This study is part of a larger user-centered design project on the design of a cognitive orthosis using pure mixed reality to support the independence of people living with NDs. Such an approach generally comprises four steps: exploration, ideation, generation, and evaluation [23,24]. This study reports the exploration phase, which aims to better understand users’ needs, motivations, and attitudes [22].

As various methods can be used to conduct the exploration phase [24], we selected a descriptive inductive qualitative research design to document the main difficulties of older adults with NDs and how assistance can be provided with pure mixed reality [25]. We collected the data through focus groups with OTs and experienced stakeholders to document their perspectives on an example of a mixed reality headset [20]. It has been documented that focus groups are relevant and appropriate for obtaining a detailed portrait of a phenomenon for which little literature exists [26]. The focus group method is particularly appropriate in this context to document the perspective of experts about the potential of using our first prototype to generate ideas because of group synergy [24].

Invitations were sent to several OTs working in various clinical settings in specialized psychogeriatrics and experienced stakeholders of a local Alzheimer association. OTs are health professionals entitled to assess the needs of people living with cognitive impairments, determine the types of interventions that can ensure safety and increase their independence in IADL, and anticipate facilitators and obstacles to the use of new technologies, such as intelligent AT [27]. Experienced stakeholders, such as those involved in associations dedicated to older adults with NDs, are closely involved in the daily life and real environment of the person and are, therefore, able to specify the main difficulties encountered by this population, as well as predict effective interventions that work with older adults in their natural environment. The inclusion criterion was participants with at least 3 years of experience in involvement with older adults with NDs. There were no exclusion criteria for this study. Participants were divided into groups of 3 to 6 in accordance with the guidelines for this method [26].

The Research Ethics Board of the Aging-Neuroimaging Research Ethics Committee of the Centre (Intégré Universitaire de Santé et Services Sociaux–Centre-Sud-de- l’île-de-Montréal) approved the project (CER VN 19-20-28). The participants provided written and informed consent to participate in the study.

Each focus group met once for 1 hour, and all sessions were held over a 3-month period. A semistructured interview guide was used (Textbox 1). Participants were asked to discuss three topics related to (1) the difficulties of older adults with NDs in everyday activities, (2) the effective interventions used to support the independence and safety of this population during meal preparation, and (3) their perspectives on the relevance of using the mixed reality headset with this population. A video describing the different features of the mixed reality headset was presented between topics 2 and 3. A member of the team (AY) acted as a facilitator and was responsible for asking questions and guiding discussions. Another member (GS) acted as an observer, took notes, and validated the discussion content with the group at the end of the discussion on each topic. All group interviews were audiotaped with the consent of each participant to facilitate the data analysis.

To ensure the validity of the data when using this type of method, words and facts were reported as accurately as possible following the focus group sessions, and we pursued data saturation. Data saturation is reached when there is sufficient information to replicate the study and when the ability to obtain additional new information has been attained so that further coding is no longer achievable. We attempted to remain close to our data, the words used, and the events described by recording the sessions and transcribing the entire verbatim [25]. We conducted inductive qualitative analysis in four stages using a thematic analysis approach [28]: (1) full transcription of the audio recordings, (2) first-order coding of transcribed data, (3) second-order coding from the first-order code list, and (4) data reduction and matrix development. To validate the data analysis, the lead author (AY) performed the coding. A list of codes was validated by an OT (PS) and a researcher from the team (NB) until a consensual integrated code list was obtained. The coding aimed to assign labels (codes) to relevant units of meaning, such as words, sentences, or paragraphs. After first-order coding, second-order codes were used to condense the data into different categories, which were then condensed into different major themes (third-order codes). Once the 2-step coding was completed, conceptual grouping matrices based on the major themes were developed to reduce the set of codes to a format that was more manageable and easier to conceptualize. All themes and matrices were validated by PS and NB.

To obtain data saturation, a total of 24 OTs from different clinical settings, including a psychogeriatric-intensive functional rehabilitation unit, long-term care, day hospital, home support, and day center (n=10, 42% of OTs had >10 years of experience; n=14, 58% had 3-10 years of experience in geriatrics) were recruited, as well as 6 experienced stakeholders from a recognized Alzheimer association. They participated in a total of 5 focus groups. Table 1 presents the participant characteristics for each focus group session.

