Technology-Mediated Non-pharmacological Interventions for Dementia: Needs for and Challenges in Professional, Personalized and Multi-Stakeholder Collaborative Interventions
Designing and using technologies to support Non-Pharmacological Interventions (NPI) for People with Dementia (PwD) has drawn increasing attention in HCI, with the potential expectations of higher user engagement and positive outcomes. Yet, technologies for NPI can only be valuable if practitioners successfully incorporate them into their ongoing intervention practices beyond a limited research period. Currently, we know little about how practitioners experience and perceive these technologies in practical NPI for PwD. In this paper, we investigate this question through observations of five in-person NPI activities and interviews with 11 therapists and 5 caregivers. Our findings elaborate the practical NPI workflow process and characteristics, and practitioners’ attitudes, experiences, and perceptions to technology-mediated NPI in practice. Generally, our participants emphasized practical NPI is a complex and professional practice, needing fine-grained, personalized evaluation and planning, and the practical executing process is situated, and multi-stakeholder collaborative. Yet, existing technologies often fail to consider these specific characteristics, which leads to limitations in practical effectiveness or sustainable use. Drawing on our findings, we discuss the possible implications for designing more useful and practical NPI intervention technologies.
1 Introduction
Dementia is a chronic syndrome characterized by a decline in cognitive function beyond normal aging, affecting memory, thinking, behavior, and social abilities over time [63]. Currently, dementia has been one of the major health concerns among older people [90], greatly impacting their health status and quality of life. Non-pharmacological Intervention (NPI), referring to therapeutic approaches and strategies that do not involve the use of medication, is being considered as an alternative option to address the needs and well-being issues of People with Dementia (PwD), encouraged by healthcare professionals, researchers, and policymakers alike [79, 90]. Research has demonstrated that NPIs can effectively reduce PwD’s responsive behaviors and co-morbid emotional disorders, maintain their functional capacity and improve quality of life (QoL) in a non-invasive and safe approach, with minimal to no adverse side effects [56, 90].
In the field of HCI, using technologies to support NPI for PwD has drawn increasing attention. With the potential promises of supporting more enjoyable and meaningful NPI activities, and enhancing PwD’s sense of self-hood, engagement, and social participation, researchers have designed and examined multiple NPI technologies (e.g. [41, 43, 69, 86]). While these studies provide the worth-expecting vision for technology-mediated NPIs, most of existing technologies are still in the prototype design and evaluation phases, involving brief testing for a short time rather than applying them in real world [40, 50, 65, 82, 85, 86]. To date, very little is published on the usage and evaluation of NPI technologies in real-world NPI scenarios beyond a limited research period. This research gap is critical because researchers and designers need an in-depth understanding of how to appropriately implement NPI technologies to increase user acceptance [86] and system uptake [54, 58]. Further, literature has increasingly noted that NPI practitioners, in particular therapists, play significant roles in practical NPI [33], and the technologies can only be valuable when the practitioners successfully incorporate them into their practical intervention practices. Yet, compared to the abundance of PwD-oriented analysis and examination [1, 22, 36, 41, 65, 69, 85, 86], issues related to how NPI practitioners experience and perceive technology-mediated NPI in practice are largely under-investigated.
This paper aims to fill these gaps by gaining an in-depth and empirical understanding of technology-mediated NPI, in particular the practical experiences and perceptions of NPI practitioners. With technologies becoming increasingly popular in practical NPIs [22, 36, 41, 43, 69, 83, 86], understanding these questions will help researchers, designers, and developers to better understand and address potential issues in fitting intervention technologies into NPI practices. We investigated these questions by conducting a multi-phase qualitative study, including observations of practical NPI practices and semi-structured interviews with 16 NPI practitioners (11 therapists and 5 caregivers). Therapist here refers to the trained professional who provides therapeutic interventions or treatments to PwDs, and caregiver refers to the professional care provider, who provides care and assistance to PwDs. During our study, we didn’t specify what kinds of technologies our participants have used so that our findings could be open to technologies with various levels in our targeted technology-mediated NPI practices.
Our study identifies 4 key aspects of technology-mediated NPI, including usage of technologies in practical NPI, challenges for PwD to adopt and engage into technology-mediated NPI, challenges of technologies in supporting professional, personalized, situated and multi-stakeholder collaborative practical NPIs, and therapists’ expected technological supports. Our findings suggest that the implementation of NPI is professional, personalized, situated, and multi-stakeholder collaborative complex process, which requires the professional workflow and comprehensive understanding of PwD. Yet, current technology-mediated interventions fail to take these factors into account, which significantly hinders NPI technologies’ practical effectiveness and efficiency. Additionally, we also identify therapists’ expectations for technologies in supporting their practical NPIs.
Our study contributes to HCI community by providing (1) an empirical study of understanding the current non-pharmacological interventions for PwD and how technologies have been used and experienced in real-world NPI scenarios, and (2) an empirical understanding of practitioners’ experiences and perceptions to existing NPI technologies, as well as their expectations of future technological support. According to the results, we (3) clarify the gaps between current NPI technology interventions and the needs of successful NPIs, and (4) propose a set of design implications for future design of technologies applied in NPIs for PwD.
2 Related Work
Our study situates into the research of technology-mediated non-pharmacological intervention (NPI) for People with Dementia (PwD) and its practical effectiveness and impacts. In this section, we first present the background of dementia and NPI. Following that, we review existing literature strands about technologies for NPI and NPI practitioners to situate our work.
2.1 Dementia and Non-pharmacological Intervention
Dementia is a neurodegenerative disorder characterized by a decline in cognitive function [49, 62]. It is a broad term that encompasses various conditions, such as Alzheimer’s disease, vascular dementia, Lewy body dementia, etc. Currently, dementia has affected a significant number of people worldwide, with the global prevalence estimated to be around 55 million individuals, and nearly 10 million new cases are reported each year [62]. This number is expected to rise in the coming years due to aging populations and increased awareness and diagnosis of the condition.
For a long period of time, pharmacological treatment methods have been widely used for treating dementia in practice, typically involving the use of neuroleptic or other sedative medications[26]. However, for PwD, these medications can lead to severe adverse effects, such as sedation, falls, and extrapyramidal symptoms [26]. Further, research has indicated that neuroleptic treatment can lead to lower quality of life and overall well-being for PwD [10], and even accelerate cognitive function decline [55].
In recent years, non-pharmacological intervention (NPI) has been widely used as the alternative treatment for PwD, due to its non-invasive and safe nature, fewer side effects, and potential to improve PwD’s function, independence, and quality of life [90]. According to Douglas et al. [26], NPI includes three main subtypes: standard therapies (e.g. behavioral therapy, reality orientation, validation therapy, and reminiscence therapy), alternative therapies (e.g. art therapy, music therapy, activity therapy, complementary therapy, aromatherapy, bright-light therapy, and multisensory approaches) and brief psychotherapies (e.g. cognitive–behavioral therapy and interpersonal therapy). These different interventions are not entirely independent; there is some overlap and intersection among them and typically they are used in a combination [10].
2.2 Technologies and Non-Pharmacological Intervention
Technology designs for NPI is an active and growing area in HCI. With the potential benefits of supporting meaningful NPI activities, adding value to the lived experiences of PwD, and improving their quality of life, various technologies have been designed and used to mediate various NPIs for PwD, including but not limited to music therapy, reminiscence therapy, daily life skills training, behavioral and psychological interventions [22, 36, 41, 43, 69, 83, 86], etc. For instance, given the significant role of reminiscence for PwD [34], researchers designed personalized digital memory book [36], reminiscence game Memory Matters [86], interactive photograph album MyStory [27], web-based digital photograph frame SharedMemories [27] etc., to support reminiscence intervention for PwD. Houben et al. [41, 43] found that sounds from everyday life can be beneficial for PwD by triggering memories and emotional responses. Building on this finding, they developed an interactive sound player for PwD, aiming to engage them in meaningful sound-based activities. Similarly, in Ruitenburg et al. [69]’s study, an intuitive step-by-step recipe tool was designed to help train PwD’s cognitive and motor skills and promote their collaboration and socialization with others.
Besides these function-specific technology designs, researchers have also examined the roles and effectiveness of different types of technologies in mediating NPI, such as devices with touch screens [65, 86], Information and Communication Technologies (ICTs) [28], interactive media [81], Virtual Reality (VR) [40, 50], Augmented Reality (AR) [89], robots [2], etc. Researchers have demonstrated that interactive media can evoke meaningful personal associations, foster a revival of identity, and encourage active participation [81]; Robotic media can help maintain attention and facilitate communication [85]; Digital devices with touch screens, such as tablets, can improve cognitive function in a more accessible way [47]; VR technologies can provide more playful, motivational, and immersive activities compared with traditional cognitive-physical exercises [50].
More recently, researchers have noticed that focusing solely on the functional needs of PwD is insufficient. Instead, technology design for PwD should support higher-level meaningful activities [42], for instance, in promoting active engagement rather than passive interactions [25], providing aesthetic and sensory enrichment and creating new experiences in everyday life and care environment [42], triggering creativity, engagement, and rich emotional expression [51], supporting empowerment and improving agency [31], promoting self-identity and personhood [7, 45], etc. Viewing dementia as a social being, Foley et al. [30] have proposed an experience-centered design framework to assist designers in integrating mutual recognition into the design process and practices, with the goal of establishing and preserving self-identity for PwD from a social and ethical standpoint.
These studies provide the worth-expecting vision for technology-mediated NPI. However, most of the existing technologies are still in prototype design and evaluation phase, involving brief testing for a short time rather than applying them into real-world scenarios [40, 50, 65, 82, 85, 86]. Given the complex process of practical NPI [20, 67] and the variety of PwD in psychological and behavioral symptoms known as Behavioral and Psychological Symptoms of Dementia (BPSD) [26], it is necessary to gain an in-depth, and empirical understanding of technology-mediated NPI. Our study contributes to bridging this knowledge gap by examining the practical experiences and perceptions of NPI practitioners when integrating technologies into their practical NPI.
2.3 Technologies and NPI practitioners
NPI practitioners, in particular therapists, play significant roles in practical NPI [33]. Before interventions, therapists first conduct comprehensive and professional assessments to gain insights into the individual’s cognitive abilities, body functions, musculoskeletal limitations, and the environmental factors that contribute to dementia symptoms [53]. According to the results, they arrange and deliver the most suitable interventions to meet the individuals’ needs and goals [9]. After interventions, they evaluate the implemented strategies and interventions [33], including accessing the outcomes, identifying the reasons for success or failure, and discerning implementation errors, etc. If symptoms of declining cognition are observed, the therapists promptly report them to physicians, and explore the possibility of other services [33].
In HCI, compared to the abundance of PwD-oriented designs [1, 22, 36, 37, 41, 65, 69, 85, 86], the research of technologies for practitioners during NPI is relatively limited, mainly focusing on the technological practices and needs of music therapists [8, 11] and education for practitioners [70]. Specifically, considering the complications in devices used by music therapists, Baltaxe-Admony et al. [11] explored the useful interfaces and proposed an iterative prototype design of a compact and intuitive device for music therapists. Shen et al. [70] and Carrasco et al. [16] examined the potential and challenges to create an empathetic environment that empowers medical workers’ and caregivers’ learning.
