The burden of proof: The process of involving young people in research

Abstract Patient and public involvement in research includes non‐academics working with researchers, on activities from consultative tasks, to joint working, and on user‐led initiatives. Health and social care funding bodies require involvement in research projects. A current debate focuses on a perceived lack of empirical “proof” to demonstrate the impact of involvement upon the quality of research. It is also argued that the working relationships between researchers and those becoming involved need to be understood more fully. These areas are beginning to be reported upon but there are few studies of young people involved in health research. This study describes the experiences of adult academics and young people, working together on a large‐scale, UK health research programme. Using qualitative interview and focus group methods, the aim was to explore participants’ perceptions about the process and outcomes of their work together. The importance of cyclical, dynamic and flexible approaches is suggested. Enablers include having clear mechanisms for negotiation and facilitation, stakeholders having a vision of “the art of the possible,” and centrally, opportunities for face‐to‐face working. What is needed is a continuing discourse about the challenges and benefits of working with young people, as distinct from younger children and adults, understanding the value of this work, without young people having to somehow “prove” themselves. Involvement relies on complex social processes. This work supports the view that an improved understanding of how key processes are enabled, as well as what involvement achieves, is now needed.


| Overview of involvement activities
"Patient and public involvement" refers to the roles for service users and members of the public in defining, delivering and disseminating research. INVOLVE (formerly Consumers in NHS Research), an organization that supports involvement in research, was set up in 1996 in the United Kingdom by the Department of Health, to guide patient and public involvement in health and social care research. 1 Involvement includes activities on a continuum from consultative tasks, through to "partnership working," to service-user-led initiatives. Involving those with lived experience is perceived as the "right" thing to do on moral, democratic and epistemological grounds. Systematic reviews suggest that it can influence all stages of research and, in broad terms, improves the "real-life" relevance of the work. 2

| "Quality" of involvement activities
One debate focuses on a perceived lack of data to demonstrate the impact of involvement upon research, [3][4][5] with a drive to outline standards of good practice. 6 The PIRICOM systematic review notes difficulties in achieving this: The poor reporting of [involvement] impact and the limited consideration of how context and process factors affect impact makes meaningful comparison across studies difficult, and so prohibits firmer conclusions about their influence. 3 Members of the PIRICOM team developed Guidance for Reporting Involvement of Patients and Public (GRIPP / GRIPP2), 7,8 calling for better reporting of practice. They echo other suggestions for a focus on how involvement works, as well as what it achieves. [9][10][11] A realist evaluation study 12 suggested that six actions support effective involvement: a shared understanding of the purposes of involvement; coordination; diversity of voices; researcher engagement; working relationships; and proactive evaluation of activities.
Importantly, involvement in health research has also been described as occupying "liminal knowledge spaces," in between established academia and novel practice, where difference, ambiguity and tensions come to the fore, creating opportunities for transformation. 13 Most studies focus on adults and less is understood about the experiences of young people involved in research.

| Children's and young people's involvement
The United Nations Convention on the Rights of the Child 14 establishes that all children have the right to be involved in decisions that affect them. Hart 15 outlines broad approaches to the participation of children and young people, and more recently, approaches for involving them in research have been proposed. 16,17 Examples include them being involved in systematic reviews, 18 expressing service preferences, 19 and commissioning decisions. 20 Guidance for involving young people and specific groups who are less frequently heard is available 6,[21][22][23][24][25][26] often drawing from experiences in other social science disciplines. 27 A literature review suggested that in research on children, with children and by children, children's perspectives can be gained. 28 In a case study review, 29 it was suggested that children and young people should be involved throughout the research process, but if this is not achievable, they can still be involved in a meaningful way, with the onus on researchers to ring-fence sections where they can collaborate and lead on tasks.
In this paper, we show that young people can meaningfully contribute to a large-scale health research programme. We suggest that adult researchers might re-evaluate assumptions about the capabilities of young people as researchers, without a burden upon them to prove their worth. However, this does not mean that adult involvement practices can be adopted uncritically. We need a continuing discourse about the challenges and benefits of collaboration with young people, as distinct from younger children and adults. We hope this work makes a timely contribution, by highlighting techniques and approaches that could be useful in working with those aged around 11-25.

| Aim
This study describes the experiences of adult researchers and young people involved in a large-scale, UK health research programme, exploring the process of working together and the outcomes of that work.

