Feasibility of Using a Wearable Device to Identify Loneliness Among Older Adults Living Alone: Mixed Methods Study.

Background: Addressing loneliness is a public health priority, with increasing interest in technological solutions for older adults. However, evidence on real-world feasibility and acceptability of such technologies remains limited. Objective: This study evaluated the feasibility and acceptability of the Conversation as Therapy (CaT) Pin, a wearable device designed to capture time spent in conversation among older adults living alone, following prototype refinement. Specifically, the study assessed whether a sufficient amount of synchronized, device-derived conversation data and self-reported loneliness data could be generated to explore the relationship between conversation time and loneliness. Methods: A 2-phase, mixed methods feasibility study was conducted. In phase 1, a Consumer Advisory Group (CAG) participated in a 1-hour discussion to inform study processes and device design, followed by a 1-week prototype test with older adults (n=3). Prototype usability, device design, and study procedures were explored through interviews and informed iterative refinement. In phase 2, community-dwelling adults aged 65 years or older living alone trialed the device for 2 weeks. Participants completed activity logs capturing loneliness, activity engagement, device wear, and conversations at 6-hour intervals. Device-captured conversation time was compared with self-reported data where available. Qualitative interviews were analyzed using reflexive thematic analysis to understand participant experiences of device acceptability and feasibility. Results: In phase 1, 6 CAG participants, including 3 prototype testers, identified usability and design issues that led to modifications before phase 2. Phase 2 involved 23 participants (15 women; mean age 79.4 y, SD 6.7 y), with 7 classified as lonely using the UCLA-3 scale. However, only 8 (34.8%) participants generated sufficient concurrent device and activity log data, preventing meaningful assessment of the relationship between conversation time and loneliness. Feasibility challenges were driven by device acceptability, namely usability, comfort, aesthetics, and integration into daily life, which limited sustained wear and data synchronization. Qualitative findings indicated that participants prioritized quality rather than quantity of social interaction, and many did not perceive personal benefit from device use, citing lack of perceived need or concerns related to functionality and design. Together, methodological and conceptual challenges were identified in using conversation time as a potential indicator of loneliness. Conclusions: This study demonstrated that, in its current form, using a wearable device to capture time spent in conversation and align these data with self-reported loneliness was not feasible among older adults living alone. Feasibility was constrained by device acceptability related to usability, participant adherence, and data synchronization challenges. Findings also suggest that conversation time alone may be insufficient as an indicator of loneliness, given participants' emphasis on the quality and personal meaning of social interaction. These results provide practical lessons for future digital health research with older adults, including the need for early co-design, monitorable data-capture systems, and more nuanced approaches to measuring loneliness.