistive robots in elderly care, the most widely studied is PARO (
Several intervention trials demonstrated promising effects of participating in PARO therapy in increasing motivation, improving mood, reducing stress, and increasing social communication in elderly people [
To date, the potential benefits of PARO therapy in Hong Kong Chinese have not been systematically examined. Given the successful application of PARO therapy, and the encouraging findings of its positive effects on mood and social interaction and communication, it is suggested that PARO therapy is both feasible and acceptable to elderly people with dementia. Therefore, in this study protocol, we describe the design of a RCT aiming to confirm the findings of our pilot data (see Piloting in the Method section), by providing an evaluation of the effectiveness and benefits of the use of PARO in dementia care in Hong Kong.
The primary objective of the present study is to 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 by conducting a methodologically rigorous RCT. Secondary objectives are to examine the effects of PARO therapy on cognitive function, depressive symptoms, and caregiver burden compared to those of psychosocial activities.
PARO, a social robotic seal.
This is a single-blind RCT comparing PARO therapy and psychosocial activities in the elderly with dementia. Treatment outcomes will be assessed at baseline (pre-), during, and post-treatment. The study will be carried out in Shatin Hospital, a geriatric day hospital located in Shatin, New Territories, Hong Kong. The protocol for this study was registered with the Australian New Zealand Clinical Trials Registry (ACTRN12614000037606) and has been approved by the Clinical Research Ethics Committee of the Chinese University of Hong Kong. The study design is detailed in
Flow chart of the study design. *Indicates the target number of subjects. CDR: Clinical Dementia Rating, CMAI: Cohen-Mansfield Agitation Inventory, and MMSE: Mini-Mental State Examination.
The study population will include 40 community-dwelling older Chinese adults with mild to moderate dementia. Recruitment will occur through clinical referrals from community dementia day care centres, geriatric outpatient clinics, nurse-led memory clinics, and day hospitals. Those who are potentially eligible will be invited to a face-to-face screening assessment including an elicitation of a medical history, medications, and hospitalization for eligibility confirmation. To be eligible for the study, subjects must meet the inclusion and exclusion criteria described in the following sections.
Community-dwelling older Chinese adults aged ≥60 years will be screened using the Mini-Mental State Examination (MMSE) [
Individuals who exhibit any behavioural and psychological symptoms of dementia (at least one item of CMAI scoring ≥2) will be excluded from the study. Subjects will also be excluded if they have severe medical conditions which limit their abilities to complete the course of treatment. Concurrent psychotropic medication will be allowed without restriction, but any change in psychotropic prescriptions over the course of the treatment period will be monitored. In addition, those who are currently participating in other studies, experimental therapies, or blinded treatments will be excluded.
For eligible individuals, the study will be introduced to the subjects and their caregivers. Written informed consent will be obtained from every eligible subject agreeing to participate, as well as their caregivers prior to the study. If a subject is unable to give written informed consent, proxy consent will be obtained.
After obtaining written informed consent, a baseline (pre-treatment) assessment will be performed. Subjects will be randomized into either the intervention or the control group. A study coordinator will randomize subjects by means of a computer-generated list of random numbers in blocks of six, stratified by gender. Treatment assignments will be concealed in consecutively-numbered sealed envelopes, which will be opened sequentially upon subject enrollment. As it is a single-blind study, the subjects and the study coordinator will not be blinded to the treatment assignment. However, the study coordinator will not take outcome measurements. All investigators and outcome assessors will be blinded to the treatment assignment.
The PARO therapy will take place for 30 minutes once a week over a six week period, and will be delivered in a quiet room that is isolated from the common unit (ward) in the Occupational Therapy Department of Shatin Hospital, with minimal environmental distractions. The number of sessions, the length of each session, and the time frame of this study were decided based on the results of our pilot study (see Piloting of this section) and other PARO intervention studies which reported small or significant changes in mood, social interaction, and communication [
The PARO therapy will be delivered in a structured, small-group approach, in which a group of three to four subjects will be arranged to sit around a table with PARO in the centre. The subjects in each group can be substituted by subjects from other groups to optimize the chance of conducting each session. A facilitator who is familiar with the PARO caregiver’s manual will deliver the PARO therapy. Research team therapists will train the facilitator and monitor the sessions.
