RTA uncovered a narrative of 5 storybook themes. Overall 3 themes described the roles participants played within the treatment situation, and 1 theme described the collaborative barriers and challenges across contextual settings. Theme 5 identified core features and collaborations based on the participants’ visions for an mHealth app. The insights from the analysis were summarized within a matrix to inform a conceptual model, identify principles for expanding the design of mHealth core features, and enable patient-parent-GP collaboration and shared decision-making.

The first theme comprised statements describing how participants experienced their emerging knee pain and the challenges related to the everyday management of knee pain. The analysis revealed how young adults described their emerging knee pain as fluctuating or something that emerged in different situations such as stair climbing, bicycling, running, sports, and gym class and affected the adolescent’s ability to engage in valued activities. The young adults described being tasked with exploring ways to cope with emerging pain and pain-related frustration, while managing the social consequences of being limited. A common theme during workshop 1 was how emerging knee pain initiated a vicious cycle in which adolescents’ efforts toward minimizing the social consequences or hiding their condition resulted in increased pain. A young adult described the cycle as follows:

One of the main challenges described, related to the invisibility of knee pain and how the adolescents were dependent on others recognizing their pain (peers, parents, and coaches) to avoid being branded as lazy, whiney, or careless. Another recurring theme related to how fear of being “benched” may result in adolescents “forgetting” or ignoring their pain to fulfill social obligations or avoid exclusion. A participant described how she considered ignoring her pain to avoid missing out of activities with friends:

Alternatively, 1 group (workshop 1) described how acceptance from others or honesty about the knee pain was important and empowered adolescents to stop hiding the pain and focus on managing the condition. A participant articulated the link between gaining parents and GP’s acceptance and managing the knee pain in other situations:

Apart from acceptance, participants described performing regular knee exercises and learning to “find the limit” with their knee pain as essential for breaking the vicious cycle and balancing self-care while performing everyday activities and how this was challenging for adolescents. Although young adults and parents highlighted adolescents forgetting their knee pain in nonimpact situations or losing faith in exercises as barriers to breaking the negative spiral, young adults and GPs emphasized how learning to differentiate between good and bad pain is essential for managing the knee pain:

Finally, young adults, GPs, and parents highlighted how adolescents may struggle to remember and expressing their pain in words when asked by parents or GPs. This posed a challenge when reaching out to parents or health care professionals for support in managing their knee pain. One group of participants (workshop 1) highlighted fear of stigma as a contributor to this problem, whereas another group’s (workshop 1) comments indicated that adolescents lacked the vocabulary for describing their pain beyond the immediate pain experience. A participant (workshop 1) described it in following way:

A recurring theme during the second workshop was parents referring to taking on the parent role to solve a problem. Although the parents generally recognized the adolescents as individuals with their own opinions, experiences, and desires, the parent role term was often used in recognition of how certain aspects of management were difficult for adolescents. Thus, parents had to step in and take control to remove barriers that inhibited adolescents’ management of their knee pain. A parent described the parent role in terms of supporting their child:

Another participant exemplified how taking the parent role also meant stepping in and setting boundaries when they felt their adolescents were not able to do so themselves. A parent described setting limits for her son’s participation in soccer:

Although parents from both groups highlighted being present, listening, taking complaints seriously, and setting boundaries as important for supporting their adolescent’s health decisions, both groups described alternating among the 3 tasks of emphasizing with the adolescents, advocating for the adolescents, or reassessing their own understanding of knee pain to support their adolescents by creating situations where adolescents were capable of self-management. Furthermore, parents described having experienced how their adolescents struggled to remember, understand, and express their knee pain in words. This complicated the parents’ task of assessing when to seek treatment, resulting in parents overlooking or negating adolescents’ attempts to express their knee pain. Thus, the parents had to learn to read between the lines, within the adolescents’ descriptions. This need was exemplified in a parent’s description of discussing the knee pain with her daughter:

