Multiple sclerosis (MS) is an inflammatory demyelination disorder of the central nervous system that is estimated to affect 2.8 million people worldwide [1]. Over a third of the people with MS have upper limb (UL) dysfunction, including weakness, tremors, and spasms in one or both ULs [2]. This can result in difficulties with activities of daily living (ADL), negatively impacting quality of life and the likelihood of remaining in employment [3,4]. Problems specifically with dexterity are related to higher health care costs [5] and a higher association with depression-like psychological measures compared to problems with lower limb function [6]. Rehabilitation and physical exercise improve motor function for people with MS [7,8]. The evidence regarding UL rehabilitation is lacking in comparison with the lower limb, despite the high frequency of UL impairments and their impact on ADL [9]. In addition, there are particular challenges in finding effective yet motivating rehabilitation strategies in MS due to the long-term, progressive nature of the disease and diversity of symptoms [10].

Virtual reality (VR) is increasing in popularity in rehabilitation research and is proposed as a possible approach to encourage long-term rehabilitation [11]. VR includes digital environments that often simulate real-world experiences with reported benefits of high motivation and engagement, with real-time feedback [12]. VR has shown promising results within MS populations, but this evidence is limited in comparison with stroke, especially regarding UL function [13]. Our systematic review, investigating the effect of VR in improving UL function in MS, found early, but limited, evidence suggesting VR has the potential to improve function in people with MS [14]. There was also a low number of dropouts in most studies within the review, supporting that VR could improve adherence compared with conventional rehabilitation; therefore, VR could be useful in conditions such as MS, where prolonged rehabilitation is required.

VR is often investigated alongside video games played within a VR setting, which can be commercially available or specifically tailored games designed with a target population in mind. Commercially available exercise games, targeted at a healthy population, can be unsuitable for disabled individuals and lead to discouragement and anxiety [15]. It is beneficial to involve a sample of target users in the creation and development of effective VR-based gamified approaches [16]. This process is known as coproduction [17]. To date, no study has systematically coproduced VR games specifically for UL rehabilitation in people with MS.

The aims of this study were to determine the views of people with MS and specialist clinicians on UL dysfunction or function in MS, challenges faced by clinicians when delivering UL therapy, barriers and motivators for exercise in MS, opinions on VR, and suggestions for development and design of VR games. These findings will guide the future development of VR applications and interventions for UL rehabilitation for people with MS.

Ethics approval for this study was provided by the School of Health and Life Sciences Ethics Committee at Glasgow Caledonian University (HLS/PSWAHS/20/002). Informed consent was obtained from the participants and clinicians.

The study aimed to recruit up to 12 people with MS and 12 specialist MS clinicians to take part in online focus groups. The sample size was determined in line with the design of other similar studies and general recommendations for qualitative analysis [18,19]. To be included in the study, people with MS were required to be aged ≥18 years and have a diagnosis of MS (self-reported) with self-reported UL impairment. Clinicians were required to have experience (any duration) in delivering MS rehabilitation within the National Health Service (NHS) or the third sector. In addition, all participants were required to have access to and the ability to operate videoconference software. There were no specified exclusion criteria. Participants with MS were identified through the MS Society UK’s research network, which advertised the study to its members. Those who were interested in participating contacted the research team directly and were emailed a participant information sheet. In terms of recruitment of clinicians, the MS Trust Therapists in MS network advertised the study to its members. Interested clinicians contacted the research team and were emailed a participant information sheet.

