Identifying and prioritising strategies to optimise community gym participation for young adults with cerebral palsy: an e-Delphi study

Abstract Purpose Identify and prioritise strategies to optimise physical activity participation in the community gym setting for young adults with cerebral palsy. Methods An e-Delphi method was implemented over three rounds with four stakeholder groups (young adults with cerebral palsy, their families, gym staff or exercise professionals, and health professionals). Strategies for change were identified by the stakeholders in round 1. In rounds 2 and 3, strategies for change were rated on the importance for implementation using a 7-point Likert scale (1 being lowest importance, 7 being highest). The consensus was achieved if ≥70% of participants identified a strategy as high importance. Results Seventy participants (20 young adults 10 family members, 21 health professionals, and 19 exercise professionals) identified 83 strategies for improving gym participation. Of these, 44 strategies met the consensus criteria. The highest priority strategies related to changing the physical environment, addressing cost barriers, gym staff training, and developing partnerships between sectors. Conclusions Addressing physical accessibility, cost of attendance and the skills of gym staff were agreed upon by the stakeholder groups as priority areas for future resource allocation and research translation. Clinicians and community leisure facilities must work with consumers to implement strategies in their local contexts. IMPLICATIONS FOR REHABILITATION The physical environment, gym staff training, and the cost of attendance are the priority areas for future interventions agreed on by key stakeholder groups Health professionals can facilitate community participation by partnering with the recreation and research sectors to create pathways to gym exercise Health professionals can play a role in developing the knowledge, skills and confidence of gym staff to support young people with cerebral palsy in the gym When implementing 1:1 social support in community gym settings, consider the preferences of young adults and the resources available


Introduction
The gym is a common leisure setting in western society and is a desired setting for physical activity participation by young adults with disability, including those with cerebral palsy (CP) [1].The community gym provides an adaptable space for exercising at all ability levels and has been shown to have physical and psycho-social benefits for young adults with CP [2,3].Meeting physical activity recommendations is important for young adults with CP to help maintain physical function and reduce the risk of secondary co-morbidities and early ageing [4].However, for young adults with cerebral palsy (CP), participation rates in physical activity remain low [1,5].Gym-based exercise is safe and feasible for people with cerebral palsy [6,7], but multiple barriers to their participation can mean gyms are perceived as exclusionary by those with disability [8,9].
To date, research has focused on identifying barriers and facilitators to physical activity participation for young adults with cerebral palsy in various settings.The social and physical environment of the gym has been shown to play an important role in influencing attendance and involvement for those with physical disabilities [3,10,11].In particular, social support, staff training, inclusive practices and the physical accessibility of the building and equipment can influence gym participation [8,12].Affordability of gym memberships and associated support or transport costs are other key barriers to this setting for people with disability, who often have reduced employment and income [13].
To optimise participation in physical activity for young adults with CP, we must shift our focus from identifying barriers and facilitators, to developing and prioritising strategies for change that can be readily implemented in practice [11].The socioecological model has been recommended to guide strategy development in healthcare across the five levels of influence of individual, interpersonal, institutional, community and policy [11].Identifying priority implementation strategies is a critical first step to enabling collaboration between stakeholders, including recreation facilities, rehabilitation services, and government.
Therefore, this study aimed to (i) identify and (ii) prioritise strategies to optimise physical activity participation in the community gym setting for young adults with CP.A secondary aim was to determine the perceived ease of implementation of the strategies identified.

