The aim of the study was to evaluate how the intervention supported the elicitation and sharing of the personal values and priorities of individuals with MCC. In this section, we describe the conjoint and respective roles of the facilitated conversations and CC, the interactive visualization tool, in supporting priorities-aligned care. We elucidate the roles in two aspects: (1) improved communication of priorities through externalization and sensemaking, and (2) improved understanding of multiple care needs and roles in patients’ everyday lives and preparation for change through iterative reflection.
6.1 Improved Communication of Priorities Through Externalization and Sensemaking
We found (1) externalization and sensemaking improved communication of priorities, and (2) our tool most closely aligned with the data-frame theory of sensemaking which allows meaningful exploration rather than focusing on actions as the necessary outcome. Our study’s findings contribute to CHI by showing how the intervention addresses design opportunities for improving communication of priorities in chronic illness care suggested by prior work [
15,
54].
As with prior studies [
15,
54], we have found that externalization helps with the sensemaking of personal values and self-management. Lim et al. challenged the prevalent assumption that patients’ personal values simply require elicitation as patients are already aware of and understand their personal values [
54]. They further emphasized the significance of externalizing personal values to bring awareness to and make new discoveries about the interconnectedness of their values, self-care tasks, and health conditions. Both Berry et al. [
15] and Lim et al. [
54] emphasized the need for trained human interventionists to guide individuals with MCC through the externalizing and sensemaking process.
Compared to prior studies, the potency of sensemaking in this study appeared to be even stronger than what prior development phase studies had discovered [
15,
54]. People were making more connections between their values, self-care tasks, and health conditions, and were exploring the connections more deeply.
We associate this new finding with the use of facilitators trained in motivational interviewing, using fully functional interactive visualization and two sequential opportunities for facilitated visualization.The conversations with trained human facilitators helped participants clarify information needs that required meaningful exploration. Creating the visualizations with facilitators helped participants overcome difficulties such as not knowing which questions to ask themselves and how to dig deeper into what they know and feel. The visualizations themselves also fortified individuals understanding that they were not dealing with many individual isolated problems, but rather interconnected problems with underlying issues. Our findings of visualization enabling the identification of underlying issues align with prior work by Goyal et al. [
35]. Our work and Goyal et al.’s both show that interacting with the visualization made the connections among individual components more apparent and the visualization may have given starting points for the investigation of finding a single underlying potential issue affecting all other issues [
35].
The unique contribution of the visualization to the sensemaking of participants echoes findings by Yi et al. [
85] on the insight gaining process, specifically “providing overview” and “adjusting”. The interactive visualization uniquely contributed to sensemaking by acting as a base map to refer to in the process of insight gaining of “providing overview” - making sense of and finding which areas needing more investigation, thereby promoting further exploration of the dataset [
85]. The facilitated conversation helped individuals with MCCs who did not know where to begin their exploration in the insight gaining process of “adjusting” – exploring a dataset by adjusting the range of selection [
85]. By grouping and aggregating information from the large, unorganized dataset (a myriad of personal values, self-management tasks, and health conditions) with trained human interventionists, individuals with MCC’s search and working memory load were greatly reduced, allowing them to use their attention to find higher-level facts and ask new questions [
21,
34]. Thus, the trained human interventionist facilitation with the interactive visualization and self-reflecting with interactive visualization respectively enhanced individuals with MCC’s sensemaking process.
Our findings are most consistent with the data-frame theory of sensemaking [45, 46]. In the data-frame theory, sensemaking focuses on understanding the mental processes at work while performing functions (e.g., problem detection and identification, forming associations, anticipatory thinking). Meaningful exploration is the goal. Our participants described this type of exploration of the relationships between personal values, self-management tasks, and health conditions [
71]. This activity differs from the representation construction model of sensemaking (e.g., Mamykina et al.’s sensemaking-based disease management model [
58]), which considers action as a necessary outcome of sensemaking. In the representation construction model, sensemaking is a process of searching for the most optimal external representations of the sought and filtered information to efficiently use them in answering questions related to a specific task [
70]. The focus is on creating the external knowledge representations for a specific task [
70,
72]. Sensemaking by our participants, in contrast, focused on exploration as the goal without a task or action as a necessary end [
46].