In terms of advocacy, parents in both groups described instances where they had stepped in and negotiated on their adolescents’ behalf and how this advocacy initially occurred in the clinical setting and extended into the parents and adolescents’ networks after consultations. Parents described that negotiation with the GP aimed at supporting adolescents in articulating their pain and ensuring that their child benefited from their consultation. Parents described how advocacy also included withdrawing from treatment or seeking alternative treatments and information sources if they felt invalidated or that the GP did not meet their needs during consultations. A parent described how her expectations had prevented her from seeking additional treatments for her daughter:

The data from workshop 3 revealed several tasks, responsibilities, and dilemmas, which GPs had to navigate when treating adolescent knee pain. GPs described taking on the role as teachers or coaches, tasked with guiding the adolescents into a positive spiral with decreased somatization; better disease management; and sustained, balanced participation in sports as their main goals when treating youths with knee pain. Through this, the GPs have to balance the tasks of managing the adolescents’ pain in situ, setting a stage for self-management in the future, gatekeeping, and navigating systemic constraints. However, the GP’s main goal was described as ensuring that adolescents learned to manage their condition, as described in the following quote:

The participants’ statements during workshop 3 indicated how treating adolescents with knee pain was a 2-step process and how ruling out serious pathologies or trauma, diagnosing the condition, identifying the right treatments, informing, and managing expectations was part of the initial step of treatment. A GP described how identifying alarm symptoms was important:

Besides momentary management, the analysis revealed how GPs developed and used different behavioral strategies in tandem with usual care, to encourage the adolescents to explore, gain insights, and gradually become better at making health decisions going forward. A GP described the strategies he used to supportive strategies:

Although one GP group highlighted how this required understanding the “full patient,” other strategies included addressing the patients’ worries and concerns to facilitate acceptance, encouraging trial and error by providing suggestions for managing pain fluctuations, exercising, motivating adherence via goal setting, and establishing working alliances with parents. However, GPs highlighted that their efforts toward supporting future self-management depended on whether adolescents felt that following the GP’s advice allowed them to better understand and control their pain. A GP described the importance of understanding the whole patient with the following quote:

GPs identified several constraints within the treatment situation that had a direct influence on the GP’s treatment decisions and possible outcomes. The long disease course of knee pain and patients slipping through the cracks were highlighted as the main concerns and challenges of the GPs. Moreover, GPs pointed to adolescents’ difficulties in articulating their knee pain and patients or parents’ misunderstanding of GP’s instructions as barriers, which contributed to adverse outcomes such as dissatisfaction, withdrawal from treatments, and parents insisting for surgery. A GP described how knowing when his message got across to the adolescents was a challenge. Other GPs suggested that forming a therapeutic alliance with parents could help facilitate the knowledge translation, avoid withdrawal, and provide GPs with the ability to monitor and adjust their treatments by proxy. A GP elaborated further on this, by stating how maintaining adherence was ultimately the patients and parents’ responsibility:

Our workshops uncovered how adolescents, parents, and clinicians engaged in different types of collaboration aimed at empowering adolescents to enter an upward spiral with increased understanding of the disease and self-management. The analysis identified several communicative barriers, which lead to tensions in parent-patient-GP communication. Young adults, parents, and GPs highlighted adolescents’ difficulties in remembering and verbalizing their knee pain as a major source of tension. GPs described how this acted as a barrier to fulfilling their role in terms of diagnosis, management education, and planning future treatments, and parents and young adults corroborated this, with young adults suggesting that adolescents’ inability to explain their knee pain may be related to lack of in-depth understanding of their knee pain. A GP suggested how pain diaries could be used to alleviate tensions, by helping adolescents to articulate developments in their pain:

Another tension source was related to adolescents or parents’ expectations of obtaining a solution to the knee pain when entering a treatment collaboration. This was corroborated by young adults and parents, who described how being told “wait and see” could lead patients and parents to conclude that the GP did not believe them or know how to treat their knee pain. A parent described feeling invalidated after receiving the “wait and see” recommendation for her adolescent, and this had affected her expectations of future consultations:

GPs corroborated this during workshop 3, by highlighting how they knew there were limits to what they could do for adolescents consulting with knee pain and how managing parents’ expectations was quintessential when gatekeeping, to avoid parents becoming frustrated and seeking other treatments prematurely. GPs also discussed using imagery to give parents something tangible, build alliances, avoid withdrawal, and prevent parents from insisting for surgery:

