Real-World Health Monitoring Using Wrist-Worn Wearables in Community-Dwelling Older Adults in China: Mixed Methods Feasibility Study

Background Real-world behavioral and physiological functioning, including heart rate, daily mobility, and sleep patterns, may provide information about aging and early neurodegenerative changes that is not captured by periodic clinical visits. Wrist-worn wearable devices offer a promising approach for continuous, passive monitoring of these domains. However, evidence regarding feasibility, usability, and acceptability in real-world observational settings among community-dwelling older adults in China remains limited. Objective This study primarily aimed to evaluate the feasibility, usability, and acceptability of using wrist-worn wearable devices to monitor behavioral and physiological function in community-dwelling older adults under naturalistic conditions. As a secondary objective, the study descriptively characterized wearable-derived behavioral and physiological measures among participants with minimal consecutive valid monitoring. Methods We enrolled 486 community-dwelling older adults from the Beijing region of the China-Aging cohort, each of whom received a Huawei Band 8 commercial wrist-worn wearable device. A valid monitoring day was defined as 10 or more hours of wear time. Participants were classified into three monitoring-pattern categories based on predefined wear-time criteria: no valid monitoring day, nonconsecutive valid monitoring, and minimal consecutive valid monitoring (≥3 consecutive valid monitoring days). Demographic characteristics, health status, and cognitive and functional measures (Mini-Mental State Examination and Barthel Activities of Daily Living Index) were compared across groups. Usability, acceptability, and perceived barriers were assessed using a structured user experience survey. Results Of 486 participants, 224 (46.1%) achieved minimal consecutive valid monitoring, 156 (32.1%) had nonconsecutive valid monitoring, and 106 (21.8%) had no valid monitoring day. Self-rated health differed significantly across monitoring patterns ( P =.005), whereas no significant differences were observed for the other measured baseline characteristics. Among participants with minimal consecutive valid monitoring, mean daily heart rate was 72.1 (SD 8.6) bpm, median daily step count was 6177 (IQR 4249‐8872) steps/day, and sleep data were available for 211 out of 224 (94.2%), with a median sleep duration of 7.23 (IQR 6.48‐8.04) hours. The user experience survey was completed by 106 out of 486 (21.8%) participants, with higher participation among those with more sustained monitoring. Respondents generally reported favorable usability and acceptability; common barriers included device discomfort, data synchronization issues, and battery life. Conclusions Wrist-worn wearable monitoring was feasible under naturalistic conditions for a substantial proportion of community-dwelling older adults in China, with generally favorable usability and acceptability among survey respondents. However, monitoring patterns varied considerably across individuals, highlighting challenges for long-term use and large-scale implementation. These findings support the potential of wearable technologies for longitudinal monitoring outside clinic-based assessments. Further research is needed to establish the stability and clinical relevance of wearable-derived behavioral measures over longer monitoring periods.