Our first objective was to understand the main difficulties encountered by older adults with NDs in their daily lives to ensure that pure mixed reality meets their needs. The participants identified the main difficulties but also specified the factors influencing these difficulties (Table 2). Factors that influence the level of disability during activity performance were disease severity, social and professional support, and characteristics of the activity (newness, structure, and complexity).

According to the participants, several interventions have been used to improve the independence of older adults with NDs. These interventions targeted the person’s abilities and task requirements or modified the environment to optimize the person’s safety, independence, and quality of life. According to the participants, these interventions had different goals, such as preventing, facilitating, adapting, and compensating, as shown in Table 3.

According to the study participants, prevention interventions mainly comprise upstream interventions to prevent undesirable situations. For example, in the case of institutionalization or delocalization, visiting the new environment and personalizing it before moving in is a relevant intervention for preventing disorientation, as well as involving the family and providing guidance, directions, and reference points. Regarding this, a participant pointed out the following:

Facilitation and simplification interventions decrease the burden of actions by reducing the global complexity of activities. This can be accomplished by targeting the person, task, or environment. The goal of these interventions is to make the activity as simple as possible to make task accomplishment easier. For example, providing physical assistance to initiate the task or providing a model that acts as an example allowing for imitation are interventions that simplify the activity as they allow the person to skip the initiation and/or planification step of the task. For example, a participant said the following:

Providing feedback, verbal assistance, and visual cues are also facilitation and simplification interventions that facilitate task accomplishment as they allow the person to reduce their cognitive burden. According to the participants, using images instead of long sentences or having a list of preprogrammed steps are effective ways of providing facilitating assistance. Regarding this, 7% (2/29) of participants highlighted the importance of such interventions:

Adaptation interventions comprise modifying an element of the activity or the environment to allow the accomplishment of the task. According to the participants, this type of intervention differs from task simplification in that it involves adding elements and steps to the task instead of reducing it. For example, adding familiar elements or highlighting essential information in the environment, using a checklist to help planification, verbalizing the steps of the task when achieving it, using a timer to achieve some activities, or establishing a routine are effective interventions to support independence and safety according to participants. For example, a participant said the following:

Finally, according to the participants, compensation interventions comprise subtracting the steps of a task when elements are already performed by external help. For example, when the caregiver performs a task for them or when they stimulate the person to start the activity, they provide assistance that subtracts the initiation step of the task. Using technologies to ensure safety or using preprepared meals are also compensation interventions as they allow the person to skip steps within the tasks. For example, using a dispill and automatic recall allows a person to skip a few steps of the medication management activity, such as planning and organizing what pills to take. The participants said the following about the technologies and safety:

Overall, participants perceived the mixed reality headset as a technological tool with great potential for older adults with NDs. They identified several facilitators and barriers related to (1) resources and technical aspects, (2) risks and ethical and social considerations, and (3) individual characteristics that may influence the use of cognitive orthosis for this population, as described in Table 5.

The purpose of this study was to describe clinicians’ and experts’ perspectives on the potential of pure mixed reality to support independence and ensure the safety of older adults with NDs in daily life. More specifically, we aimed to document (1) the main difficulties encountered by older adults with NDs in their daily life to ensure that the pure mixed reality meets their needs, (2) the most effective interventions with this population to determine the types of assistance that should be given by the pure mixed reality technology, (3) how the mixed reality headset should provide assistance to respond to clinical purposes of promoting safety and independence at home, and (4) the main facilitators of and barriers to the use of this technology among this population to develop a version ready for laboratory testing. The results suggested the following: (1) the main difficulties encountered by this population are in remembering to complete tasks, initiating the tasks, remembering where they are in a task, and in planning the task; (2) several interventions are used to improve the independence of this population, such as prevention, simplification or facilitation, adaptation, and compensation interventions; (3) the use of pure mixed reality with older adults with NDs to promote independence and safety at home is promising and may respond to 3 clinical functions identified by the participants, including assessment, assistance, and training; and (4) pure mixed reality has good potential for use with this population, with certain facilitators and obstacles, such as resources and technical aspects, risks, ethical and social considerations, and individual characteristics.