| ME THODS
The "Transition" study was a 5-year longitudinal health research programme, supported by the National Institute of Health Research, examining how health services in the UK can support young people in their move from childhood to adulthood. It comprised nine work packages, with one focusing on young people's involvement in the programme. The young people's group, formed in 2013 in the first months of the programme, met once a month to carry out their work.
They called their group United Progression (UP).
When recruited, the UP group members were all aged between 15 and 20. They had experience of accessing health-care services.
Most had experience of living with physical and/or developmental conditions, many in line with the exemplar health needs being studied within the programme. Recruitment occurred in different ways (eg health services; schools; health action groups). With membership growing steadily, the group had over 20 members, with active participation fluctuating in line with examinations and other commitments. Most meetings had around eight members present.
The group's work into the "Transition" programme was facilitated by four (adult) involvement facilitators, one of whom had additional responsibilities as involvement lead, and four peer support workers (PSWs). The PSWs were recruited from a local NHS youth group, to provide input to research tasks in the first 1-2 months, before the other young people had been recruited.
They then offered initial support to newly recruited members.
The PSWs became embedded members of the UP Group, as a natural part of the group's formation. This peer approach has since been reported elsewhere as useful for enabling the voices of those who tend not to participate. 16,29 The UP Group's role was to provide a young people's perspective, with the aim of working with the adult researchers to oversee the governance and delivery of the Transition programme. 30 The involvement lead was a member of the research team and reported to the research management meetings with the young people attending these meetings, when adults or young people felt there was a need.
At the initiation of this study, there were no pre-defined standards against which to measure involvement work (two are currently being developed 31,32 ). Therefore, it was considered that an examination of the "process factors" (eg context; change over time; relationships), as well as more concrete outcomes, was required.
Rigorous qualitative methods, carried out by members of the team with awareness of potentially significant process issues and juncture points over the years, were considered the most appropriate way to investigate the work. All authors (ie two adult researchers and three young people) collaborated to design the data collection methods. Qualitative interviews with the senior academic researchers and involvement facilitators from the Transition programme were carried out (n = 10). One person (the involvement lead) carried out the interviews. Four iterative focus groups with UP Group members took place. Six UP Group members opted in. The PSWs carried out the focus groups, with guidance from the involvement lead. They developed creative and accessible focus group methods (see Figure 1). Initial prompt questions for the interviews and focus groups were based on existing literature and developed through reflective discussions.
Interviews were recorded, anonymized and transcribed verbatim. Focus groups were transcribed in real time by two note-keepers working independently (ie the involvement lead and one of the F I G U R E 1 Methods for focus groups with (UP Group) members Focus Group 1-Design of the research: How did this influence involvement?
Super-sized board game set up on walls around the room. "Players" divided into small groups to design a human figure as a paper counter. Rolling giant dice, player answered the question landed upon and then everyone else contributed comments. Fun and interactive session covering many questions "quick-fire" but with time for additional discussion.

Example questions:
What was it about the Transition study that interested you? How do you think your work in the UP Group makes a difference to the research? What are the main things that UP has completed for the research so far?
Focus Group 2-Practical issues: How did these influence involvement?
Tabletop, paper-throwing game. "Players" wrote out and scrunched up prompt questions provided and put them into a large bowl. Took turns to pick out a question and unfold it. After giving answer they scrunched it up again, with it thrown from person to person, for all players to answer. Boiled sweet taken for every question answered. Throwing phase was timed with an egg timer, to allow the interactive element, whilst managing the length of the session.

Focus Group 3-Involvement approaches
Two life-sized body outlines drawn. Discussion prompted around the "voice" that members have and the barriers/facilitators to them using their "voice" or their "voice" being heard. Prompted a for-and-against format, to ensure debate and discussion. Answers written onto the body outlines.
involvement facilitators), and the exercise outputs were captured in written and drawn materials.
The central analysis was conducted by the involvement lead (Gail Dovey-Pearce) and a member of the research team with expertise in qualitative methods (Tim Rapley) according to the standard procedures of rigorous thematic analysis. 33 They worked independently on sections of the data, regularly coming together for discussions of their analyses, to interrogate their coding frameworks and interpretations. Techniques from first-generation grounded theory-coding and constant comparison 34 -were used, alongside deviant case analysis, 35 mapping, 36 analyst triangulation 37 and member checking. 38 The PSWs (Sophie Walker, Sophie Fairgrieve and Monica Parker) received periodic drafts of the analyses. They specifically engaged with the analysis of the focus group data. The UP Group received presentations of the focus group data analysis at the mid-point and end point, to consider if themes reflected their discussions.
The paper is informed by GRIPP2 8 reporting standards.

| RE SULTS
When academics and young people come together, formative cycles of work occur, as values, attitudes and practices develop. These cycles have the potential to increase perceived value and decrease doubts. There is also the potential for continuing doubt and active management needs to happen in order for the "work of involvement" to progress.

