In up to 85% of stroke survivors, sequelae persist in the upper extremity (UE) [1], resulting in a long-term impact on daily living activities [2,3]. To stimulate neuroplastic changes that promote motor recovery, stroke survivors should follow an intensive, task-specific, stimulating, and, above all, repetitive exercise program [4]. However, the programs offered by conventional therapies, in particular, during chronic stage, are not sufficient for people to achieve the level of repetition and intensity required for recovery [4]. In Canada, patients diagnosed with stroke receive the Graded Repetitive Arm Supplementary Program (GRASP) [5] as a home exercise program after discharge from traditional rehabilitation services [3]. The GRASP is recommended by the Canadian Best Practice Recommendations for Stroke Care [3] to provide training and increase the use of the impaired arm in daily life activities [5]. However, patients must be motivated enough to do the exercises on their own and to perform enough repetitions to achieve improvement. A minimum of 15 hours is suggested for an intervention to result in a moderate improvement in daily living activities following a stroke [6]. Numerous studies offering an intensive exercise program, in chronic stroke, reported significant improvements in UE impairment (P=.05) [7,8] and even maintenance of these changes over a period of 6 months after treatment [8].

An interesting alternative that has been proposed to exercise intensity and exploit neuroplastic properties involves the use of virtual reality. For every 30 repetitions performed in a standard rehabilitation session, the same person can perform 600 to 800 repetitions within a 1-hour session in a virtual environment [4]. This is an important difference between the two approaches that highlights the potential of virtual reality. Thanks to its design, a virtual environment can simulate the real world while maintaining total control over the parameters of the tasks to be executed within it to foster motor learning [4]. In the context of rehabilitation, virtual reality has been integrated in the form of exergames because the gaming goal is to enhance activity through various types of exercises. Among these exergames, some have been specifically designed for stroke rehabilitation, including for balance training as well as for training for both lower and upper limb deficits (eg, Caren system, Lokomat, Armeo, and Jintronix [4,6,9,10]). The nature of the exergame interface makes it possible to modify the difficulty level of the exercises via various parameters such as intensity (time and repetition), visual feedback, speed, strength, and range of motion [4]. Progressing the exercise to ensure it remains challenging may motivate the user to complete the exercises [4]. However, there is no way to ensure that the movements made in the virtual environment are performed correspond to what is being trained. When performed in clinic, a health professional can ensure that the exercises are adjusted, and the movements are executed appropriately. However, at home, monitoring by a health professional needs to be considered to prevent compensatory movements during the exercises, inadequate choice of difficulty level, inappropriate completion of the exercises, and lack of motivation to pursue exercise program. The interfaces of the exergames do not allow automatic adjustment adapted to the person’s abilities and do not sufficiently detect compensations [4,6,9,10]. Thus, the follow-up by a therapist would make it possible to optimize the exergames’ benefits for the user (such as by adapting the difficulty parameters and by choosing games that are relevant for the user) and to help transfer motor learning from the virtual environment to activities of daily living.

To increase monitoring, a telerehabilitation system could allow such a follow-up through videoconferencing sessions between the user and the therapist. Such systems are increasingly used to provide remote rehabilitation services [11]. Its efficiency, compared with the standard treatment (face-to-face interventions), has been demonstrated, resulting in similar clinical results among participants diagnosed with stroke [12], therefore increasing accessibility to stroke rehabilitation services. Combining telerehabilitation with virtual reality (VirTele) could allow live sessions in which the therapist can observe the user playing and follow the game screen, at the same time, to assess how the user manages to complete the movements requested through the exercises (detect compensations, correct pathological patterns, and directly modify the difficulty setting necessary for the smooth running of the exercise). Therefore, the VirTele technology could allow continuous adjustment to the exercises and monitoring of the user completion to create a more personalized, tailored training program. Furthermore, in the long term, the VirTele exercise program could empower users to integrate the use of their impaired UE in their daily activities, through shared decisions made with the therapist.

Given the evidence for the efficacy of exergames as well as telerehabilitation for stroke rehabilitation, but the limited evidence of combining these technologies, such as in the VirTele program, the overall goals of this study are to explore its preliminary efficacy for UE rehabilitation and examine its feasibility for use with stroke survivors at home. More specifically, the objectives of this study are to (1) determine the preliminary efficacy of VirTele on UE motor recovery, function, quality of life, and motivation in participants with chronic stroke, compared with conventional therapy (GRASP); (2) examine the feasibility of using VirTele with participants diagnosed with chronic stroke at home; and (3) identify the obstacles and facilitators for the technology use by participants diagnosed with stroke and stroke therapists and understand the shared decision-making process.

It is hypothesized that the exergame-telerehabilitation program will lead to greater UE motor recovery than usual care (GRASP) in participants with chronic stroke. We also hypothesize that the exergame-telerehabilitation program will have a greater impact on function, quality of life, and motivation in participants with chronic stroke.

This is a mixed method study design consisting of a randomized, blinded feasibility trial with an embedded multiple case study that will take place in Montreal, Canada. The randomized feasibility trial is a two-arm, single-blind trial design in which eligible participants will be randomly allocated to an experimental (VirTele for 8 weeks) or control, usual care group (GRASP for 8 weeks). The feasibility trial captures both feasibility and preliminary efficacy outcomes. Publishing the feasibility results would provide a better understanding of the context in which the efficacy data were collected and a better interpretation of the final results [13,14].