Regarding the type of difficulties encountered (objective 1), our study showed that this population faced many challenges in the accomplishment of BADL and IADL, such as eating, moving oneself, washing oneself, preparing meals, managing medication, and managing finances. The main difficulties that this population may face in these activities are difficulties in remembering to complete tasks, initiating the tasks, remembering where they are in a task (what parts are already completed), and planning the task. Difficulties in initiating and planning are part of executive functions, which have been documented as one of the main cognitive functions affecting the dementia continuum [29-31]. Impairment in executive functions puts the person at risk of errors when faced with unexpected situations and when making an appropriate decision, which potentially affects independence and safety. In general, difficulties in remembering are part of the memory components documented to be affected in this population [32]. Memory impairments can affect learning abilities and a person's daily life from simple tasks (bathing and/or moving oneself) to more complex tasks (managing finances and/or preparing meals) [22] Memory and executive function impairments in this population are highly documented in the literature [33-37], and our results confirm the evidence. According to the participants, these difficulties would require specific assistance from the pure mixed reality technology to optimize independence and safety at home. Moreover, according to our results, many factors may influence the severity of difficulties in conducting these activities, such as disease severity, presence of social and professional support, and activity characteristics. Disease severity and activity complexity were shown to be determinant factors in the severity of difficulties that affect independence in IADL and BADL. Indeed, in the earlier stages of the ND continuum, such as MCI or early stages of AD, the difficulties are more inconspicuous, appearing only at the level of very complex activities that require the coordination of several steps. Furthermore, along the continuum, the difficulties are more visible and present in basic activities [38]. In short, pure mixed reality could be useful in the early stages of the disease to overcome memory- and executive function–associated difficulties encountered in the most complex activities of daily life, such as preparing meals and managing medication and finances.

The second objective of this study was to document the effective interventions that work with this population to determine the types of assistance that should be provided by pure mixed reality technology. Our study showed that according to the participants, clinical interventions can be effective when they target the person, activity, and/or environment, which is in accordance with models of occupational rehabilitation [39,40]. According to the participants, effective interventions aim to meet four main objectives: prevention, simplification or facilitation, adaptation, and compensation. Prevention interventions mainly comprise upstream interventions to specifically prevent undesirable situations, such as falling, getting lost, or causing a fire. In the literature, the prevention interventions described are generally aimed at delaying the onset of dementia and cognitive decline or reducing their incidences, such as promoting healthy lifestyle habits and providing education, intellectual stimulation, or early screening [41]. Nevertheless, our results focused on clinical practice in rehabilitation, as we mainly involved rehabilitation experts—OTs (24/29, 83%) and experienced stakeholders (6/29, 21%); thus, in our study, prevention interventions aimed to improve the person’s independence and safety and maintain their residual capacities, such as providing guidance, directions, and reference points to help the person prevent undesirable situations such as getting lost in a new environment. Several studies have shown that the use of this type of intervention allows a person to be more independent in their environment. For example, Spector et al [42] used a reality orientation board that displayed both personal and orientation information to provide some form of continuity for older adults with AD, which has been shown to be effective in preventing the risk of getting lost in the environment and in time. Simplification or facilitation interventions decrease the burden of actions by reducing the global complexity of activities. For example, according to participants, providing a model that acts as an example is a facilitation intervention that can make it easier to accomplish some activities, especially complex activities such as meal preparation. Few interventions of this type have been documented in the literature to support independence in IADL. Rousseau and Métivier [43] proposed an intervention based on imitation for emotional management. Adaptation and compensation interventions, as described in our study, such as using a pillbox or a dispill for medication management, highlighting essential information, stimulating the person to start the activity, using a checklist, verbalizing the steps of the task when achieving the task, or establishing a routine, are interventions that are widely documented in the literature as effective in optimizing independence and safety of older adults with NDs in their BADL and IADL [44-46]. Finally, the interventions proposed by the participants allow a better understanding of what types of assistance would meet the needs identified for this population, which would help computer scientists to better design features of HMD intended to help older adults with NDs accomplish BADL and IADL.