The PARO therapy is based upon a standardised framework, and involves activities around the concepts of engaging, social interaction, and communication. There are six stages/themes, including (1) introducing PARO, (2) baby-sitting PARO, (3) grooming PARO, (4) feeding PARO, (5) making over PARO, and (6) wardrobe PARO. During each session, the facilitator will show PARO to each subject and demonstrate how PARO responds. Subjects will be encouraged to touch and hold PARO, describe the features and appearance of PARO, and help take care of PARO. The session will proceed for at least 30 minutes. With one session per week, there will be at least a total of 180 minutes of the treatment in the six week period.
Subjects assigned to the control group will be invited to practise a variety of psychosocial activities including a range of table games (eg, Chinese checkers, Jenga, board games etc). The activities will be held on exactly the same schedule as the intervention group (one 30-minute session per week for six weeks). All activities will be carried out in groups of three to four subjects, and will be facilitated by the same facilitator of the intervention group. The sessions will proceed for at least 30 minutes. With one session per week, there will be at least a total of 180 minutes of the treatment in the six week period.
Measurement of outcomes will take place at pre-, during, and post-treatment. Mood, social interaction, and communication (ie, facial expressions and reactions towards each treatment) will be the primary outcome measures. Secondary outcome measures will include cognitive function and depressive symptoms, as well as caregiver burden. Pre-treatment and post-treatment assessments will be administered by a trained research assistant who will remain blinded to group allocation. Outcome measures administered at each time point are described in
Schedule of assessments.
Assessments | Screening | Pre-treatment | Treatment session number | Post-treatment | |||||
|
|
|
1 | 2 | 3 | 4 | 5 | 6 |
|
MMSEa | √ g |
|
|
|
|
|
|
|
√ |
CDRb | √ |
|
|
|
|
|
|
|
|
CMAIc | √ |
|
|
|
|
|
|
|
|
Demographics, lifestyle, and social characteristics |
|
√ |
|
|
|
|
|
|
|
Simplified face scale |
|
|
√ | √ | √ | √ | √ | √ |
|
Observation tabled |
|
|
√ | √ | √ | √ | √ | √ |
|
CSDDe |
|
√ |
|
|
|
|
|
|
√ |
ZBIf |
|
√ |
|
|
|
|
|
|
√ |
Subjective impression questionnaire |
|
|
√ | √ | √ | √ | √ | √ |
|
Qualitative comments from caregivers, facilitator, and therapists |
|
|
|
|
|
|
|
|
√ |
aMini-Mental State Examination (MMSE)
bClinical Dementia Rating (CDR)
cCohen-Mansfield Agitation Inventory (CMAI)
dFacial expressions and reactions towards each treatment
eCornell Scale for Depression in Dementia (CSDD)
fZarit burden interview (ZBI)
gThe √ indicates at which point of the study the respective assessments will take place.
Demographic information such as age, gender, marital status, educational background, living status, lifestyle factors (eg, smoking and alcohol intake), social supporting network (eg, measures of participation in day care centre, cognitive/memory training, community activities, and supporting groups), attributes of the owners of pet (ie, animal preference, ever kept pets, and plan to keep pets) [
A simplified face scale will be used to assess mood state before and after each session. This very brief, pictorial scale of mood uses a sequence of 7 faces and does not require reading literacy [
In order to objectively examine changes in subjects’ social interaction and communication (ie, facial expressions, and reactions towards PARO or the psychosocial activities), treatment sessions will be videotaped and charted on an every-minute basis with an observation table by reviewing the videotaped clips after each treatment session. The observation table is a modified version of the one originally developed by Wada et al [
To assess the inter-rater and intra-rater reliabilities of the modified observation table, a convenience sample of 11 community-dwelling older Chinese adults with mild cognitive impairment or mild to moderate dementia with a mean age of 80.5 (range 73-88 years, SD 4.9), of which 82% (9/11) were female, and with a mean MMSE score of 16.5 (range 9-24, SD 5.9) were recruited to receive a single 30-minute PARO therapy. The therapy sessions were videotaped and observed by two raters (one occupational therapist and one medical researcher) who independently marked the subjects’ facial expressions and reactions towards PARO on the modified observation table on an every-minute basis (with standardized rating criteria), on two different occasions with an interval of two to four weeks. Intraclass correlation coefficient (ICC) was used to measure the reliability of the ratings. The ICC of the inter-rater reliability was.95-1.00 and the ICC of the intra-rater reliability was.87-1.00 [
The Mini-Mental State Examination (MMSE) is a validated scale to assess cognitive performance in both research and clinical settings [
The Cornell Scale for Depression in Dementia (CSDD) is a validated scale used to assess the signs and symptoms of major depression in patients with dementia [
The Zarit Burden Inventory (ZBI) is a validated scale to assess caregiver burden [
Subjective impression of PARO will be assessed at the end of each session of the PARO therapy by adopting four items from studies conducted by Shibata et al [
Qualitative comments from the caregivers, the facilitator, and the therapists will also be obtained from semi-structured qualitative interviews. Interviews will be conducted within one month of completing the treatment program. For the caregivers, the interviews will explore their perceptions of the impact of dementia on the subjects’ daily lives, experiences of the interventions that were designed to improve mood and encourage social interaction and communication since diagnosis, and feedbacks after the treatments. For the facilitator and the therapists, the interviews will elicit information about the subjects’ reactions towards each treatment. Furthermore, the challenges of subject recruitment, treatment implementation, as well as the factors associated with interest in engagement and adherence of the treatments will also be obtained.
