Dementia is an umbrella term for a variety of neurodegenerative diseases that most often affect memory, behavior, and communicative abilities [1]. There are approximately 47 million people living with dementia worldwide [2]. With this number set to increase to 131.5 million by 2050, it is of the utmost importance to tackle dementia’s progressive impact on the well-being of people living with a diagnosis. The World Health Organization (WHO) has called for action on dementia, presenting it as a public health priority at a global level [1]. This action includes a call for research to identify ways of supporting the needs of people living with dementia, their caregivers, and society in the context of costs, understanding, and awareness.

People with dementia (PWDs) progressively disconnect from the world; they experience loss of function, especially memory, affecting their cognition, physical activity, and verbal and nonverbal communication. A person’s ability to communicate with others progressively worsens during the course of dementia, which can lead to the individual engaging in problem behaviors as an expression of unmet needs [3,4]. This can then result in an increase in caregiver burden for family members and residential care staff.

Continuing social contact with others and participating in activities is important for maintaining quality of life. Participating in past pleasant activities has an impact on functional ability and psychological well-being for PWDs living in residential care [5]. However, care facilities often struggle to fulfil this need for engagement and active participation, especially for residents in more advanced stages of cognitive and physical impairment [6]. This lack of external stimuli in the care environment can cause PWDs to become increasingly depressed [7].

In recent years, studies have identified numerous complex needs of PWDs living in long-term care. These include management of challenging behaviors, maintenance of social relationships, involvement of people with cognitive deficits in meaningful activities, and supporting the emotional needs of all [8,9]. Emotion-oriented approaches to care have been shown to be cost-effective ways of improving psychological well-being and social behavior among PWDs [10,11]. These nonpharmacological approaches are often person-centered, focusing on the personal, social, and emotional needs of the individual. Reminiscence rooms, virtual gardens, and virtual reality forests are examples of how immersive technologies have been integrated in emotion-oriented approaches designed to create effective interventions for PWDs [12,13]. However, this area of study has called for further research in determining what works best for the individual [6]. It has recently been suggested that an individualized multisensory environment for PWDs would be a highly beneficial intervention, especially if family members are included in the selection of stimuli [14]. Our research builds on this suggestion, creating not only a personalized multisensory intervention but one that also incorporates immersive technology, all with the inclusion of family members, friends, and professional care staff.

The study is performed as part of the SENSE-GARDEN EU project, funded by the Active Assisted Living (AAL) Program Call 2016. It is a 3-year project that brings together a consortium of partners across Belgium, Norway, Portugal, and Romania. This multidisciplinary project is embracing a user-centered design approach throughout the development and implementation of the intervention. The SENSE-GARDEN intervention addresses the need for individualized approaches to dementia care by creating a multisensory environment that automatically adapts to the individual with dementia. Through integrating autobiographical music, films, images, and scent with technology, the SENSE-GARDEN is able to offer an immersive experience tailored specifically for the individual based on an individual profile including preferences in music, images, videos, and personal media, such as family photos.

The project aims at creating virtual spaces that are automatically adaptable to the personal memories and individual preferences of the users. The design of the space is shown in Figure 1. These spaces will be designed to strengthen the awareness of older PWDs by combining multisensory stimulation with physical activity and techniques from reminiscence therapy and Montessori methods. This stimulation of sight, touch, hearing, balance, and smell is expected to lead to a reconnection with reality for the PWD, resulting in an improvement in overall well-being and quality of life. The intervention will use a combination of various activities and approaches:

The rationale to include these elements in the SENSE-GARDEN intervention stems from the current evidence base of ways to improve the well-being of PWDs. For instance, there has been an increasing amount of research on combining biographical information with multimedia apps to create digital life stories. Research on multimedia biography apps for PWDs has shown numerous benefits, such as stimulating reminiscence, evoking positive emotions, stimulating social interaction with others, and improving autobiographical memory [15-17].

Research also suggests that there is potential to use technology to create immersive environments in dementia care. A recent study investigated the effects of a virtual reality forest on PWDs, in which large screen projection of scenery such as a forest or river was used in combination with movement sensors to create an immersive and interactive environment [13]. Although this environment was not based on the biographical information of the user, results showed improved levels of pleasure and alertness during the intervention.