Furthermore, GPs highlighted systemic constraints such as consultation times, subpar IT systems, and loss of communication owing to referrals as barriers leading to loss of contact with patients and parents:

Parents described how their lack of knowledge about knee pain caused tensions when assessing whether additional treatment was merited, when advocating with the GP, and when communicating their adolescents’ conditions and forming alliances with actors in their networks (teachers, coaches, and other parents). Thus, parents and young adults described how parents’ lack of knowledge meant that they risked overlooking or negating adolescents’ symptoms, unnecessarily restricting their sports participation, or accepting nonbeneficial treatments. A parent suggested how tailored patient information could help parents to know when to seek additional treatments:

Parents and young adults described an emerging dynamic, which eventually led to the formation of an alliance in which parents helped adolescents to create space for their self-management in everyday situations. Although young adults described how parental recognition made it easy to confront teachers, peers, and coaches about knee pain, reaching out to parents meant risking being dismissed or restricted from sports participation, which created tensions. In contrast, ensuring their child’s well-being was highlighted as quintessential to parents, but their lack of understanding about knee pain sometimes led them to take the wrong actions when their adolescents presented pain. Parents described how taking a more trusting approach reduced tensions and allowed them to focus on gatekeeping; advocacy during GP visits; and engaging with teachers, coaches, and other parents to create space for their adolescent’s self-management. A parent articulated this in the following way:

Finally, the analysis identified several visions for mHealth core features for enhancing collaboration and shared decision-making across collaborative spaces. The visions were distributed across 3 categories, directed toward enhancing reassurance, supporting pattern recognition and articulation of knee pain, and enhancing 2-way communication. However, participants described these core features as intersecting and needed to support different activities simultaneously for maximum effect. A GP described how his group envisioned that an mHealth app should support different tasks simultaneously:

Participants envisioned an mHealth app containing features for reassuring adolescents and helping them to test and evaluate their management decisions, when the knee pain emerged in everyday situations. Both GPs and young adults suggested how a library (videos) with trustworthy information about knee pain mechanisms, possible trajectories, and a first aid kit for managing flareups could reassure adolescents, promote self-education, and allow them to share this knowledge with peers. However, the young adults suggested that patient cases with other adolescents with knee pain were more easily relatable for youths and could provide hope for betterment. Furthermore, GPs and young adults highlighted that adolescents sometimes struggled to remember and comply with exercise programs and that adding a library with exercise videos and in-depth explanations could reassure adolescents that they were performing the exercises correctly. Ensuring that exercises were actionable (short and easily understandable) while combining them with a tracking feature could motivate exercise adherence by visualizing the short-term and long-term effects of exercises:

Participants across all groups suggested having core features that empowered adolescents to monitor, explore, and identify patterns in their knee pain. Both young adults and GPs described how a journal feature could support adolescents’ self-management by helping them in identifying activities that caused pain. Nevertheless, young adults and parents suggested incorporating reminders and predefined pain scales to reduce the burden related to monitoring the knee pain. All participants suggested that visualizing journal entries could help adolescents in overcoming their challenges by remembering pain-causing activities and articulating their knee pain when it emerged. Young adults suggested how incorporating a map visualizing common developments in the knee pain could assist adolescents and parents in assessing how the knee pain progressed and establishing treatment goals. This was corroborated by GPs who described how this feature could help adolescents in identifying activities that would not affect their knee pain:

Finally, participants envisioned how core features could be expanded to enable negotiation of meaning and shared decision-making, but this required a balance because actors had different information needs. Participants generally agreed that the journal and visualization features were central to this, by providing GPs and parents’ insights into the adolescent’s experience. Young adults and GPs described how visualizing journal entries could help resolve tensions in GPs and adolescents’ communications during clinical visits by providing a common ground for discussions, whereas GPs and parents described how visualizations could also help GPs in adjusting treatments to the patient’s needs. However, GPs described how this required visualizations to be aggregated for easy overview to avoid time loss. Another vision related to the exercise library was how incorporating a checklist with symptoms to look out for could help parents in deciding when to seek additional treatments and prepare parents for engaging with coaches, teachers, GPs, and physiotherapists. Finally, GPs and parents described how an mHealth app could facilitate information flow during transitions between treatments or when negotiating with external actors (physiotherapists, coaches, and teachers), by alleviating tensions related to parents or adolescents forgetting information obtained from clinicians between consultations. This was corroborated by the young adults, who exemplified how an app could facilitate an ongoing negotiation among multiple actors, to ensure acceptance of the knee pain. This was exemplified in the following quote:

The matrix analysis informed the construction of a conceptual model. Organizing participants’ descriptions about their roles, tasks, challenges, and interactions within a system identified how adolescents, parents, and GPs were interconnected within a triadic relationship, where all actors engaged in different modes of management behaviors (Figure 2). Considering the identified tension sources, the model outlined targets for designing mHealth core features to bridge the gap between the supporting participants’ individual management practices and collaboration across multiple contexts.

The layout with the embedded triangles illustrated that the participants’ collaboration in managing the adolescents’ knee pain unfolded at the individual and community levels across multiple contexts. A key insight was not only how participants took on different roles, tasks, and responsibilities within the collaborative space but also how these roles were often dual-sided and contradictory in nature. The individual triangles (top, left, and right) were designed to illustrate how the actors (adolescents, parents, and GPs) navigated these role-based contradictions via their management decision-making (center) in their individual contexts—an act that was obscured to other actors unless disclosed via words or observable actions. The matrix analysis identified how all actors encountered management barriers, which they could not resolve themselves (eg, obtaining a diagnosis, gaining knowledge about knee pain, and securing social support) and caused tensions in the collaborative space (Multimedia Appendix 7). To overcome these barriers, actors engaged with other actors to draw upon their competencies (as adolescents with knee pain, parents, and GPs) to expand their management capabilities (decision-making), modify contexts (eg, being excused from gym class and creating a working alliance), or adjust their roles and tasks in the collaborative space. However, participating in these exchanges meant renegotiating the participants’ individual goals, roles, and tasks to be effective (center and inner triangles). When successful, the negotiation may strengthen the actor’s individual health decision-making capabilities, articulate shared goals, and cultivate working alliances. Nevertheless, failure to negotiate was identified as having a cascading effect, leading to increased tensions and complexity in the collaborative space, inhibiting shared decision-making, and prompting treatment withdrawal if the tensions were not resolved.

Considering participants’ visions for an mHealth app, described challenges, and identified tension sources (Multimedia Appendices 7 and 8), the matrix analysis identified several touchpoints where participants interreacted to resolve individual management challenges and how these interactions contributed differently toward sustaining the collaborative situation. This informed 3 principles for organizing mHealth core features as collaborative enablers for supporting tension reduction by empowering negotiation and informing shared decision-making.

The young adults, parents, and GPs envisaged how health information collected via quantified self-tracking could support adolescent-GP communication and how their information needs differed in terms of timing, timelines, and modalities. Participants described how the act of self-tracking knee pain via, for example, pain journals, receiving tailored patient information (etiology and exercise support), and performance feedback, could help adolescents in assuming the role of explorers through the identification and articulation of patterns in their knee pain. However, this required that the delivered health information should be actionable in everyday settings to encourage exploration, compliance, and articulation. Furthermore, visualizations of aggregated self-tracked data could help adolescents and GPs to overcome communicative barriers by assisting adolescents in recalling and articulating previous developments in their knee pain, while simultaneously giving GPs a foundation for guiding the adolescents—by providing GPs an overview of the adolescents’ trajectory, the ability to monitor the effects of treatments and exercise regimes, and a starting point for discussing future treatments. However, effective presentation and delivery of the self-tracked health information were crucial to ensure GP use in complex clinical settings.