Methods
A three-round, online Delphi study (Figure 1) was conducted and is reported here in accordance with the Guidance on Conducting and REporting DElphi Studies (CREDES) [14].The Delphi method was chosen as there is a lack of evidence to determine how to improve gym participation for young adults with CP [15] and previous research suggested a collaborative, cross-sector approach was required [11].The Delphi method provides a rigorous approach to collating expert opinions from multiple perspectives while providing anonymity to individual participants, thus reducing the risk of power imbalances [14,15].The e-Delphi method has benefits for participants with different needs as they can complete the survey at a time convenient to them, take as much time as they need to respond, and there is no need to travel.
Approval for this study was granted by the La Trobe University Human Ethics Committee (HEC21144).Consent for the study was provided via an e-consent process embedded within the Round 1 survey link, prior to commencing the survey.Participants were informed during each survey round that they could withdraw at any time by exiting the web browser or informing the research team.Participants that had not responded to the Round 2 and 3 survey links were sent two automatic reminders, approximately 2 weeks apart, via Research Electronic Data Capture (REDCap) software (v12.5) and then removed from future correspondence.
Consumer involvement throughout this study is reported using the short form of the Guidance for Reporting Involvement of Patients and the Public 2 (GRIPP2-SF) [16] (Supplementary File A). "Consumers" were young people with CP (aged 16-30 years) and family members of young adults with CP.Consumer involvement was embedded in two ways.First, two young adults with CP were recruited as research partners and co-authors.Recruitment of the research partners was conducted by open advertisement and expression of interest with the criteria of (i) lived experience of cerebral palsy, and (ii) interest in the topic area of gym participation for adults with CP.The two research partners recruited had prior and varied experience exercising in community gym settings.Second, there was consultation with two consumer advisory groups made up of four participants each; one for young adults and one for parents respectively.'Experts' [15] within this study were defined as having knowledge about (through lived or professional experience) gym-based physical activity for young people with CP.Four expert groups were identified: young adults with CP, their family or caregivers, gym or exercise professionals, and health professionals.Young adults with CP were included if they were aged 16 years or older, had experience or a desire to exercise in a community gym setting, and were able to respond to the survey questions with or without practical assistance.Family members of young adults with CP were included if they cared for a young adult with CP aged 16 years or older where the young adult had experience or a desire to exercise in a community gym setting.The decision to involve those with an interest or experience in gym exercise was made to enhance the immediate utility of the findings for young adults and those working with them to access community gym settings.Gym staff or exercise professionals were included if they worked primarily in community gyms and had experience working with young adults with physical disabilities, therefore more likely to understand how young people exercised in their setting.Health professionals were included if they worked with any young adults with CP, regardless of the age or diagnosis of their primary caseload.Participants were recruited via purposeful and snowball sampling across Australia via CP support organisations, social media, community gyms, sports organisations and flyers.Target recruitment was set at 20 participants per group to allow for a final round of 10 per group if a 70% retention rate was maintained [17].Only participants who consented at commencement (R1) were included in rounds 2 and 3.
The three survey rounds were conducted via email between July 2021 and March 2022 (Figure 1).The purpose of round 1 (R1) was to brainstorm strategies that could improve community gym participation for young adults with CP.R1 questions (Supplementary File B) were based on published literature on known barriers and facilitators [10,11], developed with the assistance of consumers, and presented to participants according to the socio-ecological model.Participants provided free text responses on their ideas for improving gym participation for young adults with CP.The R1 survey was piloted with representatives of the four participant groups.The survey was delivered using REDCap.The purpose of round 2 (R2) was to evaluate the importance of each strategy, and the perceived ease of implementation.Strategies identified in R1 were presented to participants for a rating on a 7-point Likert scale [15].For importance, 1 represented 'lowest importance' and 7 represented 'highest importance' .For ease of implementation, 1 represented 'very difficult to implement' and 7 represented 'very easy to implement' .Respondents could provide free text responses with additional strategies not identified in R1.The purpose of round 3 (R3) was to determine the final consensus on the importance of each strategy.R3 included strategies that reached all-group consensus, or consensus within the young adult group in R2.
Qualitative data from R1 were analysed using a content analysis framework (Supplementary File C) and managed using Microsoft Excel software.Content analysis using data from all stakeholder groups was completed over four stages [18].In stage 1, meaning units were grouped within socioecological model (SEM) categories by two researchers (GM/AY) and cross-checked (CW/NS).In stage 2, meaning units were converted to condensed descriptions (GM) and checked to ensure the original meaning was preserved (CW/AY).In stage 3, similar condensed descriptions were grouped by topic and labelled as actionable strategies for change (GM/AY).In stage 4, the final list was reviewed and confirmed by the whole team.Consumer input (AY/FM) was sought on the phrasing of the final list of strategies for R2.Any additional qualitative statements provided in R2 were analysed using this same procedure.
In R2 and R3, descriptive statistics were calculated on the level of consensus for each statement.For importance, a consensus was determined a priori.Statements were considered to reach consensus if they were rated as "highest importance" (score of 6 or 7) by ≥70% of all group respondents based on either the percentage or an absolute number of responses AND <15% of respondents rating the item as "not important" (score of 1 or 2) [19].Consensus data were analysed within and across the stakeholder groups.Precedence was given to the ratings of young adults with CP, such that any strategies that met consensus on 'high importance' for the young adult group were retained, regardless of overall group consensus.Ease of implementation was calculated as a median and interquartile range (IQR) across all groups.Stability in the item scoring of importance across R2 and R3 was measured using the all-group median.A median change of >1 point on the 7-point Likert scale was considered to be a clinically meaningful indication of instability [20].Statistical analysis, completed using the Kruskal-Wallis test for nonparametric data, was completed on the R3 group medians to determine if there were differences between groups.