Our third objective was to determine the functions of the mixed reality headset that would be useful in assisting with the difficulties experienced by this population. Our study suggests that according to the participants, the detection, information storage or provision, and interactive features of a mixed reality headset can serve three main clinical functions: assessment, assistance, and training. In assessment, the mixed reality headset can detect or recognize the location and activity of the user and monitor them to provide data to clinicians about performance in daily activities. The use of sensor technologies has been shown to be promising in documenting the daily lives of older adults with dementia [47,48]. It has even been suggested that these technologies could be used to screen for NDs such as MCI or early AD [47]. More specifically, according to the participants, the mixed reality headset would be useful to document information on risky behaviors, routines of the person, and the number of omissions and errors during daily activities without the need for increased user interaction, which is particularly relevant to ZET principles [4,12,14]. Assistance was the clinical function that participants identified as the most useful for the mixed reality headset. Indeed, the mixed reality headset can scan surroundings to detect objects and provide assistance through visual and audio feedback to help the person accomplish their daily tasks. The mixed reality headset can also store information involved in specific daily routines and tasks to help the user perform them by providing audio and visual information such as daily reminders, options, instructions, pictograms, and a list of steps before the task or by interacting with the user during the task. These types of assistance have previously been documented in several other Assistive Technology Center design studies [49,50]; however, their use with older adults with NDs has limitations related to the nature of the proposed technology, as it often does not respect the ZET principles [12].

Finally, our last objective was to document the facilitators of and barriers to the use of the mixed reality headset among older adults living with cognitive impairments. Our study suggests that the mixed reality headset has good potential, with certain facilitators and barriers. Financial costs and maintenance resources were identified as the main barriers to the use of pure mixed reality. The affordability of technology is often identified by clinicians when discussing the use of technology with this population [21,51]. However, the use and early introduction of technology could delay the institutionalization of older adults with NDs for up to 8 months, when the technology is efficient and adapted to the person’s needs [52,53]. Thus, from a long-term perspective, it can be argued that the use of pure mixed reality at a more mature stage could support functional independence and, therefore, aging in place, as well as optimizing health care costs [52,54].

In terms of individual characteristics, some barriers to the optimal use of the mixed reality headset were reported by participants, including the risks of nausea, confusion, and loss of contact with the person's actual environment while wearing the headset. However, it has been documented that there is little to no virtual reality sickness in mixed reality or AR as there is no loss of contact with the real world; thus, it would be important to better inform people about this type of technology for optimized future adoption [55]. Social stigma may also be a barrier to the use of the mixed reality headset according to the participants, as it could be difficult to encourage this population to wear it in its current version. Considering the increasing technological advances in the domain, it will be possible to consider more conventional eyeglasses in the future [56]. In contrast, in earlier studies, the simplicity of use has been suggested to be an important factor that may facilitate the use of the mixed reality headset within this population, which reinforces the need to conduct usability testing in the future to document how to adapt and make the use of the device simpler, as required by the ZET principles [4,12,14,57]. In addition, other factors documented in our study, such as familiarity with the technologies and interest in using them, are shown to be important in facilitating the use of the mixed reality headset [21]. These factors are indeed part of the Technology Acceptance Model, which is a key model for understanding predictors of human behavior toward potential acceptance or rejection of the technology [58]. Considering the growing increase in technology use by older adults [51,59,60], it is possible to predict that the mixed reality headset will be democratized and implemented as an AT for home support services in the future.

However, we acknowledge that the current development of mixed reality headsets has several limitations that could prevent the democratization of such technology to people with cognitive disorders and, more widely, to the general public for several years [13]. Despite positive preliminary studies, the high weight of this device might affect the user’s experience [13,61,62]. Moreover, immersion is not optimal because of the small field of view of the screens [13]. The high price of the device is also an obstacle to its mass adoption [13]. The design of 3D graphical user interfaces and interactions is also important and should be carefully considered [63]. The literature suggests paying attention to the specific needs of older people with cognitive disorders during the design process because of their cognitive, perceptual, or physical limitations [13,64-68], which is supported by the conclusions of this study. For example, it is suggested to limit cognitive overload by limiting possible options [66], and the size of icons should be large enough as small targets might be difficult to reach [68].