| Hopes and doubts
The adult researchers understood the rights of young people to be involved and the importance of avoiding tokenism. The work of managing hopes and doubts was not just central in the initial phase but was returned to over time. Values and attitudes appeared to shift steadily as the work occurred. The ideas of those that initially championed involvement became part of the way the group began to think. As an adult researcher outlined, through "experiencing the difference and value of what young people bring in," a transformation occurred. They noted that: [S]ome of the team members, in the anxieties they had about young people being involved, or being given responsibility, or whatever, well that's in the past because the project team has grown with it, and has learnt from it (AR3: [28][29][30][31] However, this was not about a key moment of "conversion" but rather a process of learning over time. An involvement facilitator described an element of the cyclical process: Little cycles all the time where you think… "we've achieved collaboration"… back to… "we are being con- To support such a coming together, the facilitators required experience of involvement methods and an ability to effectively mediate between stakeholders, to accommodate the different ways of thinking but also the emotional and social processes that played out.

| Negotiating the work
Interactions and negotiations between and within groups were cen- During this early period, the involvement lead was prioritizing initial tasks with the adult researchers and "commissioning briefs" were identified as a useful way to communicate and begin working together.
The involvement facilitators reflected on this iterative process in their team discussions, feeling they had to demonstrate, at different times, degrees of resilience (eg to be able to constructively challenge and assimilate various views); sensitivity (eg to convey outcomes that did not meet with initial expectations); negotiation skills (eg to manage points of power, responsibility and decision making); reflexivity (ie to move between representing the young people's views and expressing their own opinions); and pragmatism (eg balancing the desire for a priori conceptual alignment with getting on with the work of task delivery). They were aware of the importance of their debriefing, reflective discussions as a team, to enable them in this work.
It was very much about keeping a strong foundation about the core beliefs around involvement, but actually being pragmatic as to how we were going to deliver within the timescales the (adult researcher) management group needed. (AR8: 118-121) The young people reported that they felt they were given an overall aim and structure but with the freedom to work within that. Data

| Witnessing involvement work
One of the first tasks of the young people was to contribute to the programme's launch event. This was a visible role at an early stage. In this way, we see a shift from involvement work rendered as potentially tokenistic practice to being positioned as something that framed the direction of that event.
Beyond the launch, contact between adult researchers and young people was often mediated by a range of proxy actors. The young people were always able to decide how they took tasks forward, often delivering on things in a different way than might have been anticipated by the adults. They also had a work stream that they led on, around the scope and utility of health passports as a tool for young people using health-care systems. They were supported in this latter work by the involvement facilitators with little direct input from the research team. All aspects of the young people's work were outlined at the formal research management meetings, by the involvement lead as a proxy in the early years and then increasingly, with young people attending themselves. This was a space that attempted to build an overview and coordinate the diverse elements of the whole research programme and was attended by the senior researchers.
A junior researcher also acted as a proxy when they met with the young people to discuss a key data collection tool. The young people's input was seen as central to shaping the tool, and this was reported back to the adult researchers. The young people decided to use the work they had done on this tool, to develop an interactive learning resource for professionals, which has been made available via one of the voluntary organization partners of the research programme. This demonstrates the flexibility that was required to support young people not just with the tasks that were more predictable but with the unexpected opportunities and the added impacts that can emerge. The PSWs reflected that at such times, they seemed able to "surprise" the adult researchers, as they demonstrated the value of the young people's work.
However, this transformation in perception seems to have oc- The young people also highlighted the importance of face-toface interactions. When asked if they thought the adult researchers appreciated their input, one young person noted that "Yes, they make the effort whenever they see us to make us feel that way" (UP 3; FG4: 68-69). Looking to the final stages of the programme, they described wanting further opportunities for directly working together. One young person outlined that: We would like the managers to come and tell us about their work, or make a short video for us, like we did for them, to see at one of our meetings. (UP 4; FG4:

83-85)
Even once involvement processes are occurring and working well, it remains important to be aware of different stakeholder views and the need for continued, two-way dialogue.

| Appraising value
In witnessing young people's input, the adult researchers appeared to place an increasing value on it, particularly on a sense of the "authentic voice" conferred upon the research. They highlighted that young people offered a "real-life" view, with a multidimensional narrative being achieved: "The whole is greater than the parts.
By their angle coming in as well, it makes the whole thing much more interesting and relevant" (AR1: 469-471). They noted how the young people could offer a clarity and directness of message.
They reflected on the evolving nature of the process and the importance of not taking a mechanistic approach, with overdefined, a priori goals.
There's a lot more trust and faith in [the young people] than I think people would've imagined possible …and whatever our ambitions at the start, we succeeded.
Whether or not it looks like what it was meant to look like (pause) but I think that's fine with public engagement … I think we should have that more emergent agenda, rather than a "we will do this and then we will do this." (AR6: 677-671) In this way, their vision of how "good" involvement could and should be organized was transformed away from a more consultative approach. They all felt that they would engage with the process again, with a renewed sense of the possibilities. They also stressed that such

