Adapting Experience-Based Co-Design to Disability Research: Co-Producing the CycLink Co-Design Study
ABSTRACT
Introduction
Participatory methods like experience-based co-design (EBCD) can be used to develop complex interventions, but may need adaptations when co-designers include young people with disability, parents and community partners. We aimed to adapt EBCD through co-production by involving people with lived experience of disability as co-researchers. This paper reports the co-produced protocol and reflects on co-researchers' contributions.
Methods
Guided by a six-stage co-production process, we formed a team of co-researchers, academic researchers, co-design convenors and evaluators. A five-person steering group, comprising three co-researchers and two academic researchers, led decision-making and project oversight. We communicated via videoconferencing, phone and email. Briefing documents, meeting minutes and diaries supported our reflections and reporting.
Results
We adapted EBCD to include people with disability through creative online methods and co-produced a two-part ‘CycLink Co-design Study’ protocol. Part 1 proposed using EBCD to design principles for a community cycling intervention (CycLink). Part 2 planned a mixed-methods evaluation of our adapted EBCD. Co-researchers influenced participant choice and accessibility by developing phased involvement options, inclusive consent processes and adapted research materials. Interpretative support during qualitative analysis improved the relevance and reflexive rigour of findings. However, resource constraints limited co-researcher involvement in conducting EBCD activities.
Conclusion
Co-production enabled us to adapt EBCD for people with diverse support needs and invite under-represented populations (e.g., young people with childhood-onset disability) to co-design. Cumulative adjustments resulted from our disability expertise, guidelines and approaches facilitating co-designers' opportunities to engage. Future studies should consider early and ongoing co-researcher involvement within both processes.
Patient or Public Contribution
Two adults with disability and a parent of a young child with disability joined our team as co-researchers. Co-researchers valued flexible involvement, which ranged from consultative to collaborative. Co-researchers' experiential expertise influenced the relevance of project materials and qualitative findings. We reported on co-researcher involvement through the Guidance for Reporting Involvement of Patients and the Public Version 2 Short Form (GRIPP2-SF) [1] (Supplemental File S1—Section A, Table S1).
1 Introduction
Partnership-based approaches [2] to intervention development, such as co-design, co-creation and co-production, offer collaborative ways to involve consumers and community partners in developing complex interventions. Complex intervention development is a broad process that includes design as a distinct time point [3]. Design involves conceptualising intervention elements [4], including conceptual goals, components and principles. To date, people with disability have lacked opportunities to collaborate on health, fitness or active leisure intervention development or design [5].
Many children and young adults (herein referred to as young people) with disability face additional barriers to learn and participate in cycling [6, 7]. Barriers can include individual or environmental challenges and/or access to assistive technology and personnel [6]. These factors lead many young people with disability to seek interventions to support cycling goals. Evidence suggests that beginner programmes develop emergent skills [8-11] but rarely link riders to community practice [12]. Strategies to develop advanced skills (e.g., navigate foot, bike or car traffic) and sustain community participation are lacking [6, 10]. Therefore, we planned to collaborate with consumers and community partners to co-design a community-based programme.
In our context, consumers include people with disability and immediate caregivers (e.g., parents). We identified people with disability as those with ‘long-term physical, mental, intellectual or sensory impairments, which, in interaction with various barriers, may hinder their full and effective participation in society on an equal basis with others’ [13]. Community partners included cycling advocacy representatives and cross-sector providers in allied health, education, sport and recreation. We acknowledged that the terms co-production, co-creation and co-design are often used interchangeably [14] within partnership-based approaches. While these approaches always intend to collaborate, recent reviews suggest variations in the origins and applications of the respective processes [14, 15]. We differentiated between co-production and co-design as two distinct processes and used each process with people with lived experience of disability for different purposes.
We defined co-design as a solution-oriented process which brings consumers, community partners and researchers together to ‘solve a particular problem or challenge’ [16] or enhance the ‘human experience’ [17, 18]. We envisaged using co-design to design the principles of a novel cycling programme alongside young people with disability (aged 8–30 years), parents and community partners. Given the paucity of evidence exploring young people's experiences of learning to cycle or using beginner programmes, we sought an experience-based approach. Experience-based co-design (EBCD) is an established process originating from healthcare quality improvement [19]. EBCD uses qualitative methods conducted over ‘two essential phases’ of experience gathering and co-design [20]. A preceding planning phase and celebration event are also important, but often overlooked [20]. Several steps are embedded into each phase, ranging from observations, interviews, touchpoint identification (i.e., an interaction or feature of a service or product that affects overall experience), a feedback video and workshops [21]. EBCD aims to identify emotional touchpoints in a consumer's journey and generate change initiatives through consumer and provider collaboration. Traditionally, EBCD has been used to improve health services or care pathways, and researchers [20, 22] have recently identified potential applications for intervention design.