The focus groups for people with MS and clinicians were conducted separately with a maximum of 5 people per focus group. To comply with COVID-19 pandemic regulations at the time, focus groups were held online using Zoom (Zoom Video Communications) or Teams (Microsoft Corp) videoconference software; this also provided an opportunity for recruitment of participants from across the United Kingdom and Ireland. The focus groups were conducted in a semistructured style using a focus group schedule split broadly into three sections important for the development of VR interventions for UL problems in people with MS: (1) UL dysfunction and exercise or therapy; (2) opinions on VR; and (3) suggestions for development and design of any developed VR games (Multimedia Appendix 1). In addition, clinicians were asked what information and feedback they would want from a patient’s VR therapy session. The questions included prompts that allowed more targeted responses from participants regarding their experiences and views [20]. Within the focus groups, participants were shown three videos demonstrating different commercially available head-mounted devices (HMDs) and hand-tracking devices: (1) a nonimmersive VR set up using a Leap Motion controller and computer monitor, which is a hand motion capture device that allows users to visualize their hand movements and interact with virtual environments; (2) immersive VR using the Oculus Rift HMD with a mounted Leap Motion device for hand tracking; and (3) immersive VR using the Oculus Quest, with in-built hand tracking (Figure 1 [21,22]; Multimedia Appendix 2). Videos were shown as participants were unable to try these devices since the focus groups were online due to the COVID-19 pandemic. These videos attempted to contextualize and demonstrate the different VR and motion capture devices in terms of users interacting with environments, possible hand movements, and previous games developed from prior research. After watching the videos, participants were encouraged to share their initial thoughts on each of the technologies. The focus groups involving people with MS and clinicians lasted approximately 90 minutes and 60 minutes, respectively. The focus groups were facilitated by a female researcher (AW) who had been involved in the recruitment of participants and an additional senior, female researcher (LP) attended.

All focus groups were audio recorded and transcribed verbatim. Qualitative analysis of the data was performed based on theme-based content analysis (TBCA) as described in the study by Neale and Nichols [23]. This qualitative method groups responses into content-related themes to enable researchers to view the user preferences more easily and has been used to influence the development or evaluation of a VR environment [23-25]. TBCA is a flexible qualitative data analysis method that involves five key steps: (1) data collection, (2) data collation, (3) raw theme definition and classification, (4) higher order theme selection, and (5) presentation of classification matrix [23]. Owing to the large number of higher order themes, we added an additional step by grouping the higher order themes into main themes. The raw themes were assigned independently by 2 researchers in the transcripts of people with MS (AW and LF) and clinicians (AW and LP). After agreement on the raw themes, the responses were then independently grouped by 2 researchers (AW and LP) into higher order themes. Any discrepancies in assigning the themes were resolved through consultation with a third reviewer, if necessary. Once the higher order themes were determined, the main themes were determined by 2 researchers (AW and LP). The main themes with their associated raw and higher order themes are presented in tables. The raw and higher order themes were quantified manually within the matrix based on the number of responses necessary to display popularity or consensus [23], and example quotes for each higher order theme were included. Focus groups of people with MS and clinicians were analyzed separately to allow comparison of the findings between the 2 groups.

A total of 10 people with MS were recruited to the study and took part in 1 of 2 focus groups, each of which had 5 participants. Most participants with MS were female (7/10, 70%), with a mean age of 56.4 (SD 16.5) years and a mean time since diagnosis of 14.4 (SD 12.3) years. Participants had varying MS types (Table 1).

A total of 8 clinicians were recruited (5 physiotherapists, 2 occupational therapists, and 1 MS specialist nurse). Among them, 6 participants worked in the NHS and 2 worked in other settings. There were 2 focus groups for clinicians with 4 participants in each group. All clinicians were female, with a mean age of 46.2 (SD 9.6) years, and the mean length of experience was 17.9 (SD 10. 2) years.

This study aimed to explore the views of people with MS and clinicians on UL impairment associated with MS and the potential role of VR as a rehabilitation approach to address this impairment. The discussion focuses on the combined findings from the 2 groups of participants: people with MS and clinicians (Figure 2). Figure 2 is a visual representation of the principal findings based on the higher number of responses assigned, which should inform the development of VR applications and interventions aiming to improve the UL function for people with MS and how VR could tackle challenges of existing UL exercise raised by clinicians and people with MS in this study. The findings agree with those of previous studies that people with MS commonly have UL impairments that impact function, including problems with dexterity and ADL, which leads to loss of meaningful activities [26-28]. Despite UL difficulties, UL exercise was neglected due to MS symptoms, such as fatigue, lack of motivation, and dislike of exercise, as well as the challenges clinicians faced regarding time constraints and finding appropriate therapies that were not childlike or boring. Lack of focus on UL rehabilitation has been reported previously in MS [9] and in other long-term neurological conditions such as stroke [29]. The progressive and unpredictable nature of MS was raised by both groups, and consequently, clinicians raised the importance of setting realistic expectations with therapy, sometimes focusing on maintenance of function rather than improvement.

