Stroke can cause chronic sequelae in the hemiparetic upper extremity, requiring long-term rehabilitation care [1]. In Canada, only 11% of stroke survivors have access to outpatient and community-based rehabilitation after discharge from inpatient rehabilitation [2]. Stroke rehabilitation can help minimize impairments and promote independence during both the acute and chronic phases of stroke [3,4]. Hence, innovative community-based strategies are needed to provide adequate postrehabilitation services to stroke survivors in the community when these resources are not available. Consequently, virtual reality (VR) exergames are a promising alternative to traditional rehabilitation services, allowing continuous access to challenging exercises [5]. When combined with a telerehabilitation app [6], a follow-up with a clinician may be provided to adjust the difficulty level of the exergames and choice of exercises, according to the stroke survivor’s abilities and goals. The VirTele program—a program that combines VR exergames and a telerehabilitation app—was developed to provide stroke survivors with residual upper extremity deficits the opportunity to participate in a personalized rehabilitation program.

Motor and functional gains can be achieved during the chronic phase of stroke. However, it remains difficult to maintain these gains throughout time as they are essentially dependent on adherence to exercise guidelines and the availability of community-based resources, such as access to rehabilitation services, programs, and clinicians with stroke recovery expertise [4]. Therefore, rehabilitation programs developed for chronic stroke survivors should include strategies to encourage and motivate the use of the affected upper extremity through self-directed exercises and everyday activities [4,5]. Behavior change techniques (BCTs) were therefore incorporated into the VirTele program to empower users to continue exercising and using their upper extremities in everyday activities (eg, brushing their hair, getting dressed, and eating) [7]. Furthermore, there is increasing evidence that having a strong theoretical basis and using BCTs improves the efficacy of programs aimed at changing behaviors [7-9].

In light of current knowledge, the self-determination theory [10] was used to guide the clinician-stroke survivor interaction during videoconference sessions in the VirTele program. This theory stipulates that people tend to change their behavior when three of their needs are respected, including autonomy (the ability to organize their behavior according to their own aspirations and values), connectivity (the sense of belonging and the desire to be supported), and competencies (belief in their skills and abilities) [10]. Thus, the clinician was trained to consider these three psychological needs when interacting with a stroke survivor. In addition, BCTs (eg, 1.1. Goal setting, 1.4. Action planning, and 1.5. Review behavior goals) [7] and motivational techniques (open-ended questions, affirmation, reflective listening, and summary statements) [11] may be used to enhance stroke survivors’ motivation and adherence to VirTele exercises [12,13].

As we intend to implement the VirTele program in stroke survivors’ communities, factors that could influence behavioral intention and use behavior regarding the intervention must be explored. This study was conducted in the context of a larger study before starting a randomized clinical trial to inform the development of evaluation and intervention protocols [14]. The clinical trial was registered at ClinicalTrials.gov (NCT03759106).

The objectives of this study are (1) to determine the feasibility of using a telerehabilitation app combined with an exergame for a remote upper extremity rehabilitation program in a chronic stroke survivor; (2) explore the preliminary efficacy of VirTele on upper extremity motor function, the amount and quality of upper extremity use, and impact on quality of life and motivation; and (3) explore the determinants of behavioral intention and use behavior of VirTele along with indicators of empowerment.

A case study using a mixed methods design was conducted before the start of a randomized clinical trial. A quantitative approach was used to collect feasibility and preliminary efficacy data before and after the VirTele intervention and during a 1- and 2-month follow-up period. A qualitative approach was used to explore the factors that influenced the use of the technology and its impact on the participants’ empowerment. This study was approved by the research ethics boards of the Center for Interdisciplinary Research in Rehabilitation of Greater Montreal (June 28, 2018).

A 63-year-old male stroke survivor (3 years) from the United States participated in this study. His upper extremity score on the Chedoke McMaster Stroke Assessment Scale indicated moderate impairment (arm stage 6). The participant was no longer receiving rehabilitation services and had an active lifestyle (swimming, walking, and fitness training). He was very comfortable with computers as he had a professional information technology background and used a computer one or more times a week.

