Stroke is a worldwide health issue and one of the main causes of death and disability (1). In Canada, it is estimated that half of stroke survivors are left with disability that requires intensive rehabilitation, long-term assistance for daily activities and support in the community (2). As populations live to older ages, stroke prevalence and the long-term deficits associated with this chronic condition are also expected to increase considerably (3). These long-term deficits result in limited participation, mostly regarding mobility outside the home and participation in leisure and community activities (4). Restrictions in participation is a major concern post-stroke as it is fundamental to an individual's well-being and life satisfaction (5, 6). Participation in physical activity following stroke was shown to improve independence in activity of daily living, general health, and quality of life (7, 8). Despite these known benefits, however, most individuals with stroke, including those who sustained a mild stroke, do not return to pre-injury, high-demand leisure activities that involves physical activity (9), ultimately leading to an engagement in more ‘passive’ activities and sedentary lifestyle (10, 11).

The physical benefits of cycling as a physical activity have been well researched. In the literature, stationary cycling is one of the predominant exercises used among stroke survivors to improve aerobic capacity (12, 13) as well as motor control and muscle strength in the paretic lower extremity (14, 15). Cycling outdoors, however, differs in many ways from stationary cycling, not only in terms of sensorimotor and cognitive demands, but also as a multifaceted leisure activity and a mode of transportation (16, 17). In Canada, where the study presented in this manuscript was conducted, outdoor cycling is the second most popular outdoor leisure activity (18), with approximately 40% of the Canadian population aged 12 or older riding a bicycle at least once a year (19). In some European countries such as the Netherlands, where there are 1.3 times more bicycle than people, 28% of the population use a bicycle as its primary mode of transportation (20). In that same country, older adults aged between 65 and 74 years cycled on average 24 km per week in 2020 (21). Beyond the positive effects of exercise on mood and well-being (22, 23), the open environment, fresh air and scenic views associated with outdoor cycling offer a more dynamic and engaging experience and were suggested to be factor that also contributes to the well-being of older adults practicing outdoor cycling (24). It was also shown that more time spent outdoor at daytime is associated with better mood and sleep (25). Furthermore, outdoor cycling presents the advantage of continuously challenging balance by negotiating multiple and changing environmental factors (such as terrain variation, wind resistance and road condition), requiring increased efficiency and adaptability in terms of postural adjustments compared to stationary cycling. In fact, compared to non-cyclists, outdoor healthy adult cyclists have been shown to display enhanced postural strategies, leading to higher stability (26). Cycling outdoor could therefore be beneficial to improve balance following stroke. Another significant benefit of cycling outdoor is the active engagement in a leisure activity, which was shown to contribute to increased psychological well-being, including quality of life and mood in stroke survivors (27). Cycling within a group, when possible, could also increase social interactions as well as community participation, which is associated with reduced depression and isolation (28, 29). Group intervention is also an opportunity to learn from others. In fact, evidence show that the mirror neuron system can be activated in response to observation of actions performed by other people and therefore promote motor function recovery (30). Watching and imitating someone performing a cycling skill could therefore potentially stimulate one's learning abilities. Altogether, these finding advocate for group cycling outdoor as a mean to promote physical and mental health, both of which are affected following stroke (31). Such activity further has the potential to be sustained over the long term, thereby maximizing the emergence and maintenance of health benefits.

Currently, existing protocols and guidelines for outdoor bicycle training target healthy adult populations (32, 33). However, there is a major gap in the literature regarding such interventions for individuals with stroke. Meanwhile, interventions geared at improving cycling proficiency were successfully applied to various populations such as typically-developing children (34) and children with intellectual disability (35), as well as healthy adults primarily aged between 25 and 74 years (16). Individuals with stroke, however, present unique challenges due to persistent sensorimotor, cognitive, and perceptual impairments. These deficits notably impact posture, balance and movement coordination, as well as the ability to respond to environmental demands (e.g., obstacles, curb, etc.) in a timely fashion, all of which are essential for safe outdoor cycling (36). As such, these impairments not only limit the ability to return to cycling but also represent key rehabilitation targets that outdoor cycling could help address. Engaging in a structured outdoor cycling program may therefore provide a task-specific and ecologically-valid rehabilitation context to improve post-stroke impairments, while also facilitating reintegration into leisure and community activities. Individuals with stroke would therefore need a tailored program with even greater emphasis on safety precautions, given the consequences of falls on participation, quality of life and even mortality (37, 38). Moreover, unlike children learning the basic skills of cycling for the first time, stroke survivors build on previously acquired skills and must adapt their strategies to their current level of ability. Given these specific challenges, there is a need for a real-world outdoor cycling intervention that is grounded in established rehabilitation principles and tailored to the unique needs of individuals following stroke. Such an intervention should be informed by recognized principles of motor learning, including repetition, task specificity, progression of difficulty, feedback and saliency, all of which have been shown to support neuroplasticity and motor recovery (39). It should also align with the International Classification of Functioning, Disability and Health (ICF) by targeting not only impairments in body functions (e.g., balance, motor control), but also limitations in activity performance (e.g., cycling abilities) and participation (e.g., community mobility and leisure involvement). The present study was therefore designed to develop a structured, stroke-specific outdoor cycling protocol that integrates these theoretical foundations to promote both functional recovery and meaningful engagement in everyday life.

