Research agenda setting with children with juvenile idiopathic arthritis: Lessons learned

Abstract Aim The aim of this qualitative study is to understand the research priorities of Dutch children with juvenile idiopathic arthritis (JIA) as well as researching how children can be involved. Background Several health research agendas have successfully been developed with adults but rarely with children. Children are still seldom recognized as possessing credible knowledge about their own body and life. This research project with focus group discussions and interviews with children with juvenile idiopathic arthritis (JIA) was an innovative addition to a nationwide prioritization of research questions of patients with JIA, their carers and health care professionals, based on the James Lind Alliance (JLA) methodology. Results Children with JIA appreciated being invited to give their opinion on JIA research prioritization as knowledgeable actors. They have clear views on what topics need most attention. They want more insight on how to medically and socially treat JIA so that they can better fulfil their aspirations at school, later in work and with their relationships. Conclusion We have identified the Top 5 research priorities for children with JIA. Most priorities are unique and differ from the priorities of the adolescents and young adults, parents and healthcare professionals in the main JLA priority setting exercise. Ultimately, two of the children's priorities were included in the final JLA Top 10.

. It is also underpinned by the democratic right to influence decisions affecting one's life (Thompson et al., 2014). So far, several research agendas have successfully been developed with adults but rarely with children (Hart et al., 2016).
Unfortunately, children are still seldom recognized as possessing credible knowledge about their own body and life. The lack of agenda setting with children is probably also triggered by the limited insight and expertise in how best to include their voice (Bate et al., 2016;Bird et al., 2013;McDonagh & Bateman, 2011;Thompson et al., 2014) and the idea that medical professionals and adults know best (Gibbs et al., 2018).
Social science research has shown that even young children are willing and able to be involved in research (Alderson, 1992;Christensen & Prout, 2002;Dedding, 2009;James & Prout, 1997;Nap-van der Vlist et al., 2019) and that they offer a unique perspective (Lems et al., 2020;Sarti et al., 2017;Schalkers et al., 2014;van Bijleveld et al., 2020). Together with the promotion of the international rights of children for their opinion to be taken into account (UN General Assembly, 1989), this has led to their stronger direct engagement in research (instead of their guardians representing them) and to a shift from researching on children to researching with children (Bird et al., 2013). Therefore, the question is not whether children could be involved in research but how to do this in a meaningful way. That is, working together with respect for their needs and competencies, and their voice to be heard and acted upon (Dedding et al., 2013;Gibbs et al., 2018).
This project with children with juvenile idiopathic arthritis (JIA) was part of a nationwide prioritization of research questions of patients with JIA, their carers and healthcare professionals, based on the James Lind Alliance (JLA) methodology. The JLA was founded in 2004 to facilitate Priority Setting Partnerships (PSPs) with patients, carers and clinicians. Their methodological approach is used worldwide, albeit rarely with children. We found one JLA PSP that worked with children only, namely, children requiring elective surgery for conditions affecting the lower limbs (Vella-Baldacchino et al., 2019) and one in progress, concerning children with cancer. 1 Further, one research project on research priority setting in the United States was conducted with children with rheumatic diseases and their carers living in the United States. However, children from the ages below 13 years were excluded, and only caregivers attended focus groups (Correll et al., 2020).
The research protocol (Schoemaker et al., 2018) and the outcome for the nationwide research patient priorities have been described elsewhere (Verwoerd et al., 2021); a process evaluation of the whole project is also published separately (Jongsma et al., 2020). This article solely focuses on the outcomes and the qualitative research methods that were additional to the standard JLA approach. The aim of our study was to explore the research priorities of young children with JIA, and to reflect on the process and methodologies used in order to guide future JIA research and funding to the issues that matter most to them. Furthermore, we aim to improve our understanding of how to meaningfully involve children in agenda-setting processes.

| Generating questions
This qualitative study used focus group discussions (FGDs) of 1 h each and interviews of approximately half an hour to involve children, aged 9-13, with a diagnosis of JIA, to formulate and prioritize research questions. We recruited our participants via convenience  During both the FGDs and interviews, we started with a warm-up activity to help the children to familiarize themselves with the researchers and with each other and to feel comfortable and included.
To ensure that the children were not alienated from their own experiences and lives in the process, we started by using images that represented different domains of their lives.

