Patient co‐design of digital health storytelling tools for multimorbidity: A phenomenological study

Abstract Background Interest in both narrative medicine and electronic health records has increased over the past 30 years. However, electronic health records are mainly written by and for clinicians, and the patient narrative and voice are not incorporated. Recent studies within the United Kingdom have indicated that there is a need to incorporate patient stories into health records, to improve quality and continuity of care. This is particularly important when treating people with multiple long‐term health conditions (multimorbidity), whose health stories can be particularly complex. Objective To understand the goals and requirements of people with multimorbidity for digital health storytelling tools. Methods The methodology uses narrative within a phenomenological approach to inform a process of co‐design. Results The findings indicate that people living with multimorbidity would use health storytelling tools to understand and reflect on their journeys, convey their experiences to others and advocate for themselves against scepticism. Conclusion Outputs from the project give insight into the lived experience of multimorbidity, as well as understanding the goals of people living with multimorbidity for using health storytelling tools as part of treatment and self‐management. Future research could explore other areas such as collaborative health storytelling or the technical implementation of tools. Patient or Public Contribution Five adults with multiple long‐term conditions participated in the project, and research was carried out in three stages. First, semistructured interviews were used to understand each participant's health story. Second, each participant worked with the researcher to co‐design a visual representation of their story. Finally, digital prototypes based on their health story were reviewed with each of the participants.

of people with multimorbidity are also steadily rising, making their treatment a pressing concern in healthcare. 10,11 The interactions between different conditions mean that the health stories of people with multimorbidity are necessarily complex, encompassing a wide spectrum of contexts beyond the clinical setting. 12 Understanding these stories is therefore a crucial part of treatment. 13 Within the past 5 years, several projects within the United Kingdom have identified a need for health stories to be included alongside patients' electronic health records, as well as exploring in a limited way how such tools might look and function. [14][15][16] This project builds on this previous work, but also looks at how health stories can be used more broadly: as a means of creative expression and personal empowerment, and as a tool for people to shape and reflect on their care. Adopting a phenomenological approach of dialogic, relational research, 17,18 participants worked closely with the researcher in cycles of co-creation, narrative analysis and reflexive digital design practice. The work was carried out as a year-long Masters of Research project, funded by Scotland's Digital Health & Care Innovation Centre. The aim of the project was to understand the attitudes of people with multimorbidity towards health stories, and their goals for using health storytelling tools.

| Reflexivity statement
The researcher on this project has lived experience with long-term health conditions, and professional background as a digital interface designer and developer.

| MATERIALS AND METHODS
The fieldwork used reflexive, semistructured interviews to drive an overall process of co-creation with the research participants ( Figure 1). Each participant worked with the researcher individually, meaning that no participant ever met another. Fieldwork was organized into three cycles of activity. Each cycle commenced with a participant interview. Narrative analysis 19 was used to extract the overarching themes in dialogue with participants, which then informed a reflexive practice of digital interaction design by the researcher. The output from the prototyping was then carried forward to form the foundation of the next cycle. The final outputs of the work were individual digital prototypes which reflected the health stories of each participant, alongside overall findings on their perspectives on the uses of a health storytelling tool.

| Participant recruitment
Participants in the project were adults with multimorbidity residing in Scotland. Following the phenomenological approach, the project recruited five participants, with multiple interactions with each participant.
Participants were recruited via patient support groups for conditions with a high incidence of multimorbidity. Interested participants were then F I G U R E 1 Methodology comprised of three cycles of fieldwork and analysis directed to a webpage where they could review the project information (Supporting Information: Appendix A). One participant (Hazel) found out about the project through word of mouth. All participants are identified by assigned pseudonyms throughout.
The final cohort was comprised of one man and four women between the ages of 31 and 53 and included the following conditions: borderline personality disorder, chronic migraines, depression and anxiety, early menopause, fibromyalgia, irritable bowel syndrome, myalgic encephalomyelitis (ME), polycystic ovary syndrome, posttraumatic stress disorder, postural tachycardia syndrome, psoriatic arthritis, Raynaud's disease and relapsing-remitting multiple sclerosis.

| Co-creation
The entire dialogic research process within this project can be viewed as a form of co-creation, defined as 'any act of collective creativity, i.e., creativity that is shared by two or more people'. 20 The cocreation took two forms. First, the research approach allowed 'dialogic co-creation' 18 of the research findings, which emerged organically from the relationship which developed between the researcher and the participant. Second, the project also explicitly used the method of co-design, 20 a process in which the researcher acts as a facilitator, creating toolkits that the participants can use to design their own preferred outcomes. The focus in co-design is therefore not on what participants say but on what they make, making co-design a useful way of gaining insight into designing for the future. 21 These activities also incorporated elements of elicitation, in that the co-designed artefact also served as a focus for discussion (described further in the following section).