Specifically, the first facilitated conversation supported individuals with MCC to perform the above-mentioned functions to define the initial information needs for meaningful exploration. Also, it supported rendering of the initial frame and subsequent iteration of frames composed of personal values, self-management tasks, health conditions, and the interconnections of the three components. The sensemaking process then moves to refining informational needs by elaborating, preserving, or questioning a frame and comparing frames while making further associations between the frame components [
45,
46,
71]. We incorporated the process of iteratively refining the initial frames created in the first intervention by employing a second intervention which followed Lim et al.’s suggestion of providing ample opportunities to reshape the individuals with MCC’s information needs continuously [
54]. Berry et al. claimed that reflection on values and health in the future allows people with MCC to clarify their healthcare priorities and articulate those with their providers [
15].
Prior to this study, manifestation and refinement of frames through sensemaking in sequential interventions had not been attempted. In our intervention, participants were able to connect the frames to a clinical visit more directly as they felt more confident about their priorities and had the opportunity to think further about how they could better articulate them to their providers. We acknowledge the intervention would not have succeeded without the participants’ agency to meaningfully explore and communicate with the interventionists. We attribute their active engagement and agency to having the provider visits after the first facilitated conversation and the participants knowing that the results they had selected would be shared with their providers [
15].
This study demonstrates the viability of the design opportunities suggested by prior work [
15,
54] on improving communication of care priorities. Through sequential interventions with interventionists, patients were able to (1) meaningfully explore their values and their connections to self-management tasks and health conditions and (2) connect sensemaking to a clinical visit more directly.
6.2 Improved Understanding of Multiple Care Needs and Roles in Patients’ Everyday Lives and Preparation for Change Through Iterative Reflection
In section 6.1, we elucidated on how the first and second facilitated conversations followed the data-frame theory of sensemaking which focuses on meaningful exploration rather than regarding action as the necessary outcome [
45]. However, many participants also unexpectedly talked of changing their behaviors following reflection on the connections between values, self-management work, and health conditions. This was due to the intervention helping individuals with MCC develop a holistic and contextualized understanding of their multiple care needs and roles in their everyday lives and being able to pinpoint the specific issues they wanted to address. Participants who embarked on change also used the second intervention as an opportunity for iterative behavior change and reflection over time. These unexpected findings contribute to CHI by showing how the intervention could reinforce self-management technology design opportunities.
Talk of change often started with deeper reflections on causal or correlational relationships established in the visualization and conversation with the interventionist, especially in the first facilitated conversation where participants uncovered how a base issue interplayed with other connected issues. This finding relates to Saksono et al. [
74]’s Experience-Reflection-Insight framework, which illustrates that abstract conceptualization (interpreting the reflected-upon experience) affects dialogic reflection (exploring data relationships and new perspectives), which leads to causal insights. It is also corroborated by Fleck and Fitzpatrick’s taxonomy of five different levels of reflection (R0-R4) [
31]. After describing the data laid out in CC without justification or reasons for actions related to the data (R0), participants then examined the relationships between two or more data points (R1). They established casualty or correlation between their previous experiences and their data in dialogic reflection (R2). Then, participants developed a new perspective for reassessing their orientation to perceiving, feeling, or acting in transformative reflection (R3). Participants’ reflections further encompassed critical reflection (R4), which involves taking into consideration aspects that transcend the immediate context, such as socio-cultural contexts and ethical or moral issues. These included participants reflecting on deep-seated beliefs of taking on the burden of managing their health and well-being themselves and internalizing problems they face in self-management tasks. Reflections from R1 to R4 were highly valuable in that they led to frequent and unexpected talk about changing behaviors.
The intervention was meant primarily to support care planning conversations with healthcare providers rather than patients considering and acting on changes in self-management of their health and wellbeing. The unbounding of the reflection process from a planned provider visit or agenda may have enabled reflection that led to considerations for behavior change. In prior development work for CC, Berry et al. emphasized balancing outcome-oriented reflection and exploratory reflection [
15]. They stressed that once the outcome was set as visit preparation, pre-existing perceptions of what the patient thinks they could share with their providers were reinforced. However, in this study, rather than making the goal of the interventions to be composing a visit agenda, we asked the patients to help their providers learn about what is important to them (e.g., personal values). This setting supported a reflection process that gave enough room for the participants and the facilitators to ask questions and explore values, feelings, motivations, and behaviors.
