The economic burden of stroke on health care systems is substantial due to large numbers of people affected by stroke and the high costs associated with acute treatment and long-term care. This treatment is necessary due to the diverse range of poststroke impairments, including reduced upper-limb function [1]. In 2019, over 12.2 million new stroke cases were reported worldwide, marking a 70% increase in stroke incidence since 1990 [2,3]. Due to population aging, this upward trend is expected to continue in the coming years [4], leading to even higher costs [5]. Furthermore, a shortage of medical personnel can arise due to the expected increase in stroke cases. To reduce costs and personnel, a shift in focus toward home-based rehabilitation (HBR) has recently emerged [6]. HBR offers additional benefits including reduced travel time for stroke survivors, the possibility to increase therapy intensity, increased accessibility allowing users to perform the activities throughout the day, and longer therapy duration [7].

Since higher training intensity leads to greater improvements in most patients [8,9], a critical element for the success of HBR is adherence. However, it is challenging with HBR to achieve adherence consistently, especially given that in general, treatment adherence tends to decrease over extended periods [10]. This decrease is often the result of low acceptability and usability, the inability to stay engaged, or integrate the practice into daily life [10]. Nonetheless, adherence to treatment is crucial for patients with chronic stroke, since behavioral changes in arm use require a substantial training dosage schedule [11]. Adherence rates in HBR for stroke vary significantly, ranging between 13% and 140% [10]. Several factors positively influence the adherence rate in HBR, including rehabilitation programs with game-like training exercises with multiplayer modes, sufficient variability in games, and adjustable difficulty levels. Moreover, the availability of monitoring and feedback options to track users’ progress and motivate users also increases adherence [10,12]. Technical support for issues, such as malfunction of the system, and the possibility to customize the system to meet individual users’ needs are also important considerations [10,12]. Implementing these factors in HBR can increase adherence and thus improve its overall success.

When considering HBR, multiple end users can be identified, such as stroke survivors, health care professionals, and caregivers. All these end users have a different relation to individuals with stroke. These different relations could result in different perspectives; for example, a health care professional might be more objective, whereas the caregivers and stroke survivors are more subjective. However, Demir et al [13] found that stroke survivors, health care professionals, and caregivers generally share the same priorities during stroke rehabilitation, focusing on improving the health condition post stroke, arm and leg mobility, and increasing independence. Nonetheless, HBR might require other demands compared to usual care. In home settings, caregivers and stroke survivors might prioritize ease of use, whereas health care professionals might focus primarily on health outcomes. As a result, health care professionals might have specific requirements for outcome visualization and monitoring, whereas caregivers are more concerned with the practical challenges of implementing rehabilitation therapy in a home setting. These potential differences underscore the importance of a flexible and adaptive system design that meets the needs of diverse users and should be carefully considered in developing a successful HBR system. However, the different viewpoints of these end users on user requirements for HBR is still unknown.

One advantage of HBR is its potential for scalability and implementation across multiple countries, making it accessible to a diverse range of populations. However, differences in health care systems and cultural practices across Europe must be considered. Across Europe, there are several clusters with comparable health care systems, with Spain, Italy, and the Netherlands belonging to the same cluster [14]. However, Italy and Spain present a higher overlap with each other compared to the Netherlands [14]. Additionally, health care systems in Italy and the Netherlands rely more on inpatient care, while Spain has a stronger outpatient care tradition compared to Italy and the Netherlands [15], suggesting that Spain may have greater experience with home-based exercises. There are also some cultural differences. For example, Italy and Spain are more collectivistic, while the Netherlands is more individualistic [16]. Thus, the Netherlands might be more receptive to HBR, since this is a more individualistic therapy form. Studies have even shown that countries across Europe use different cognitive emotion regulation strategies [17]. These differences, both in health care systems and cultural practices, can potentially influence the requirements of an HBR tool. To accommodate these differences in system design, a more prominent role for the coaches in countries less familiar with HBR and enhance group interaction features where social engagement is valued could be integrated. However, there is currently no information available on these possible differences.