Results
Seventy participants were recruited from across Australia (Table 1).
The free text R1 survey data from all groups were analysed together, resulting in 914 data points, and a final 83 strategies for inclusion in R2.In R2, 54 strategies reached consensus and were retained for re-rating in R3; 47 by all group consensus plus 7 based on young adult consensus (Supplementary File E).No new strategies were identified by participants in R2.The ease of implementation of all 83 strategies was rated as 'moderately' to 'very easy' to implement (median >3), although most strategies had an IQR ≥2 indicating variability between participant ratings (Supplementary File D) [21].In R3, 44 strategies reached consensus; 41 by all group consensus plus 3 based on young adult consensus.There was stability in participant responses between R2 and R3, except for the item 'celebrating achievements' (Supplementary File D).
The 44 Strategies that reached consensus in R3 were categorised according to the socio-ecological model (Figure 2).The majority (n = 20) related to the gym environment, followed by strategies relating to policy (n = 12), individual (n = 6), community (n = 4) and interpersonal (n = 2) levels.The top 10 consensus strategies are outlined in Table 2.
At an individual level, providing information on exercise and the gym, developing routines, goal setting and tracking progress was highly important to all groups.

"Program[s] linked to clear goals so [they] can see achievements and outcomes"
"Teaching adults with CP that their bodies are robust and will respond well to physical activity.Dispel the myths that they're vulnerable and need to be sheltered.""Empowerment of young people to advocate for the equipment, space and support they need to overcome these barriers" (R1 free text data) At the interpersonal level, all groups identified the skills and knowledge of support workers as important and provided information to parents of young adults with CP (Figure 3).

"Making sure the family understands the importance of exercise and interaction with their peers." "Education of support workers to understand the importance of PA [physical activity] for health for all individuals and their role in promoting health to clients in their care, e.g supporting gym access, home programs and healthy eating"(R1 free text data)
Within the gym environment (Figure 4), all items relating to staffing and the physical environment were considered highly important across all stakeholder groups.However, strategies relating to social support (instrumental, informational, emotional or appraisal support) [22] in the gym were more varied.While having 1:1 support reached a consensus as important, there was no consensus on who should provide social support, although support from gym staff did reach a consensus within the young adult and family groups (Supplementary File E).Strategies relating to planned social activities or identifying with a disability (e.g., therapy groups led by clinicians, CP groups, disability hours in the main gym, and integration aids) did not reach a consensus (Supplementary File D).
"Having 1:1 instruction with someone who also understands the mechanics of the individual's CP would be most beneficial"   Partnerships and ongoing support between the health and recreation industries were highly important.Strategies relating to sharing and finding information on gym programs were mixed however a 'directory of trainers and health professionals with a special interest in CP' and 'distributing information through registries and databases' did reach a consensus (Figure 5, Supplementary File D).

"Partnerships between organisations creating pathways for participants to cross between." "Creating a network of health professionals, personal trainers or support workers who can be called upon" (R1 free text data)
Five strategies addressed gym membership costs (Figure 5).These are related to gym policy (reduced or waived memberships), public health funding (e.g., Australian Medicare service), and disability funding (e.g., Australian National Disability Insurance Scheme, NDIS) (Supplementary File D).Of these, NDIS funding for gym memberships was the only strategy that reached consensus across all groups.