We anticipated that adaptations would be required to extend EBCD to our research and create an inclusive setting for co-designers with diverse abilities. For us, inclusivity meant enabling people from diverse backgrounds to participate meaningfully [23]. We operationalised inclusivity by considering roles, methods and approaches that enable participation and offer a sense of belonging [24]. Collaborative roles empower consumers to become ‘makers and shakers’ [25] rather than passive recipients. Methods include democratic procedures, accessible materials and flexible involvement opportunities. Our approach to disability was influenced by inclusive research methods [26, 27], strengths-based practice [28] and contextually important frameworks [29, 30]. Therefore, we superimposed co-production [31] as an inclusive method [26] across our entire research process. We were guided by Strnadová's co-production definition, as a ‘process of collaboration and collective decision making’ where ‘all consumers have a role in knowledge creation’ [31]. In co-production, consumers join the research team as co-researchers offering ‘an equal but different contribution’ to research planning, conducting and reporting [31]. Involvement spans a continuum from consultative feedback to collaborative contributions and consumer-led roles [26]. Impact helps describe the value of co-researcher contributions on research decisions and outputs [32]. Co-production is driven by the principles of diversity, power sharing, accessibility, reciprocity, flexibility and transparency [31]. Using co-production as our overarching process, we perceived advantages in preparing EBCD resources, offering steering group oversight and supporting qualitative methods.
- A.
how EBCD was adapted to our context;
- B.
the co-produced EBCD study protocol and
- C.
the involvement and impact of co-researchers throughout co-production.
2 Materials and Methods
2.1 Context
This study forms the development phase of a larger project called ‘CycLink’. CycLink aims to connect young people with cycling goals to local cycling opportunities through a participation-based programme. Early conceptualisation of CycLink was initiated by academic researchers (R.T. and J.J.C.) and drew upon a literature review, cross-sector engagement and a randomised controlled trial for young people with cerebral palsy [33]. While the trial intervention was effective at achieving two-wheel cycling goals [9], gaps were evident for adapted bike users and consumer involvement. Therefore, we engaged with five consumers (n = 2 adults with disability and n = 3 parents) to gauge the suitability of EBCD and apply for a grant. This early engagement informed us that involving consumers should extend to ‘all parts of the research’, including both the research process (i.e., co-production) and CycLink's design (i.e., co-design). Consumers valued EBCD's narrative focus and ‘structured’ process, which offered ‘different options for involvement’. Choosing to participate in ‘some or all’ EBCD phases (i.e., phased involvement) was considered particularly suitable to young people and families who described busy schedules and competing priorities. We chose an online setting to protect the study from Covid-19 restrictions and to support flexibility for participants located in geographically dispersed areas.
2.2 Team composition
Our co-production team included eight academic researchers and three co-researchers with lived experience of disability. Roles included lead researchers (J.J.C. and R.T.), co-researchers (F.O.K., M.Y. and H.S.), co-design convenors (V.P. and J.B.), evaluators (C.I. and M.W.) and senior researcher support (A.S. and N.S.). Lead researchers and co-researchers formed a five-person steering group which led decision-making and project oversight. Convenors, evaluators, senior researchers and external experts offered consultative support with methodological decisions.
2.3 Study design
We used Strnadová's six stages of co-production [31, 34] to guide our participatory research (see Table 1). This paper represents the team's reflective account of co-production.
| Co-production stage | Co-production processes | Project steering group online activity schedule |
|---|---|---|
| Stage 1: Initiating | Team formation |
|
| Setting project scope and costs |
|
|
| Briefing and training |
|
|
| Stage 2: Planning | Deciding on research methods |
|
| Ethics application preparation |
|
|
| Stage 3: Doing | Supporting recruitment and project administration |
|
| Preparing for qualitative methods |
|
|
| Stage 4: Sense-making | Interpreting qualitative analysis |
|
| Refining qualitative findings |
|
|
| Stage 5: Sharing | Preparing for presentations |
|
| Presentation of findings |
|
|
| Stage 6: Reflecting | Critical reflection on involvement and impact |
|
| Deciding on future steps |
|
2.3.1 Study Procedures
Co-production was conducted through videoconferencing, email and asynchronous document review (e.g., comments and track changes). Phone calls and text messages supplemented communication. Decision-making was informed by project objectives, discussion, consensus [35] and, where required, voting. Our field notes, attendance log, researcher diaries and meeting minutes formed a co-production audit trail. Minutes were itemised according to co-researcher contributions, project leads' contributions and noted decisional flow from proposals to actions. We voted when proposals met consensus, but reservations remained. Votes were cast via online polls (within meetings) or email. Each steering group member held an equal vote, and a majority of three votes was required to carry a proposal.
2.3.2 Stage 1: Initiating
Consumers were invited to express interest in co-researcher roles through an open advertisement hosted on a REDCap form [36]. We approached known networks, including individuals (n = 5), disability organisations (n = 3) and research groups (n = 1) by phone or email. Applicants received a project overview and position description informed by co-production guidelines [31]. We considered suitability criteria, time capacity and cycling interest when selecting co-researchers. After three co-researchers were onboarded, we consolidated project objectives by discussing external factors (e.g., prospective grants), our EBCD deliverables and project constraints. Project objectives included (i) developing an inclusive EBCD process, (ii) overseeing the EBCD process and (iii) involving co-researchers across all research stages. We communicated roles and responsibilities through terms of reference (Supplemental File S1— Section A). Co-researchers participated in introductory training led by J.J.C. and R.T. on research ethics, EBCD and consumer involvement.