Both groups (people with MS and clinicians) were optimistic about the use of VR and believed VR could be a solution to their exercise challenges. Positive comments, including avoiding traveling, being accessible, and being engaging or fun addressed the identified barriers for UL rehabilitation. This concurs with previous VR studies [30,31] and specifically in UL rehabilitation in MS, with a recent home-based, feasibility study using the Oculus Quest 2 VR headset in which participants described VR as fun, interesting, and innovative [32]. Participants in the study by Kamm et al [32] suggested adding difficulty levels and scoring to their exercises, competitive elements previously described to be motivating by people with MS using nonimmersive exercise games delivered through the Nintendo Wii [33]. In this study, both groups were especially positive regarding the immersive approach of the Oculus Quest. Participants thought the escapism properties and visualization of movements could potentially “disguise exercise,” which may occur with the “fun” element of VR reducing the perception of exertion during exercise [34], therefore encouraging more UL therapy.

Negative views about VR were also expressed, mainly the potential for cybersickness. Cybersickness is thought to be caused by conflict of stimuli, leading to nausea, disorientation, and pain in the eyes and head [35]. Women are more susceptible to cybersickness [36], which is relevant in MS, with a higher number of women affected. Although cybersickness with VR has been reported previously in people with MS [37], there are development strategies for reducing cybersickness, such as designing VR activity with less overall movement within the virtual environment. Cybersickness is, however, thought to reduce over time with exposure to VR [38]. There were unnecessary concerns raised for those wearing glasses as the HMD can accommodate glasses, but there were valid concerns about the weight of the HMD for some users. Disengagement was another concern both groups expressed, with limited data on long-term adherence to VR in MS rehabilitation. Exercise is a behavioral intervention, and long-term adherence to exercise can be supported by evidence-based behavior change techniques [39]. These behavior change techniques, such as goal setting, rewards, and feedback, can be incorporated into VR games or activity to support long-term engagement in UL exercise. While VR can be more engaging than other methods of exercise [40], frequent performance, feedback on progress, and adjusting levels of difficulty can maximize VR engagement for those with long-term neurological conditions [41]. Finally, clinicians had specific concerns regarding digital poverty, the technical ability of people with MS, and insufficient technical services to support VR.

Considerations for VR game development align with user-centered design principles for VR in motor rehabilitation in survivors of stroke, such as being fun, tracking progress, having an element of competition, challenging oneself, and providing feedback [42], and are not specific to any clinical population. Participants raised that VR development should be mindful of the different end users (people with MS) who may differ in ability and preferences. Clinicians suggested VR would appeal to younger individuals with MS, whereas people with MS felt older people with MS might need more basic gameplay. While there is some, albeit limited, evidence for lower usability scores for older VR users compared with younger users, there can be higher user enjoyment [43], and there is moderate evidence for good usability of VR in older populations [44]; therefore, this concern may be overly cautious.

Consideration of the end user links to the importance of choice when designing VR interventions, with a variety of games to appeal to as many as possible. Participants felt the games should include different movements, levels of immersion, level of difficulty, or feedback on performance. Accommodating individual preferences is a key element for the design of VR games for rehabilitation, as it increases user engagement [45]. However, our previous systematic review found that a choice of games was rarely included in VR interventions in MS [14].

There were differing views in terms of the type of feedback people wished from VR. Some people with MS wanted to track scores and visualize results, which is supported by reward theories for users during both entertainment and serious games [46]. Conversely, concerns were raised about feedback potentially being discouraging or demotivating, especially given the variable nature of MS. As an example, countdown timers provide slight pressure to motivate players to increase engagement [47]; however, in this study, people with MS felt they could be stressful. Feedback on the duration of exercise completed was appealing to people with MS, as reported previously [19]. As well as the quantity of VR exercise, clinicians also wished feedback on the quality of movement when performing the games. Rehabilitation often involves highly repetitive movements to stimulate neuroplasticity; however, stroke specialist therapists have also previously reported concerns that quality of movement in VR rehabilitation for UL maybe sacrificed for a good gaming outcome [18], although this has not been explored in people with MS. Both groups were interested in the reported outcomes of using VR approaches which, if positive, would increase engagement.