More recently, researchers have increasingly recognized the significance of therapists and other stakeholders (e.g. formal and informal caregivers) in practical NPI [16, 18, 43, 68, 69, 75, 76], indicating the need to have a comprehensive understanding of the whole roles in current non-pharmacological interventions. Some researchers have also begun to conduct usability analyses of technology from the perspective of other stakeholders. For example, Flynn et al. [29] have analyzed the experiences and perceptions of VR technologies for PwD from the viewpoint of formal and informal caregivers. Our paper adds to this literature by deeply exploring the practical opportunities and challenges of technology-mediated NPI from the perspectives of NPI practitioners, which are invaluable in better incorporating technologies into practical interventions and improving the effectiveness of technologies in real-world NPI scenarios.
3 Methodology
In this paper, we aim to gain an in-depth and empirical understanding of technology-mediated NPI in real-world scenarios. We are particularly interested in: 1) the current practices and process of real-world NPIs, 2) NPI practitioners’ experiences of using technologies to support NPIs, the benefits and inconvenience they experienced, if any, and 3) NPI practitioners’ attitudes, perceptions and expectation to technology-mediated NPIs. To investigate these questions, we conducted a multi-phase qualitative study, including observations [19] of in-person NPI practices and semi-structured interviews with 16 NPI practitioners (11 therapists and 5 caregivers). We now present details on the types of data we collected, participants and recruitments, and the data analysis process.
3.1 Data Collection: Observations
We began with the observations of in-person NPI activities, with the purpose of obtaining firsthand insights into NPI practices for PwDs in real-world NPI scenarios. Table 1 shows details of the observed five activities. The duration of each activity was approximately 40 minutes.
Table 1:
ID
Intervention activities
Scale of older adults
Practitioners
Dementia degree of older adults
Settings
N1
Music therapy
6
1 therapist
Mild and moderate
Nursing home
N2
Music therapy
5
1 therapist
Moderate
Nursing home
N3
Art therapy
16
1 therapist
Mild
Community center
N4
Horticulture Therapy
6
1 therapist
Mild
Community center
N5
Pet intervention
12
1 therapist
Mild
Community center
Table 1: Information about the observed NPI activities
3.1.1 Recruitment and Criteria.
The observed activities were recruited through two approaches: 4 from Izhaohu1, a well-known dementia care center in China that specializes in providing technologically mediated care and intervention services specifically tailored for older adults with dementia, and 1 from a WeChat official account2, named Pet Intervention Studio.
The first author, a native Mandarin speaker, contacted Izhaohu through its call for cooperation3 and disclosed our intention of researching technologically mediated NPIs, requirements (including observations of in-person NPIs they conducted and interviews with the engaged practitioners), research ethical approval, and commitments to data privacy and safety. The manager of Izhaohu was very supportive of our research and provided us with the schedule of NPI activities they conducted in Shanghai over the course of one month, allowing us to choose and observe the activities that interested us, and interview the engaged practitioners.
To ensure the diversity and comprehensiveness of the collected data, we chose four activities (N1-N4 in Table 1), covering different intervention activities (including music therapy [73], reminiscence therapy [84], occupational therapy [35], palm therapy, art therapy [24], horticulture therapy [6], and aromatherapy [32]) (see Appendix 7.2 for the specific descriptions of these intervention methods), older adults with varying levels of dementia (including mild and moderate cognitive impairment), different healthcare settings (2 in Nursing home and 2 in community senior service center), and different scales of activities (ranging from 5 to 16). The activity N5 was learned from Pet Intervention Studio, in which an event notification of a public pet intervention (also known as animal-assisted intervention, referring to an intervention approach that involves interactions between humans and trained animals to promote physical, emotional, cognitive, or social well-being [60]) for older PwD in one community service center for the elderly was posted. The first author signed up for, participated in, and observed this event.
3.1.2 Data Collection.
During the practical NPI, the first author stayed in the corner of the activity site, and observed and recorded the whole process, without disturbing the entire process. She closely documented various aspects, including the intervention contents, procedure, engaged people, their practices, interactions, and communications, as well as the techniques and materials employed by practitioners. With the participants’ permission, she also used photographs, audio, and videos to record data. After the observation of these five activities, we identified the general NPI procedure, contents and workflows, the engaged people, their collaborations, interactions and communications, as well as the used technologies and materials during the NPI delivering process. This data served as essential evidence to support our qualitative findings on the one hand, and guided the following interview study on the other hand.
3.1.3 Ethical Considerations.
We obtained research ethical approval from the Ethics Committee of the authors’ institution to conduct all the procedures of our study involving human subjects. Throughout the research process, we took careful steps to protect user rights and privacy, and also got feedback from peers inside and outside our faculty to validate the ethical perspective of our research. Before each activity, the first author introduced herself and the intention of researching the practical NPI process and the possible opportunities and challenges of technologies in mediating NPIs to all participants, including therapists, their assistants, caregivers, as well as the older adults participating in the intervention, and obtained their consent.
Table 2:
ID
Gender
Work Experience
Education Background
Workplace
P1
F
3 years
Rehabilitation
Care facility
P2
M
8 years
Social support and rehabilitation
Care facility
P3
F
2 years
Geriatric nursing
Care facility
P4
F
3 years
Nursing
Mental health center
P5
M
> 10 years
Computer science
Pet intervention center
P6
F
4 years
Geriatric nursing
Community aged-care center
P7
F
4 years
Social work
Nursing home
P8
F
8 years
Nursing and social work
Hospital
P9
F
> 20 years
Psychology
Mental health center
P10
F
7 years
Social work
Nursing home
P11
F
5 years
Nursing
Care facility
C12
F
4 years
-
Hospital
C13
F
7 years
-
Care facility
C14
F
6 years
-
Care facility
C15
F
4 years
-
Community aged-care center
C16
F
1 years
-
Community aged-care center
Table 2: Participants’ demographic information. P1-P11 denote the therapists, and C12-C16 denote caregivers.
Taking into account the cognitive abilities of older participants, we adapted the informed consent process to accommodate their comprehension and decision-making capabilities. To be specific, in N1 and N2 conducted at nursing homes, some of the participating older individuals had moderate cognitive impairment, lacking sufficient capacity to comprehend our research intention and provide consent. For them, we explained our study to and obtained consent from their families accompanying them to the intervention. In N3, N4, and N5 conducted at the community senior service centers, the older participants were primarily ones with mild cognitive impairment. They exhibited only mild cognitive challenges but retained basic abilities for independent living, communication, and decision-making. They independently attended the intervention activities. We therefore directly communicated with them, informed them of the purpose of our research, and obtained their consent.
We explicitly informed all participants that our main objective was to observe the practical intervention process and explore potential applications of techniques. We wouldn’t collect any personal information from them, or interfere with any of their behavior. We promised to strictly abide by the confidentiality agreement, and the collected data was only used for this research. we also carefully made it clear to stakeholders that we would not proceed if they were unwilling to participate. We were grateful that older adults and their families were very supportive of our research and appreciated our attention to elderly individuals with cognitive impairment. They also expressed expectations for the NPI technologies we were studying, hoping to extend assistance to a broader elderly population.
3.2 Data collection: Semi-structured Interviews
To deeply understand NPI practitioners’ practical experiences of using technologies to support NPIs, and their attitudes, perceptions and expectations of technology-mediated NPI, we then conducted semi-structured interviews with 16 NPI practitioners (11 therapists and 5 caregivers). The reason we interviewed these two roles was that, they were two primary practitioners involved in practical NPI based on prior observation. Therapists were primarily responsible for providing interventions, while caregivers were mainly responsible for the health, safety, and emotional well-being of PwD during NPI. Their perspectives were helpful for us to better examine our research questions.
3.2.1 Recruitment and Criteria.
We recruited our participants using two approaches: 5 therapists and 5 caregivers were recruited during the observation, and 6 therapists were recommended by the interviewed therapists. After the observation of each activity, the first author actively reached out to the practitioners of the activity and asked whether s/he was willing to participate in our interview study. 5 therapists and 5 caregivers were recruited through this way. After interviews with 5 therapists, we also invited them to help recommend the other therapists they knew. Through this snowballing way [64], we recruited the other 6 therapists. The demographic information of our interviewees is shown in Table 2.
Of the 16 participants, 2 were male and 14 were female. All 11 therapists had relevant professional educational backgrounds, and worked as therapists for two to more than ten years. Their work organizations covered various healthcare settings, including care facilities, community care centers, mental health centers, nursing homes, and hospitals. 5 care workers were all female, working in care facilities and community care centers. They mentioned a wide range of NPIs that they had conducted, including music therapy [73] (N=4), art therapy [24] (N=2), reminiscence therapy [84] (N=3), horticulture therapy [6] (N=2), aromatherapy [32] (N=2), animal-assisted therapy [60] (N=1), reality orientation [71] (N=2), exercise therapy [61] (N=3), and occupational therapy [35] (N=1). Details of these therapies can be found in Appendix 7.2.
Providing more accessible reminiscence activities for PwD using photos as input to generate questions about their life
Table 3: Probes used in interviews
3.2.2 Data Collection.
The interview questions primarily consisted of two parts. The first part was about participants’ daily NPI, including 1) their daily NPI practices, procedures, targeted groups, 2) used materials and technologies, if any, and 3) encountered challenges. For 10 interviewees recruited through the first way, we also asked 4) the follow-up questions of their practical interventions based on our observations, such as the intention of some practices.
The second part was mainly about participants’ experiences, attitudes, perceptions, and expectations of technologically mediated NPIs. Through the observations, we found the technologies used by therapists and caregivers in practice were limited. To better stimulate interviewee’s thinking about technologically mediated NPIs, we first promoted them through several existing technologies designed to support NPIs, including VR-supported reminiscence intervention and memory training [17], AR-supported behavioral intervention [4], tangible device for music intervention [41], screen-based occupational intervention [68], robot-supported cognitive intervention [21], dialogue-based reminiscence therapy [15]. Details could be found in Table 3. In the initial literature review phase of our research, we identified various types of intervention techniques (see details in Section 2.2). With the selection criteria of covering as many different types of intervention content and techniques as possible, we chose these six probes as interview probes.
Based on these, we promoted participants to think about and discuss the following questions: 5) their practical experiences of using technologies to support NPIs, the benefits and inconvenience they experienced, if any, and 6) their attitudes, perceptions, suggestions, and expectations to future technology-mediated NPIs. All questions were designed to be general, so that the questions could be inclusive to participants to share various levels of experiences with different technology-mediated NPI practices.
All interviews were conducted in Mandarin by the first author, lasting 30 to 60 minutes. 10 were conducted face-to-face in activity sites and 6 were conducted through online Tencent Meeting. All interviews were audio-recorded after obtaining their consent and transcribed into Chinese verbatim after the interviews for analysis. In addition, we also collected other forms of data to triangulate the interview responses, such as related photos or videos they recorded and activity documents, if any.
3.2.3 Ethical Considerations.
Before we started interviews, we further informed participants of our intention and background information, promised the collected data was only used for this research, and got permission from them. All data collected during our study was used in an anonymized way, i.e., there was no link between the collected data and an individual user.