After reading about our registered protocol at ClinicalTrials.gov, he approached our research team via email. At that time, the development and testing phases of the rehabilitation program and technology with VirTele were still ongoing. The participant was enthusiastic about playing an expert role in the study and testing the VirTele remote rehabilitation program in a setting similar to that of the proposed research study. He was eager to help identify the best way to combine telerehabilitation and videogames for home use to optimize upper extremity functional and motor recovery following a stroke. In doing so, he would assist in developing a final VirTele rehabilitation training that could be used as an intervention model for our future randomized clinical trial. Given the remote nature of the VirTele program, we were able to conduct the study with the participant remotely. The participant agreed to contribute to preliminary data collection and will not be included in future clinical trials. He provided informed consent before starting the case study, as per the Centre for Interdisciplinary Research in Rehabilitation of Greater Montreal’s institutional ethics review board.

Motor and functional gains can be achieved during the chronic stroke phase; however, for long-term maintenance, rehabilitation programs should include strategies to encourage and motivate the use of the affected upper extremities through self-directed exercises and everyday activities [4,5]. Therefore, motivational interviewing based on BCTs was incorporated into the videoconference sessions to ensure that shared decision-making and empowerment were consistently integrated into the VirTele intervention. The BCTs included were as follows: 1.1. Goal setting, 1.4. Action planning, 1.5. Review behavior goals, 4.1. Instruction on how to perform the behavior, 8.1. Behavioral practice, 6.1. Demonstration of the behavior, 2.2. Feedback on behavior, 2.7. Feedback on outcomes of behavior, 1.6. Discrepancy between current behavior and goal, 1.2. Problem solving, 2.3. Self-monitoring of behavior, 9.2. Pros and cons, 5.1. Information about health consequences, and 15.1 Verbal persuasion about capability [7]. Motivational interviewing techniques included open-ended questions, affirmation, reflective listening, and summary statements [11].

In addition to exergames, the program included supplementary daily activities to meet participants’ individual goals, such as hand dexterity (eg, grasping, holding, and pinching) using the resources available within the participant’s home.

Thus, many tools available through the telerehabilitation app (document sharing) and the exergame (computerized participant logs such as active time, compensation, and scores during each game) were used by the clinician for encouragement, goal setting, tracking of upper extremity use in daily activities, monitoring of exergame adherence (duration, intensity, and type), and tailoring the exergames to the participant’s abilities.

In accordance with the self-determination theory [10], the clinician was trained to consider the three psychological needs when interacting with the stroke survivor by including him in every step of the decision-making process related to the treatment plan, respecting his autonomy, and valuing his skills and ability to achieve his goals.

The clinician received training in motivational interviewing before starting the intervention and used standardized interview guidelines during each videoconference session. The motivational interview guidelines were standardized to facilitate their use in further research and better monitor the use of BCTs. In that sense, predefined questions were prepared and applied concerning the stroke survivor’s three psychological needs (autonomy, competence, and connectivity). The choice of questions varied depending on the stroke survivor’s behavior state (total adherence or nonadherence to VirTele). The questions were provided as examples, but the clinician was free to reformulate them. Further details can be found in Multimedia Appendix 2.

Descriptive analyses were conducted for performance-based measures and self-reported data. The interview was audio-recorded, transcribed verbatim, and analyzed through the lens of the self-determination theory [10] and the UTAUT [40]. Content analysis was applied to identify the factors that influenced behavioral intention and use behavior of VirTele and significant indicators of empowerment that could result from the use of the VirTele program, including motivational strategies. The identified empowerment indicators were categorized into capacities and behaviors based on the conceptual model of patient empowerment [42]. When relevant, themes identified through the qualitative analysis were used to deepen and explain the quantitative findings.

For the scientific rigor of qualitative data, the principles of Lincoln and Guba [43] were applied. Audit trail and verification were performed to ensure confirmability. External verification by team members was applied to ensure credibility. Reliability was achieved through verification by 3 coders of a part of the data. For transferability, reflexive notes and a detailed description of the context of the intervention were provided.