| D ISCUSS I ON
In this study, a need for cyclical, dynamic and flexible approaches to involvement working is suggested. The face-to-face work of building relationships is highlighted, along with the need to focus on the emotional, as well as the practical issues that arise. We suggest that this is likely to be relevant to all involvement work, but a central challenge is to understand how approaches might need to be adapted when working with young people in research, as distinct from younger children and adults.
Acknowledging and working with difference was a central finding in this study. Differences were presented as questions and doubts: for example, Will young people be scientific enough? Will I be able to take part in meetings? and Will we be able to effectively bring together adult researchers and young people? The questions had an emotional resonance, with concerns that differences might derail or block processes rather than enhance them. Participants described potential ways to acknowledge and navigate the challenges and stated a desire to not resort to potentially tokenistic ways forward. They demonstrated an understanding that avoiding doubts and difference, no matter how appealing a way forward this seemed at times, was not likely to be helpful.
Exploring the utility of critical discourse in social science, Burman 40 argues that by assuming universalities and overlooking differences, rather than exploring ambiguity and variety, we reach a position of "banality" with seemingly shared, yet narrow understandings. Similarly, Cook 41 acknowledges the importance of "messiness" in research: In research, having multiple view-points, where each new view and theory is a springboard for further reflection, is an important way of finding new ways of seeing.
Our findings suggest that in order to achieve such discourse, continued reflexive appraisal is required to realize potential difficulties and respond with creative solutions. The current lack of involvement evidence means that issues specific to a project and other local factors are likely to be as important in defining possible ways forward. In this study, descriptions of ways to acknowledge difference and promote discourse included roles for mediators and proxies; central tasks for involvement facilitators; the PSW role; and using commissioning briefs and other ways to support "leaps of faith." Other mechanisms have been described in studies of adult involvement 12,42 and the emerging young people's involvement literature. 16,29 In this way, instrumental involvement actions and guidance are being proposed, 7,8,42,43 but we also need to consider the processes within which these actions need to be enabled. Current involvement practice can be seen as occurring within transformative "liminal spaces" where fundamental contradictions can arise, requiring communicative rather than instrumental action. 13 An overly prescriptive use of guidance to pre-specify the structure of involvement work should be guarded against, as it does not show us how to manage all various challenges and emergent opportunities that involvement working presents, 44 especially when working with young people.
We might look to agile models of working 45 adopted in technology and other industries, where an awareness of local problems without a known solution, and the potential skills needed, quickly brings people with varied expertise together, to work within sprint cycles. There is an openness to acknowledge that some elements of the work might succeed and some might not. Any learning is taken forward into the next sprint cycle. Such a cyclical pattern was described within our findings, and the involvement facilitators The openness and transparency required for such agile and liminal ways of working is articulated in this study, with a suggestion that academics need to let go of aspects of the "professionalized self." It is suggested that involvement work requires a critically curious standpoint, rather than being fixed on maintaining supposed existing positions that are likely to be based on assumptions and generalizations. 40 They are also likely to be based upon pre-existing power relationships, as described in an examination of how experiential capital gained by patients might be recognized alongside the academic capital held by researchers. 47 Face-to-face working is suggested in this study as a key mechanism for exploring the "liminal knowledge spaces" between teenagers and adults, and between service users and academics.
Constructivist approaches to adult learning 48 stress the importance of recurring experiential opportunities, and not just knowledge acquisition, in the on-going transformation of thinking and practice. Similarly, the learning of children and young people is reported as a social and cyclical process, based on co-operation and interdependence, with face-to-face working as a key mechanism. 49 Within our own study, we have seen how direct social and verbal exchanges facilitated deeper levels of understanding and the mutual negotiation of meaning. We need to come together to negotiate a balance between: "ill-informed social experiments where any [involvement] practice is legitimate… [and] …the determinism of top-down control by experts". 50 However, it is suggested that currently, no easy consensus will be reached: Both literature and practice remain mired in a 'conceptual muddle'…and the principles underlying the why, whom and how of (involvement) remain confusing and contradictory. 49 To strengthen the debate, we need to move beyond a focus on proving the worth of involvement outputs, and consider involvement as a complex social process. 13,51-54

| CON CLUS ION
Adult service users may have a range of work-based skills or experience of other structures that will shape their approach to a research role, yet questions would likely still arise about their research "credentials" in the liminal space currently occupied by involvement work. In our study context, the "burden of proof" seemingly needed to justify the efforts of meaningfully involving young people in research may have been heightened by them being aged 15-25, by perceptions about their lack of professional and work-based experience. This may have also added to initial anxieties and the sense of "surprise" when adult researchers witnessed the young people's work. We suggest that the findings of this study add a valuable insight into work with young people in research and that adult academics might need to reappraise their assumptions about the capabilities of young people as researchers.

ACK N OWLED G EM ENTS
The authors would like to thank Caroline Potts, Norma Cardill and Molly Pledger; The UP Group and staff team; and the Transition Programme Collaborative Group for their support.