3.3 Data Analysis
We applied the thematic analysis approach [13] for analysis. Three of the authors participated in data analysis process. We started coding and analysis while the data was being collected, and iterated through generating codes and themes from collected data. During the open coding phase, three authors independently reviewed the data, and generated codes pertaining to our research questions. We had weekly meetings to discuss ideas and ensure reliability. After this stage, we generated the initial code list, capturing participants’ practices, procedures, and challenges related to their daily NPI activities; their usage and experiences of technologies in practical intervention, as well as their perceptions, concerns and expectations of the technologically mediated NPIs. Based on this code list, we then re-focused our analysis at the broader level of themes, sorted and synthesized different codes into potential, higher-level overarching themes. Through several rounds of analysis and discussions, we reached an agreement and generated our final satisfactory thematic map of the data, which included four primary themes. In the following section, we present the full details of these themes. We use representative quotes from our participants, translated from Chinese into English, to illustrate our points.
4 Results
According to our research questions, we present the key findings from the following four aspects (as shown in Fig. 1): (1) activities and existing used technologies in practical non-pharmacological interventions (NPI), (2) challenges for PwDs to adopting and engaging into technology-mediated NPI, (3) challenges of technologies in supporting professional, personalized, situated, and multi-Stakeholder engaged NPIs, and (4) practitioners’ expectations for technologies that can better support their practical NPI process.
Figure 1:
Figure 1: Diagrams of Key Findings and Structures
4.1 Activities and Use of Technologies in Practical NPIs
4.1.1 Overview of Practical NPI.
Figure 2:
Figure 2: Observed non-pharmacological intervention sessions. (a) N1 (b) N2 (c) N3 (d) TV used to display materials for reminiscence
According to our participants, the practical NPI could be divided into one-on-one and group interventions, each with its own advantages and disadvantages. Generally, one-on-one interventions could be tailored to individual issues of PwD, support more personalized interventions, but often require better manpower and economic costs. Compared to that, group interventions couldn’t support the personalized intervention very well, but could provide PwD a small community and support their social interaction and better participation. Our participants told us that, what they mostly conducted were group interventions, due to limited practitioner resources and PwD’s economic constraints. All five activities we observed followed the group intervention format, with scales ranging from 5 to 16. All of them were carried out collaboratively with multiple stakeholders (including therapists, assistants, nurses, and caregivers). The details of these five activities could be found in Table 1 and Fig. 2.
According to our observations and the conversations with the practitioners, the workflow of NPI included four main stages:initial assessment, pre-intervention planning and preparation, execution, and post-intervention evaluation. Fig. 3 shows this four-stage workflow of NPI. To be specific:
Initial Assessment. Initial assessment was primarily for understanding PwD’s personal situations and current ability levels, based on what to generate careful and scientific intervention plans. Initial Assessments were often conducted by therapists, and multiple stakeholders, such as PwD’s caregivers and family members, were involved through providing various information of PwD. The concerned information included but was not limited to PwD’s physiological, psychological, cognitive, and ecological factors.
Pre-Intervention Preparation. Following the initial assessment and understanding their specific circumstances, therapists would tailor specific kinds of (or joint) therapies targeting PwD’s conditions with unique strategies, not simply organizing or arranging playful activities. In the five activities, four adopted a joint NPI approach (see Table 1). Based on the tailored therapies, therapists then prepared intervention materials, and outlined the intervention procedures.
Intervention Execution. Next, therapists implemented the NPI. The practical NPI process was flexible, situated and dynamically adapted to PwD’s responses. Handling the dynamic and situated progress and guiding PwD to share more thoughts and feelings were commonly experienced by the therapists in all kinds of NPI. We took the process of N1 as an example to illustrate this process. At the beginning of N1, the therapist first asked all PwD to sing a song as a warm-up beginning, i.e. music therapy; next, the therapist led a reminiscence therapy session, where she invited all PwD to sit in a circle (Fig. 2-a), talk one of their favorite childhood food and then share with others in the group. During the sharing session, the therapist would ask questions to facilitate PwD to self-reflect their childhood memories. Multiple stakeholders, such as caregivers, nurses, and families, were involved in the delivery process. They observed and recorded all PwD attendees’ conditions, provided nudges for them to answer questions, if needed, assessed their engagement levels, and handled emergencies, if any.
Post-Intervention Evaluation. After NPI, therapists monitored the progress and performance of PwD to assess the effectiveness of this NPI, and also decided the PwD’s take-home assignments. Moreover, therapists also tried to gather feedback, from PwD if possible, as well as other stakeholders involved in the interventions, for refining future intervention plans.
4.1.2 Existing Usage of Technologies.
Generally, our findings suggested that existing use of technologies in practical NPIs was very limited. In the five NPI activities we observed, the intervention process primarily relied on traditional materials such as printed papers (e.g. Fig. 2-c), old objects, and outdated machinery, with very limited use of what was commonly referred to as technology. The only technology used in the observed five NPI activities was a projector for displaying images, as shown in Fig. 2-d. In the interviews, all participants reported the same result: very few technologies have been used in their daily NPI practices. Three therapists (P2, P3, and P10) reported they only used projectors and computers for playing audio, video, or other visual materials, such as slides. One therapist (P7) had used arm-strength training machine in physical exercises and another one (P6) mentioned gamified exergames, such as Nintendo Wii. Besides, P6 reported she used a “smart sofa” that could play music for PwD with sleeping disorders. Beyond these, no other mobile applications or emerging technologies were reported by our participants. When further discussing the underlying reasons, they elaborated on a series of challenges that hindered the implementation of technology-mediated interventions in the NPIs from both practitioners’ and PwD’s perspectives, which will be detailed in the following sections.
4.2 Challenges for PwDs to Adopting and Engaging in Technology-Mediated NPIs
In technology-mediated NPI scenarios, PwD’s adoption and engagement are the most critical factors that directly impact the practical effectiveness of technology-mediated NPI. Yet, older adults’ adoption and use of technologies have been inherently complex and situated [74]. For older PwD, these situations are more complicated, which challenges the practical effectiveness of current intervention technologies.
4.2.1 “ It is Challenging for PwDs to Adopt and Engage into Technology-Mediated NPIs Proactively”.
According to our participants, most PwD they’ve met had restricted and fixed thought processes, making it a significant challenge for them to adopt complex intervention technologies. Most of our participants (N = 7) responded that older PwDs usually experienced significantly decreased sense of autonomy to experience technology-mediated NPIs, even if they knew the interventions could be beneficial. For instance, P3, a NPI therapist, commented that, “compared to traditional interventions, older PwD always perceive technological elements as overly complex, and are unwilling to accept and experience them.” She considered the technologies we probed as “I didn’t quite understand some of them, let along the older PwD.”. C13, who had worked as a formal caregiver of PwD for over 7 years, supported this viewpoint with her practical experience of taking PwD to experience VR-based reminiscence training techniques: “These elderly individuals won’t actively engage or experience these things; they lack that awareness. If you put it on them, they won’t feel fresh or enjoyable, and might even be afraid.”
Meanwhile, given the probes we showed that considered PwD as the primary stakeholders, many participants (N = 4) indicated that this kind of technology would be hard to be successful in mediating practical NPIs. In their professional perspective, the actual physiological and cognitive characteristics of older PwD often prevented them to use these technologies independently, particularly for those with moderate to severe cognitive degeneration who required assistance with their daily activities. As P1 explained,
“We usually do not consider older PwD to be independent individuals who can take care of themselves. For PwD, their caring and intervention processes become long-term and inseparable with us and their caregivers. Therefore, the intervention technologies would become useless without taking us or caregivers into consideration.” (P1)
Besides, our participants (P1, P10) emphasized that most older PwD often had multiple Behavioral and Psychological Symptoms of Dementia (BPSD) [26], such as being violent, hallucination, and distrusting others, which greatly stopped them from adopting and engaging into technology-mediated intervention. P1 expressed that, only a very limited number of educated PwD in their mild stage, or PwD with sufficient support and good socioeconomic standing would smoothly adopt it as a daily activity.
4.2.2 “It is Hard for PwD to Trust Technologies”.
In addition to a lack of proactive engagement and participation awareness, our participants also reported that establishing trust with NPI technologies and, in turn, encouraging active involvement from the elderly, posed a significant challenge in the practical implementation of technology-mediated NPI for PwD. They (N= 7) expressed that the practical effectiveness of NPI targeting PwD heavily relied on the trust of PwD towards the practitioners. Only when they trusted the practitioners, better intervention outcomes could be achieved. Particularly, for PwD, the process of building trust was extremely challenging. Our participants mentioned that in their practical NPI processes, they often invested a significant amount of time and effort in establishing trust with older PwD. P6, for instance, gave an example of how she initiated the trust-building process with older PwD:
“We begin our sessions by communicating with PwD to build a rapport with them, without claiming any clear goals. I usually guide PwD gradually to do something. For instance, let them listen to music and relax, and ask them their feelings. Once they feel calm, safe, and comfortable about me and the environment, I then move forward to my interventions.” (P6)
Many of our participants (N = 5) even believed older PwD decided to try and engage with NPI usually because they trusted or were familiar with their closed therapists or caregivers, not because of trusting the interventions per se or related components. C14, working as a caregiver in the dementia care facility for more than six years, told us they (caregivers) were very important in practical NPIs because they were the persons PwD are most familiar with and trust the most.
“We are not professional therapists, but therapists need us very much when they conduct NPI. Because the elderly may not be familiar with therapists, but they trust us the most. In the beginning, we help therapists communicate with the elderly and encourage their participation. If the elderly encounter issues during NPI, such as emotional instability, we are the most capable of comforting them.” (C14)
However, when technologies served as part of or full “practitioners”, it became difficult for older PwD to establish the emotional trust relationship with these technologies, which directly affected older PwD’s adoption of and engagement with intervention technologies, consequently impacting the actual effectiveness of technology-mediated NPIs.
4.2.3 “It is Potentially Risky when Adopting Technologies to Mediate NPIs for PwD”.
While most existing literature considered the practical effectiveness of intervention technologies in improving PwD’s cognitive or spirit conditions as the main target (e.g. [30, 41, 43, 45]), most of our participants (N = 9) ranked PwD’s safety as their priority when considering applying any intervention technologies for PwD. They expressed their concerns about potential safety issues and risks PwD would face if technologies we probed were applied to their intervention workflow.
For instance, for VR technology (T1 in Table 3), although most of our participants (N=7) saw potential in adopting it for reminiscence therapies, many of them (N = 5) expressed their concerns about putting PwD inside VR helmets would cause safety issues, such as falling over, mental disorders and so on. C13, who had taken PwDs to experience VR-based reminiscence training techniques, shared her perception: "After some older adults experienced it, their clothes were completely soaked. Some even couldn’t walk steadily. This is very dangerous, especially for those with high blood pressure levels." C15 who worked in a community-based care center concerned PwD was not able to discern virtual and real worlds given that they might already have hallucinations: “Sometimes, when PwD see a dark block on the ground, they get scared, let alone in a virtual environment.”.
Furthermore, while most existing technologies were designed to provide PwD the means for independent use, with the aim of reducing the workload on therapists and caregivers [2, 15], our participants expressed their concerns about the potential risks when older PwD used these technologies all by themselves.