"Making the cost of gym membership more affordable for young adults with CP and their families and having the cost funded through NDIS. Usually gym membership is not funded." "My biggest issue was the NDIS did not recognise how beneficial a good gym program is. When I go to the gym I need to have two support workers to help me as I do lift quite heavy weights." (R1 free text data)
Analysis of R3 data comparing group 'high importance' ratings showed statistically significant between-group differences for 3 items that reached all-group consensus (Supplementary File E).The strategy of 'encouraging health professionals to consider the gym from an early stage as part of usual care' was rated lower by the young adults compared to the gym and health professionals (p = 0.021).The strategies of 'having a knowledge sharing process between gym staff' (p = 0.038) and the strategy of "develop videos of adapted versions of common exercises" (p = 0.034) were rated differently between gym professional and young adult groups, with young adults rating a knowledge sharing process lower, and adapted exercise videos higher than the gym professionals.There were several strategies that did not reach overall consensus in R2  (e.g., a central portal to find information, integration aides, key contact staff members, specialist CP groups, and specific gym memberships) that were considered highly important within the parent group (Supplementary File E).

Discussion
The highest priority strategies identified for optimising physical activity participation in the gym for young adults with CP related to the institutional (gym) and policy levels of influence within the socio-ecological model, particularly the physical environment, funding, gym staff training, and developing partnerships between sectors (Table 2, Figure 2).
Accessibility to the physical environment continues to be a barrier to physical activity participation for people with disability [8].Our findings highlighted the physical environment as a priority area for change that was moderately easy to implement, raising questions as to why the physical accessibility of community gyms has not yet been universally addressed.One possibility is the diverse nature of the disability and what accessibility means to different people.Engaging people with disability in the development of inclusive facilities (identified in this study as an important strategy) may enable change at a facility-level.Further, there are existing resources designed to support gyms to eliminate accessibility barriers such as the Accessibility Instruments Measuring Fitness and Recreation Environments (AIMFREE) [23].Gyms should be encouraged to use these tools to review their facilities and, in consultation with consumers, make inclusive decisions around facility development, renovation and equipment selection and replacement.
Cost barriers to physical activity participation are frequently reported by people with disability [3,11], who often have lower employment rates and lower incomes [13].All stakeholders in this study considered external funding for gym memberships (for example via disability schemes or public funding) as important to support physical activity participation in community settings.Gym-based intervention research studies involving people with disability frequently subsidise the cost of attendance [7,24,25], however long-term participation outside of research requires sustainable funding models.There may therefore be a role for disability-based (for example the NDIS identified by our study participants) or public funding models to eliminate cost barriers to gym attendance to improve or lessen the decline in physical function for those with disability.Further evaluation of the cost-benefit of subsidised gym access is warranted.
Staff training on disability awareness, exercise assessment and exercise prescription for young adults with CP was considered highly important to all groups.The knowledge and attitudes of gym staff has previously been identified as a barrier to gym participation, influencing both the practical support and the sense of belonging within the gym setting [8,9].However, gym staff have shown an overall positive attitude towards disability inclusion and a desire to gain more skills in this area [26,27].This is supported by our findings where staff training strategies were rated highest by the gym professionals themselves.Our study did not identify all the ways in which staff training could occur however a 'train the trainer' model, 'employing gym staff with disability', and 'learning by doing' were all highly important to the gym staff and reached an overall consensus.This suggests that, in addition to formal courses and organisational-related traininone-to-oneone interaction with people with disability may be an important mechanism to promote positive attitudes and skill development among gym professionals [26].
We found a lack of consensus on which social support strategies should be prioritised in gym settings.Social support is a key facilitator of physical activity for people with disability [8,10] and 'access to 1:1 support within the gym setting' was considered highly important to facilitating participation, particularly to young adults and families in this study.However, a consensus was not reached on who should provide social support, based on options suggested by the stakeholder groups themselves.This may reflect the individual nature and preferences relating to the provision and use of social support.Offering a variety of social support strategies could allow young adults to choose about what works for them [3].While 'exercising with a paid gym staff member' did not reach overall consensus, this strategy was considered highly important to young adults and their families indicating young adults may prioritise opportunities to exercise in the community gym as their peers would -with the support provided by the gym.
Partnerships between organisations and ongoing support from health professionals were found to be highly important indicating interest in collaboration across academic, rehabilitation support in the gym was important to young adults however future studies are needed to identify how this is best operationalised.Developing and maintaining partnerships between the research, health and recreation sectors is desired by all stakeholders and would facilitate the implementation of priority strategies.and recreation sectors.Physical activity interventions are often developed and tested in clinical or rehabilitation settings [28].Given young adults with disability report a preference for community-based physical activity [1], future studies should investigate collaborative partnerships to co-design, implement and evaluate interventions within real-world community recreation and leisure contexts [11,29].
In this study, we included only gym and exercise professionals who had experience working with patrons living with a disability.This may have affected the identification and prioritisation of strategies when compared to gym staff with no previous disability experience, particularly on the perceived ease of implementation if they came from inclusion focussed settings.However, the prioritisation of strategies by the gym professionals included in our study was reflective of known barriers to physical activity participation in gym settings [3,10,11], including staff knowledge and skills, the built environment, and funding which lends weight to the findings.Similarly, although recruitment was open to any young adult or family member with an interest in gym exercise, all the participants in these groups had previous or current experience exercising in the gym.This implies participants had already overcome any barriers to commencing the exercise.This may have influenced the prioritisation of strategies related to getting started in the gym that did not meet consensus, for instance 'having a key contact staff member' , 'clinician led groups' , and 'online portals to find opportunities, or share experiences' .An important contextual consideration is that there was not always a consensus between the young adult and parent groups.Several strategies relating to finding opportunities or information, and accessing gyms were rated highly by the parent group but did not reach an overall consensus.In addition, recruitment of the parent group was limited and did not reach target numbers, despite efforts to connect with this group through multiple avenues, making it difficult to determine the practical relevance of these differences.Development of future interventions should treat these groups as separate entities, identify their specific priorities, and consider that families often remain key to facilitating physical activity participation even as young people with disability transition to adulthood [10].
The inclusion of consumers as both research partners and research advisors within this study ensured the research remained grounded in the experiences of the end-users -young adults with CP and their families.Another strength is that the results stem from the lived experiences of key stakeholder groups, enhancing the credibility and transferability of the results.
Many strategies that did not reach consensus across groups were still considered highly important to some stakeholder groups.It is likely then that implementation strategies need to be chosen in collaboration with the target audience(s) and in their local contexts, to ensure that the strategies chosen are reflective of their particular needs.However, the strategies reaching consensus of highest importance in this study address many previously reported barriers to physical activity participation, providing a practical starting point for intervention planning.