Key messages
• The children's top research priority is the influence of JIA on future opportunities regarding school results, work and relationships.
• Adding qualitative research methods to the JLA approach enabled children to (collectively) reflect on their lives and to familiarize themselves with the role of agenda setting, research and the broad field it encompasses.
• Dialogue between children, young people, caregivers and healthcare professionals needs to be carefully facilitated to ensure all perspectives are mutually understood and acted upon.

| Analysing questions
First, we analysed the transcripts of the first two FGDs and the interviews using Microsoft Word comments and reached agreement on the questions that the children wanted to be studied, and delivered the list of questions to the PSP steering group. The steering group received questions from patients of all age ranges, carers and healthcare professionals and categorized these into summary questions (Verwoerd et al., 2021). The summary questions based on questions of children were categorized into main themes for the children and used as a starting point for the discussion about their research priorities.

| Interim session
The next step involved organizing two sessions for the children's interim agenda setting (CD and KA), again at the annual patient information day (November 2019). The main themes were written on envelopes and presented in two separate meetings to the children.
First, each child shared in the group what their personal preferred theme was and why (see Figure 2). The children were then able to open envelopes that contained the corresponding summary questions to help them to narrow down their preferred research questions. We then invited them to jointly discuss and decide the Top 5 themes of the whole group, leading to a final Top 5 of research questions for F I G U R E 1 A flow chart of the participation of children, in addition to the model of Verwoerd et al. (2021, fig. 1) showing the regular James Lind Alliance (JLA) process in grey and in blue and white the additional research activities. FGDs, focus group discussions F I G U R E 2 Children during the interim agenda setting. They view the envelopes with the main themes and write down their personal preference each group. Next, the two groups compared their respective Top 5 and then, through facilitated discussion, formulated a final list in order of preference. Briefly after the FGDs, the children physically handed this final list in a golden envelope to the PSP steering group.
Afterwards, we analysed the chosen themes and how they supported their choice so that we could select the best corresponding summary questions (see Table 3).

| Final analysis
The list of the children's research priorities, and their arguments as articulated during the process, was used for a basic content analysis (Hsieh & Shannon, 2005) (KA and CD), in order to give context for their prioritized issues.

| RESULTS
In total, 40 children (24 girls and 16 boys, 9-16 years, living with JIA for between 4 months and 9 years) participated in the FGDs and interviews (see Table 1). Further, 42 children of the same age bracket filled in the surveys. On the first online survey of this PSP, no more than 23% of responding patients were younger than 16 years, and only 16% attended primary school. On the second survey, no more than 2.4% of the responding patients attended primary school (Verwoerd et al., 2021). By undertaking the phases of the JLA PSP, saturation was reached on the identified research questions of the children.
In Part A, we first describe the Top 5 research priorities identified by the children, followed by an explanation of why this is of importance to them. In Part B, we reflect on the process and lessons learned.
3.1 | Part A: Children's priorities for research The first two FGDs and the interviews provided 81 questions. These questions were added to the PSP question database of the survey as part of the JLA methodology. The steering group received in total 604 questions of patients, carers and healthcare professionals and categorized these into 53 summary questions (Verwoerd et al., 2021).
Twenty-three of these 53 summary questions included the questions that were upraised by the children. Furthermore, the children who had participated in the online or hardcopy survey also delivered additional unique research questions, which were categorized under mostly the same summary questions. Five extra summary questions were added.
We felt we needed to focus the discussion. Therefore, we omitted the 25 summary questions that were not based on original questions from children. As a result, the children did not have to discuss 'uninteresting' summary questions. This led to a total of 28 summary questions based on concerns raised directly by children. Further, we thought 28 questions were too many to discuss. Therefore, we grouped them into 12 themes (see Table 2) and reworded them into child-friendly language so that children could easier oversee it and make priorities. Table 3 shows the results of the interim priority setting sessions where children with JIA chose their final Top 5, followed by an explanation of why these matter to them.