The current study also emerged from a request from rehabilitation therapists at our clinical site who, in the absence of a standardized training protocol and guidelines, were unsure of how to effectively and safely address the spontaneous demands of individuals with stroke to get support to return to outdoor cycling. The primary objective of this study was thus to develop and refine a group-based training protocol aimed at enhancing outdoor cycling skills in individuals with post-stroke hemiparesis. The secondary objective was to test the feasibility of the protocol in terms of its acceptability and the limited efficacy of the intervention on participant's personal cycling goals, self-reported confidence in cycling skills, balance, mobility, and participation in leisure activities. The constructs of acceptability and limited efficacy were identified as key areas to determine whether the intervention is worth testing in a future efficacy study (40).

A standardized cycling training protocol was successfully developed by our research team and further refined during the second phase of the study by adding a fourth module to address more complex tasks requiring additional training from the participants. The detailed printable version of the program is presented in Supplementary Appendix B. Characteristics of participants who were included in the study are presented in Table 3. The sample consisted of individuals aged between 25 and 76 years (mean age of 63 years), predominantly males (n = 4), who had a stroke 4–43 months previously which, in the majority, took place in the right hemisphere (n = 4). Chedoke-McMaster Stroke Assessment scores ranged between 3 and 6 for the arm and hand, and between 3 and 7 for the leg and foot, indicating a wide range of severity of motor deficits. Berg balance scores at study enrollment varied between 49 and 56 (average of 52), reflecting moderate to high balance abilities (45, 55). Out of the five participants, only two had attempted bicycling after their stroke, one of them using a recumbent bike.

The primary purpose of this study was to develop a standardized cycling group intervention protocol for individuals who had a stroke, and to make this protocol available to the clinical community and researchers. The protocol developed includes a detailed description of the tasks to be accomplished, which are presented in a logically progressive order, along criteria for progression. This protocol, developed in response to the needs of both rehabilitation therapists and their patients with stroke, proposes a short-duration, intensive group intervention tailored to the reality of the rehabilitation hospital with an outpatient stroke program. It can be easily adapted in terms of format, frequency, and duration to suit patient needs, endurance levels, and available treatment time.

This study additionally examined the feasibility of the bicycling intervention. Overall, results support its feasibility, as demonstrated by the excellent adherence, whereby participants attended every single session, with the exception of one participant who missed one session due to unexpected circumstances. Feasibility is further supported by the high acceptability of the intervention by all participants who gave the highest satisfaction ratings to all elements of the intervention they were questioned on, including elements such as training procedures, progression, patient/therapist ratio, frequency, intensity and length of the intervention. The intervention was also perceived as providing the right challenge to participants, which aligns with the fine progression of participants through various tasks of increasing complexity during the intervention. In fact, given that some of the participants progressed through the modules at a rate that was faster than anticipated, and given the need to address more complex cycling skills, a fourth module was created and tested.

Results also show that by the end of the intervention, four out of the five participants biked outdoor independently in a safe and effective manner and would not require supervision, which further support the feasibility of the intervention. The progression noted during the intervention was further corroborated by significant improvements in self-perceived confidence in bicycling skills and the fact that four out of five participants exceeded their personal cycling goal post intervention. Importantly, gains in terms of confidence and goal achievement were maintained at follow-up. Current literature on motor recovery after stroke supports the importance of developing training approaches that provides sufficient repetition to induce training effects, while being meaningful, challenging and tailored to the individual's capacity (56, 57), and we believe these elements, which were also highlighted in the participants’ feedback, were key to the success of the intervention. The group setting was another potential contributing element as it was brought up as a positive aspect of the intervention by most participants, especially when it comes to receiving encouragement from other participants and observing others which was considered inspiring and uplifting. While group interventions have been suggested to improve adherence and motivation, consistent evidence supporting this assertion, as well as the superiority of group vs. individual interventions in terms of health outcome improvements following stroke, has yet to be demonstrated (58).