Considering the promising results of these studies, the innovative approach of SENSE-GARDEN has the potential to provide numerous benefits to its users. Together, it was anticipated that the integration of these components and various stimuli would complement each other to create an immersive environment that the PWD can connect with, for example, pictures of forest scenery could be combined with the sound of singing birds and the odor of a pine forest. Other scenarios may include seas, streets, the beach, urban, or rural areas according to the most meaningful memories of each individual PWD. Having this immersive environment tailored to the individual memories of each user will create a connection to the more active areas of the memory; such stimuli may result in various types of emotional states.

Another important aspect of SENSE-GARDEN is the aim to support PWDs in sharing their life story with a caregiver. The theoretical underpinnings of this aim can be linked to previous study conducted on dementia and narrative. The literature has commented on the important role of digital media in conveying narratives and supporting meaningful conversations for PWDs and their caregivers [18]. By interacting with the various stimuli together with a caregiver, it is hoped that the PWD is able to reminisce on their past and become engaged in the present moment. The SENSE-GARDEN encourages PWDs to exercise at both mental and physical levels and takes them back into places they feel connected to. They can, for example, cycle or walk in a well-known space and feel like they are going home. Such experiences may have an effect on invigorating their identity and helping to recover their sense of self.

Relatives will have a key role in the initial adaptation of the space by providing information regarding the past of the user. After this initial setting, the SENSE-GARDEN will automatically adjust to the individual person. Feedback during the sessions will allow the SENSE-GARDEN system to learn the preferences of each user, meaning that the sessions will become increasingly personalized with each visit.

The study will test operational SENSE-GARDENs installed in the 3 study sites: Belgium, Norway, and Portugal. With these countries, a large coverage of the European context is achieved, allowing to study cultural, social, economic, and legal differences between the countries and European regions. The primary objective of this study is to determine whether the delivery of the innovative technology developed in SENSE-GARDEN can improve well-being in older adults with intermediate to advanced dementia. The WHO defines health as the “state of complete physical, mental, and social well-being and not merely the absence of disease or infirmity” [19]. This study will adopt this holistic approach to health and well-being by measuring well-being through primary outcomes related to behavior, activities of daily living, cognitive function, quality of life, self-identity, communication, and social presence.

The study will examine 2 quantitative and 2 qualitative primary outcomes for the PWD, namely, reduction in behavioral and psychological symptoms in dementia (BPSD), improvement in cognitive function, increase in the feeling of social presence, and an increase in self-awareness and engagement. The primary outcomes for the informal caregiver (family caregivers) are a reduction in carer burden, improvement in carer coping and relief from stress, an increased quality of visits from the informal caregivers to the PWD, and an improvement in the quality of the relationship with their relative with dementia. The primary outcome for the professional caregiver is a reduction in caregiver burnout.

Secondary outcomes for the PWD are changes in the prescription of medication, International Classification of Functioning, Disability, and Health (ICF) scores, mortality rate compared with previous existing records, number of hospitalizations comparing control and intervention groups, and an improvement in physical function and balance.

Some interesting outcomes will be measured with an exploratory aim, but because of their nature as explained ahead, they may not allow for clear conclusions to be drawn. Instead, the purpose of these outcomes is to provide preliminary insights that can be built upon in future studies. These outcomes are as follows: a reduction in depressive symptoms and loneliness, relief from feelings related to the care burden, and improvement in quality of life.

The outcomes regarding depression and relief from feelings related to care burden are measured by means of semistructured interviews. The inclusion of people in later stages of dementia means that there may be challenges in applying these measures to all participants in the study.

A total of 4 months is a short amount of time, and therefore, it is not expected that a large improvement in the quality of life will be observed during this period. However, a small change in the quality of life may provide rationale for future longitudinal studies of the use of SENSE-GARDEN in dementia care, in which quality of life can be examined to a greater extent.

Physiological data will also be collected during the SENSE-GARDEN sessions. The Empatica E4 [20] will be used to collect information on electrodermal activity (EDA) and heart rate. These will be assessed during SENSE-GARDEN visits, as reaction to different stimuli. Data will be collected from a subgroup of participants, depending on their tolerance to use a wristband-mounted device. The device is validated for clinical use. Previous research on nonpharmacological interventions has used the E4 wristband to measure engagement and arousal states in people with mild to moderate dementia [21].

To ensure that the SENSE-GARDEN meets the needs of the users, the project has adopted a user-centered design approach to the development of the intervention. This user-centered design process is divided into 3 phases. An overview of the phases is shown in Figure 2.

