Adolescents and adults with cerebral palsy (CP) are less physically active and spend more time sedentary than their typically developing peers.¹ This is associated with greater risks for physical, metabolic, and social ill-health.¹ Being physically active as an adolescent with CP doubles the probability of being physically active as an adult,² which may reduce the risk of ill-health in adulthood. A systematic review of 49 studies exploring the impact of leisure time physical activity for children and adults with CP reported improved health, fitness, and physical function outcomes following a range of physical activity interventions.³ However, the available evidence was gleaned primarily from ambulant children (those classified in Gross Motor Function Classification System [GMFCS] levels I–III).⁴ Adults, wheelchair users, and those with motor function classified in GMFCS levels IV and V were vastly under-represented,³ with only seven of the 49 reviewed studies including adults, and only 5% of the 1513 participants having gross motor classifications in GMFCS levels IV or V.

Given the substantial lifespan effects of inactivity on motor, metabolic, cognitive, and psychosocial health, it is important that appropriate physical activities can be accessed safely by people with CP of all ages and complexities. Adults with CP classified in GMFCS levels IV and V are at even greater risk of chronic health conditions such as cardiovascular disease and diabetes mellitus,⁵ and frequently experience premature mobility decline.⁶ It is therefore imperative to identify suitable physical activity interventions in settings acceptable with this group, to address their significant health needs.

A recent mixed-methods systematic review synthesized the barriers and facilitators of physical activity participation for adolescents and adults with childhood-onset physical disability.⁷ The authors concluded that participating in physical activity was enhanced through positive social connections, social support, and an appropriate and accessible physical environment.⁷ However, it was noted people with physical disabilities expressed doubts about their capabilities, fear of injury and pain, concerns about the energy they require to be physically active, and the impact of exercise on fatigue.⁷ Unsurprisingly, those with more complex physical disabilities, such as people with CP classified in GMFCS levels IV and V, described not engaging in physical activity because of their own concerns around their safety and health, or because of the concerns around perceived fragility expressed by their immediate social networks or other stakeholders (e.g. family, personal trainers). Health professionals have also acknowledged fear as a barrier to physical activity participation in adults with disability,⁸ which can perpetuate the perception that people with severe disability should not participate in physical activity owing to unspecified risk. While all physical activity incurs some risk, identification of the nature of any risk, and potential strategies to ameliorate such risk, would be useful to enhance physical activity opportunities and engagement for people with complex CP.

Key to physical activity participation for young people with disability are inclusion and involvement in the community.⁷ For people with disabilities, this fosters positive mental health, enhances physical health benefits (such as optimizing functional independence), provides opportunities for social relationships, and improves quality of life. Furthermore, young adults with CP express a desire to move away from engaging with therapy interventions in a clinical environment⁹ to participating in non-healthcare settings, to maximize their social interactions and facilitate social connectedness. However, ensuring safety while exercising in community settings for those with complex disability is important to counter fear and perceptions of fragility. Understanding the initiatives that can be used to support people with complex forms of CP to participate safely in community-based physical activity interventions can inform best practice. This will lead to better health outcomes and increased community-based physical activity participation for those with complex CP. The aims of this scoping review were therefore to identify implementation strategies and safety outcomes (i.e. presence or not of adverse events) following community-based physical activity interventions for adolescents and adults with complex CP.

METHOD

This scoping review was guided by the methodological framework proposed by Arksey and O'Malley,¹⁰ the key stages of which are (1) identification of the research question; (2) identification of relevant studies; (3) study selection; (4) data charting; and (5) collation, summarizing, and reporting of results. This protocol was registered with the Open Science Framework on 26th May 2022 and is reported here according to the PRISMA Extension for Scoping Reviews guidelines.¹¹

RESULTS

A systematic search returned 3419 items for screening (Figure S1). Once duplicates were removed and abstracts screened, 85 full-text articles were assessed for eligibility. Seventeen studies with 262 participants, 124 (47%) females, and 160 (61%) participants with gross motor function classified in GMFCS levels IV or V were included (Table 1). Six of these studies included only participants classified in GMFCS levels IV and V while the remaining 11 studies also included participants with less severe impairments (GMFCS levels I, II, and III) (Table 1). There were 11 studies where participants had a mean age of 18 years or younger; one study where participants had a mean age in young adulthood (20 years); and five studies where participants had mean ages in middle adulthood (range 36–50 years).