4.3 Challenges of Technologies in Supporting Professional, Personalized, Situated, and Multi-Stakeholder Collaborative NPIs
The other category of challenges identified in our study primarily pertains to the limitations of technologies in supporting professional, personalized, situated, and multi-stakeholder collaborative NPIs. Our study provides evidence that practical NPI is a highly professional process that requires professional practitioners to design intervention plans based on the characteristics of PwD; Real-time perception of PwD’ participation status is necessary during the execution of NPI to make targeted adjustments to the plans; Multiple stakeholders are involved in the practical NPIs. However, current technologies have largely overlooked these characteristics, which makes it challenging for them to have a meaningful impact in practical NPI scenarios.
Figure 3:
Figure 3: Workflow of Non-pharmacological intervention
4.3.1 Challenges in Supporting Professional NPIs.
Our observation of practical NPIs and all interview data demonstrated that NPI was a very professional practice which required therapists to follow a professional four-stage workflow (see details in Fig. 3). Therapists (N = 5) in our study considered that such a professional intervention process required comprehensive and professional knowledge. Unfortunately, in the opinion of our participants, the majority of intervention technologies we probed were focused on developing new “games”, algorithms, or mobile applications, and lacked an in-depth understanding of and supporting to this professional procedure. As a result, these technologies were completely incapable of supporting professional NPIs with practical utility. Many of our participants considered existing intervention technologies as “more like entertaining games rather than effective tools” (P2). P9 took technology-mediated reminiscence intervention as an example to illustrate this opinion.
“You can design a new tool, for example, setting a vintage enamel cup and asking the individual to reminisce if it belongs to the 60s or 70s. Designing such a tool is very easy, but it is just one aspect of the entire system. Reminiscence intervention requires a professional procedure, starting with a professional assessment of PwD and followed by standardized validation. The usage process also requires a professional evaluation to determine which types of PwD the specific scenario and difficulty level are suitable for.” (P9)
Similarly, P2 expressed his dissatisfaction with T3 in Table 3 and other similar technologies for music therapy, stating that these tools resembled a music class for preschoolers rather than a professional music therapy program for PwD.
“Music therapy is not just about singing or playing songs; it involves many professional elements and emphasizes a structured process. Currently, many tools, such as what you just showed me, seem to have transferred kindergarten or elementary school music classes to nursing homes and labeled themselves as music therapy applications.” (P2)
In addition, the professional and comprehensive nature of NPI required interventions to be carried out for long-term to observe PwD’s changes which were then used for iterated NPI plans to maximize intervention outcomes. However, several therapists (N = 6) in our study addressed most technologies or technological interventions were not reusable or sustainable due to a lack of consideration for long-term implementation.
4.3.2 Challenges in Supporting Flexible, Situated, and Multi-Therapies Combined NPIs.
After probing existing technologies, most of our participants (N=7) expressed that the practical NPIs did not adhere to a fixed process and a single intervention method, as the technologies were designed. Instead, it was flexible, situated, and multi-therapies combined, which challenged the practical effectiveness of these NPI technologies.
To be specific, NPIs were flexible regarding PwD’s ways of participation and practitioners’ approaches in organizing intervention activities. Therapists (N = 9) in our study said that in practical NPIs they never required PwD to stay for the entire sessions or follow the strict process. Instead, PwD could choose to leave a part/stop and then join back for the rest based on their preferences. Meanwhile, the practical NPIs was situated based on PwD’s personal backgrounds and their on-going participation conditions of moods, engagement, environments, and current conditions. During practical NPIs, therapists needed to keep real-time attention to these factors of PwD, and provide constant and situated interactions with them. P6, for instance, introduced her situated interaction with PwD as follows.
“If we notice that a patient is having difficulty with a particular movement, we might say, ’you did that wonderfully’ to facilitate his/her confidence. If a patient seems unhappy with participating in the activity, we might encourage him/her, or switch him/her to an alternative intervention approach.” (P6)
Therapists thought that such synchronous interactions were critical to providing each PwD affirmation that PwD’s participation was worthwhile and their efforts were respected, so PwD would be motivated and continue their participation. Further, our participants (P7, P9) also responded that, PwD’s engagement conditions could constantly change during NPI, thus practitioners would observe their performances and adjust intervention content and methods accordingly. However, most of the existing intervention technologies we probed used fixed interaction and were unable to real-time perceive PwD’s condition and adjust intervention plans accordingly. As P9 said,
“During our practical intervention, the physical and emotional responses of PwD are taken into account. We make adjustments based on their reactions. If the frequency is too high, we decrease it or take a break. If the difficulty level is too low, we increase it. Any issues that arise require adjustments. This process relies on the observation of practitioners. However, when it comes to machines, how can machines achieve this observation and adjustment?” (P9)
The flexibility and situationality of NPI were also reflected in the selection of actual intervention strategies. According to therapists, in practice they usually took composite approaches involving a variety of intervention therapies rather than relying on a singular intervention modality. Our observations also identified this feature (see details in Table 1). Our participants expressed that the amalgamation of diverse intervention modalities often yielded more favorable outcomes, yet most intervention technologies could only support one kind of therapy.
4.3.3 Challenges in Supporting Highly Personalized NPIs.
As most of the existing technologies were designed towards supporting NPI for people with coarse-grained cognitive levels (e.g. mild, moderate, severe cognitive impairment) (e.g. [44, 47, 87]), our participants expressed that these based on coarse-grained cognitive categorization were unlikely to be effective in real-world NPI scenarios, because each PwD exhibited highly personalized characteristics, encompassing physiological, psychological, and cognitive traits, behavioral and psychological symptoms in dementia (BPSD), personal histories and education background, personalities, and other demographic factors. As P7 stated, “each Pwd is unique”. Thus, in therapists’ NPI workflow, they started with customized assessments to develop personalized intervention plans, and later performed individualized outcome evaluations. P7, who worked in the nursing home as a therapist, shared her personalization approach in somatic interventions,
“We will carefully select activities for each intervention. For instance,if one PwD has hemiplegia symptoms or Parkinsons, they might fall down if they stand. For her/him,we will not carry out somatic interventions because of the high risk. We will customize the activities and let PwD sit around a table to do handcraft, which is also called occupational therapy.” (P7)
However, almost all therapists (N=9) in our study considered technologies we probed and the others they learned were hard to work effectively in real-world NPI, because they didn’t take into account the personalized characteristics and uniqueness of each PwD. P2, for instance, commented most tools we probed in Table 3 as “more like toys” due to “they totally don’t understand PwD”. He used T4 shown in Table 3 as an example to explain the toy-like nature:
“Guiding PwD through this design step by step is very challenging. Some PwD may have poor hand motor skills; some may have poor vision and can’t see the screen clearly; some may have cognitive impairments and can’t understand your instructions. Each person’s BPSD symptoms are different.” (P2)
According to P2 and most other therapists (N = 9), “Only PwD that align with these required characteristics had the potential to achieve the intended outcomes.” Even for applications that claimed to design for personalization, such as T1 in Table 3, our participants still considered, achieving a high degree of personalization could be challenging. For instance, P9 questioned how T1 could customize their tasks for the unique characteristics of each PwD: “When designing personalized VR scenes, how can this personalization be achieved given that each person has different familiar scenes and interests?”
4.3.4 Challenges in Supporting Multi-Stakeholder Collaborative NPIs.
The intervention execution usually involved collaboration among several roles (see details in Fig. 3). The therapist played the most crucial role in the practical NPI process. They guided the entire intervention process, interacted with PwD, and executed and adjusted activity plans. Besides therapists and PwD, other stakeholders’ participation in NPI had also been considered to be essential. The assistants, for instance, were mainly responsible for observing PwD’s real-time responses and supporting therapists handling a few PwD who did not follow the procedures, especially in group sessions. Caregivers and family members could provide more detailed personal information in the pre-intervention assessment stage that allowed therapists to gain in-depth understanding of the PwD and prepare personalized intervention plans. C13, for instance, told us before each intervention, the therapist would communicate with them to learn PwD’s personal backgrounds, interests, lifestyle, and preferences, and so on, because “we are the ones who understand the older PwD the most.”.
Active engagement and positive support from these stakeholders potentially facilitated PwD’s high-level participation and improvements. Therapists (N = 7) in our study highly emphasized the importance of these stakeholders in the actual NPIs, stating “without assistants and nurses, I can not be able to independently carry outanNPI activity.” They believed successful interventions had to be collaborative around multi-stakeholders. They thus felt eager to build a "therapists-stakeholders-PwD" interactive network for close connections and collaborations among multi-stakeholders for better PwD’s intervention outcomes, instead of a "technologies-PwD" interaction as what existing interventions did.
4.3.5 Challenges in Supporting Care-Centered NPIs.
Our participants emphasized that NPI targeting PwD needed to be care-centered, and put PwD in the center of making decisions or preparing detailed plans for intervention content. During the interviews, they showed us many specific practices on how they carried out PwD-centered care and interventions. For instance, P2 told us, during the practical NPI process, he put himself at the same horizontal eyesight level with PwD and hugged them during communications to value PwD’s feelings and build emotional connection with them. P9, who worked in the mental health center, expressed the core of NPIs was accompanying, supporting, and listening, instead of "treating or healing". She thought this PwD-Centered NPI absolutely couldn’t be supported by technologies, because "Machines lack warmth and are incapable of establishing trust and emotional connections with PwD" (P9).
Yet, in their minds, current technologies merely replicated the process of NPIs while overlooking PwD’s feelings and emotional needs. This, in turn, affected the level of adoption and engagement of PwD and the practical effectiveness of these technologies. P3 took horticulture therapy as an example to illustrate this viewpoint:
“Horticulture therapy is not just about putting seeds in the soil; it’s a process in which we need to pay attention to PwD’s experiences, observe their reactions, and allow them to feel the soil or the scent of plants. Observing their responses is a complex process. However, technology only knows asking PwD to plant the seed and then fertilize and water it daily.” (P3)
Participants (N = 4) suggested that technology-mediated NPI should involve the focus on individuals during the execution of interventions, rather than solely on processes or data. They also suggested taking NPI to be a sophisticated ecosystem, covering varying aspects, including PwD’s personal and unique conditions and other surrounding factors, such as the current environments and intervention content, ongoing circumstances during interventions, and community and stakeholders’ engagement. Every component had to seamlessly align for successful human-centered interventions with PwD.
4.4 Therapists’ Expected Technological Supports in NPIs for PwD
Despite these challenges our participants reported in using technologies to mediate the NPIs, they expressed numerous expectations and ideas to technological supports in NPIs for PwD. We summarized therapists’ expectations as the following three aspects.
4.4.1 Supporting NPI Practitioners’ Learning and Training.
The first technological expectation expressed by our participants was to use more realistic and immersive interactive technologies to assist NPI practitioners’ learning and training. Our participants indicated that currently there were very limited NPI practitioners, which couldn’t meet the increasing demand of NPIs for PwD. They hoped that more stakeholders could learn basic intervention techniques, so they could conduct NPI for PwD at their workplaces and homes too. However, due to the specialized and complex nature of NPI, traditional learning and training methods often struggled to be effective. P3 took horticulture therapy as an example to explain the challenges and difficulties she encountered when training caregivers.