The first phase focuses on collecting an initial impression of the user experience with nonfunctional low-fidelity prototypes of the SENSE-GARDEN (eg, mock-ups). Small groups of users will be invited to each test site to give their feedback on the overall concept of SENSE-GARDEN. These user groups will include professional care staff, family caregivers, and people living with mild cognitive impairment. Including these individuals at an early stage of the project will ensure that their views are incorporated into the development of SENSE-GARDEN.

The second phase focuses on creating experiences for individual users and aims at gaining a deeper understanding of the users’ needs and requirements. An Alpha version of the SENSE-GARDEN system will be available for this phase of testing. This initial prototype of the SENSE-GARDEN will be tested in a controlled environment. Following the test, a semistructured interview will be conducted with the user to gain a rich insight into their experience.

The third phase involves feasibility testing of the system in the form of pretrials. A small number of users will be recruited at each site (1-2 users). The SENSE-GARDEN test will be tailored to each individual user by using personalized media content, such as family photographs, videos from holidays, and favorite music. The purpose of these tests is to not only ensure that the system works but also define the process of creating individualized experiences for each user.

For the full trials, a controlled before-after study design will be used. The timeline for the study is shown in Figure 3. All participants eligible for the study will receive the SENSE-GARDEN intervention for 3 months. For the final month of the study, half of the participants (the intervention group) will continue to receive the SENSE-GARDEN intervention. The other half of the participants (the control group) will discontinue visits to the SENSE-GARDEN, receiving only normal care. The randomization procedure will use sealed envelopes and will be performed by the local researcher at each test site. The total number of envelopes will be prepared before the recruitment period. A sealed envelope will be assigned to each participant at the time of her or his enrolment. This randomization is to determine whether visits to the SENSE-GARDEN have lasting effects.

The study will start in August 2019, with a recruiting period of 6 months. The intervention period for each participant will be 3 months, followed by 1 month in either the intervention or control condition. The total study period will be 10 months.

There will be 3 sites in 3 trial countries: Belgium, Norway, and Portugal. With these countries, a large coverage of the European context is achieved, allowing to study cultural, social, economic, and legal differences between the countries and European regions.

The test site in Belgium will be Aan de Beverdijk care home based in Hamont-Achel. This care home is part of the VULPIA group, a care provider that comprises 22 elderly care homes. Aan de Beverdijk focuses on providing tailored care based on the individual needs of its 90 residents. The test site in Norway will be Odda sjukeheim, a municipality-based care home for the elderly based in the center of Hardanger. The test site in Portugal will be the Lar Santa Joana Princesa care home for the elderly. This care home is a part of the Santa Casa de Misericórdia de Lisboa (SCML). SCML operates according to a humanitarian goal, and its care homes focus on promoting the quality of life of its residents.

Although it is inevitable that there will be differences among care homes, the SENSE-GARDEN space at each site is being custom-built for the specific purposes of this project. Each SENSE-GARDEN will have the same equipment, the same software and will be in a closed, private room. Having a controlled space at each test site will help reduce the amount of variability among the 3 countries.

This study will recruit a total of 55 PWDs living in care facilities, in a multisite trial with 3 sites. Details for each site are as follows: (1) Belgium—1 study site, 25 PWDs, 25 informal caregivers, and 3 professional caregivers, (2) Norway—1 study site, 15 PWDs, 15 informal caregivers, and 3 professional caregivers, and (3) Portugal—1 study site, 15 PWDs, 15 informal caregivers, and 3 professional caregivers.

Participants will be recruited at each study site by a professional caregiver from the SENSE-GARDEN team. The caregivers will inform potential participants about the study with the use of an information leaflet about the project. The potential participants who are interested in the study can then voluntarily express their willingness to participate. The participants who do not express an interest will not be contacted by the project team.

Professional caregivers included in the study will have at least 6 months experience in caring for dementia patients and a background in either nursing, occupational therapy, or physiotherapy or another relevant background when supported by long-term professional experience as a caregiver.

Before the study, the professional caregiver will have a 1-week training in SENSE-GARDEN use. This training comprises guidance on how to collect information from the participant and/or their family members for creating a user profile, upload media contents to the SENSE-GARDEN system, create workflows of media contents for individual SENSE-GARDEN sessions, and control the numerous elements in the SENSE-GARDEN space.

Each potential participant will undergo an assessment by a health professional for the evaluation of inclusion and exclusion criteria. For the participants fulfilling the criteria, a randomization to the intervention or control group will be performed.