2.3.3 Stage 2: Planning
Next, we prepared methods and a protocol for the University Human Research Ethics Committee's review. Before meetings, lead researchers sent preparation material detailing agenda items, meeting goals and preparatory activities (e.g., reflexive exercises and pilot testing). Co-researchers were paid for 1 h of preparation time per meeting. We pilot-tested two participant questionnaires from the Public and Patient Engagement Evaluation Tool (PPEET) Version 2.0 [37] (Optional Demographics and Module B Ongoing/Long-term Initiative), the Patient Engagement In Research Scale (PEIRS) [38] and study-specific surveys. The PPEET and PEIRS are publicly available surveys used to evaluate or describe engagement in consumer involvement initiatives. Piloting enabled feedback on language, usability, completion time and contextual adaptations. We developed easy-read information using guidelines [39] and prepared recruitment material to be accessible via multiple sources (e.g., video and easy read).
2.3.4 Stage 3: Doing
Co-researchers supported recruitment by sharing advertisement material and selecting participants using our sample frame (Supplemental File S1—Section B). One co-researcher (M.Y.), who had skills in project management, was employed as a project administrator. The administrator scheduled participant activities, sent preparatory material and collected quantitative data (e.g., attendance log). We prepared for qualitative methods by observing local cycling programmes via video archives. Observations gathered our personal assumptions of early cycling opportunities through an EBCD-informed template.
2.3.5 Stage 4: Sense-Making
During experience gathering, adapted methods were used to analyse photo-elicitation interviews. Lead researchers developed candidate themes from transcripts using Steps 1–3 from reflexive thematic analysis [40]. Co-researchers refined candidate theme wording and offered personalised interpretations of preliminary findings. A three-part photo-elicitation interview analysis described photo intentions, interpretations and summated stories from the photo-text product. Co-researchers sorted photographs into categories and identified core photograph sets [41, 42]. A digital story [43] triangulated the candidate themes and categorised photographs through a script and video. Digital storytelling was led by a co-researcher (F.O.K.), who was skilled in video production. Lead researchers developed the script, and F.O.K. produced, edited, annotated and narrated the video. During co-design, we synthesised updates from convenors and suggested priority objectives (e.g., intervention elements [4]) for the combined workshop. Before the celebration event, we discussed the convenors' co-design report, our methods and study limitations.
2.3.6 Stage 5: Sharing
Sharing involved preparing visual presentations, presenting findings and contributing to this article. We decided on the celebration event's format, roles, invitation and an event gift. Each member of the steering group prepared slides and presented findings. ‘Meet the researcher’ videos and an infographic were prepared as knowledge translation outputs.
2.3.7 Stage 6: Reflecting
Reflections discussed the level of involvement, impact and future directions for co-production research. We used involvement matrices [26, 44] to describe the perceived level of involvement at each co-production stage. An impact taxonomy [32] aided descriptions of co-researcher impact. We reported barriers, facilitators and limitations in the GRIPP2-SF [1] (Supplemental File S1—Section A, Table S1).
2.3.8 Ethical Considerations
While completed in an ethical manner, this part of the research did not require formal ethical approval as research data was not collected. Consumers from our early engagement and co-researchers granted permission to use quotes and findings gathered during co-production.
3 Results: Outcomes of Co-Production
Our results are presented in three sections: (1) EBCD adaptations; (2) our primary co-production output—the CycLink Co-design Study protocol and (3) co-researcher involvement and impact in co-production.
3.1 Adaptations to EBCD
We adapted EBCD to our population's diverse support needs by tailoring methods for inclusivity, intervention design and the online setting (see Table 2). Adaptations sought to facilitate experience sharing, choice-making and engagement of all co-designers. We chose design principles rather than an entire complex intervention as a feasible EBCD deliverable for this development phase. Setting a desired deliverable early enabled us to plan around our constraints of time frames, convenor availability and funding. We retained EBCD's phases and steps, but moved through an accelerated process, whereby workshops were shorter and conducted over 8 weeks. Design activities focused on early cycling experiences and intervention elements. We aimed to enhance engagement through preparation materials, a digital display tool and breakout rooms. Shorter workshop durations (i.e., 1–2 h) were considered more practical for retaining online engagement. Delivering EBCD online required several ethical safeguards. For example, to protect online safety, we used a study-specific email address, trained participants in photo-taking, agreed-upon expected etiquette and separated adults and children into different workshops. We screened participants' online capabilities by phone/survey (e.g., ability to use email, surveys and videoconferencing), promoted use of a familiar device (e.g., personal iPad), remained on one website during videoconferencing and offered technical support (e.g., tip-sheet) as required.
| Phases | Traditional EBCD steps [21] | Adapted EBCD for CycLink Co-Design Study |
|---|---|---|
| Setting up | Project steering group formation
|
Initiating and planning for research
|
| Experience gathering | Step 1: Site-based observations
|
Step 1: Virtual observations
|
| Step 2: Staff interviews
|
Step 2: Photo-elicitation interviews with community partners
|
|
| Step 3: Patient and family interviews
|
Step 3: Photo-elicitation interviews with young people and parents
|
|
| Step 4: Videos edited, development of touchpoints and trigger film
|
Step 4: Co-production of a digital story
|
|
| Co-design | Step 5: Staff feedback event
|
Step 5: Co-design workshop with community partners
|
| Step 6: Patient feedback event
|
Step 6: Co-design workshop with lived experience group (young people aged > 18 years, parents and carers)
|
|
| Step 7: Joint workshop
|
Step 7: Combined workshop with lived experience group and community partners
|
|
| Step 8: Small co-design teams
|
Step 8: Children's advisory group (aged 8–15 years)
|
|
| Celebration | Step 9: Celebration event
|
Step 9: Celebration event
|
- Abbreviations: AEBCD = accelerated experience-based co-design, EBCD = experience-based co-design, rTA = reflexive thematic analysis, SMART = Specific, Measurable, Attainable, Realistic and Timed.