“Horticulture therapy is a process that we want PwD to experience and feel the soil and the smell. We need to observe their responses and provide guidance. However, I’m concerned about how to ensure the quality of training so that anyone can achieve the same results. I feel they (caregivers) might have learned the operational procedures, but it is challenging for them to effectively perceive and learn these details.” (P3)
She, as well as many other therapists (N = 6) expected to have technologies that provided training support to them and other trainees, such as ‘‘using XR to simulate the intervention scenarios” and “using mobile applications to provide standardized intervention procedures”. In addition, our participants also expressed a strong desire for a knowledge-sharing platform specifically dedicated to NPI, which would provide them with access to more NPI information and knowledge.
4.4.2 Providing Semi-Immersive Senses and Rich Materials for NPIs.
Meanwhile, our participants (N = 4) also expected that technologies could assist in preparing materials and offering multi-modality sensations to enrich intervention content. According to therapists, the higher modality and richer immersion the intervention could offer, the better outcomes reminiscence therapy would be. Our participants (N=5) described that in practice, what they often used were 1D materials, such as pictures, which in their minds could not provide PwD a strong sense of immersion that recalled their memories. For addressing this, P3 brought old objects, such as old military boots and red scarves for her patients to replace, and P7 got true plant pots, seeds, and soil for horticulture therapy.
Our participants told us that such preparations cost huge time and effort, and sometimes therapists only had limited access of resources to handle the diverse personalized PwD and intervention scenarios. P3 and P9, for instance, told us that their time and effort left for preparing each intervention were usually limited and they had to prepare many materials and tools based on the intervention types and content, which also required financial expenses. Thus, they expected to have supporting technologies that offered customized and immersive materials, such as images created by generative AI models, and enriched senses, such as immersive VR environments and simulations.
4.4.3 Assisting in Evaluating PwD’s Status and Changes.
The therapists in our study expressed that they met extreme challenges in evaluating PwD’s real-time status, and intervention outcomes. Many of them (N = 6) evaluated the post-intervention outcomes based on their subjective experiences and impressions, or reports from caregivers and family members, knowing that this assessment approach was limited. Consequently, they expected to have some forms of technologies that could support the post-intervention assessment.
Besides, our participants (N = 4) also strongly desired technology that could assist them in observing and perceiving the emotions and state of PwD during the execution process, which then enabled them to provide more personalized interaction and interventions tailored to the specific needs of PwD. P2, for instance, expressed he hoped technologies could help read PwD’s facial expressions, body gestures, emotional feelings, etc., through which to assist him in conducting more PwD-centered therapy. He also hoped technologies could automatically map the music according to PwD’s conditions, then he could focus on interacting with them.
5 Discussion
This research aims to gain insights into the practices and challenges of technology-mediated NPIs for PwD, taking into account the distinct nature of dementia conditions and the complex intervention processes involved. Our findings revealed that implementing NPIs required a professional workflow and a comprehensive understanding of the subject. However, current technology-mediated interventions were limited by the types of technologies used, practical use cases, and involvement levels, which were often constrained by various factors. In addition to PwD’s cognitive impairment status of adopting technologies, their trust in technology, and practitioners’ concerns about the potential risks associated with technologies all hindered the effectiveness of technology-mediated NPIs. Further, we summarized that successful NPIs for PwD typically presented unique characteristics, such as being professional, personalized, situated, and multi-stakeholder collaborative. Yet, existing intervention technologies often failed to support NPI with these characteristics. According to our findings, we now delve into practitioners’ perceptions and expectations of technology-mediated NPIs and propose several crucial design implications for future research.
Our findings from Section 4.3.2 and 4.3.3 revealed that practical NPI was highly personalized, in which practitioners needed to prepare materials and NPI content according to PwD’s personal experiences (e.g., their personal histories, educational backgrounds, personalities, etc.) and personalized physiological, psychological, and cognitive traits (e.g., BPSD symptoms). Such personalized NPIs can serve as better memory triggers, and lead to better changes in agitated behavior and emotion [3, 52]. However, although prior research [22, 36, 41, 43, 69, 83, 86] developed various technology-mediated interventions or prototypes for supporting NPI for PwD, as well as various design methods, trying to improve PwD’s adoption of technologies and motivate their long-term sustainable use via a human-centered strategy, such as co-design [43], experience-centered design [72], and participatory design approaches [69], most of this work adopted a simplified categorization of PwD’s cognitive impairment levels, such as mild, moderate, or severe [41, 43], overlooking personalized requirements for practical NPI [9].
Further, even for the studies that addressed the importance of tailored NPI [14, 44, 47], they only offered tasks with limited range of difficulties accompanying one aspect of PwD’s cognitive skills at a group level. Even if the difficulty of the exercises varies, these interventions remain standardized for big groups of PwD, that is, everyone receives the same training [47]. Based on our findings, such kinds of technology-mediated interventions - designing for the general PwD group without considering individual-level factors - will meet significant practical challenges in PwD and practitioners’ adoption and sustainable use, even leading to potential risks and safety issues.
We therefore suggest HCI and related communities to explore ways in supporting individual-level customization from content and interaction aspects. At its most fundamental level, technologies for NPI should be designed considering beyond the coarse-grained mild, moderate, and severe cognitive impairments to a more diverse range of physiological, psychological, and cognitive traits of PwD. For example, given the common challenges faced by older PwD, such as visual impairments and difficulty distinguishing between virtual and real features, the design of touchscreens [48] and virtual reality (VR) interfaces [17] needs to carefully consider the applicable user groups. Further, the professionals in our study highlighted that most PwD not only experienced the decline in cognitive abilities but also exhibited various types of BPSD symptoms. These BPSD symptoms were crucial considerations in their daily formulation and implementation of NPI, which should also be taken into account by technological interventions.
Despite the benefits brought by personalized technological interventions, we suggest carefully considering the possible privacy and ethical issues when increasingly incorporating technologies into practical NPI. With the conventional technological strategies, the personalization process often implies more personal data sensing and collection, inevitably giving rise to corresponding data privacy and ethical concerns, particularly when dealing with a unique group like PwD. We, as researchers adhering to human-centric technological development, do not strongly advocate for the development path of “technological determinism” and “data determinism”. Instead, we suggest a thoughtful examination of the role boundaries of technology in NPI for PwD. As articulated by our participants, NPI for PwD is a human-centered intervention, where technology should serve as a better assistant to therapists rather than a replacement for them.
Design Implication: Achieving Personalized Interventions through PwD-centered Design with Both PwD and Stakeholders. We suggest that designers and researchers in HCI domain adopt a PwD-centered design approach to customize technology-mediated NPI based on PwD’s personal traits through activities with both PwD and professional stakeholders, such as co-design workshops, participatory design, and user testing and feedback collection at different stages of development. Given that therapists usually use multi-therapies which are selected according to PwD’s states, technology should be able to meet the varied needs of different PwD while allowing therapists to flexibly switch and adjust interventions during the process, leaving space for practitioners to customize tasks and activities, for different individuals at different stages. In addition, recognizing that PwD may face emotional and psychological challenges (e.g. lack of trust) and safety issues, the design of elements such as interface, graphic elements, and physical interactions should be emotionally warm, supportive, and encouraging as well. Further, personalization should extend beyond content to include the interaction forms, taking into account the cognitive and behavioral characteristics of the specific older adults and their therapists.
5.2 Technology-Mediated Professional NPIs for PwD: Engaging and Empowering Professional Stakeholders at Multiple Intervention Stages
Prior work identified that non-pharmacological management of neuropsychiatric symptoms in clinical settings took place within multidisciplinary teams, and occupational therapists play a significant collaborative role in the care of PwD [33]. Our results in Section 4.1.1 extended this finding and highlighted the roles of professional practitioners in specific intervention stages (Fig. 3). Although practitioners’ critical roles have been commonly acknowledged in general dementia care research [33], few work has investigated their specified roles across complicated intervention stages. Specifically, our findings identified that the practical NPI process and practices are highly comprehensive and professional. Thus, practitioners and other key stakeholders’ involvement are essential in PwD’s NPIs at all intervention stages,e.g., understanding PwD’s conditions and background in the pre-intervention assessment stage, personalizing intervention materials in the preparation stage, monitoring PwD’s performances in the execution stage, and adjusting intervention plans based on feedback from post-intervention evaluation.
While technology holds great potentials to be applied at the above-mentioned stages of the professional NPI workflow (Fig. 3), existing technology-mediated NPI research mostly developed to be used at the execution stage, interacting with only PwD to deliver training tasks [22, 36, 41, 43, 69, 83, 86]. According to our findings (in Section 4.2.1, 4.2.2, and 4.2.3), such design and technological solutions are less likely to address PwD’s practical challenges without considering professional practitioners’ engagement and inputs (or, designing for these practitioners at the same time). Our findings (in Section 4.3.1, 4.3.2 and 4.3.3) revealed therapists’ concerns and skeptical attitudes about the practical usability and sustainability of research prototypes and technology-mediated interventions. Additionally, our findings (in Section 4.2.2) proved therapists function as the "invisible" personnel who implemented these technologies in practical scenarios, guiding PwD’s usage, and assessing potential risks. Few PwD’s intervention processes can happen independently in real-life without the presence of therapists and other potential stakeholders, regardless of PwD’s cognitive impairment levels. Further, our findings also showed that PwD had a high demand for empathy and trust in human-human interactions (in Section 4.2.1, 4.2.3, and 4.3.5). Since therapists can provide "humanistic care" [38] that technologies hardly offer, PwD’s independent use of technologies could be an unattainable goal to achieve during technology-mediated NPIs. These barriers that NPIs are facing demonstrate the critical roles of PwD stakeholders and the necessity of inviting them to participate in the design and research process. Consequently, these technologies are likely to encounter substantial challenges when transitioning from labs to real-world scenarios involving PwD with different personal backgrounds.
Design Implication: Including Professional Practitioners in Design and Research Processes, and Engage them at Multiple Intervention Stages. We recommend future HCI and CSCW researchers and designers to (1) incorporate professional practitioners in their design and research processes, (2) take therapists’ perspectives and consider them as one of the end-users, and (3) support the multiple stages in NPI workflow (Fig. 3) in developing future technology-mediated NPIs. In addition to the intervention execution stage, we encourage future work and research to create technologies for the initial assessment, pre-intervention preparation, and post-intervention evaluation stages. For instance, in the post-intervention evaluation stages, technology can offer more flexible and engaging assessment methods, such as game-based assessments [46], to better assist therapists in adjusting plans based on feedback.