During the 3 months of the intervention period, the PWD will visit the SENSE-GARDEN on an average of 3 times per week. Each participant should visit the SENSE-GARDEN a minimum of 25 times during the intervention period. Any participant who visits less than 25 times will be considered a dropout. On each visit to the SENSE-GARDEN, the PWD is accompanied by 1 caregiver (formal or informal). It is desirable that approximately one-third of the visits are performed with an informal caregiver.

Logs of interaction with the system will be collected continuously, including visit time and duration and system feedback given by the caregivers (through a tablet interface).

Observations will be made at 3 of the SENSE-GARDEN sessions during the intervention period. During this time, measurements on social presence and physical activity will be collected. Video recording will be required during the 3 observation sessions. The recordings will solely be used for data analysis and will not be shared. They will be accessible only to authorized researchers in charge of data analysis. Video recordings will only be made during sessions involving participants that have given their informed consent to being recorded.

For the control period, visits from the family will be logged during normal care.

The methods include the following types of data collection: self-reported measures, questionnaires, interviews, and observations. For interviews and questionnaires requiring the participation of the PWD, the capacity of the PWD to provide information will be considered. In the case of participants that are not competent to provide input, the tests will be performed with a proxy who will provide information on the patient’s behalf. In some cases, the presence of both the PWD and proxy will be considered.

Quantitative data will be analyzed using SPSS Statistics version 25 (IMB Corp). Stratification by gender, age, and type and stage of dementia will be used to analyze the data. Repeated measures analysis of variance (ANOVA) will be used to assess whether the SENSE-GARDEN intervention has an effect on participants over time. Given the novelty of the SENSE-GARDEN intervention, it is not possible to assess the previous literature for clinically meaningful differences. However, Cohen d has been used to determine effect size in studies focusing on the effects of nonpharmacological interventions on behavioral outcomes in PWDs [43].

Cohen d is a measure of standardized difference between 2 means [44]. It uses standard deviation units to express the magnitude of difference between the 2 means, indicating the importance of the difference. Cohen has defined a small effect size as d=0.2, a medium effect size as d=0.5, and a large effect size as d=0.8.

For this study, repeated measures ANOVA will be performed on the data using SPSS to assess the effects of the intervention over the numerous time points (T0-T3). We estimate that a sample size of 55 will have over 80% power to detect a medium difference between means at a .05 significance level.

Qualitative interview data will be analyzed using thematic analysis [45]. Thematic analysis is a method of identifying patterns of prevalent ideas or responses and can offer rich insight into the attitudes and beliefs of participants. This aspect of the study will be vital for understanding the users’ experiences of the SENSE-GARDEN.

Applications for ethical approval have been submitted by the 3 study sites in accordance to the national regulation. This study has been submitted to Norway’s Regional Committee for Medical and Health Research Ethics and is currently under assessment (document ID 1094463).

This paper has outlined the objectives, study design, and preliminary results of an AAL-funded study, SENSE-GARDEN. To our knowledge, SENSE-GARDEN is the first intervention to combine multisensory stimuli, virtual technology, and autobiographical material to create an immersive and personalized room for PWDs. Our progress thus far has indicated the value of involving user groups in the initial stages of intervention development and has shown that these users have a positive outlook toward SENSE-GARDEN. The evaluation summary report has been provided in Multimedia Appendix 1.

A limitation of this study proposal is its small sample size. However, although the study may not be able to produce generalizable results, it will provide important insights into the use of a novel technology, such as SENSE-GARDEN for PWDs. The results will provide foundation for further study in the use of immersive, individualized environments and how these environments can be used to facilitate communication and person-centered care in residential care home environments.

A further limitation is the potential variance between the test sites. Implementing this study protocol in 3 different countries may be difficult in terms of ensuring that the SENSE-GARDEN intervention is delivered in the same manner in each care home. Cultural differences between the test sites may mean that methods of facilitation from care home staff could affect the overall SENSE-GARDEN experience, therefore, influencing the results from the study. However, all professional care staff will be given the same training on how to use the SENSE-GARDEN. This training should help to ensure that the intervention is conducted in a similar way at each test site.

Despite these limitations, this study has the potential to provide important contributions to current research on dementia care. The interdisciplinary nature of this project will allow us to evaluate the SENSE-GARDEN from multiple perspectives, such as technical, sociological, and psychological. The findings from the full trials have the potential to offer numerous implications on future research in dementia care and also in promoting person-centered care in practice.