The FMA-UE (motor function component) exhibited a significant increase in the total score from baseline (T1) to postintervention (T2; total score difference 6; Table 1), which is within the estimated minimal clinically important difference (MCID) range of 4.25 to 7.25, specifically for the upper extremity motor function component [44]. The FMA-UE total score was maintained during the follow-up period at T3 and T4 and may suggest a slight increase, although there were no significant changes beyond the postintervention results at T2 (Table 1).

The total score on the MAL–amount of use increased to 2.45 at T2, close to the score of 2.5 (Table 1), predicting 50% or greater perception of recovery in the affected upper extremity (based on a scale of 0% to 100% of SIS–section perceived recovery) [45]. During the follow-up period, improvement in the amount of use scale was maintained and exceeded the score of 2.5 at T3 and T4. The change in the MAL–quality of use at T2 (total score difference 0.86) was not clinically relevant (MCID is between 1.0 and 1.1 [31]) but was significantly different from baseline at T3 (total score difference 1.22) and T4 (total score difference 1.52; Table 1).

The SIS revealed improvements in hand function and activities of daily living dimensions with a meaningful difference from baseline to T4 (Table 1), exceeding both the MCID range of 9.4 to 14.1 [33]. There was no improvement in the SIS mobility dimension, with a decreasing tendency over time (Table 1).

The TSRQ showed a high score on the autonomous motivation subscale (score=42/42), which was maintained during the follow-up period at T3 and T4. The amotivation subscale showed a low score (score=3/21) from baseline to the follow-up period at T3 and increased by 1 point at T4 (score=4/21). The external regulation subscale showed an increase in score from baseline (T1) to postintervention (T2; score from 6 to 10/28), followed by a decrease at T3 (score=4/28) and an increasing tendency at T4 (score=8/28). The introjected regulation subscale showed a moderate score (score=8/14), which was maintained from baseline to postintervention T2 and decreased during the follow-up period.

We addressed three objectives in this case study: (1) determine the feasibility of using a telerehabilitation app combined with an exergame for a remote upper extremity rehabilitation program in a chronic stroke survivor; (2) explore the preliminary efficacy of VirTele on upper extremity motor function, amount and quality of use, impact on quality of life and motivation; and (3) explore the determinants of behavioral intention and use behavior of VirTele and indicators of participant empowerment.

During the 2-month intervention, the participant demonstrated complete adherence to VirTele, completing 48 sessions or 33 hours and 15 minutes of active games, surpassing the minimum 20 hours planned for the program. A minimum of 15 hours is suggested for an intervention aiming at a moderate improvement of activities related to daily living following a stroke [46]. The 48 sessions included six supervised game sessions with a clinician and 42 autonomous game sessions. In addition to the exercises provided, the participant participated in eight videoconference sessions with a clinician for monitoring and empowerment purposes.

Although the participant encountered some technical challenges related to interoperability, internet connection issues (interruption of the video or sound, poor connection), and the login procedure (forgotten password), he was able to find a solution and use both the exergame and the telerehabilitation app throughout the 2 months. Furthermore, the participant suggested some improvements for the avatar and background, which he felt would render the game more meaningful. In addition, he was more drawn to realistic scenes and avatars, which is consistent with the findings of a previous study that indicated that players preferred simulated real-world games [47]. The sense of presence in a virtual environment or the degree to which an individual perceives the environment as real may affect the participant’s engagement and motivation [48,49]. Thus, it is important to consider certain elements, such as the background, avatar design, and the scene’s realism, when developing a VR system for rehabilitation use [48]. Fortunately, the lack of these elements did not decrease the participant’s compliance with or commitment to performing the exercises, given that he spent 11 hours using the Space Race game.