- * File folder: This adapted and semi-structured method [46] helps scaffold photo-taking for people with intellectual disability. The method involves posting the participant a Polaroid camera and four card envelopes, which are pre-labelled with suggested themes.
3.2 The Outputs of Co-Production: The CycLink Co-Design Study Protocol
We developed a two-part study to use (Part 1) and evaluate (Part 2) co-design. We received ethics approval from the Human Research Ethics Committee at the University of Melbourne (reference: 22588) and collected data from July 2022 to April 2023. We aimed to recruit young people, parents and community partners via targeted advertisement. Potential participants accessed the study via an expression of interest before progressing to consent procedures. Inclusion criteria sought young people with disability aged 8–30 years with: (i) cycling goal(s); (ii) recent experiences of learning to cycle; (iii) residence in Victoria, Australia; and (iv) intentional communication (reliable yes/no response and ability to express opinions) in English. Eligible parents and community partners needed to have experience supporting young people with disability to cycle.
3.2.1 Ethical Procedures
Three consent pathways were developed to accommodate three distinct study populations: (i) adults without intellectual disability; (ii) children and young people aged < 18 years and (iii) young people aged > 18 years with intellectual disability or complex communication needs. Adults without intellectual disability were invited to provide informed consent via an online consent form. We sought parental consent, alongside the young person's consent, for children and adolescents aged < 18 years through a similar online form. Young people with intellectual disability or complex communication needs were invited to provide verbal consent with a guardian or familiar communication partner by videoconference. The verbal consent process followed similar steps to those described by Arscott et al. [49]. We modified this process to screen-share easy-read information and consent options via videoconferencing (Supplemental File S1—Section B). The first step identified how a young person communicated a reliable yes or no response. Next, we shared the easy-read information. Then, we screened their understanding of the study via a series of yes/no questions. The final step allowed young people who understood the study's nature, risks and benefits to provide verbal consent. Supported decision-making strategies were used with young adults who could not provide verbal consent but wanted to participate. In these instances, assenting young people needed third-party consent from an appointed individual (e.g., guardian, advocate or supportive attorney) and could nominate a preferred communication partner for dyad responses. Communication partners were briefed on strategies to maximise young persons' participation in sharing perspectives (Supplemental File S1—Section B). In cases where assent was not possible, parents/carers could participate as ‘experts by experience’ offering proxy perspectives [50].
As consent is an ongoing process, we reminded participants of the voluntary nature of the research before each activity. During workshops, one convenor observed for behavioural cues that may indicate dissent, refusal or withdrawal. Behaviours included actions (e.g., repetitive yawning) or communication strategies (e.g., expressing desire to leave or looking to exit points). Where dissent/withdrawal or coercion (e.g., from a support person) was suspected, researchers arranged a follow-up call or breakout room to collaborate on a participation plan. Planning for subsequent activities considered amended roles, additional supports and/or reassessing ongoing involvement (e.g., withdrawal).
All participants had opportunities to debrief by phone and access a resource on ‘minding your wellbeing’ during co-design. Participants were reimbursed with a $25 AUD voucher per research activity lasting ≤ 60 min for both parts of the study.
3.2.2 Part 1: Use of Co-Design
Part 1 used multiple qualitative methods guided by our four-phase EBCD process (see Figure 1). Colour and visuals oriented participants to the process. All participants were invited to complete a ‘characteristics survey’ to describe demographics, identify support needs and aid selection for research activities through maximum variation sampling (Supplemental File S1—Section B).
3.2.2.1 Experience-Gathering Phase: Photo-Elicitation Interviews
Participants. Experience gathering aimed to invite at least 15 participants (8 young people, 2 parents and 5 community providers) to photo-elicitation interviews. Data collection. Photo-elicitation interviews asked participants to provide three photographs of early cycling experiences that explored barriers/facilitators to community cycling. Participants could take new photographs or choose pre-existing photos. Before interviews, participants received a ‘photo-tip sheet’ (Supplemental File S1—Section B), guiding them on ethical considerations for photo-taking, such as asking for consent and including recognisable features. A QR code or weblink enabled participants to obtain third-party consent from people who featured in photographs. Photographs were used as conversation-starters during photo-elicitation interviews and complemented our semi-structured interview guide. Data analysis. We used adapted methods (see Methods—‘Sense-Making’) to develop a script and co-produce a digital story.
3.2.2.2 Co-Design Phase: Workshops
The co-design phase involved a total of five online workshops for adults and children. Participants. We aimed to recruit 16 adult participants and up to 10 children to attend two workshops each. Data collection. Both adults' and children's workshops included icebreakers, online agreements, structured agendas and regular breaks. Adult workshops were facilitated by two experienced convenors (V.P. and J.B.) who developed co-design activities iteratively based on co-designers' needs and study objectives [51]. A digital display tool was used to present activities and collate co-designers' inputs via sticky notes. Separate adult workshops were held for community partners and people with lived experience. These workshops were followed by a combined workshop bringing the two adult groups together. Researchers J.J.C. and R.T. facilitated two separate children's workshops with 8–15-year-olds. Children's workshops were informed by adult workshops and aimed to consolidate emerging ideas from co-design. Creative methods, small group sizes and adjustments (see Table 2) helped facilitate children's participation. Data analysis. Convenors synthesised key findings into conceptual designs to illustrate experiential goals and component parts, informing the CycLink's principles. EBCD outcomes will be published separately.