Moreover, having a dedicated time and space to externalize their personal values, self-management practices, and health conditions may have made participants more inclined to make behavior changes as it gave them the support with a trained facilitator they needed in knowing what to change and making the changes for themselves [
54]. The importance of this type of support in individual behavior modification is consistent with the concept of autonomy support within self-determination theory (SDT), which posits that an “autonomy supportive” environment is essential for establishing care partnerships that lead to patients’ autonomous motivation for healthful self-management through developing individual competence (ability to master environmental challenges skillfully) [
73]. Although SDT informed the design of CC, the intent of SDT in the intervention was to change how patients and providers interacted together in care planning, rather than directly influencing an individual’s self-management of their health. Although improving self-management was not the primary aim, facilitated reflection with CC also provided individuals with MCC an opportunity to think about the workload associated with treatment and the impact of the workload on everyday activities and patient identity [
28]. New insights from these reflections led naturally to talk of how individuals wanted to change their lives to be consistent with what was most important to them.
Although this study did not primarily aim to improve self-management, our results support the potential of design opportunities in self-management technology in the HCI field [
24,
30,
64,
65,
81].
First, our results support design that incorporates a holistic understanding of the multiple care needs and roles in individuals with MCC’s everyday lives. Complexity and individuality result from the number of different roles that patients hold in their everyday lives as a person and outside of the clinical setting [
4,
64,
65,
81] Our intervention design elicited both personal values and the key types of daily self-management work in those with chronic illnesses including illness, biographical and everyday work [
26,
27,
56]. This approach provided a holistic understanding of patients’ everyday lives which has been strongly supported by prior work on self-management technology design in the HCI field [
4,
30,
64,
65,
81]. In our study, we found that individuals with MCC having the opportunity to view aspects of their life beyond the clinical context led to an extended understanding of their different roles and responsibilities in their everyday lives. For example, beyond the role of medially managing their own diabetes, P12 had the role of a mother and sister who wanted to bring joy to her family through cooking holiday food. By understanding how each of their various roles and corresponding responsibilities interacted and affected one another, participants were able to prioritize their needs for more sustainable self-management of their overall well-being and health.
Second, our results also suggest potential for design that supports iterative behavior change and reflection over time in individuals with MCC [5, 15, 54]. We found that the second facilitated conversation and visualization helped most patients further clarify connections to underlying issues and deepen reflections on changes they are considering in their lives related to their health. Tools for reflection may support keeping track of how individuals with MCC’s perspectives evolve or any discoveries they make about misalignments or desired changes [
54]. However, relying on passive automation and predefined presentation of personal data hinders reflective thinking about the past and speculative thinking about the future [
5]. Thus, Lim et al. [
54] suggested providing ample opportunities to modify after initial elicitation to allow individuals with MCC’s perspectives to evolve. In this study, we conducted two sequential interventions which were not attempted in prior prototype development studies [
15,
54]. In the second intervention, participants who had acted upon the changes discussed in the first conversation reflected on their changes. After the second conversation, they further made modifications to their values and/or the connections between their personal values, self-management tasks, and health conditions. Such possibility of iteration seemed to have affected participants’ behavior change execution after defining the base issue affecting all other issues and changes they would need to make during the first intervention. With ample opportunities to engage in mindful and reflective thinking through designing and completing their visualizations, participants were able to further speculate about the future and even modify the changes they made to their lives to be consistent with what was important to them [
5]. Thus, although not all participants have executed a considered behavior change, the possibility to examine and iterate on the effects of the changes through a second facilitated visualization proved valuable to participants.
Lastly, our results demonstrate the potential for designers to use interactive and guided visualization to help individuals managing MCC identify and work through important issues one at a time and in order of priority. Participants identified and prioritized specific issues to address rather than dealing with all issues at once. Through guided conversations and the interactive visualization tool, participants could understand the problem they chose to focus on, what it interacts with, and how to resolve it. As patients were not previously able to pinpoint what was really making them feel burdened [
28] and not aware of their priorities and personal values [
54], having the opportunity to identify and choose a specific issue to address one at a time may have decreased the overwhelmingness of MCC care.