| Reflexive interviewing and participatory visual methods (PVMs)
Reflexive interviews are semistructured, conversational and episodic, with periods of reflection by both parties in between. 22 This approach allows the participant to verify the researcher's analysis, creating a 'shared intelligibility' 23 which adds further rigour to the research. Interviews for this study were carried out remotely via Zoom during the COVID-19 lockdown and were recorded with the participants' consent. Each participant was interviewed three times over a 2-month period (Supporting Information: Appendix B).
PVMs 24 have been praised as giving participants an empowering voice within the research process. In this project, PVMs were used alongside the interviews to provide a focus for the discussion. The first interview used an adaptation of Marini's health storytelling prompts, 25 which have been developed to elicit a comprehensive health story from participants using only basic English. This was combined with a form of mixed object elicitation and photovoice, 26 in which participants were asked to select and photograph an object which represented their health to discuss with the researcher. In the second interview, participants were shown a variety of visuals, some of which had been created by the researcher through reflexive analysis. This was then used as graphic elicitation 27 during the interview. In the third interview, participants were shown an interactive digital prototype of a health storytelling tool based on their story. Again, this was used as graphic elicitation.

| Analysis
The analysis took place following each stage of the fieldwork (Supporting Information: Appendix C).
Narrative analysis 19 examining the holistic meaning and form was first used to develop an understanding of each participant's story.
Following this, further analysis was done to pull out themes from the findings, first within each individual story, and then across all the participants. At each stage, a written summary of the findings was taken back to the participants following a dialogical approach. 17 This allowed participants to review and dynamically consent 28 to their inclusion in the final output, as well as validate the researcher's analysis.
Using the findings from the narrative analysis, the researcher designed digital images and interactive online prototypes representing a health story tool for each participant. These were used during the second and third interviews as another form of graphic elicitation.
Although the design of the prototypes was a creative process, within this study it was also used as a form of analytical practice. This follows the argument of Creative Analytical Practices 29 whereby analytical output may take creative forms such as artwork or poetry.

| RESULTS
The outputs of the project were individual prototypes for health storytelling, co-designed with each of the research participants, as well as findings on patient requirements for a health storytelling tool. The following sections summarize the findings for each participant along with a description of their individual prototype. These were developed through the analysis done in dialogue with the participants. The final section summarizes the overall findings across all participants, which was done separately by the researcher after the conclusion of the fieldwork.

| Hazel
Hazel's goal was to use self-management to control her condition so that she could live her life as normally as possible. For her prototype, we used a sport metaphor to demonstrate a health story which is told through self-management activities ( Figure 2). Individual selfmanagement activities take on the role of players who can be selected by Hazel to form a strategy which will help her overcome particular obstacles. Strategies can be ongoing (e.g., dealing with daily fatigue) or short-term (e.g., planning for a social event).
Hazel can also record player profiles in the prototype which record detailed information about each activity and the costs and benefits of

| June
June often feels the need to spend time alone, a tendency that she attributed to her mental health conditions. However, emotional connections with others were also important to her, and she viewed her support network as critical for self-managing her conditions. She wanted to be able to share parts of the story with others, but only in a controlled way.
June's love of enclosed spaces, combined with her desire to control and separate the information she presented to others, resulted in the idea of a prototype based on a burrow with individual story caves ( Figure 3). Each story cave could be composed of separate episodes grouped together under a single heading. Caves could contain different groups of collaborators, playing on ideas of togetherness and also moving through a physical space.
Privacy was very important to June, so we included multiple ways for her to control content sharing. At the default level, content is simply visible to others. June could also have private stories which are only visible to herself or shared with trusted friends and family members. On the other side, June may have parts of her story where she is actively seeking input or advice from others. The prototype also included a way for June to seek guidance by sending out a call for help from collaborators-an idea inspired by her interest in spiritual guidance.

| Alison
The primary theme of Alison's story is memory. It is important to Alison that she remembers everything that has happened to her, and that she accumulates as many good memories as she can during her lifetime. She feels that she has a 'finite amount of time', as her primary condition, multiple sclerosis, could progress to the point where she is no longer able to do things. Alison wanted friends to be able to communicate with her about her story, but only in a general way. We represented this by allowing friends to post emoji reactions on her story (e.g., a heart), and also message her directly using a contact form.