Clinicians and people with MS felt VR activity had to be related to the patient’s personalized and meaningful goals, which is known to increase motivation in physiotherapy settings [48]; however, this is often neglected in VR regimes [14]. Goals need to be adjusted over time in a progressive condition, such as MS, and to avoid disengagement as raised earlier. Participants with MS frequently stated that their goals were related to not only improvement but also the maintenance of ability and the prevention of further deterioration. In terms of suggestions for VR activities, the groups differed with clinicians suggesting ADL or hobby simulations and people with MS being more ambivalent, stressing activity to be fun with a variety of real-life and abstract VR games. Previous studies of VR have often involved ADL activities such as cooking or other kitchen activity [49,50]. Although VR can provide a safe environment to practice ADL for people with mobility issues [51] people with MS in this study were less interested in ADL, especially kitchen simulations. Both groups suggested an “end result,” such as creating a drawing, or learning a new skill would be positive and facilitate a feeling of accomplishment. There were also suggestions from people with MS to incorporate haptic activities, such as grabbing and gripping. However, the user is not able to receive tactile feedback when interacting with a virtual environment, and handheld controllers may need to be considered for some VR activities [52]. Another solution could be to incorporate pseudo haptics, the use of different stimuli such as visual or auditory stimuli, to mimic a variety of haptic properties in a virtual environment [53]. This is an emerging field that could be explored in VR for people with MS. Similarly, as many of the participants suggested finger-related exercises, it is important that VR systems use good hand-tracking motion capture devices to allow visualization of the movement of fingers and wrists within a VR setting.

Many people with MS were supportive of VR for home use, as being more convenient and accessible. However, there was recognition that users needed demonstration of the technology and a level of clinician supervision. Assessing quality of movements and monitoring of patient progress are reported challenges for VR home use [54]. A recent study with a small number of participants found VR to be feasible for home-based UL rehabilitation in people with MS, after 3 supervised sessions [32], but larger studies of home-based VR for UL rehabilitation are required. There was agreement in both groups that an element of social interaction could be considered in the development of VR games. Generally, there is a lack of evidence on the effect of socialization within UL therapy, but it may improve adherence and motivation [55] and provide better outcomes [56]. Specifically in relation to VR, there is some evidence that social aspects increase motivation through competition [57], but participants in our study were more interested in self-competition rather than competing against others. This is similar to a study of a walking app for MS where users were less interested in sharing their goals or achievements with others [58].

Recruiting participants through online sources may result in a biased sample, as those comfortable with technology and access to online services are more likely to take part. Being online allowed the involvement of people with MS with varying abilities and clinicians who worked in the NHS and the third sector across the United Kingdom. However, the online nature meant it was not possible for participants to physically test the VR equipment and explore their reactions. While it can also be challenging to engage all participants in online focus groups, this was resolved by asking questions using participants’ names or by getting participants to use the raise hand function within the videoconference software and encourage discussion between participants.

The TBCA methodology groups responses into themes to quantify them but does not allow consideration of the interaction between participants. Participants had a number of specific questions, such as the long-term outcomes of using immersive VR, the optimal target users for VR (level of disability), and the extent of translation of VR activity into “real-life” function. However, there is currently a lack of literature to provide responses to these questions, which highlights areas for future research.

This is the first study exploring the views of people with MS and clinicians in terms of VR for UL rehabilitation for people with MS and has highlighted the current challenges in UL rehabilitation even though UL impairment is common and impacts meaningful activity. Overall, people with MS often found dexterity-related activities difficult, which impacted multiple ADL and challenges faced in therapy related to motivation, lack of resources, and difficulty finding interesting UL exercises. There was positive support for VR for UL exercise. Overall, to improve engagement and satisfaction for the user, this study suggests any VR games developed for people with MS should (1) be fun and engaging; (2) have clear aims related to the individual user’s goals; (3) offer personalization, such as a variety of games (abstract and ADL based), different movements, levels of difficulty, and methods of feedback; (4) monitor quality as well as quantity of movement during gameplay; (5) incorporate design features to reduce the potential for cybersickness; (6) consider if the games can incorporate education or skill development; (7) incorporate aspects of social interaction; and (8) consider including haptic properties. The findings support the need for the creation of bespoke serious games rather than using commercially available exercise games, which can discourage users with motor dysfunction [15,59]. Overall, future development of VR games for UL rehabilitation should focus on a personalized and customizable approach to encourage long-term engagement to improve meaningful outcomes for people with MS.