Recently, within an international project involving partners from Italy, Spain, and the Netherlands, a new HBR solution was defined, called the Neurorehabilitation Ecosystem for Sustained Therapy (NEST). As of January 2021, this solution is still in the design phase and has not been developed. This innovative rehabilitation system offers an integrated rehabilitation solution for stroke survivors, featuring the following components: (1) the RGS (Rehabilitation Gaming System) app for smartphones, which offers gamified exercises using augmented reality for upper-limb training, focusing on finger, wrist, elbow and shoulder movements while holding a smartphone; (2) the RGS-wear, a smartwatch-based system for monitoring upper-limb movements during daily activities; and (3) a virtual coach, the agent for well-being assistance (AWA), integrated into both the RGS app and RGS-wear, providing motivational feedback on training performance and daily arm use as well as assistance in the use of the gamified exercises. The combination of these three components, which were found relevant in previous literature [10,18,19], makes NEST a unique HBR tool. However, whether specific user requirements for NEST exist and if the system fulfills these needs remains unclear.

In summary, no specific user requirements are known for a NEST-based HBR system, including optimization of gamified training exercises, monitoring of arm use, and virtual coaching. More insight into user requirements specific to NEST could help to further optimize and develop NEST. In addition, there is a lack of knowledge about its implementation across different end users and multiple countries. It is essential to conduct research with multiple end users and across different countries to gain more insight into user requirements. Therefore, the first aim of this study was to define the user requirements for a newly developed upper-limb training system that includes the three aspects of NEST for the end users: stroke survivors, health care professionals, and caregivers. Our second aim was to identify possible differences in user requirements among these end user groups and across three European countries (The Netherlands, Italy, and Spain).

The first objective of this study was to identify the user’s needs of a newly developed upper-limb training system that includes gamified exercises, monitoring, and a virtual coach. This was investigated with focus groups and interviews with stroke survivors, caregivers, and health care professionals in clinical centers in Italy, Spain, and the Netherlands. Participants across all user groups and countries highlighted the relevance of each component of NEST, the RGS app, RGS-wear, and the AWA coach. The participants recognized the potential of NEST, including these elements, as an HBR tool. In order for this combination to be successful, motivational elements such as gaming, feedback, and reminders are important. In addition, during the focus group sessions and interviews, several improvements were suggested, mainly concerning adding motivational components and enhancing system customization. If these suggestions are integrated, the system has the potential to maintain high adherence and effectively train upper-limb function. In addition, these aspects may help to make HBR more goal-oriented, potentially enhancing stroke survivors’ awareness of arm use in daily life.

In line with previous studies, this study highlights the need for HBR systems to have three demands in particular: (1) customization to emphasize the need to tailor HBR technology to individual patients [10,19,25-27]; (2) motivational elements, including features such as reminders [19], feedback [10,12,18,19,25-28], a variety of games or levels [7,10,12,19,26,29,30], and the ease of use [7,10,19]; and (3) active engagement of the therapist to ensure proper execution, progress monitoring, and provision of feedback [7,10,19,25,28-30]. These general recommendations for HBR should apply not only during exercises but also in daily life. HBR should be customized to support daily life goals, motivate patients to use the affected upper limb in daily activities, and provide insights into daily-life patterns that are useful for therapists. Since the primary goal of upper-limb rehabilitation is to promote active use of the affected limb in everyday life, these requirements and factors help facilitate the transfer of rehabilitation gains to daily life.

Several studies have underscored the importance of active engagement of the therapist during HBR (ie, active monitoring) to enhance patient adherence and progress [7,10,28-30]. Additionally, HBR can highlight specific goals and therefore provide useful insights for therapy [31]. This study investigated the presence of a virtual coach in detail, highlighting the importance of its interaction with the patient. It can help the patient with daily structure by using reminders and feedback [32]. On a more detailed level, this study found that measuring both the quantity and quality of movements during daily life and exercises and providing goal-based feedback are crucial, as this may help to control compensatory movements and provide input for therapists to further tailor rehabilitation. Only two studies investigated user requirements for home monitoring in detail [19,33]. However, Langerak et al [19] investigated home monitoring using sensors and Cavuoto et al [33] used a smartphone, whereas this study focused on the use of a smartwatch. Still, all three studies revealed similar results. This indicates the need for careful consideration of monitoring methods and the type of feedback provided to stroke survivors.