3.2.2.3 Celebration Phase: Online Presentation
All consenting participants were invited to attend an online celebration event where the study's findings were presented.
3.2.3 Part 2: Embedded Process Evaluation of the CycLink Co-Design Study
We co-produced a process evaluation of our EBCD (see Figure 2) involving a convergent parallel mixed methods design [52]. Participants. All participants from Part 1 were eligible to participate. Data collection was embedded into co-design methods (e.g., attendance records and check-in surveys) and gathered after the final workshops (e.g., questionnaires or interviews). The PPEET (Module B Long term/Ongoing Initiative Questionnaire), a study-specific young person's evaluation questionnaire, routine data (e.g., workshop invitation/attendance) and an EBCD [20] fidelity checklist formed the quantitative dataset. Our young person's evaluation questionnaire used subsections of the PPEET [37] and included topic sentences that reflected concrete events, tasks or people within the study. Consultative advice from speech and language pathologists identified visual rating scales aligned to Talking Mats™ [53] (e.g., ‘Yes/No/Unsure’ and ‘Like/Don't like/Unsure’) and corresponding Pictorial Communication Symbols®. The fidelity checklist rated adherence (completed vs. uncompleted/unclear) to 10 improvement activities identified by Green et al. from the PoCF toolkit [20]. Qualitative data included workshop recordings, open-ended surveys (factors to stop, start and continue during co-design) and semi-structured interviews. An evaluation team, independent of the co-design process, emailed surveys, conducted semi-structured interviews and audited the fidelity check. Data analysis. We plan to analyse qualitative and quantitative datasets separately, before merging the findings to gain a deeper understanding of engagement quality and co-design mechanisms [54]. Evaluation findings will be published in a subsequent paper.
3.3 Level and Impact of Co-Researcher Involvement in Co-Production
3.3.1 Co-Researchers Engaged
Three co-researchers joined our project steering group and met 18 times over 2 years. All were new to co-researcher roles, of Caucasian ethnicity and had concurrent commitments (e.g., employed or studying full-time).
3.3.2 Participation Attendance and Involvement in the Process
Steering group attendance was high (see Table 3). One co-researcher was unable to attend several joint meetings due to clashes with work (n = 5) and health appointments (n = 1). Follow-up phone calls (n = 5) accommodated their updates and contributions.
| Co-production stage | Online engagement | Co-researcher attendance (n/N) | Lead researcher attendance (n/N) | Timea (h) |
|---|---|---|---|---|
| Initiating | EOIb process | 3/3 | 2/2 | 0.6 |
| Meeting 1 | 3/3 | 2/2 | 1.0 | |
| Training 1 | 3/3 | 2/2 | 2.0 | |
| Training 2 | 0/3 | 2/2 | 5.0 | |
| Training 3 | 0/3 | 1/2 | 8.0 | |
| Planning | Meeting 2 | 3/3 | 2/2 | 1.1 |
| Meeting 3 | 2/3 | 2/2 | 1.0 | |
| Meeting 4 | 3/3 | 2/2 | 1.1 | |
| Meeting 5 | 3/3 | 2/2 | 1.1 | |
| Meeting 6 | 2/3 | 2/2 | 1.1 | |
| Doing | Administrationc | 1/1 | 2/2 | 30.0 |
| Meeting 7 | 2/3 | 2/2 | 1.1 | |
| Meeting 8 | 3/3 | 2/2 | 1.1 | |
| Sense-making | Meeting 9 | 3/3 | 2/2 | 1.1 |
| Meeting 10d | 1/1 | 1/1 | 30.0 | |
| Meeting 11 | 2/3 | 2/2 | 1.2 | |
| Meeting 12 | 3/3 | 2/2 | 1.2 | |
| Sharing | Meeting 13 | 3/3 | 2/2 | 1.1 |
| Presentation 1 | 3/3 | 2/2 | 1.3 | |
| Presentation 2e | 1/1 | 2/2 | 0.5 | |
| Reflecting | Meeting 14 | 2/3 | 2/2 | 1.0 |
| Meeting 15 | 2/3 | 1/2 | 1.1 |
- Note: Attendance recorded as n/N, where n = number attended, N = total invited.
- a Time spent attending online meeting, note this figure does not include the 60 min of preparation time before most meetings.
- b EOI = Expression of Interest.
- c Administration: individual co-researcher involvement over several months.
- d Individual co-researcher activity relating to digital story production.
- e Individual co-researcher representation during the community of practice presentation.
Planning and sense-making stages were our most time-intensive periods. Decision-making was iterative and co-researchers were influential in reducing jargon and enhancing the clarity of proposals (see Figure 3, Supplemental File S1—Section A, and Figures S1 and S2).
