As Artificial Intelligence (AI) conversational agents become widespread, people are increasingly using them for health information seeking. The use of off-the-shelf conversational agents for health information seeking could place high metacognitive demands (the need for extensive monitoring and control of one’s own thought process) on individuals, which could compromise their experience of seeking health information. However, currently, the specific demands that arise while using conversational agents for health information seeking, and the strategies people use to cope with those demands, remain unknown. To address these gaps, we conducted a think-aloud study with 15 participants as they sought health information using our off-the-shelf AI conversational agent. We identified the metacognitive demands such systems impose, the strategies people adopt in response, and propose considerations for designing beyond off-the-shelf interfaces to reduce these demands and support better user experiences and affordances in health information seeking.

OBJECTIVES: To understand healthcare providers' experiences of using GlucoGuide, a mockup tool that integrates visual data analysis with algorithmic insights to support clinicians' use of patientgenerated data from Type 1 diabetes devices. MATERIALS AND METHODS: This qualitative study was conducted in three phases. In Phase 1, 11 clinicians reviewed data using commercial diabetes platforms in a think-aloud data walkthrough activity followed by semistructured interviews. In Phase 2, GlucoGuide was developed. In Phase 3, the same clinicians reviewed data using GlucoGuide in a think-aloud activity followed by semistructured interviews. Inductive thematic analysis was used to analyze transcripts of Phase 1 and Phase 3 think-aloud activity and interview. RESULTS: 3 high level tasks, 8 sub-tasks, and 4 challenges were identified in Phase 1. In Phase 2, 3 requirements for GlucoGuide were identified. Phase 3 results suggested that clinicians found GlucoGuide easier to use and experienced a lower cognitive burden as compared to the commercial diabetes data reports that were used in Phase 1. Additionally, GlucoGuide addressed the challenges experienced in Phase 1. DISCUSSION: The study suggests that the knowledge of analytical tasks and task-specific visualization strategies in implementing features of data interfaces can result in tools that lower the perceived burden of engaging with data. Additionally, supporting clinicians in contextualizing algorithmic insights by visual analysis of relevant data can positively influence clinicians' willingness to leverage algorithmic support. CONCLUSION: Task-aligned tools that combine multiple data-driven approaches, such as visualization strategies and algorithmic insights, can improve clinicians' experience in reviewing device data.

Engaging with multiple streams of personal health data to inform self-care of chronic health conditions remains a challenge. Existing informatics tools provide limited support for patients to make data actionable. To design better tools, we conducted two studies with Type 1 diabetes patients and their clinicians. In the first study, we observed data review sessions between patients and clinicians to articulate the tasks involved in assessing different types of data from diabetes devices to make care decisions. Drawing upon these tasks, we designed novel data interfaces called episode-driven data narratives and performed a task-driven evaluation. We found that as compared to the commercially available diabetes data reports, episode-driven data narratives improved engagement and decision-making with data. We discuss implications for designing data interfaces to support interaction with multidimensional health data to inform self-care.

Tools for self-care of chronic conditions often do not fit the contexts in which self-care happens because the influence of context on self-care practices is unclear. We conducted a diary study with 15 adolescents with Type 1 Diabetes and their caregivers to understand how context affects self-care. We observed different contextual settings, which we call contextual frames, in which diabetes self-management varied depending on certain factors - physical activity, food, emotional state, insulin, people, and attitudes. The relative prevalence of these factors across contextual frames impacts self-care necessitating different types of support. We show that contextual frames, as phenomenological abstractions of context, can help designers of context-aware systems systematically explore and model the relation of context with behavior and with technology supporting behavior. Lastly, considering contextual frames as sensitizing concepts, we provide design direction for using context in technology design.

Patient-provider communication is crucial to the quality of patient care. To avoid communication breakdowns, existing studies often aim to improve patients' health literacy or providers' interpersonal communication skills. Little has been done to look at the interaction between the two parties or the situated health practices where the communication occurs. To understand the barriers and the underlying factors leading to challenges in the patient-provider communication, we conducted an ethnographic study with patients and providers in an emergency care setting. We report three different communication difficulties between patients and providers: 1) conflicting perceptions of how the ED functions, 2) different understandings of the alternative information sources available in the ED, and 3) different perceptions of what information is critical to ED care. Highlighting the importance of ED practice literacy and balanced information sources, we discuss design opportunities for systems to support effective patient-provider communication during emergency care.

Making effective use of patient-generated data (PGD) is challenging for both patients and providers. Designing systems to support collaborative and individual use of PGD is a topic of importance in CSCW, considering the limitations of informatics tools. To inform better system design, we conducted a study including focus groups, observations and interviews with patients and providers to understand how PGD is interpreted and used. We found that while PGD is useful for identifying and solving disease-related problems, the following differences in patient-provider perceptions challenge its effective use - different perceptions about what is a problem, selecting what kinds of problems to focus on, and using different data representations. Drawing on these insights, we reflect on two specific conceptualizations of disease management behavior (sensemaking and problem-solving) as they relate to data specific activities of patients and providers and provide design suggestions for tools to support collaborative and individual use of PGD.