This was also the first study that took the perspectives of various end users on the user requirements of NEST into account. Previous studies mostly focused on stroke survivors or held focus group sessions or interviews with the stroke survivor and caregiver together, indicating the need for insights from therapists and caregivers independently, free from influence of the stroke survivor [10,25]. Studies that have included stroke survivors, caregivers, and health care professionals have generally found minimal differences in user requirements between these groups [7,34]. However, despite the significant overlap in user requirements among end users, some minor differences were noted. This suggests the need to pay attention to these differences when further developing the NEST system, as tailoring the systems to accommodate these nuances could enhance its effectiveness and user satisfaction. Caregivers did not express opinions on the early introduction of the system, quality measurement, or goal-oriented feedback. Instead, they emphasized the importance of maintaining their role as a partner rather than a therapist. The absence of these therapeutic considerations from the caregivers’ answers suggests a more social perspective of their role, most likely because they are not directly involved in the therapy process. Consequently, caregivers may view themselves primarily as supporters rather than as system users. Previous reviews have showed that social support positively influences both the quality of life and physical recovery of the stroke survivor, underscoring the need to consider caregivers in HBR [35,36]. Interestingly, the user requirements mentioned by stroke survivors were also mentioned by therapists and caregivers, suggesting that these groups have a good understanding of the stroke survivors’ needs.

In line with the high overlap between end users, there was a strong consensus on user requirements across countries. A previous review showed an agreement across 15 countries on the importance of place, relationships, and therapy [25], though it only focused on HBR without incorporating technology. In addition, a study comparing HBR with technology between Bangladesh and the United Kingdom found similar user requirements but noted more sociotechnical challenges in Bangladesh [37]. In this study, only subtle differences in addressed topics were noted, including privacy concerns raised at the Spanish site, negative caregiver influence at the Dutch site, and self-awareness at the Italian site. The minor differences may reflect the inclusion of sites exclusively from Western countries, where major cultural variation is less likely. Although only small, these differences should be taken into account when further developing the NEST system. Specific social aspects, such as social stigma and relation dynamics among users, were not explicitly addressed nor raised by end users during the focus groups in any country. However, these topics might be more country-specific and could contribute to national differences. The variations in this study may stem from the distinct participant characteristics between the countries. Dutch participants in the chronic phase of stroke recovery likely offered more comprehensive insights into experiences with caregivers at home, unlike the acute stroke survivors at the Italian and Spanish sites. The literature supports that stroke over time affects the experience of being at home, for example, a change of feeling in bodily routines and relationships [38]. In addition, the high motivation observed among Italian stroke survivors may have highlighted the self-awareness. These findings indicate that clinical characteristics, home environments, and coping mechanisms influence the perceptions of HBR by stroke survivors.

One limitation of this study was the predominance of male stroke survivors and the age imbalance among participants across different clinical sites. This could have influenced the needs of the stroke survivors for NEST, since age and sex can influence the severity of stroke [39,40]. Also, women more often live alone after hospital discharge [39]. Additionally, the inclusion criteria for this study, including minimal cognitive impairments, no aphasia, and smartphone usage, likely do not represent the entire stroke survivor population. These inclusion criteria may have contributed to a more positive attitude toward NEST. However, despite the imbalance, our results are aligned with those of previous research, and there was a high degree of overlap in findings across all countries. Another limitation was the participants’ difficulty in conceptualizing NEST, which was still in development. To address this, videos of the RGS app were shown, and when participants asked for more clarity, they were asked to describe their ideal scenario. This approach often led to open discussions about potential features and improvements. The study also faced limitations due to the use of varied interview formats (focus groups, online focus groups, and interviews) imposed by the constraints of the COVID-19 pandemic. Interviews may generate a broader range of items, potentially enhancing depth, whereas focus groups may elicit more personal information due to a peer environment [41]. Furthermore, the online focus groups may have had constrained discussion due to the less natural setting. Nevertheless, previous research showed that content generated in either interviews or focus groups have a high overlap [41], and the same applies to physical or online focus groups [42-44]. Furthermore, the use of identical interview guides ensured consistency in discussion topics. The high degree of overlap in results suggests that the same questions were effectively addressed across the countries. Therefore, this aspect was not considered further in the data analysis. Lastly, no specific questions on between-country or end user differences, such as culture or care pathways, were included, as the focus remained on the primary aim, including user requirements for system development. These topics could have highlighted potential differences; yet, they were not raised by participants, suggesting limited perceived relevance. Therefore, more research is needed on this topic.

In conclusion, all end users across the participating countries emphasized the importance of integrating gamified exercises, monitoring, and a virtual coaching component into an HBR system. The user requirements for such a system can be categorized into three key areas, customization, motivational elements (including ease of use), and feedback elements, which are essential for maximizing user engagement and effectiveness. These areas extend beyond HBR exercises and are also important for daily-life activities. Users’ needs found in this study are largely similar; however, subtle differences between end users (eg, social perspective caregivers) and western countries (caregiver burden, privacy concerns, and self-awareness) should be considered when further developing the NEST system.