Figure 4 represents the ‘ebb and flow’ of co-researchers' involvement throughout co-production. Collaborative involvement was most evident for choosing methods, adapting materials and developing the digital story. Co-researchers felt empowered to offer ‘unsolicited advice’ and be equal decision-makers. However, at times they felt ‘sidelined’ in leading research activities and wanted greater involvement in collecting ‘cycling-related’ data. Ultimately, co-researchers valued flexibility, role clarity and adequate time to meaningfully participate in co-production.
3.3.3 Impact
The impacts of co-production are described in Table 4. Our co-production resulted in an inclusive and safe online process that could empower co-designers with diverse abilities to participate. Co-researchers reflected on a ‘person-centred’ impact, illustrated through simplified involvement options, accessible materials and relatable knowledge translation outputs. They strengthened the relevance and trustworthiness of the study's findings by modifying wording and challenging academic researchers' assumptions. Co-researchers enjoyed sharing cycling expertise and valued developing knowledge and skills in research. Personal successes resulted in one co-researcher securing ongoing employment in academia and another writing a cycling blog.
| Co-production stage | Consumer and co-researcher contributions (elements/activities) | Evidence to support decisions and impact (e.g., research team quotes) (metric/evidence) |
|---|---|---|
| Stage 1: Initiating |
|
Expression of interest attracted 11 applicants:
|
| Identified phased involvement as a pragmatic way for young people/families to participate | ||
| Endorsed EBCD as a method that centralises phased involvement and storytelling: ‘Experiences are key … EBCD offers us a blueprint’ | ||
| Stage 2: Planning |
|
Identified three phased involvement options: (1) photo-elicitation interview, (2) co-design workshops and (3) evaluation survey and/or interview. |
| Used colour, visuals and ‘clear expectations’ (i.e., time/task demand) to illustrate options | ||
| Included ‘all disabilities’ in eligibility criteria: ‘broad is best … cycling shouldn't be limited by your diagnosis’. Emphasised recruiting diverse experiences over specific diagnoses | ||
| Identified preferred terminology for project materials and surveys (e.g., PPEET): ‘…“patient” makes it sound like I am unwell and sick … it should be people with disability.’ | ||
| Emphasised participation (e.g., factors that ‘challenge’/enable) and learning journey as foci of photo-elicitation interviews | ||
| Decided that all co-designers should be invited to evaluate EBCD either informally (e.g., via check-in survey) or formally (e.g., as Part 2 participants). Embedded check-in surveys were perceived as useful: ‘they give us the chance to change course if we need.’ | ||
| Offered feedback on the PPEET's demographics modulea:
|
||
| ‘I'm interested but…’: The decision was made not to be involved as both co-researchers and co-design participants. The team discussed time burden, role commitments (i.e., already involved in dual roles) and power imbalance (e.g., in analysis) as important factors | ||
| Stage 3: Doing |
|
Shared recruitment material with n = 4 organisation leads and n = 2 community of practice groups using personal networks |
| Populated the recruitment flow chart and attendance log with prospective quantitative data | ||
| Targeted advertisement led to n = 1 father joining the study as a co-designer following the identification of a recruitment gap | ||
| Co-researchers asked ‘what programs or resources are already out there?’ and ‘who's missing out?’ This led to a list of local cycling opportunities being shared with co-designers | ||
| Reflected on the ‘different hats’ (i.e., perspectives) worn in qualitative analysis and identified a positionality that:
|
||
| Stage 4: Sense-making |
|
One co-researcher related strongly to the candidate themes developed by lead researchers: ‘As a parent of a child at the pre-biker phase … the [candidate] themes address all the issues I've come up against’ |
| Modified a candidate theme from ‘family helps keep the wheels in motion’ to ‘supporters help keep the wheels in motion’ to be cognisant of young adults who accessed support from allied health professionals or coaches | ||
| Led the co-production of the digital story: ‘We need to create a succinct narrative and core script … 250 words at the most … needs to highlight the strengths and the issues’ | ||
| Emphasised meaningful and relatable outcomes that could be used readily by riders with disability and their families | ||
|
Emphasised further sharing of the digital story: ‘It offers a clear way to communicate’ | |
| Stage 5: Sharing | All (n/N = 3/3) co-researchers presented content at the celebration event | |
| Advocated sharing local opportunities/practical implications, in addition to study findings: ‘What now? … some parents came in [to co-design] wanting something for their family or child … need to share principles … but also social media stuff on local cycling…’ | ||
| Stage 6: Reflecting |
|
Supported the development of Figure 4: ‘the dotted line shows it wasn't a clear rigid path… aspect of learning as you go … we had to carve our path … it was iterative’ |
| Developed an ongoing commitment to research:
|
- Note: We adapted the PPEET's Optional Demographics Questions (Participant Questionnaire) aand Module B: Ongoing/Long-Term Engagement Initiative bfor our context.
- Abbreviations: AUD, Australian dollar; CALD, culturally and linguistically diverse; EBCD, experience-based co-design [19]; GRIPP2-SF, Guidance for Reporting Involvement of Patients and the Public Version 2 Short Form [1]; PEIRS, Patient Engagement in Research Scale [37]; PPEET, Public and Patient Engagement Evaluation Tool (Version 2) [37].