5.3 Technology-Mediated Actionable and Practical NPIs for PwD: Enabling a Collaborative Caring Ecosystem Among Multi-Stakeholders
The collaboration and coordination among key stakeholders are also essential. We extend discussions in Section 5.2 about engaging professional practitioners to further analyze the involvement of other key stakeholders in this section. As suggested in Section 4.3.4 and shown in Fig. 3, besides practitioners, successful NPIs are collective efforts within a whole caring system, where therapists, caregivers and nurses, and family members contribute collaboratively. For example, caregivers and nurses can be involved in monitoring PwD’s health and well-being conditions to prevent any unforeseen incidents through and outside interventions, and offer personalized information and intervention feedback; family members assist therapists by providing essential information related to PwD individuals, and they may also care for PwD at home. While every stakeholder makes a unique contribution, they collectively shape the dynamic influences on context-driven NPI. Thus, research on technology-mediated NPI should take a holistic approach and involve stakeholders within the caring ecosystem, instead of only focusing on PwD or their specific disorder symptoms and skills. It involves considering how various roles perceive and influence the implementation of technology and how technology might impact and empower these roles in NPI. To our knowledge, few literature has comprehensively addressed stakeholders’ roles mentioned in our ecosystem throughout the technology development process.
Besides stakeholders’ collaborations, the intervention process is a dynamic and situated workflow that can also be influenced by various contexts as well as social and environmental factors. Focusing on designing technology fulfilling a function would not be sufficient enough to address PwD’s problems within a holistic environment. Thus, it is crucial to evaluate the contextual situations and performances of PwD. While the monitoring of PwD’s states during NPIs usually relies on therapists’ observations, technology holds the potential to capture, visualize and document their changing states and contextual performances to help better adjust NPI plans and enable the collaboration in dynamic caring ecosystem for practical and actionable use.
However, current research mostly focuses on technologies for supporting PwD, neglecting other stakeholders’ needs in communication and collaboration with PwD from the caring system. Our finding in Section 4.4 elaborated our participants’ needs and expectations to technologies, which are far beyond PwD-centered NPI technologies, but rarely be focused on by existing work. Taking ICT for practitioners’ training and education as an example, given the increasing number of the aging PwD population and the severe shortage of professional therapists [75, 76], researching how to use technology to assist therapist education becomes a crucial and meaningful research direction. However, current efforts in this area are limited and not sufficiently in-depth. Face-to-face methods remain the most common education methods for healthcare providers [88]. Existing technologies mainly looked into providing formal and informal care providers with e-learning and training tools through Internet and electronic devices [12, 23, 39, 57, 66], and most are video-based [59]. Research found that these education tools and resources encounter problems including low completion rates [78], poor outcomes in knowledge gains and attitudinal change [77], and inefficiencies due to limited interactivity [78]. In summary, technologies to support other stakeholders in the PwD caring ecosystem are insufficient and urgently needed for communication, collaboration, and education purposes.
Design Implication: Designing for Collaboration, Communication, and Education Among Multiple Stakeholders in the Ecosystem. We suggest that future work carefully consider how the collaborative dynamics of multiple stakeholders within the complex caring ecosystem impact technology adoption, usage, and effectiveness. Future work could explore how to create technologies that empower multi-stakeholders’ engagement across multiple workflow stages in NPI design and research. Further, in the technology deployment phase, we suggest providing comprehensive education, training, and support to the stakeholders. This may includes the development of resource libraries, training on technology use, and feedback systems to ensure multi-stakeholders can make full use of technological tools. While prior VR/AR applications encountered adoption issues due to the adverse effects reported by PwD [5], our findings from Section 4.4.1 suggested that these technical systems have the potential to assist in the training sessions for therapists and caregivers by simulating care scenarios to help them acquire and practice necessary skills. Moreover, we recommend future work focus on technologies that can facilitate multiple stakeholders’ communication and engagement (e.g., between family and professionals) throughout the intervention process to achieve a collaborative caring ecosystem and establish timely and long-term communication. Possible directions could be a private data-sharing and collection platform or application connecting practitioners and PwD’s families, which visually displays PwD’s personal and intervention status and accessible to only the PwD and his/her stakeholders.
5.4 Limitations and Future Work
In summary, our research offers a comprehensive and empirical investigation of NPI practitioners’ experiences, encounters, and perspectives regarding technology-mediated NPIs. We delve into various tensions reported by our participants when utilizing technology to facilitate NPIs for PwD in practical settings. It’s important to acknowledge a major limitation of our study concerning the participants. Most of our participants had limited experiences with technology-mediated NPIs. While we employed pictures and video prompts to demonstrate current technologies designed for NPIs, and stimulate their discussions, we believe that practitioners with more hands-on experiences with a particular technology over a substantial period may provide more nuanced, specific, and constructive insights into our targeted scenarios. Actually, we initially set recruitment criteria, seeking NPI practitioners who had practical experience using at least one type of technology in their NPIs via multiple sampling approaches. However, as the research progressed, we encountered challenges in recruiting such practitioners. Consequently, we adjusted the recruitment criteria to include NPI practitioners who had learned about technology-mediated NPIs. Future work could focus on NPI practitioners with hands-on experiences in technology-mediated NPIs to gain a more comprehensive understanding of their long-term usage experiences.
Additionally, our study primarily focused on research sites situated in nursing homes and community care centers. Beyond these two settings, the home environment constitutes another vital context for NPIs. Notably, a substantial number of individuals with mild cognitive impairment continue to reside in their own homes [80], underscoring the importance of investigating technology-assisted interventions within ecosystem, i.e., at the home setting. Future work could delve deeper into how environmental and spatial aspects may affect the adoption and sustainable use of NPIs for PwD.
6 Conclusion
While non-pharmocological intervention (NPI) for dementia has drawn great interest in HCI, most existing technology designs are still in the prototype and evaluation phases, and perspectives of NPI practitioners are under-investigated. This paper presents an empirical understanding of technology-mediated NPI, in particular the practical experiences and perceptions of NPI practitioners. Through observations and interviews with NPI practitioners, we find that current technology-mediated interventions are constrained by various factors such as the technology types, case scenarios, involvement levels, PwD’s cognitive impairment and trust in technology, as well as the potential risks associated with intervention technologies. We also demonstrate the gap between technology-mediated NPI and the needs of practical successful NPIs including being professional, personalized, contextually situated, and collaborative among multiple stakeholders. We end by proposing several crucial design implications for future technology-mediated NPI research.
Acknowledgments
We thank the Kunshan Government Research Fund for funding this research (23KKSGR031). We deeply appreciate Shanghai Izhaohu Co., Ltd. for their supports to our research study. We are also grateful for the valuable feedback provided by the anonymous reviewers. The contributions of Zhennan Yi were made during her role as a Research Assistant at Duke Kunshan University.
A Interview Questions
I. Recall of Recent Non-Pharmacological Intervention Sessions:
1. Can you introduce the most recent or impressive non-pharmacological intervention that you organized, Including its duration, frequency, and the experience and feedback from the older individual at that time?
2. In your opinion, was this intervention effective? Why or why not? What were the key factors contributing to its success? How long did it take to see results?
3. How do you usually evaluate the effectiveness of such interventions?
4. What do you think are the current challenges with this intervention, and how can they be improved?
5. Would you recommend this intervention to older adults or their families? Why or why not?
6. Do you involve their family members or caregivers in this intervention? What are the benefits and drawbacks of such involvement?
7. Do you personalize the intervention content? If so, based on what characteristics? Can you provide an example of personalization and its benefits or drawbacks?
(Show prompts)
II. Current Use of Intervention Techniques:
8. Have you used any techniques in intervention? If yes, can you briefly describe one of the most effective or memorable intervention processes and the feedback and experiences of the participants? How was its effectiveness evaluated?
9. How do you like the use of technology in memory training for older adults with cognitive impairments? What are the potential benefits, challenges, and suggestions for improvement?
10. Do you think this application is effective? Why or why not? (In terms of effectiveness in memory training outcomes/reminiscence, emotional wellbeing, social interactions, etc.)
11. How do you think the effectiveness of technological involvement in non-pharmacological interventions should be assessed?
12. What do you think are the current issues with this application, and how could it be improved?
13. Would you recommend this application to older adults or their families? Why or why not?
III. Expert Perspective on Technology Use in Non-Pharmacological Interventions:
14. What is your attitude towards technology used in non-pharmacological interventions? (Is it a useful method/effective tool? What factors influence its usefulness/effectiveness? Suggested evaluation methods, etc.)
15. Would you be willing to use it? Would you be willing to recommend it to family members? Why?
IV. Usage Scenarios and Processes:
16. How can this type of tool be used? (scenarios, processes, how to use)
17. What potential issues do you think this type of tool might face? What are possible solutions?
18. Who do you think this type of tool is suitable for in dementia cases? Can older adults or their families accept/understand it? (In terms of software operation and hardware devices)
19. If we are planning to develop a tool to support non-pharmacological interventions for people with dementia, what suggestions do you have?
20. Do you have any specific requirements or expectations for such tools?
B Non-Pharmacological Intervention Methods Mentioned by Therapists
Table S1:
Types
Descriptions
Therapists who use the intervention
Music therapy
Music therapy (MT) offers a promising non-pharmacologic solution for addressing BPSD by systematically utilizing musical instruments to enhance communication, with a growing focus on managing symptoms like agitation and aggressiveness [73].
P2, P3, P5, P10
Art Therapy
Art therapy is a way that utilizes art to help people with dementia express and communicate emotions and thoughts, benefiting individuals with dementia by promoting personal growth and well-being through creative expression [24].
P6, P10
Reminiscence Therapy
Reminiscence therapy is a way to encourage individuals to recall past events and experiences, using memory triggers like household items, objects from the past, photos, and music [84].
P5, P6, P7
Horticulture Therapy
Horticultural therapy involves engaging participants in horticultural activities under the guidance of a registered therapist to attain specific goals within a structured treatment, rehabilitation, or vocational program [6].
P2, P3
Aromatherapy
Aromatherapy involves the use of natural essential oils derived from fragrant plants such as peppermint, sweet marjoram, and rose, to alleviate health concerns and improve one’s overall well-being [32].
P3, P8
Pet Intervention (Animal-assisted Intervention)
Pet Intervention (Animal-assisted intervention) usually includes the interaction between PwD and a specifically trained animal under the supervision of a human handler, to relieve symptoms and enhance social engagement [60].
P5
Reality Orientation
Reality orientation is a technique to improve the quality of life of confused elderly people through the presentation of repeated and meaningful time-place-person orientation stimulation [71].
P2, P3
Exercise Therapy
Exercise-based interventions are systematic approaches that involve engaging in regular physical movements according to a routine recommended by a trained therapist, which may encompass a variety of programs, such as walking, dancing, Yoga, to martial arts [61].
P3, P6, P10
Occupational Therapy
Occupational therapy focuses on supporting PwD to perform everyday activities and enhancing independence and participation in social activities [35].
P7
Palm Physical Therapy
Palm Physical Therapy uses a "hands on" treatment for many conditions, usually as a result of muscle dysfunction such as minor soft tissue injuries that our bodies experience due to our occupations and lifestyle.
P2
Table S1: Non-pharmacological Intervention Methods Mentioned by Therapists
WeChat official account, embedded WeChat APP, is a venue for organizations and businesses to connect with their target audience. They can post their stories, information, notifications etc., or redirect followers to a website or e-commerce.
Lucy A. Sparrow, Martin Gibbs, and Michael Arnold. 2021. The ethics of multiplayer game design and community management: industry perspectives and challenges. In Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems. 1–13.