Results also revealed that the intervention had no significant impact on outcomes of balance and mobility, with the notable exception of the CogTUG, and no impact on overall leisure participation. The absence of improvements on the Mini-BESTest and the single task version of the TUG could be due, in part, to the lack of alignment between the tasks involved in those assessments (largely related to locomotor and/or static skills) and the bicycling skills targeted in the intervention, to the short duration of the intervention, but also to the fact that participants involved in this study generally showed good balance and mobility skills at baseline (see participant characteristics in Table 3), with three participants displaying the maximal score on the Berg Balance Test and three displaying a TUG time within normative values (e.g., ≤10 s) (59, 60). As for the significant improvements observed on the CogTUG over the short duration of the training, it is reasonable to assume that they were induced by the multiple cognitive demands imposed by the practice of outdoor cycling during the training (i.e., awareness of the surrounding environment, planification of routes, mental flexibility to adapt to unexpected constraints, etc.). While the lack of improvement on the Nottingham Leisure Questionnaire may not be that surprising, considering the wide large variety of leisure activities covered by this questionnaire (from watching TV to practicing sports), it is worth mentioning that four participants continued to bike between the end of the intervention and the follow-up evaluation, which suggests that the intervention has the potential to lead to sustainable biking practice.

Last but not the least, this feasibility study intended to monitor adverse events. Two falls with minor sequels were reported, which while unfortunate was not entirely unexpected given the nature of the training. Safety was a major concern when designing this intervention and strategies were incorporated in the training to minimize falls, collisions and injuries. These include, amongst others, a safety checklist for therapist(s) to use, constant supervision and/or assistance of one or two therapists and the use of safety walking belts as needed. The choice of outdoor training environment also received special considerations and, for the present feasibility study, the parking lot (dedicated area delimited by cones, free of cars or unwanted obstacles) and surrounding area of the rehabilitation hospital was found to be a safe area yet sufficiently challenging to train the participants. This choice of environment may vary across different rehabilitation settings and would need to be determined on a case-by-case basis. To maximize safety during training but also for the independent practice of outdoor cycling, loss of balance and fall avoidance techniques were incorporated in Module 4. Participants were also screened in terms of their motor recovery, visual-perceptual abilities, cognitive function and balance capacity (see inclusion and exclusion criteria) to minimize the risk of falls. From a more global perspective, and with respect to the practice of cycling as an activity in everyday life, studies suggest that health benefits of cycling in the general population largely outweigh the risks associated with traffic accidents and inhaled air pollution (61) The increased accessibility to electrical bicycles to the general consumers, including older adults with or without stroke, may to some extent mitigate those benefits and pose new challenges in terms safety and approaches from clinicians.

This study has limitations, including a small sample size and a gender imbalance in recruited participants (4 males, 1 female), as well as conservative inclusion criteria in relation to the secondary objective of the study on feasibility, which limits the generalizability of the findings to the broader stroke population, particularly those with more severe sensorimotor and/or cognitive impairments who could potentially benefit from the activity with the appropriate adaptations (e.g., adapted bicycle, additional assistance and supervision). The sample size was limited by seasonal considerations in Canada and the availability of participants wanting to resume cycling during that period. The latter was likely due to a lack of awareness about the possibility of resuming cycling following stroke and the novelty of the intervention, which was then largely unknown from potential participants. Further studies involving larger sample sizes will be needed to support elements such as the intervention's efficacy, and to identifying who are the potential responders to this intervention. In addition, while the study shows that a three-week training was sufficient for participants to improve their confidence in their cycling skills and to achieve or even surpass their cycling goals, it cannot be ruled out that a longer intervention duration might have yielded even better outcomes.

In response to a need expressed by clinicians and their patients, as well as a significant gap in the literature, this study developed and tested a structured bicycle training program for stroke survivors with hemiparesis, which is now available to the community through this manuscript. The training protocol was found to be feasible, as demonstrated by participant adherence, progression, minimal occurrence of adverse events, and improvements in cycling-related outcomes. Further research could focus on determining the longer-term impact of outdoor cycling on health-related and functional outcomes, as well as identifying the minimum criteria for inclusion in such a training program following a stroke.