4 Discussion
This reflective account describes how co-production benefited the planning, conduct and reporting of adapted EBCD. We reflected that EBCD [19, 21] offered us a flexible structure to anchor our co-design study, but adaptations were essential to include young people with disability. Co-production [31] harnessed our collective expertise in disability, enabling us to anticipate, plan and prepare for meaningful participation by diverse co-designers. Choice, access and creativity shaped our adaptations. Co-production required additional time (8 months) and resourcing, but enabled us to develop an ethically approved protocol and invite under-represented groups to co-design. During EBCD, co-researchers positively influenced the quality of findings and digital storytelling by contributing lived experience interpretations into a fast-tracked EBCD process. Here, we discuss considerations for inclusive co-design and critically reflect on co-researchers' involvement throughout our co-production.
Similar to studies that have used [50, 55, 56], or planned to use [57, 58] EBCD with people with disability, we found that consumer-led steering groups and established disability approaches supported inclusivity. Like Pozniak et al. [59], we valued the strength-based approach [28] for its emphasis on individuals' interests, capabilities and future goals, over deficits and problems. Combining co-researcher expertise with disability approaches was illustrated through the subtle re-framing of terminology within evaluation tools and EBCD. We chose terms like ‘people with disability’ rather than ‘patient’ and ‘digital story’ rather than ‘trigger film’. By making these changes, we aimed to retain the emotive foundations of EBCD but avoid tragic or heroic portrayals that can encapsulate ableist views of disability [60]. This was particularly relevant to the digital story, which required several iterations to develop a strength-based script and video.
We found existing toolkits [27, 61] and guidelines [31, 34, 39] useful for adapting our methods and supporting person-based and environmental needs. Like Pickering [62], we anticipated that young people's perspectives would be enriched by gathering multiple creative data sources and layering them with parents' and professionals' perspectives. Consequently, photo-elicitation interviews were chosen over more common experience-gathering methods like semi-structured interviews or surveys [20]. Photo-elicitation interviews can enhance young people's ability to share perspectives and develop rapport [63], and we found the method was readily adapted for online use. Other co-design studies [55, 56] have demonstrated the need for visual methods and flexibility in workshops. Working online enabled us to readily share visually engaging material (e.g., photographs and Pictorial Communication Symbols®) via digital display tools and video. In workshops, co-designers could participate flexibly via open discussion, sticky notes and the videoconference's ‘chat’ function. Tailored communication strategies [48, 64] and iterative changes enhanced our preparation materials and facilitation skills based on support needs. Other groups [55, 56] have facilitated workshops using topic cards and Talking Mats™, respectively, which we could have explored with further training. Inviting guest speakers has also sparked workshop discussion [55]. Guest speaker roles could have supported the meaningful involvement of our co-researchers during workshops. Identifying potential mechanisms on ‘how’ our adjustments impacted participation will be explored in subsequent publications.
Like other researchers [65-67], we faced an intrinsic tension resulting from our co-production's academic-initiated source. A power imbalance resulted as the desired deliverables, time frames, funding, grant stipulations and accountability ultimately sat with the lead researchers. We addressed power imbalances by having a higher representation of co-researchers on the steering group and attaining consensus on methodological decisions. We ‘handed over power’ by sharing defined roles, illustrating the movement of ideas/proposals/decisions and fostering a collaborative environment which valued experiential knowledge equally.
We found time frames and funding constraints exacerbated power imbalances by compressing opportunities for collaborative decision-making and involvement. Our time pressures related to co-researchers' availability for involvement, the complexity of preparing creative participatory methods, the rapid turnaround time between EBCD phases and external deadlines. Our strategies to address these challenges included flexible meeting times (e.g., lunchtime/weekend), resource reallocation (e.g., additional time/budget for digital story production), adapted methods (e.g., triangulating analytic products) and timeline extensions (e.g., 12-month extension for external grant). The impact of time pressures meant co-researchers sometimes appeased the team, ‘doing what's most needed and most useful’, and lead researchers sought consultative inputs (e.g., review documents or interpretations) rather than collaborative contributions. For example, we sought feedback on pre-analysed candidate themes rather than empowering co-researchers to undertake familiarisation and/or coding. This decision was based on our need for a rapid translation of experience-gathering findings to digital story production to maintain the momentum of EBCD. Seeking consultative support on pre-analysed qualitative data is not uncommon [68], but we missed an opportunity for deeper engagement. Reflections from other groups suggest that formal training [69, 70], scaffolded roles [68, 71], collaborative analytic methods (e.g., framework method) [72], methodological flexibility [73] and documenting analytic decisions are important starting points for co-analysis. We reflected that offering flexibility [68, 71] for preferred involvement levels [26, 44] in different research phases may have led to deeper co-researcher involvement. Such phased involvement may be a consideration for future co-production projects where resourcing and time constraints are significant issues. Reciprocating substantive research opportunities would have particularly benefitted our younger co-researcher who felt ‘less involved’ than peers who held responsibilities in video production and project administration, respectively.
Reflecting upon other co-production principles, we noted markers of success and areas for improvement. We incorporated diverse views into our decision-making, but acknowledge that our expertise and connections were situated in childhood-onset disability (e.g., cerebral palsy, Down syndrome and autism). Other principles, such as accessibility, flexibility and transparency, were largely met by our procedures. Co-researchers raised the University's pay and email systems as barriers due to unfamiliarity and infrequent access. This was addressed through a video training tutorial and increased phone contact. Another challenge was managing the tension of age eligibility. We decided to include young people aged 8–30 years based on our grant stipulations, the need to offer developmentally appropriate materials and cater for contextual differences in paediatric and adult systems. This decision was accepted with reservations by co-researchers, some of whom felt disappointed that research often excludes older adults with child-onset disability. Ultimately, the team's strong relationships and collective vision for accessible cycling meant that reservations or standing aside (i.e., agreeing to compromise on a proposal) enabled decisions to flow and collaboration to occur.