Hojjat Abdollahi, Ali Mollahosseini, Josh T Lane, and Mohammad H Mahoor. 2017. A pilot study on using an intelligent life-like robot as a companion for elderly individuals with dementia and depression. In 2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids). IEEE, 541–546.
Soraia M Alarcão, André Santana, Carolina Maruta, and Manuel J Fonseca. 2022. Developing Assistive Technology to Support Reminiscence Therapy: A User-Centered Study to Identify Caregivers’ Needs. arXiv preprint arXiv:2201.02418 (2022).
Luis Duarte Andrade Ferreira, Sofia Cavaco, and Sergi Bermúdez i Badia. 2018. Feasibility study of an augmented reality system for people with dementia. (2018).
Lora Appel, Suad Ali, Tanya Narag, Krystyna Mozeson, Zain Pasat, Ani Orchanian-Cheff, and Jennifer L Campos. 2021. Virtual reality to promote wellbeing in persons with dementia: A scoping review. Journal of rehabilitation and assistive technologies engineering 8 (2021), 20556683211053952.
American Horticulture Therapy Association 2017. American horticultural therapy association: Definitions and positions paper. Seattle, WA: American Horticultural Therapy Association (2017).
Anna N Baglione, Michael Paul Clemens, Juan F Maestre, Aehong Min, Luke Dahl, and Patrick C Shih. 2021. Understanding the technological practices and needs of music therapists. Proceedings of the ACM on Human-Computer Interaction 5, CSCW1 (2021), 1–25.
Alex Bahar-Fuchs, Anthony Martyr, Anita MY Goh, Julieta Sabates, and Linda Clare. 2019. Cognitive training for people with mild to moderate dementia. Cochrane Database of Systematic Reviews3 (2019).
Clive G Ballard, John O’Brien, Ian James, and Alan Swann. 2003. Dementia: management of behavioural and psychological symptoms. Nordic Journal of Psychiatry 57, 2 (2003), 159–160.
Leya Breanna Baltaxe-Admony, Tom Hope, Kentaro Watanabe, Mircea Teodorescu, Sri Kurniawan, and Takuichi Nishimura. 2018. Exploring the creation of useful interfaces for music therapists. In Proceedings of the Audio Mostly 2018 on Sound in Immersion and Emotion. 1–7.
Michael W Bentley, Rohan Kerr, Margaret Ginger, and Jacob Karagoz. 2019. Behavioural change in primary care professionals undertaking online education in dementia care in general practice. Australian journal of primary health 25, 3 (2019), 244–249.
Grigore Burdea, Kevin Polistico, Amalan Krishnamoorthy, Gregory House, Dario Rethage, Jasdeep Hundal, Frank Damiani, and Simcha Pollack. 2015. Feasibility study of the BrightBrainer™ integrative cognitive rehabilitation system for elderly with dementia. Disability and rehabilitation: Assistive technology 10, 5 (2015), 421–432.
Mariona Carós, Maite Garolera, Petia Radeva, and Xavier Giro-i Nieto. 2020. Automatic reminiscence therapy for dementia. In Proceedings of the 2020 International Conference on Multimedia Retrieval. 383–387.
Romina Carrasco, Felicity A. Baker, Anna A. Bukowska, Imogen N. Clark, Libby M. Flynn, Kate McMahon, Helen Odell-Miller, Karette Stensaeth, Jeanette Tamplin, Tanara Vieira Sousa, 2020. Empowering caregivers of people living with dementia to use music therapeutically at home: design opportunities. In Proceedings of the 32nd Australian Conference on Human-Computer Interaction. 198–209.
Ruiqi Chen, Shuhe Wang, Xuhai Xu, Lan Wei, Yuling Sun, and Xin Tong. 2023. Design and Evaluation of a VR Therapy for Patients with Mild Cognitive Impairment and Dementia: Perspectives from Patients and Stakeholders. In 2023 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops (VRW). IEEE, 597–598.
Daphne Sze Ki Cheung, Shuk Kwan Tang, Ken Hok Man Ho, Cindy Jones, Mimi Mun Yee Tse, Rick Yiu Cho Kwan, Kit Ying Chan, and Vico Chung Lim Chiang. 2021. Strategies to engage people with dementia and their informal caregivers in dyadic intervention: a scoping review. Geriatric Nursing 42, 2 (2021), 412–420.
Malgorzata Ciesielska, Katarzyna W Boström, and Magnus Öhlander. 2018. Observation methods. Qualitative Methodologies in Organization Studies: Volume II: Methods and Possibilities (2018), 33–52.
Peter Craig, Paul Dieppe, Sally Macintyre, Susan Michie, Irwin Nazareth, and Mark Petticrew. 2008. Developing and evaluating complex interventions: the new Medical Research Council guidance. Bmj 337 (2008).
Dagoberto Cruz-Sandoval, Arturo Morales-Tellez, Eduardo Benitez Sandoval, and Jesus Favela. 2020. A social robot as therapy facilitator in interventions to deal with dementia-related behavioral symptoms. In Proceedings of the 2020 ACM/IEEE international conference on human-robot interaction. 161–169.
Jiamin Dai and Karyn Moffatt. 2020. Making space for social sharing: Insights from a community-based social group for people with dementia. In Proceedings of the 2020 CHI conference on human factors in computing systems. 1–13.
Emma Dixon and Amanda Lazar. 2020. Approach matters: Linking practitioner approaches to technology design for people with dementia. In Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems. 1–15.
Simon Douglas, Ian James, and Clive Ballard. 2004. Non-pharmacological interventions in dementia. Advances in psychiatric treatment 10, 3 (2004), 171–177.
James Edmeads and Oussama Metatla. 2019. Designing for reminiscence with people with dementia. In Extended Abstracts of the 2019 CHI Conference on Human Factors in Computing Systems. 1–6.
Emily Fisher, Danielle Proctor, Luke Perkins, Cerne Felstead, Joshua Stott, and Aimee Spector. 2023. Is Virtual Cognitive Stimulation Therapy the Future for People with Dementia? An Audit of UK NHS Memory Clinics During the COVID-19 Pandemic. Journal of technology in behavioral science (2023), 1–8.
Aisling Flynn, David Healy, Marguerite Barry, Attracta Brennan, Sam Redfern, Catherine Houghton, and Dympna Casey. 2022. Key Stakeholders’ Experiences and Perceptions of Virtual Reality for Older Adults Living With Dementia: Systematic Review and Thematic Synthesis. JMIR Serious Games 10, 4 (2022), e37228.
Sarah Foley, John McCarthy, and Nadia Pantidi. 2019. The struggle for recognition in advanced dementia: Implications for experience-centered design. ACM Transactions on Computer-Human Interaction (TOCHI) 26, 6 (2019), 1–29.
Sarah Foley, Daniel Welsh, Nadia Pantidi, Kellie Morrissey, Tom Nappey, and John McCarthy. 2019. Printer Pals: Experience-centered design to support agency for people with dementia. In Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems. 1–13.
Lene Thorgrimsen Forrester, Nicola Maayan, Martin Orrell, Aimee E Spector, Louise D Buchan, and Karla Soares-Weiser. 2014. Aromatherapy for dementia. Cochrane Database of Systematic Reviews2 (2014).
Joyce Fraker, Helen C Kales, Mary Blazek, Janet Kavanagh, and Laura N Gitlin. 2014. The role of the occupational therapist in the management of neuropsychiatric symptoms of dementia in clinical settings. Occupational therapy in health care 28, 1 (2014), 4–20.
Jesus Gonzalez, Teresa Mayordomo, Marta Torres, Alicia Sales, and Juan C Meléndez. 2015. Reminiscence and dementia: a therapeutic intervention. International Psychogeriatrics 27, 10 (2015), 1731–1737.
Maud JL Graff, Myrra JM Vernooij-Dassen, Jana Zajec, Marcel GM Olde-Rikkert, Willibrord HL Hoefnagels, and Joost Dekker. 2006. How can occupational therapy improve the daily performance and communication of an older patient with dementia and his primary caregiver? A case study. dementia 5, 4 (2006), 503–532.
Anis Hashim, Riaza Mohd. Rias, and Muhamad Fairus Kamaruzaman. 2013. The use of personalized digital memory book as a reminiscence therapy for Alzheimer’s disease (AD) patients. In Advances in Visual Informatics: Third International Visual Informatics Conference, IVIC 2013, Selangor, Malaysia, November 13-15, 2013. Proceedings 3. Springer, 508–515.
Anis Hasliza Abu Hashim, Annas Najhan Ismail, Riaza Mohd Rias, and Azlinah Mohamed. 2015. The development of an individualized digital memory book for Alzheimer’s Disease patient: A case study. In 2015 International Symposium on Technology Management and Emerging Technologies (ISTMET). IEEE, 227–232.
Liliana Vanessa Lúcio Henriques, Marilia de Assunção Rodrigues Ferreira Dourado, Rosa Cândida Carvalho Pereira de Melo, and Luiza Hiromi Tanaka. 2019. Implementation of the Humanitude Care Methodology: contribution to the quality of health care. Revista Latino-Americana de Enfermagem 27 (2019).
John V Hobday, Joseph E Gaugler, and Mary S Mittelman. 2017. Feasibility and utility of online dementia care training for hospital staff: The CARES® Dementia-Friendly Hospital™ Program. Research in Gerontological Nursing 10, 2 (2017), 58–65.
James Hodge and Kellie Morrissey. 2020. Sharing a virtual world with people living with dementia. HCI and Design in the Context of Dementia (2020), 237–253.
Maarten Houben, Rens Brankaert, Saskia Bakker, Gail Kenning, Inge Bongers, and Berry Eggen. 2020. The role of everyday sounds in advanced dementia care. In Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems. 1–14.
Maarten Houben, Rens Brankaert, Emma Dhaeze, Gail Kenning, Inge Bongers, and Berry Eggen. 2022. Enriching everyday lived experiences in dementia care. In Sixteenth International Conference on Tangible, Embedded, and Embodied Interaction. 1–13.
Maarten Houben, Rens Brankaert, Gail Kenning, Inge Bongers, and Berry Eggen. 2022. Designing for everyday sounds at home with people with dementia and their partners. In Proceedings of the 2022 CHI Conference on Human Factors in Computing Systems. 1–15.
Lee Hyer, Ciera Scott, Mary Michael Atkinson, Christine M Mullen, Anna Lee, Aaron Johnson, and Laura C Mckenzie. 2016. Cognitive training program to improve working memory in older adults with MCI. Clinical Gerontologist 39, 5 (2016), 410–427.
Wijnand IJsselsteijn, Ans Tummers-Heemels, and Rens Brankaert. 2020. Warm technology: a novel perspective on design for and with people living with dementia. HCI and Design in the Context of Dementia (2020), 33–47.
Jittrapol Intarasirisawat, Chee Siang Ang, Christos Efstratiou, Luke William Feidhlim Dickens, and Rupert Page. 2019. Exploring the touch and motion features in game-based cognitive assessments. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies 3, 3 (2019), 1–25.
Eider Irazoki, Leslie María Contreras-Somoza, José Miguel Toribio-Guzmán, Cristina Jenaro-Río, HenriëTte Van der Roest, and Manuel A Franco-Martín. 2020. Technologies for cognitive training and cognitive rehabilitation for people with mild cognitive impairment and dementia. A systematic review. Frontiers in psychology 11 (2020), 648.