If these issues are addressed, mixed reality could have several benefits, as well as limitations, because of their nature compared with other AT. In contrast to smart environments, mixed reality headsets do not require any modifications to the user’s environment. In contrast, the user is required to wear the device to receive assistance. This might be inappropriate in certain situations, for instance, in the case of night wandering [69]. In contrast to smart environments, which are stationary, mixed reality headsets can deliver assistance at any time and place. However, they cannot act directly on the user’s environment, for example, to prevent the evolution of a dangerous situation. Embedded technologies, such as the Cognitive Orthosis for Cooking, can turn off power to the stove if unsafe use by a person with a cognitive disorder is detected [70,71]. The mixed reality headset will also not be able to monitor the user’s health, in contrast to body sensors or some smart environments [66].

Smartphones are popular devices that offer an alternative to delivering AR or mixed reality apart from headsets [13]. This technology is inexpensive and socially accepted [13]. However, smartphones offer less mobility than a headset, as the user needs to hold them with one hand and point the device in the direction where the assistance will be located in the space [72]. A mixed reality headset provides assistance in front of the user, allowing the user to keep their hands free [72]. Headsets also offer a more immersive experience [72]. Finally, projectors can be used to free the user from wearing a device and to use both hands [13]. However, in most cases, projectors are stationary [13] and, consequently, cannot offer assistance at any time or place.

Several future research paths can be suggested to continue advancing knowledge about the potential of mixed reality with individuals with NDs. First, observational studies can be undertaken to refine the users’ requirements based on the directions proposed by health care professionals. Second, ZET mixed reality headset prototypes could be developed to be useful for the target population by following a user-centered design approach. In particular, optimal interactions and graphical user interfaces should be explored. Regular usability testing with users should also be undertaken to maximize the usefulness of the prototype. Third, specific mixed reality ZET principles could be developed for this population by completing the already existing guidelines [65]. Fourth, coupling mixed reality headsets with both sensors to monitor the user’s health and smart environments to manage critical situations could be explored. Finally, evaluating whether those results can be transposed to other populations, such as people with traumatic brain injuries or children with neurocognitive development disorders, could help in generalizing the applicability of mixed reality to other populations in need of ATs.

This study had some limitations. First, >75% of our participants were female, which may represent a gender issue. However, this reflects a reality in health care settings where women represent most health care professionals to which our participants belonged. Second, participants were recruited from clinical settings within a single city, which may limit the generalizability of the results. However, our focus groups were homogenous [73], as recommended by qualitative research guides, as all participants were OTs or experienced stakeholders from diverse settings in psychogeriatrics. This diversity allowed for in-depth documentation of our assumptions regarding the functional profiles of older adults with NDs across the continuum of care. Third, data saturation was noted as early as in the fourth focus group, reinforcing the credibility of the results obtained from the analysis. Finally, this study does not directly document the end user perspective but instead involves them indirectly, which is a limitation to the applicability of the results currently. However, we decided to initiate the first step of the user-centered design cycle (exploration) by involving clinical experts as our rationale was that older adults living with NDs could have had difficulty answering our research questions because of a lack of abstraction abilities required to discuss an intangible topic [74,75]. It has been previously documented that the low maturity of technology is a barrier to the initial intention for use and may result in the rejection of the prototype in the future. Therefore, we took the possible obstacle into consideration when choosing our first design step [76]. Our intention was to document the needs of older adults with NDs from the perspective of experts to design a version that can later be tested by them. Our next step will be to design a prototype that meets the recommendations identified in this study (ideation and generation) and then test its usability with older adults living with NDs and their caregivers (evaluation).

This study aimed to describe experts’ perspectives on the potential of pure mixed reality to support independence and ensure the safety of older adults living with NDs. The results suggest that a mixed reality cognitive orthosis may help older adults with NDs face difficulties in everyday activities, such as remembering to complete tasks, initiating the tasks, remembering where they are in a task (what parts are already completed), and planning the task. Thus, the use of mixed reality cognitive orthosis in older adults living with NDs to overcome these difficulties and promote independence and safety at home is promising and may respond to several clinical functions identified by the participants, including assessment, assistance, and training. Finally, the mixed reality headset has good potential for use with older adults with NDs, with certain facilitators and limits. Future studies should address usability testing in this population to develop a usable and implementable prototype to support aging in place.