| Impact on (future) school results, work and relationships
The children stressed the impact of JIA on school results, work and relationships. First, almost all the children experienced a strong impact from JIA on their schooling, as it hindered their mobility at school: T A B L E 1 Characteristics of focus group participants (n = 31) and interview (n = 9)

| Safe practice of sports
The fifth question that the children prioritized was about sports: 'What is the best way to practice your favorite sports safely?' Eva (10) described being comforted by friends if she is tired during sports: they tell her to stand on the side for a moment, and they sometimes join her, 'I think that's sweet.' Bas (10)

| Part B: Reflections on the methodology
Most of the children were pleasantly surprised that they were asked to think about what should be researched for children with JIA. It made them feel important and recognized as serious stakeholders.
Although some spontaneously shared some first thoughts, most needed time to get acquainted with the purpose of the consultation.
Creating an informal and relaxed atmosphere, starting from their own experiences and lifeworld, the supporting tools (the cartoon and the template) and the interaction with the experienced facilitators and their peers, helped them to get started.
We know from our work in the field of participatory health research that participation is most meaningful if it starts from the lifeworld of participants and the process is carefully facilitated, with room for flexibility (Dedding et al., 2013). Therefore, we did not simply collect questions but jointly generated questions, jointly because this enabled children to develop their voice in a safe space of their own peer group. Through dialogue, they could reflect on their (common) issues and impact of JIA on their lives, and what matters most to them. Working together with peers, feeling understood, mobilized their energy and sense of empowerment.
Although we originally aimed to work with children aged 10-13, in practice, we broadened the criteria. This was partly because the patient organization did not want to exclude enthusiastic members but also because it touches upon key principles of participatory health research, namely, shared ownership, being locally situated and inclusion (the right to have a say) (Abma et al., 2019;Dedding et al., 2013; International Collaboration for Participatory Health Research  Table 4).

| DISCUSSION
The objective of this project was to engage children in a national JIA research agenda-setting process to improve the relevance of JIA research. In this additional qualitative trajectory, 40 children with JIA shared their personal experiences and what they think is the most important to study, namely, the impact of JIA on their (future) opportunities regarding school results, work and relationships. They look at their JIA not only as an inhabitant of their body but also as a person who wants to belong and take part in normal everyday life. This may not come as a surprise, as young people with JIA in earlier studies showed that JIA impairs children's capacity for social participation (Tong et al., 2012) and triggers fears for the future, feeling unsure of the physical, psychological and social impact of JIA in their lives (Eyckmans et al., 2010) including being rejected if they tell others about their JIA (van Gulik et al., 2020).
Three questions in the children's Top 5-numbered 14, 16 and 19 in Table 4-were ranked much lower by the other three groups.
This illustrates the added value of the focus groups (Verwoerd et al., 2021).  (Verwoerd et al., 2021)  The following questions were also discussed and put in order of priority at the workshops:

11
What are the long term physical consequences of JIA?  (Correll et al., 2020). The authors mention that the differences between the results in the United States and the United Kingdom could be related to disease-and culture-specific outcomes. However, this may also be related to differences in age and lived experience between the children and parents, as is the case between the children and parents in the Netherlands (Verwoerd et al., 2021). It may also be due to differences between the research approaches, as children attended

| STRENGTHS AND LIMITATIONS
Most of the involved children were members of the patient society.
Working closely together with a patient organization provided the opportunity to bring children from various parts of the country together at the annual event, although it also created a selection bias.
We therefore deliberately reached out to interview children in hospitals. However, these individual children were not part of the final working sessions, as these took place at one of the annual events. By undertaking the phases of the JLA PSP, saturation was reached on the identified research questions of the children. However, future research could explore these research questions in more depth.
The difference between the agenda of the children and the final JIA PSP Top 10 research priorities cannot solely be attributed to their age. The difference in approach and methods might also be a reason.
The advantages of interviews and FGDs were multiple. First, parents were not involved in the discussion, whereas for the survey, they were likely to be supervising the children's participation. Second, through the discussions, the children had the opportunity to learn more about the broad field of research. Third, in the FGDs, they could reflect together with peers, which strengthened their confidence in articulating their priorities. Fourth, both methods were deliberately designed to start from their lifeworld-ensuring genuine participation to share their experiences. Finally, carefully listening to them provides context for their prioritized topics. These reflections might also be relevant for guiding agenda-setting processes with adults next to the current approach. Signifies that a question is ranked in joint place with another question.