The research designs used in the reviewed studies were randomized controlled trials (n = 5), non-randomized controlled trials (n = 1), randomized crossover trial (n = 1), single-subject design (n = 5), pre–post test studies (n = 4), and a repeated measures study (n = 1). Study sample sizes were generally small: the largest trial enrolled 36 participants, 13 of whom had gross motor classifications in GMFCS level IV.¹⁴ The community settings for the studies included schools (n = 4), participants' homes (n = 3), gymnasia (n = 2), swimming pools (n = 2), an indoor track and field facility (n = 1), a day programme (n = 1), and other community settings (n = 2). Sixteen studies were completed in developed countries, and one in a developing country (Brazil).¹⁵ Most studies had been published in the previous 5 years (10 of 17 studies).

All included studies specified exclusion criteria, with one exception.¹⁶ Exclusion criteria were predominantly related to recent or upcoming medical management, specific medical comorbidities, and predicted ability to participate. Potential participants were often excluded if they had received surgery in the previous 6 to 12 months (n = 12 studies), botulinum neurotoxin A in the previous 3 to 6 months (n = 4 studies), or had sustained a fracture in the previous 6 to 12 months (n = 4 studies). Three studies excluded those who had upcoming scheduled surgery (e.g. single-event multilevel surgery or correction of scoliosis) or another planned intervention (e.g. serial casting or injections of botulinum neurotoxin A).

Medical comorbidities relevant to exercise participation were exclusion criteria in 11 studies. Comorbidities included known cardiovascular or cardiorespiratory disease (n = 4 studies), with two studies specifying exclusion if the condition was severe (e.g. symptomatic, requiring medication, or hospitalization in the previous 6 months) and posed a safety risk during exercise training; medical comorbidities common in CP such as uncontrolled epilepsy (n = 4 studies), selective dorsal rhizotomy, intrathecal baclofen, or metal and other implants (e.g. spinal rods, plates, screws, ventriculoperitoneal shunt, percutaneous endoscopic gastrostomy, stoma); medications that could affect physiological responses to exercise (e.g. beta blockers); and other medical conditions (e.g. reflux, vomiting, hernia, retinal detachment, prolapse, haemorrhagic conditions, haemophilia). Eight studies specified musculoskeletal impairments or pain as exclusion criteria, particularly where it was believed it might prevent exercise participation such as severely reduced range of motion (joint contractures), joint instability, or dislocation, particularly of the hip (migration percentage > 30%), deformity of the lower limbs, atlanto-axial instability, spondylolisthesis, hypermobility, and severe spasticity. However, one study did not exclude participants with musculoskeletal impairments as long as they were able to walk 10 m.¹⁷

Potential participants were also excluded on the basis of safety concerns, specifically where there was concern about their inability to reliably understand or follow directions (n = 8), or if they had a severe communication impairment (n = 1) or psychiatric impairment (n = 1). Two studies where the setting was a swimming pool specified exclusion criteria based on safety; these were ongoing compromised skin integrity, significant dysphagia or oromotor impairments, or a hearing impairment. A further two studies specifically targeted novice exercisers by excluding those with previous exercise experience.^{17, 18}

Most study interventions targeted body functions (e.g. cardiorespiratory fitness or muscle strength; n = 10 studies) and/or activity limitations (e.g. gross motor function, transfers, walking; n = 11 studies). A smaller number of studies targeted participation in physical activity (n = 2 studies), improvements in pain and fatigue (n = 1 study), social skills (n = 1 study), or specific physical activity skills (e.g. swimming; n = 2 studies). The modes of exercise used were static, recumbent, or adaptive cycling (n = 5 studies), treadmill (n = 2 studies), weight machines/free weights (n = 2 studies), swimming (n = 2 studies), dance (n = 2 studies), rowing (n = 1 study), race running frame (n = 1 study), trampolining (n = 1 study), physical activities either embedded in daily life (n = 1 study) or performed as part of a group exercise class with weights and exercise bands (n = 1 study). The median intervention duration was 10 weeks (range 1–52 weeks), the median frequency was three times per week (range two to seven times per week), and the mean activity session duration was 30 minutes (range 25–90 minutes). Exercise was progressed during the intervention period in all studies where this was reported and overall attendance at scheduled sessions was high (Table 2).