Phil Joddrell, Arlene J Astell, 2016. Studies involving people with dementia and touchscreen technology: a literature review. JMIR rehabilitation and assistive technologies 3, 2 (2016), e5788.
Cees Jonker, Mirjam I Geerlings, and Ben Schmand. 2000. Are memory complaints predictive for dementia? A review of clinical and population-based studies. International journal of geriatric psychiatry 15, 11 (2000), 983–991.
Sukran Karaosmanoglu, Sebastian Rings, Lucie Kruse, Christian Stein, and Frank Steinicke. 2021. Lessons learned from a human-centered design of an immersive exergame for people with Dementia. Proceedings of the ACM on Human-Computer Interaction 5, CHI PLAY (2021), 1–27.
Minmin Leng, Yajie Zhao, and Zhiwen Wang. 2020. Comparative efficacy of non-pharmacological interventions on agitation in people with dementia: a systematic review and Bayesian network meta-analysis. International Journal of Nursing Studies 102 (2020), 103489.
Lori Letts, Mary Edwards, Julie Berenyi, Kathy Moros, Colleen O’Neill, Colleen O’Toole, and Colleen McGrath. 2011. Using occupations to improve quality of life, health and wellness, and client and caregiver satisfaction for people with Alzheimer’s disease and related dementias. The American Journal of Occupational Therapy 65, 5 (2011), 497–504.
Hanuma Teja Maddali, Emma Dixon, Alisha Pradhan, and Amanda Lazar. 2022. Investigating the potential of artificial intelligence powered interfaces to support different types of memory for people with dementia. In CHI Conference on Human Factors in Computing Systems Extended Abstracts. 1–7.
Rupert McShane, Janet Keene, Kathy Gedling, Christopher Fairburn, Robin Jacoby, and Tony Hope. 1997. Do neuroleptic drugs hasten cognitive decline in dementia? Prospective study with necropsy follow up. Bmj 314, 7076 (1997), 266.
Claudia Meyer and Fleur O’Keefe. 2020. Non-pharmacological interventions for people with dementia: A review of reviews. Dementia 19, 6 (2020), 1927–1954.
Anne Moehead, Kathryn DeSouza, Karen Walsh, and Sabrina W Pit. 2020. A web-based dementia education program and its application to an Australian web-based dementia care competency and training network: integrative systematic review. Journal of medical Internet research 22, 1 (2020), e16808.
Kevin Muirhead, Leah Macaden, Keith Smyth, Colin Chandler, Charlotte Clarke, Rob Polson, and Chris O’Malley. 2021. Establishing the effectiveness of technology-enabled dementia education for health and social care practitioners: a systematic review. Systematic Reviews 10 (2021), 1–26.
Lena Nordgren and Gabriella Engström. 2014. Animal-assisted intervention in dementia: Effects on quality of life. Clinical nursing research 23, 1 (2014), 7–19.
Akin Ojagbemi and Nnenna Akin-Ojagbemi. 2019. Exercise and quality of life in dementia: A systematic review and meta-analysis of randomized controlled trials. Journal of Applied Gerontology 38, 1 (2019), 27–48.
Harleen Kaur Rai, Justine Schneider, and Martin Orrell. 2020. An Individual Cognitive Stimulation Therapy App for people with dementia: development and usability study of thinkability. JMIR aging 3, 2 (2020), e17105.
Nicholas V Resciniti, Weizhou Tang, Masroora Tabassum, Joseph Lee Pearson, Sharon Melinda Spencer, Matthew C Lohman, Diane K Ehlers, Dana Al-Hasan, Maggi C Miller, Ana Teixeira, 2020. Knowledge evaluation instruments for dementia caregiver education programs: A scoping review. Geriatrics & gerontology international 20, 5 (2020), 397–413.
Juleen Rodakowski, Ester Saghafi, Meryl A Butters, and Elizabeth R Skidmore. 2015. Non-pharmacological interventions for adults with mild cognitive impairment and early stage dementia: An updated scoping review. Molecular aspects of medicine 43 (2015), 38–53.
Yvon Ruitenburg, Rens Brankaert, Maarten Houben, Minha Lee, and Gert Pasman. 2023. Happje: Stimulating Collaborative Cooking for People with Dementia. In Extended Abstracts of the 2023 CHI Conference on Human Factors in Computing Systems. 1–9.
Yvon Ruitenburg, Gert Pasman, and Rens Brankaert. 2022. One Step at a Time: Evaluation of a Step-By-Step Recipe Tool Designed for People with Dementia. In Dementia Lab Conference. Springer, 77–92.
Ximing Shen, Yun Suen Pai, Dai Kiuchi, Kehan Bao, Tomomi Aoki, Hikari Meguro, Kanoko Oishi, Ziyue Wang, Sohei Wakisaka, and Kouta Minamizawa. 2023. Dementia Eyes: Co-Design and Evaluation of a Dementia Education Augmented Reality Experience for Medical Workers. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems. 1–18.
Aimee E Spector, Martin Orrell, Stephen P Davies, Bob Woods, Cochrane Dementia, and Cognitive Improvement Group. 1996. Reality orientation for dementia. Cochrane Database of Systematic Reviews 2005, 3 (1996).
Sandra Suijkerbuijk, Henk Herman Nap, Wijnand A Ijsselsteijn, Mirella MN Minkman, and Yvonne AW de Kort. 2023. ‘I already forgot half of it’–Interviewing people with dementia for co-designing an intelligent system. Human–Computer Interaction (2023), 1–32.
Oksfriani Jufri Sumampouw, Jeini Esther Nelwan, and Adisti Aldegonda Rumayar. 2019. Outline Images Download Cite Share Favorites Permissions ORIGINAL ARTICLE Socioeconomic Factors Associated with Diarrhea among Under-Five Children in Manado Coastal Area, Indonesia. (2019).
Yuling Sun, Xianghua Ding, Silvia Lindtner, Tun Lu, and Ning Gu. 2014. Being senior and ICT: A study of seniors using ICT in China. In Proceedings of the SIGCHI Conference on Human factors in Computing Systems. 3933–3942.
Yuling Sun, Xiaojuan Ma, Silvia Lindtner, and Liang He. 2023. Care Workers’ Wellbeing in Data-Driven Healthcare Workplace: Identity, Agency, and Social Justice. Proceedings of the ACM on Human-Computer Interaction 7, CSCW2 (2023), 1–29.
Yuling Sun, Xiaojuan Ma, Silvia Lindtner, and Liang He. 2023. Data Work of Frontline Care Workers: Practices, Problems, and Opportunities in the Context of Data-Driven Long-Term Care. Proceedings of the ACM on Human-Computer Interaction 7, CSCW1 (2023), 1–28.
Claire A Surr and Cara Gates. 2017. What works in delivering dementia education or training to hospital staff? A critical synthesis of the evidence. International journal of nursing studies 75 (2017), 172–188.
Claire A Surr, Cara Gates, Donna Irving, Jan Oyebode, Sarah Jane Smith, Sahdia Parveen, Michelle Drury, and Alison Dennison. 2017. Effective dementia education and training for the health and social care workforce: a systematic review of the literature. Review of educational research 87, 5 (2017), 966–1002.
Theresa Thoma-Lürken, Michel HC Bleijlevens, Monique AS Lexis, Luc P de Witte, and Jan PH Hamers. 2018. Facilitating aging in place: A qualitative study of practical problems preventing people with dementia from living at home. Geriatric Nursing 39, 1 (2018), 29–38.
Myrte Thoolen, Rens Brankaert, and Yuan Lu. 2020. AmbientEcho: exploring interactive media experiences in the context of residential dementia care. In Proceedings of the 2020 ACM Designing Interactive Systems Conference. 1495–1508.
Maho Tominari, Ryuji Uozumi, Carl Becker, and Ayae Kinoshita. 2021. Reminiscence therapy using virtual reality technology affects cognitive function and subjective well-being in older adults with dementia. Cogent Psychology 8, 1 (2021), 1968991.
Daniel Welsh, Kellie Morrissey, Sarah Foley, Roisin McNaney, Christos Salis, John McCarthy, and John Vines. 2018. Ticket to talk: Supporting conversation between young people and people with dementia through digital media. In proceedings of the 2018 CHI conference on human factors in computing systems. 1–14.
Robert T Woods, E Bruce, RT Edwards, R Elvish, Z Hoare, B Hounsome, J Keady, ED Moniz-Cook, V Orgeta, M Orrell, 2012. REMCARE: Reminiscence groups for people with dementia and their family caregivers-Effectiveness and costeffectiveness pragmatic multicentre randomised trial. Health Technology Assessment 16, 48 (2012).
Ryuji Yamazaki, Masahiro Kochi, Weiran Zhu, and Hiroko Kase. 2018. A pilot study of robot reminiscence in dementia care. International Journal of Biomedical and Biological Engineering 12, 6 (2018), 257–261.
Fang Yu, Michelle A Mathiason, Kari Johnson, Joseph E Gaugler, and Dan Klassen. 2019. Memory matters in dementia: Efficacy of a mobile reminiscing therapy app. Alzheimer’s & Dementia: Translational Research & Clinical Interventions 5 (2019), 644–651.
Natasha Yuill and Vivien Hollis. 2011. A systematic review of cognitive stimulation therapy for older adults with mild to moderate dementia: an occupational therapy perspective. Occupational therapy international 18, 4 (2011), 163–186.
Yayi Zhao, Li Liu, and Helen Yue-lai Chan. 2021. Dementia care education interventions on healthcare providers’ outcomes in the nursing home setting: A systematic review. Research in Nursing & Health 44, 6 (2021), 891–905.
Chen Liang Yuru Huang Zisu Li, Li Feng and Mingming Fan. 2023. Exploring the Opportunities of AR for Enriching Storytelling with Family Photos between Grandparents and Grandchildren.Proc. ACM Interact. Mob. Wearable Ubiquitous Technol 7, 3 (2023), Article 108. https://doi.org/10.48550/arXiv.2309.03533
Chiara Zucchella, Elena Sinforiani, Stefano Tamburin, Angela Federico, Elisa Mantovani, Sara Bernini, Roberto Casale, and Michelangelo Bartolo. 2018. The multidisciplinary approach to Alzheimer’s disease and dementia. A narrative review of non-pharmacological treatment. Frontiers in neurology 9 (2018), 1058.
Technology-Mediated Non-pharmacological Interventions for Dementia: Needs for and Challenges in Professional, Personalized and Multi-Stakeholder Collaborative Interventions
Human Aspects of IT for the Aged Population. Design, Interaction and Technology Acceptance
Abstract
World Health Organization and local governments recommended that older adults self-isolate due to the elevated risk for adverse health outcomes faced when contracting COVD-19. Technology offers better access to virtual communications for social ...
DIS' 20 Companion: Companion Publication of the 2020 ACM Designing Interactive Systems Conference
Research in design and HCI is investigating the role of design interventions in adding value to the quality of life of people with dementia. However, the sustainable use of design interventions in care environments remains challenging due to the complex ...
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected].
Yuling Sun, 
Xin TongAuthors Info & Claims
Best Paper