| Gillian
Gillian discussed how difficult she found it to 'stay on course' with her treatments for secondary breast cancer. This included understanding her feelings towards her health and that of those around her, as well as keeping track of treatment options. We represented this using the concept of cloud cover, a screen which can be pulled up or down on the main page. Using this would let Gillian control how much detail she sees, and how far into the future her map will extend. When cloud cover is enabled, it could also flag to friends and healthcare professionals that Gillian does not want to have detailed discussions about her health.

| John
The genre of John's story is probably closest to Arthur Frank's idea of the quest, 30 in that he has a specific goal: that one day, there will be a treatment or cure for his primary condition, ME. A reoccurring theme in our discussions was hope. John felt that it was easy to 'burn up' one's hope in the search for potential treatments, and that this must be carefully cultivated and conserved for the future. A healthcare professional could use these to learn about something John is interested in and to understand John's approach towards his illness (as in Kleinman's idea of explanatory models 31 ). He is very involved with patient advocacy, so we also considered how the rutter might operate as a joint work with several contributors.

| Overall findings
Participants identified three primary audiences for a health storytelling tool: family and friends, healthcare professionals and the participants themselves (Table 1). They also identified a number of goals/uses for the tool ( Table 2). These can be distilled into three overarching goal categories: (1) Goal 1 (G1): Seeing the big picture; (2) Goal 2 (G2): Conveying the illness experience;  (Table 3). These can be roughly divided into two categories: the ability to record information about oneself, and the ability to share information with others. Appointments with healthcare professionals tended to be infrequent, so remembering details of daily activities was also practically useful. Participants related this to evidencing (G3), saying that they felt they sometimes needed to 'prove' what they had been doing to professionals.
Recording information was also important for participants to understand their conditions and ongoing treatments (G1.4). This was particularly noted by Gillian (who was undergoing numerous treatments for cancer) and John (who felt dissatisfied with the treatment options suggested by professionals).

Hazel June Alison Gillian John
Family/friends X X X X X Healthcare professionals X X X X X Self X X X X X  Participants also related knowledge sharing to patient advocacy (G3.3). Stories were seen as something which could be used to advocate for change-both within their immediate support circles and on a societal level.

| Desired functionality
The functionality of a health storytelling tool identified by participants can be divided into two categories: F1. Tracking information about oneself; F2. Sharing with others.
Within the first category (F1), the most important feature was tools for tracking mental health (F1.1), which was mentioned by every participant. In contrast, only two participants mentioned tracking physical health (F1.6), which may be due to the prevalence of existing tools for this purpose.
Privacy (F2.1) was the most important feature in the second category (F2). In some cases, participants wanted information to be visible only to themselves. While all the participants wanted to share their stories with others, they also wanted to limit how collaborators could interact with them (F2.2 and F2.3)

| DISCUSSION
The findings from the project show that the health stories of multimorbidity are complex, with implications far beyond a clinical frame. 32 Although ostensibly stories about health, the stories that participants told were really about their lives. Much of the impact that participants described sat outside of the normal scope of clinical treatment-for example, John mourning the loss of his boat and his ability to sail. Every participant had experienced difficulties with mental health, not uncommon among people with multimorbidity, 33 and there was a desire to have this emotional dimension reflected in their health records. Participants also rarely spoke of their conditions separately. 13  privacy showed that they would still want tight control over how this information was shared, and in many cases might only want to share general rather than specific information. June, for example, discussed not wanting to share her story with professionals that were dismissive of her concerns.

| Addressing inequality and power disparities
Participants also linked telling their stories to personal advocacy and addressing power imbalances. All of the participants discussed the difficulties of trying to make those around them (even close friends and family) understand the lived experience of their health. Some participants related scepticism to the stigma surrounding their conditions-this was particularly notable amongst participants who had ME, which is consistent with similar research. 36 Participants also said they faced inequality due to factors such as age and social background. In many cases, participants felt that they understood their conditions better than the professionals they were working with, reinforcing the idea that patients may become experts based on their lived experiences. 8 Participants commented very positively about the experience of sharing and seeing the representations of their stories, again demonstrating the positive effects of health storytelling on personal well-being. 3,4 This suggests that an alternate approach to future work might be reproducing or adapting the methodology used here for use in treatment and care.

| CONCLUSION
Previous research has identified a need for health stories to be included Centre.

CONFLICTS OF INTEREST
The authors declare no conflicts of interest.

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.