Prior to the study, a pilot study was performed to explore the feasibility and potential benefits of PARO therapy in older adults with dementia. Using a pre-post single group design, community-dwelling older Chinese adults with mild cognitive impairment or mild to moderate dementia were recruited to receive six sessions of the PARO therapy (one 30-minute session per week for six consecutive weeks) in Shatin Hospital. The PARO therapy was delivered in a structured, small-group approach, in which a group of 3-4 subjects were arranged to sit around a table with PARO in the centre. An occupational therapist familiar with the PARO caregiver’s manual delivered the PARO therapy (
PARO therapy.
Sample size calculations were based on the effects on mood in the pilot study: a sample size of 15 subjects per group will allow us to detect differences among the mean values of face scale scores (0.63, SD 0.74) between pre- and post-interventions using one sample
Double data entry, consistency check, and data cleaning will be performed prior to data analysis. An intention-to-treat (ITT) analysis will be carried out, in that all available data will be included, without considering the subjects’ compliance to the allocated treatment. Mean and standard deviation will be used for continuous variables while frequency and percentage will be used to describe the distribution of ordinal and categorical variables. Unpaired tests will be used for the primary analysis. Differences between the intervention and control groups in relation to outcome measures will be compared using an analysis of variance (ANOVA) or independent
Recruitment to the pilot study began in 2014 and the last subject is expected to complete the post-treatment assessment in 2015.
The proposed study is, to our knowledge, the first RCT of the use of PARO in improving mood, and stimulating social interaction and communication compared to psychosocial activities in older Chinese adults with mild to moderate dementia in a day care setting in Hong Kong. The results could have particular importance given the rise in the prevalence of dementia in our society.
The proposed study has several notable methodological strengths. The use of videotaped observations will allow us to capture subjects’ facial expressions and reactions towards each treatment during the treatment sessions, of which standard questionnaires or proxy interviews may miss. The modified observation table assesses the degree to which people with dementia will respond with the treatment by videotaped observation, which has the advantage in that ratings were specifically developed in the context of social interaction and communication between elderly patients with dementia and PARO or the psychosocial activities and other subjects, the facilitator, and therapists. Inter-rater and intra-rater reliabilities of the table have been developed. Another strength of the proposal is the measurement of various variables including caregiver burden, and qualitative components from the caregivers, facilitator, and therapists, which could reflect mood and social behaviours in another perspective.
There are several limitations in the protocol. Although behavioural and psychological symptoms of dementia occur frequently in people with dementia, we will confine our study population to mild to moderate dementia, and will exclude those with behavioural and psychological symptoms of dementia, limiting the generalizability of the results to a wider population. It has been suggested that interventions for those with behavioural and psychological symptoms of dementia should tailor the person’s specific needs and capabilities [
This proposed study will provide an evaluation of the effectiveness and benefits of the use of PARO in older Chinese adults with mild to moderate dementia in a day care setting. Results of this study would showcase a novel activity to improve mood, and stimulate social interaction and communication in community care of older people with dementia, as well as provide an evidence base for the use of such social robots. Further research is warranted to examine the use of PARO in managing behavioural and psychological symptoms of dementia using individualized approaches.
analysis of covariance
analysis of variance
clinical dementia rating
Cohen-Mansfield Agitation Inventory
Cornell Scale for Depression in Dementia
diagnostic and statistical manual of mental disorders
Intraclass correlation coefficient
intention-to-treat
Mini-Mental State Examination
randomized controlled trial
Zarit Burden Inventory
Takanori Shibata is the developer of PARO.