All included studies implemented safety strategies as part of their intervention protocols (Table 3). Exercise was delivered as an individual programme (n = 10 studies), as a group programme (n = 6 studies), or with both group and individual components (n = 1 study).¹⁹ The intervention in 13 studies was either delivered by a health or exercise professional (e.g. physiotherapist or exercise physiologist) or overseen by a health professional. Most studies reported the involvement of an additional assistant(s) (e.g. student, family member, or volunteer) for implementation. In-session physiological monitoring was a feature of the interventions in nine studies, most completed by heart rate monitoring. Only one study formally evaluated fatigue using a recognized fatigue survey (Pediatric Quality of Life Inventory multidimensional fatigue scale),²⁰ and three studies included formal evaluation of pain.^20-22 Other implemented safety strategies were pre-exercise screening, a familiarization phase, supervision, education, and providing physical assistance (Table 3).

Attrition in trials was relatively low (0–23% where reported) (Table 4). Eleven studies explicitly reported whether adverse events occurred, and whether these events were related to physical activity (Table 4). Two of these studies reported no serious adverse events, injuries, or minor incidents.^{23, 24} Several studies reported minor incidents of soreness (n = 4 studies), fatigue (n = 4 studies), skin irritation (n = 2 studies), discomfort due to a harness that resolved by adjusting it and adding extra padding (n = 1 study), muscle spasms that resolved with stretching (n = 1 study), and a low-speed non-injurious fall (n = 1 study). A few adverse events that affected attendance or trial participation were reported. Five studies reported instances where participants described experiencing pain during training: in one study, two participants reported increased pain during training that did not require review by a physician;²¹ in four studies, participants reported pain that required them to cease exercising for a short period¹⁷ or resulted in them withdrawing from the trial.^{20, 25, 26}

DISCUSSION

Our findings indicate that participation in physical activity interventions for most adolescents and adults with CP classified in GMFCS levels IV and V is safe and feasible. The findings endorse the capacity of people with more severe CP to participate in community-based physical activity across different modes and in a variety of community venues. Their attendance at scheduled sessions was high (around 80% or more) and attrition relatively low (most commonly 0%). Furthermore, most interventions were designed to be delivered a median of three times a week, for a median of 10 weeks, for around 30 minutes per session; a dose that aligns with exercise and physical activity recommendations for people with CP.²⁷

An important finding was the low number of adverse events reported and that most participants did not report an increase in pain or fatigue after physical activity. Only 4 of 160 participants (<2%) either withdrew from an exercise trial (three participants) or ceased exercise for a short period (one participant) because of pain. Previous studies have reported participants with CP describing a fear of injury, or new or increased pain or fatigue associated with physical activity, limiting or preventing their engagement. As many as 70% of adults with CP report living with chronic pain,²⁸ and around 40% experience daily fatigue,²⁹ so this hesitation to increase their physical activity is understandable. There is a high likelihood that anyone who is new to exercise will experience muscle soreness post-training initially, and this was the case for those with CP classified in GMFCS levels IV and V; six studies reported ‘post-training muscle soreness’ or ‘minor symptoms of soreness’ which quickly resolved without medical review and did not preclude continuing participation. Similarly, four studies reported some participants experiencing ‘minor symptoms of fatigue’ but this did not affect attendance or participation in the intervention. This low frequency of minor adverse events may suggest that factors other than the physiological response influence physical activity participation of people with complex CP. Adults with CP have a substantially greater risk of cardiovascular and respiratory disease than adults without CP³⁰ and this risk is even further elevated in those with more complex disability.³¹ The review findings suggest physical activity, when implemented with specific safety strategies, is feasible to address the elevated health risks in people with complex CP.