Outcome measures will be assessed for both groups on four occasions: at baseline (T1), at the end of 2-month intervention (T2), after a 1-month follow-up period (T3), and after a 2-month follow-up period (T4; Figure 1).

Randomization and evaluations will be performed by research assistants who are not involved in the study. All participants will provide informed written consent before enrollment. This study is registered at clinicaltrials.gov (NCT03759106) and has received ethics approval from the Research Ethics Boards of Centre for Interdisciplinary Research in Rehabilitation of Greater Montreal (June 28, 2018).

At study entry, we will administer the Chedoke-McMaster arm component [15] to get evidence of UE impairment. Only participants with a score of 2 to 6 will be eligible. We will also verify balance maintenance during sitting position and detect any UE mobility restrictions limiting the ability to play (restricted shoulder movements because of pain) through active and passive mobilizations.

Stroke survivors who have residual UE deficits and are no longer receiving rehabilitation services will be eligible for study participation if they fulfill the eligibility criteria. The inclusion criteria are as follows: first-time unilateral ischemic or hemorrhagic stroke or no residual deficits from a previous stroke and able to use the Jintronix system (eg, able to move the game avatar with impaired limb). The exclusion criteria are as follows: being medically unstable (eg, uncontrolled cardiac condition), severe cognitive or communication deficits, visual impairments limiting the ability to use the games, and UE mobility restrictions limiting the ability to play (eg, restricted shoulder movements because of pain).

We intend to recruit participants from the community and from the archives of rehabilitation centers (offline) situated in the Montreal, Sherbrooke, and Laval areas (Quebec, Canada).

Study therapists (physical therapists or occupational therapists) from the different participating sites will also be included in the study to explore their experiences with the technology and comprehend the shared decision-making process underlying their choice of games and difficulty levels.

To date, no studies have reported on the use of combining such technologies, such as in the VirTele program so that this randomized, blinded feasibility trial will provide evidence for the effect size. However, as the first estimate of effect size, a sample size of 52 participants has been calculated using the Fugl-Meyer Upper Extremity Assessment as a primary outcome and G*Power 3.1 [16]. We assumed a medium effect size of 0.2, which was reported in a study with chronic stroke survivors with a 2-arm randomized clinical trial [17] and accounting for 20% retention issues (alpha=.05 and power=80%). This sample size corresponds to recommendations for pilot efficacy and feasibility trials [18] and is realistic, given time and budgetary constraints as well as recruitment potential. Thus, for the randomized, blinded feasibility trial, 52 participants will be recruited and randomly allocated to the experimental (n=26) or control group (n=26). A block randomization with a block size of six will be used, given time and material constraints.

For the multiple case study component, four therapists and the first 10 participants from the experimental group will be invited to participate in interviews. However, the final sample size of the participants diagnosed with stroke will be adjusted depending on the qualitative data saturation [19].

This paper describes the research protocol for a mixed method study, including a randomized, blinded feasibility trial with an embedded multiple case study. The aims of this study are to (1) determine the preliminary efficacy of VirTele on UE motor recovery, function, quality of life, and motivation in participants with chronic stroke, compared with conventional therapy (GRASP); (2) examine the feasibility of using VirTele with participants diagnosed with chronic stroke at home; and (3) identify the obstacles and facilitators for the technology use by participants diagnosed with stroke and stroke therapists and understand the shared decision-making process.

We have chosen a mixed study design because the use of both qualitative and quantitative approaches makes it possible to construct a more complete image of the studied phenomenon. The combination of the two methodologies should be approached not from the point of view of their differences but from the complementarities they can bring to the study [41]. If the feasibility trial examines the content of the intervention to see if it is effective and feasible, the qualitative approach examines the context of the intervention to see if it can be accepted and applied in clinical practice and explain some of the quantitative findings [42]. Feasibility trials are important to ensure that larger randomized clinical trials are rigorous and feasible and economically justifiable [13].

Participants will be recruited from different sites to increase the representativeness of the target population in the region. The control group will not receive any motivational interviewing or follow-up to keep the standard aspect of therapy that corresponds most to the clinical reality. This will enable us to identify the added value of the VirTele program compared with GRASP. After outcome measures collection in T1 and T2, to compare the effect of each intervention within and between groups, we will collect additional measures at T3 and T4 to evaluate the retention of gains.

The multiple case study provides an in-depth description of stakeholders use experience and potential use of VirTele program. In this study, we will have two case groups: participants diagnosed with stroke and study therapists (physical therapists or occupational therapist). Each of these cases experiences the intervention differently, and it is therefore essential to report them through interviews. The variation of the cases makes it possible to increase the variation of the experiences and thus to increase the robustness of the qualitative results [19]. UTAUT and SDT will serve as a basis for establishing certain links between the concepts that will emerge (Multimedia Appendix 1).

The results of this study will allow to verify if all elements of the protocol work well together to conduct a broader future study. We have tried to provide as much detail as possible about the various processes and steps of the protocol to facilitate its reproduction by other studies that seek to develop tools for the remote management of chronic diseases.

This project will also provide preliminary evidence of the efficacy of VirTele on motor and functional recovery of the UE following chronic stroke for future guidelines review, although studies caution us about solely using results from feasibility studies to establish intervention efficacy [35].

Extending rehabilitation following a stroke with remote services may be a promising strategy to overcome the limited resources in the health system. The VirTele program is a new approach that may provide stroke survivors continuous and remote access to rehabilitation services. This paper describes the protocol underlying the study of this technology to better understand how it can be used among different stakeholders and explore its preliminary efficacy in a chronic stroke population.