Regarding the combined use of a telerehabilitation app and an exergame, the results of this case study suggest that the VirTele intervention and study protocol could be feasible with stroke participants, although it is important to ensure that it will be accessible to participants who may be less motivated and technologically savvy. Moreover, the participant appreciated the follow-up by the clinician when playing the exergames and described this room for maneuver in VirTele as the time to reflect. In fact, while the clinician was supervising, the participant was able to identify his errors during the exercises (compensatory movements), discuss the solutions, adjust the difficulty level as needed, and correct his posture and upper extremity movements. This may have prevented him from reproducing a pathological pattern of movement when playing alone. The participant exhibited clinically meaningful improvements at T2 in the motor function dimension of the FMA-UE and the hand function dimension of the SIS. Not surprisingly, there was no improvement in the SIS mobility dimension, which addresses the impact of stroke on lower extremity mobility, further indicating the impact of the intervention on upper extremity motor function. The FMA-UE (motor function) and the SIS (hand function) scores were maintained at T3 and T4, followed by a relevant increase in the SIS score (daily living activities), suggesting maintenance of motor gains throughout time, contrary to the findings of a previous study using VR exergames without supervision and a motivational approach [6].

During the follow-up period, there was a meaningful change in the MAL–amount of use subscale, suggesting more involvement of the affected upper extremity in daily activities, along with a meaningful improvement in the SIS activities of daily living dimension at T3 and the SIS hand function dimension at T4. These results suggest that a transfer pattern occurred between the virtual and daily life activities, as the participant showed meaningful improvements in daily life activities and amount of use of the upper extremity following 2 consecutive months after completing the VirTele program.

Although we had predetermined that 20 hours of active games within the VirTele program should be performed, the participant decided to do more (33 hours and 15 minutes). A much higher number of repetitions may have positively affected the upper extremity motor function. In fact, a meta-analysis of 20 randomized clinical trials, including 2686 stroke patients (acute, postacute, and chronic), supports the presence of a dose-response relationship between intense practice (therapy time ranged from 132 minutes to 6186 minutes) and better outcomes [50]. However, it is important to note that all users may not use VirTele at this level of intensity. More research is required to determine the amount of time required to obtain optimal results.

The MAL–quality of use subscale showed no change in movement quality at the end of the program but exhibited a meaningful improvement during the follow-up period. In fact, during the 2-month training, the participant did not use his affected hand sufficiently during daily life activities (pain and fear) and was mostly manipulating virtual objects during the exergame, with no force or touch feedback, which may have limited the integration of somatosensory information that contributes to motor learning [5]. During the follow-up period, the participant stopped using the exergame and was more involved in daily life activities, more frequently using his upper extremity to manipulate real-life objects, which involves somatosensory feedback that may favorably affect the movement quality of the affected upper extremity [5]. These results highlight the necessity of transferring virtual activities to real-life activities to avoid a potential gap in somatosensory information affluence.

The participant exhibited a lot of enthusiasm from the day he contacted our research team and demonstrated a high level of autonomous motivation (score=42/42) throughout the entire intervention, which may partly explain the high level of adherence to the VirTele rehabilitation program. The TSRQ showed a low level of external regulation, meaning that the participant made autonomous choices regarding his use of VirTele and completion of the exercises and was not influenced by certain consequences (“I want others to approve of me”) [5] or pressure (“I feel pressure from others to do so”) [34]. However, the external regulation score increased from 6 to 10 during the VirTele intervention. This could have possibly resulted from the desire to satisfy the research team or the clinician, leading to short-term changes in the participant’s behavior (eg, better adherence to VirTele and higher motivation), improving upper extremity motor scores. However, the long-term maintenance of motor gains involves continuous adherence to exercises (after the end of VirTele) and maintaining motivation despite the lack of external factors (eg, encouragement by the clinician).

Similarly, the TSRQ showed a medium level of introjected regulation (score=8/14), meaning that the participant partially introjected and self-identified with the values linked to the VirTele rehabilitation program. This suggests that the participant possibly integrated the potential benefits of the exercises with his affected upper extremity and identified aspects of the VirTele program that could match his own values and beliefs, potentially affecting the long-term behavioral use of the technology.

Performance, effort, and social influence were meaningfully weighed in the behavioral intention use of VirTele. The relative advantages of the VirTele program compared with standard therapy and the perceived improvement in daily life activities facilitated the use performance of the VirTele program. The participant’s level of comfort with the technology positively influenced the effort determinant. The additional constructs of the UTAUT model were not identified as direct determinants of behavioral intention or use behavior. According to Venkatesh et al [51], attitudinal constructs may indirectly affect behavioral intention and use behavior through performance and effort constructs.