5 Strengths and Limitations
The strengths of our paper include a clear demarcation between processes, a co-production audit trail and illustrative materials. However, it should be noted that all of our research team were adults and no one had an intellectual or cognitive impairment. Power sharing and role reciprocity were areas for improvement in co-production. With a larger budget, greater resourcing capacity and longer time frames, we would have supported co-researchers to undertake more meaningful roles in qualitative data collection and analysis. This may have included scaffolded [68] learning (e.g., observe, co-lead or lead interviews/workshops) [71], targeted training in co-analysis and/or phased involvement in preferred research activities. Such involvement may have achieved a greater depth of co-researcher influence.
A strength of our protocol was developing an embedded co-design evaluation. Process evaluation has been identified as a gap in previous EBCD projects [20]. Ethical safeguards were another strength, which are shared through our supplemental file. Our protocol also had limitations. While aiming to include people with diverse disabilities, we could only include symbolic communicators (i.e., people who can share intentional messages via verbal or augmentative and assistive communication with or without a support person). This decision was influenced by ethical and methodological requirements for informed consent and rich informative accounts. This meant people with severe-profound intellectual disability, or non-intentional communicators, were excluded. We attempted to redress this exclusion through parent-proxy representation. Time limited our ability to pilot or validate the young person's evaluation survey. However, survey content was influenced by a standardised outcome measure [37] and consumer evaluation with a similar target population [74]. Lastly, our accelerated EBCD may have impacted the quality of co-designers' engagement. We detailed several strategies to retain high-quality engagement and will explore their effectiveness through our process evaluation.
6 Conclusions
Our reflection may guide researchers, quality improvement leaders and consumers in exploring inclusive co-design in disability or marginalised community groups. We found that involving co-researchers in EBCD through a superimposed co-production process enabled us to prepare the online setting, oversee EBCD processes and implement EBCD activities. Our experience highlighted the importance of identifying feasible deliverable(s) early, setting shared project objectives and mapping meaningful co-researcher roles across both processes. We stretched co-researcher involvement across a prolonged period and faced funding constraints, which sometimes led to more consultative roles. However, by sharing our process outputs, we envisage that others can advance co-researchers' involvement in ‘doing’ EBCD activities, such as interviewing participants or facilitating workshops. Future studies should plan for early consumer involvement, offer phased involvement options and share resources to accelerate the possibility for more consumer-led research.
Author Contributions
John Carey: conceptualisation, methodology, software, validation, analysis, investigation, resources, data curation, writing – original draft, writing – review and editing, visualisation, project administration, funding acquisition. Alicia Spittle: conceptualisation, methodology, analysis, investigation, writing – review and editing, visualisation, supervision. Christine Imms: conceptualisation, methodology, analysis, investigation, resources, writing – review and editing, supervision. Nora Shields: conceptualisation, methodology, investigation, writing – review and editing, supervision. Margaret Wallen: conceptualisation, methodology, validation, analysis, investigation, resources, writing – review and editing. Finn O'Keefe: conceptualisation, methodology, analysis, investigation, resources, writing – review and editing, visualisation. Miriam Joy Yates: conceptualisation, methodology, analysis, investigation, writing – review and editing, visualisation, project administration. Holly Skilbeck: conceptualisation, methodology, analysis, investigation, writing – review and editing, visualisation. Rachel Toovey: conceptualisation, methodology, software, validation, investigation, resources, data curation, writing – review and editing, supervision, project administration, funding acquisition.
Acknowledgements
We acknowledge Professor Victoria Palmer and Dr Jennifer Bibb from the Co-Design and Implementation and Effectiveness node of the Methods and Implementation Support for Clinical and Health research hub (MISCH), who were contracted to provide co-design convenor services and resources. They generously offered in-kind support for conceptualisation and qualitative methodology.
Thank you to the five consumers (Campbell Message, Lucy and Simon Crunden, Finn O'Keefe and Miriam Yates) involved in early consumer involvement who helped gauge the suitability of EBCD and contribute to a grant application.
Thanks to speech pathologists Megan Walsh and Annabelle Scotney for their expertise on verbal consent processes, communication supports and rating scales.
The project received a seeding grant and PhD student top-up funds from Melbourne Disability Institute (MDI) at the University of Melbourne. These funding sources had no role in study design, collection, analysis or interpretation of data, report writing or submission for publication. Open access publishing has been facilitated via the Council of Australian University Librarians, which includes an agreement between Wiley and the University of Melbourne. This study will contribute to John Carey's PhD thesis.
PCS® (Picture Communication Symbols®) is a trademark of Tobii Dynavox LLC. All rights reserved. PCS® has been used with permission. Open access publishing facilitated by The University of Melbourne, as part of the Wiley - The University of Melbourne agreement via the Council of Australian University Librarians.
Conflicts of Interest
The authors declare no conflicts of interest. The research team catered to the potential conflicts of interest of financial reimbursement for MISCH's co-design services by citing their involvement in participant materials and publications.
Open Research
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