Young adults highlighted how acceptance and adapting an honest perspective about knee pain was important for facilitating adolescents’ transition to self-management and how parental support could help adolescents to take on the role as explorers. The analysis revealed how different types of static information (patient cases, lists of symptoms, exercise videos, and patient information) could promote safety in making management decisions (individual level) by providing reassurance, along with vocabulary and expert information that adolescents could include when explaining their knee pain to peers, coaches, and GPs to avoid stigma. For parents, static information delivered with the app (eg, leaflets or webpages targeting parents and adults) could empower them to create space for adolescents’ exploration of their knee pain (decision-making) and remove management barriers through negotiations with other parents, teachers, coaches, and GPs. This included enhancing parents’ knowledge about knee pain symptoms and treatment types, while providing them with guidance and tools for how to engage and educate other actors, coaches, teachers, and other parents about knee pain and the management needs of their child.

Finally, the analysis outlined how communicative difficulties between GPs and parents could lead to tensions and parents deciding to withdraw and seeking other forms of treatments and how this was driven by parents not feeling seen or heard when consulting GPs. Both parents and GPs suggested incorporating core features that could help parents and GPs in entering negation and building alliances. Including a checklist for parents with symptoms and questions for GPs could limit tensions by ensuring that parents felt heard during consultations, while providing GPs space for addressing parental expectations to treatments. Furthermore, providing parents with information materials (folders and webpages) about the adolescent’s symptoms, treatment options, and prognosis could help them to adjust their expectations, while cultivating a sense of co-ownership and forming treatment alliances with the GP.

Qualitative studies outline how managing knee pain is complex and involves adolescents balancing several activities including understanding their pain, maintaining function in everyday situations, care seeking, self-reflecting, and adjusting to a life with pain [56,59]. Our analysis identified targets for designing mHealth apps for supporting adolescents’ efforts toward managing their knee pain between consultations. For increasing adolescents’ understanding of pain, participants suggested incorporating a first aid kit feature, with information about the etiology of knee pain and actionable advice for pain alleviation. This would provide adolescents with reassurance, while promoting reflections about pain triggers and pain relief strategies, similar to what Rathleff et al [12] documented in clinical trials with their activity management tool.

Regarding participating in care activities, participants highlighted how remembering and sustaining correct performance with exercises was a barrier, as observed by Faber et al [79] in youths. Participants suggested that incorporating a video feature with patient information, exercise instructions, and prompts for reminders could reduce maladaptive beliefs and empower exercise adherence. This aligns with the findings of Selhorst et al [80] and Holt et al [81], who showed that video-based handouts with patient education and exercise support limit maladaptive beliefs and boost adherence in youths with musculoskeletal pain. In addition, Riel et al [82] documented how live feedback increased the compliance of adolescents with knee pain with digital exercises. Studies exploring the use of gamification elements (leader boards, goals, minigames, leveling, esthetics, feedback, and rewards) to sustain long-term adherence to health interventions have shown positive effects such as enhancing medical adherence, disease understanding, and physical activity in adults and older people with chronic conditions [83-85]. Additional user-centered studies are needed to ensure that gamification elements align with adolescents’ needs to exert control of their knee pain [86] and are experienced as useful, meaningful, and inspiring for self-management in everyday contexts [83].

Finally, recall and articulation of knee pain were highlighted as barriers when seeking support from GPs and parents. Participants suggested that goal setting, quantified self-tracking, and pain journals could help adolescents in overcoming this barrier, by registering pain triggers, identifying pain thresholds, and assessing the value of behavior change. This resonated with the findings by Slater et al [69] about how adolescents were capable of setting personal recovery goals and the descriptions by Moon et al [72] and Lalloo et al [87] about quantified self-tracking's potential for enhancing adolescent-GP communication during clinical encounters, by allowing the adolescents to capture, explore, and articulate connections among management experiences, reflections, and outcomes [88]. However, our findings expanded upon this by illuminating the gap between ensuring that the self-tracked data are actionable to the adolescents and integrate into the GP’s decision-making process to facilitate mutual articulation. In total, our insights extend the current knowledge base and inform which individual management challenges should be addressed by mHealth core features in future mHealth concepts.

The young adults highlighted how GP and parental support had helped them to accept their knee pain and take on the role as explorers [34] and how experiencing having their requests rejected—real or perceived—increased doubts, dissatisfaction, and stagnation and prompted withdrawal, as observed in adults and youths with chronic pain [89-91]. Our analysis revealed how entering into treatment led to the emergence of a complex triadic relationship, where patients, parents, and GPs took on different roles, tasks, and responsibilities, similar to what Brooker [58] and Hohmann [57] outlined in the collaborative care triangle and what Kanstrup et al [92] presented as a complex interplay.

