Paper Session 4: Digital Health, Care and Wellbeing
Date: Tuesday 3 Nov 2026
Time: 11:00–12:00
Add this session to your calendar, or click a talk’s time to add just that talk.
| Time | Presentation | Abstract | Authors | Contribution type |
|---|---|---|---|---|
| 11:00-11:10 | Constructing Metric Ecologies: How Fitness Tracker Users Distinguish Personal Health and Well-Being Metrics | Fitness trackers increasingly promise insight into both health and well-being. However, these constructs are hard to operationalise given they are not directly measurable via a single metric. Instead, people must interpret combinations of metrics—or metric ecologies—to make sense of their health and well-being. Yet, little is known about what metrics form these ecologies in practice, or how people distinguish health from well-being. To address this gap, we conducted an exploratory study with Garmin fitness tracker users (n = 21) using gaze-cued retrospective think-aloud sessions. Our findings show that participants treated health and well-being as related but distinct reflective constructs. Health was commonly understood as bodily and measurable, making it easier to interpret through objective metrics. Well-being, by contrast, was understood as emotional and contextual. We contribute empirical insights into how users define health and well-being and the relationship between these constructs, the metrics included and excluded within ecologies, and discuss opportunities for personal informatics systems to better support reflection of health and well-being. | Kersting, Rey, Vredenborg, Bentvelzen | Paper |
| 11:10-11:20 | Shift-Integrated mHealth for Safety-Critical Work: Formative Evaluation of a Police Wellbeing App | Designing mHealth tools for police personnel requires attention to irregular schedules, interrupted breaks and operational unpredictability. This paper reports a qualitative formative usability evaluation of a shift-integrated wellbeing app developed for police shift workers. Seven police personnel with 8–33 years of shift-work experience and three Human-Computer Interaction (HCI) experts participated in a 90-minute guided demonstration and focus group evaluation. Transcripts, facilitator notes and expert usability observations were analysed thematically. Findings show that the usability of a shift-integrated wellbeing tool was shaped by four occupational conditions: operational unpredictability, fragmented interaction windows, institutional trust concerns and shift-linked wellbeing needs. These conditions generated formative design requirements across four areas: operational adaptability, low-burden interaction, learnability and discoverability, and trustworthy schedule-aware wellbeing support. The paper proposes a formative analytic framing of the shift schedule as usability infrastructure in policing and related shift-work settings, describing how shift schedules may shape the timing, adjustability and perceived relevance of wellbeing support when operational work disrupts planned routines. | Ayorinde, Dogan, Adedoyin, Jiang, Bitters, Dempsey | Paper |
| 11:20-11:30 | Examining Personality-Aware Active Listening Strategies in Support Agents | Active Listening (AL) is a foundational social behaviour in support and coaching, yet its impact depends on how it is experienced by the person being supported. While prior work links personality—especially Openness and Agreeableness—to how people enact AL, less is known about how personality shapes preferences for receiving AL from conversational agents (CAs) in coaching contexts. To address this gap, we conducted a vignette-based human-subjects study (N=105) where participants evaluated text-based CA coaching dialogues that varied in AL intensity (low, moderate, strong), built using LISTEN-R, a theory-grounded framework that operationalises AL intensity for language-based coaching. We examine how Big Five traits moderate preferred AL intensity and perceived outcomes (personalisation, working alliance, satisfaction), and analyse demographic differences, revealing age- and gender-linked patterns. Our findings show that AL in conversational coaching is not one-size-fits-all, motivating adaptive agents that calibrate listening intensity to user characteristics and provide user-facing control over listening style. | Hussain, Keegan, Ross | Paper |
| 11:30-11:40 | Cognitive Empathy Toolkit: A Co-design Exploration with Adolescents with MID | Cognitive empathy is an essential skill in adolescents’ social and emotional development. Adolescents with a mild intellectual disability (MID) face additional challenges in this development. Interventions and school programs tailored to this target group are scarce. We present a co-design study with adolescents with MID (age 13-14) and their teachers, resulting in a design concept for a cognitive empathy toolkit. We contribute 1) a conceptual design of a digital toolkit for cognitive empathy development and 2) how to meaningfully involve adolescents with MID in a participatory design process. | Ooms, van Boxtel, Hauptmann | Paper |
| 11:40-11:50 | Personalising TaHaMi: Tangible Haptic Mindfulness Interface for Novice Meditators | While most HCI research on mindfulness meditation focused on audio and visual feedback to support attention regulation, haptic interfaces have been less explored. We present the design, and exploration with 19 novice meditators of a novel haptic interface integrating vibration and soft haptic. The study also engages users in the personalisation of their haptic patterns. Findings highlighted key experiential qualities of the two modalities and their potential to support or hinder mindfulness states. In particular, the result supports novice meditators’ phenomenological experience of mindfulness, facilitates grounding through deep breathing and promotes subtle, non-gamified interaction. We conclude with four design implications for haptics-based mindfulness technologies. | Asgari Tappeh, Sas | Paper |
| 11:50-12:00 | Intrusive by Default? Navigating Privacy by Design in LLM-Supported Care Documentation | In an LLM-based documentation support system for residential nursing homes, audio recordings of care worker–resident interactions are used to generate brief care reports. Drawing on ethnographic observations, we show that inconsistent privacy practices reflect a privacy calculus shaped by concerns about workplace surveillance. We examine the limits of operationalising Privacy by Design in a context where privacy concerns are intertwined with distrust toward both technology and the care organisation. As a technology probe to explore the technical potential for privacy protection, we developed an LLM-based privacy detection pipeline. While approximately three-quarters of pipeline-identified segments were confirmed as privacy-sensitive, unreliable speaker diarisation remains a major challenge. We argue that care workers should retain autonomy over their privacy preferences rather than relying on predefined technical safeguards. Our findings suggest that technical Privacy by Design measures alone are insufficient for establishing trustworthy AI documentation support systems in situated care work. | Kletter, Koeszegi | Paper |