In fact, the participant’s attitude toward VirTele, the affected upper extremity, and information technology indirectly influenced behavioral intention through a positive and negative impact on effort and performance determinants. Owing to the participant’s experience in information technology, he may have been more sensitive to some aspects of the exergame design (eg, avatar and background) and the operating system within the technology, although these aspects did not affect his compliance with the VirTele program.

The participant demonstrated considerable empowerment on the behavior and capacity levels, which may have positively affected his adherence to the VirTele rehabilitation program and explain the meaningful improvement in many clinical outcomes. The self-determination theory was used to inform the VirTele program by identifying the main constructs that target the desired behaviors (increase in adherence to the VirTele program and increased use of the affected upper extremity in daily life activities) and providing the means for developing component intervention techniques such as motivational interviewing. Health behavior change literature mentions numerous advantages of applying theory to interventions [52-54], including stronger effects [52], which may corroborate the findings of our study.

Other components of the VirTele program that may have also contributed to the participant’s empowerment include the autonomous access to a personalized training program, the motivational aspect of VR exergames (eg, augmented feedback and challenging exercises), shared decision-making using the combination of VR and telerehabilitation, health counseling, and the ability to get real-time feedback during direct or web-based supervision (promoting the desired behavioral result). The extent to which the different VirTele program components played a role in the improvements observed in the clinical outcomes or empowerment indicators is still unclear.

First, the FMA-UE results should be carefully interpreted, as the remote administration of the scale without the presence of an evaluator on site with the participant has not been validated in the literature. However, the FMA-UE (motor function) results in this case study were consistent with the self-reported findings (SIS and MAL) and provided an indication of the affected upper extremity motor function pattern during the intervention. The literature lacks a scale that remotely assesses the motor function of the arm in stroke survivors and is extremely relevant to the current need for telerehabilitation interventions.

Second, the participant was already highly motivated when he started the intervention. This may have naturally increased his adherence to the VirTele rehabilitation program and, consequently, favor his clinical outcomes. However, this level of adherence may not be representative of all participants.

In addition, with only 1 participant, it is difficult to capture all the factors that could influence the behavioral intention and use behavior of VirTele, such as technology familiarity, motivation, and previous experiences. The findings gathered so far are limited to the participant’s own experiences, thereby limiting their transferability to other contexts.

The participant was also very comfortable with the technology and easily handled the technical difficulties that occurred during the intervention. This may not represent future users, although the participant’s input during the development phase allowed the team to address potential future technological issues. This is crucial because it is highly likely that many participants affected by a stroke would be older and possibly less comfortable with new information technologies [55].

Finally, we cannot predict how well other participants would manage technical problems or how comfortable they would be with the technology used in the VirTele program. The technical difficulties are considered a direct determinant factor of intervention adherence and may influence a participant’s willingness to use the technology. Technical problems should indeed be considered as a part of technology use, and end users should be technically (where to click and what this signal means) and emotionally (do not panic, you do not need to be frustrated, you did nothing wrong, and that is not your fault) prepared to better manage various situations. Accordingly, a video and informative pamphlet, including simple, concise instructions on how to open and use the exergames and videoconference platform, could be prepared to support VirTele use among future participants.

It is important to explore the extent to which technological knowledge and experience play a role in using the latest technology. In addition, further research is needed to explore the role of shared decision-making and empowerment in exercise program adherence and progression and behavior modification for upper extremity use.

The data collected so far may not be representative of stroke survivors aged 63 years. However, the feasibility and preliminary efficacy results are promising and support the use of the VirTele rehabilitation program and have contributed to improving the intervention and study methodology. Further studies are warranted to corroborate these preliminary findings and test the feasibility and efficacy of VirTele in a diverse stroke population.

The results of this case study suggest that it is relevant to continue investigating the use of exergames provided by telerehabilitation to improve upper extremity motor function and enhance the upper extremity quality and amount of use in activities of daily living.