Nonetheless, our findings expanded upon this by describing how all participants were engaged in individual reflective processes and navigated both proximal and distal goals, as illustrated by Ryan and Sawin [93] in their family self-management concept. Furthermore, the emerging triadic relationship shared several properties with the communities of practice by Lave and Wenger [94] in terms of how participants organized, collaborated, and shared knowledge and how successful participation solidified the collaborative relationship. Ensuring that adolescents entered into an upward spiral acted as a shared domain of interest [95], with our analysis indicating how this acted as a cornerstone for shared health decision-making, as described by Makoul and Clayman [96]. Still, this required that said goals were articulated to be effective.

In terms of legitimate peripheral participation, our analysis outlined how adolescents were engaged in a situated learning process, which was driven by their explorative approach to managing knee pain [95]. In contrast, parents and GPs took on the roles as masters by observing, encouraging, diagnosing (GP), informing, educating (GP), and setting boundaries to facilitate the adolescents’ transition to self-management, as observed by Cha et al [43]. However, the absence of a shared vocabulary and repertoire for addressing self-management challenges made it difficult for participants to establish a working collaborative relationship. A future mHealth app should focus on creating a shared language for addressing the processual, proximal, and distal goals of participants [93] to support shared learning and avoid early treatment withdrawal.

A key insight was how participants navigated the collaborative space through 2 modes of management behaviors, echoing the descriptions of Brooker [58] about members alternating between professionalism and providing care. Although adolescents, parents, and GPs navigated individual management challenges via their management decision-making, our matrix illustrated how participants solicited the expertise of other actors to overcome individual or contextual management barriers—an act that was collaborative in nature and required renegotiation of goals, roles, and tasks to succeed. Literature highlights that negotiation is an essential component in shared health decision-making [60,96]. Our matrix analysis identified how adolescents’ articulation, memory for exercises, parent’s knowledge, the long trajectory of knee pain, and GP’s ability to engage with adolescents inhibited negotiation and shared learning and increased tensions within the collaborative space. On the basis of these insights, our conceptual model presented 3 principles for designing or consolidating mHealth core features to act as enablers for mediating between participants to support negotiation, reduce tension, and create a basis for shared decision-making across multiple contexts. The model identified targets for enhancing the interpretive flexibility of core features to accommodate multiple user needs simultaneously [97] to enhance patient-GP communication, support parents in facilitating the adolescents’ transition, and help parents and GPs in building mutual trust and enhancing patient alliances.

Our exploration of mHealth literature related to this study failed to uncover mHealth concepts that incorporated all 3 principles for supporting negotiation and shared decision-making simultaneously; 3 designs were identified, which included 1 or 2 of the previously mentioned principles. The PainApp described by Koumpouros [98] supports patient-clinician communication, by using quantified self-tracking and aggregated data to empower carers’ clinical reasoning when negotiating treatments with adults with musculoskeletal pain. Furthermore, the PainApp uses self-tracking notifications as an actionable component to encourage patients to track and reflect on treatment effects at home. The ICanCope concept by Stinson et al [70] used a hybrid design to support adolescents with musculoskeletal pain to transition to adult care. The ICanCope app uses theory-informed interventions (self-tracking, goal setting, coping skills training, and social support) that are actionable to enhance adolescents’ symptom exploration and management decision-making at home. The web component included education features with self-tracked data to support adolescents’ articulation and features (discussion boards, goal sharing, and self-advocacy skills) that we interpret could enhance adolescents’ safety during patient-clinician communication and when negotiating roles and space with parents. As no mHealth tools were identified with features that directly supported parent-adolescent collaboration, the diabetes management plan tool [99] was included to exemplify how mHealth core features may bridge adolescents’ needs for safety from sanctions and parents’ needs to frame and facilitate their child’s self-management when making shared management decisions. By allowing users to choose and share items representing adolescents and parents’ tasks and obligations, the tool inspires role negotiation and shared decisions by articulating the tasks and responsibilities of both parties.