Adolescents and adults with cognitive and communication impairments were frequently excluded from trials of physical activity interventions. Eight studies specified minimum cognitive or communication skills, or a requirement to be able to follow instructions as an inclusion criterion. Intellectual disability has been reported as being present in 45% of people with CP, increasing to around 83% in people with CP classified in GMFCS levels IV and V.³² Given the focus of this review was on identifying the implementation and safety of physical activity interventions in those with CP classified in GMFCS levels IV and V, potentially many people were excluded from the interventions evaluated. Similarly, the prevalence of epilepsy and other medical complexities is also highest in people with CP in GMFCS levels IV and V,³² and these were common exclusion criteria, even though physical activity is not contraindicated in epilepsy and many other medical conditions. In the limitations of the current literature (17 studies with 262 participants), much of which has systematically excluded those with significant cognitive and communication impairments, epilepsy, or other medical conditions, the risk of serious adverse events from community-based exercise was very low (<2%). We recommend future studies ensure their samples are generalizable to the whole population of people with CP classified in GMFCS levels IV and V, to advance understanding of the benefits of physical activity in these groups and to ensure opportunities for participation in physical activity programmes.

In all studies, supervision or physical assistance was included as a component of the intervention, which meant the adolescent or adult with complex CP was rarely an independent exerciser. Overall, the nature of the supervision or physical assistance supports actually required was poorly reported, other than the professional background of the person providing the support (often a physiotherapist) and in some instances the staff to participant ratio. What remains unknown is whether supervision is a requirement for safe participation; that is, can the low numbers of adverse events reported be attributed to the specialist physical assistance and supervision provided or does the unspecified ‘risk’ remain low without assistance and supervision? This is particularly important as young people with CP have expressed their desire to exercise in community environments, where there may be few support staff available, and where staff may have less disability expertise than health professionals. We support that people with complex CP be afforded the dignity of risk in community settings; that is, the opportunity for personal growth through exploration and engagement in self-determined, reasonable risk-taking.³³ Overall risk associated with exercise may be reduced through supervision, but the opportunity to take reasonable risk should not be precluded through exclusion. We advocate for future research to extend understanding in this area.

This review demonstrated successful access to physical activity programmes offered in educational facilities, community swimming pools and gymnasia, an indoor track and field facility, people's homes, and other community environments for adolescents and adults with complex CP. These facilities were generally accessible to the wider public. The physical environment has been reported as a major barrier to participation in community-based physical activity for people with CP.⁷ A recent qualitative study³⁴ reported ‘accessible and accommodating environments’ took many forms, including physical access, practical supports, reasonable accommodations/adjustments, and positive attitudes towards disability. Another recent study described the supports required for people with complex CP to successfully access a community-based exercise facility, and highlighted significant additional time requirements from all stakeholders.³⁵ Adapted equipment is frequently needed by people with disabilities which may limit their ability to engage without initial assistance or financial support to explore or implement equipment solutions or both. Options for sustainable community physical activity for adults with complex CP therefore requires not only physically accessible venues but also those where reasonable adjustments are made to the environment, and where staff and other participants are accepting, supportive, and not time-limited.

Studies that included participants classified in GMFCS levels IV and V but were completed in a health setting or a non-community venue such as a research laboratory were not included in this review. Additionally, grey literature was excluded. It was estimated that fewer than 10 participants (enrolled in three included studies) had not yet reached adolescence (i.e. were younger than 10 years of age at the time of enrolment); however, this represents fewer than 4% of all participants enrolled in the 17 included studies. Finally, all studies in this review reported time-limited physical activity programmes. Any additional adverse events or risks associated with engaging in sustained physical activity programmes throughout the lifespan are unknown.

This study confirms that among previously sedentary adolescents and adults with CP with significant motor-impairment, progressive physical activity can be performed in community settings with supervision, without excessive post-exercise pain or fatigue, or serious adverse events. We posit that the known health benefits of physical activity, including limiting the development and progression of secondary multisystem disease, outweighs the small risk of adverse outcomes in this group. We encourage further physical-activity-based research in this population to maximize representation and generalizability of findings.

Funding information

SC is supported through the Centre of Research Excellence by the National Health and Medical Research Council (grant APP1171758).