Conclusion
We worked with four stakeholder groups to identify priority implementation strategies for improving participation in community gym settings for young adults with CP.Stakeholders agreed that strategies targeting the gym and policy levels of influence were of highest importance.Addressing the physical accessibility of the gym environment, the costs of attendance, and the skills and support provided by gym staff should be prioritised in future resource allocation and implementation studies.Individual social

Figure 1 .
Figure 1.Flowchart of the Delphi study.

Figure 2 .
Figure 2. strategies meeting consensus mapped to the socio-ecological model levels of influence and the level of agreement achieved across all participant groups.large circles represent >90% agreement, medium circles represent 80-90% agreement, and small circles represent 70-80% agreement.

Figure 3 .
Figure 3. Consensus strategies act at the individual and interpersonal levels.

Figure 4 .
Figure 4. Consensus strategies acting at the gym environment level.

Figure 5 .
Figure 5. Consensus strategies acting at the Community and Policy levels.
^other including medical professions and transition service workers.×healthprofessionalsincluded if they had any caseload of young adults with CP.Due to funding and healthcare models in australia, few allied health providers would have primarily adult CP caseloads.#exercisephysiologistsworking in gym settings.*Programsincluding'access and inclusion' , 'support program' , 'diversity and inclusion' ."Accessibleparkingand environment including toilet and shower facilitiesConsideration of space around equipment to allow the use of mobility aids.Having a variety of equipment available to encourage people of all abilities to exercise" "The knowledge and skills of gym staff to be able to support those with CP.Disability specific knowledge and awareness, and adapting exercise programs for people with disability" (R1 free text data)

Table 2 .
top 10 strategies for improving gym participation for young adults with CP.
nDis: australian national Disability insurance scheme (government-funded scheme for supports associated with disability).