Quantitative Evaluation of a Community DX Health Promotion Initiative Using Wearable Sensor Data: A Case Study in Older Adults

Community-based digital transformation (DX) initiatives are increasingly used to promote healthy aging, although their effects are often evaluated using subjective self-reported measures. This study aimed to quantitatively evaluate a community DX health promotion initiative with Takashimadaira as its primary target area in Japan using wearable sensor data. A total of 500 community-dwelling older adults participated in the initiative, and wearable sensor data were obtained from 355 participants during the five-month observational period. The initiative combined wearable device distribution with event-information dissemination via the messaging platform. Daily step count and heart-rate indices were analyzed. Within-participant analyses showed significant increases in daily step counts in the information-provision groups, whereas the primary linear mixed-effects model showed no significant information-provision × time interaction for daily step count. Information provision was associated with differences in mean and minimum daily heart rate in cross-sectional analyses. Adherence and dropout also varied by residential area. Age, BMI, and exercise frequency were associated with several health-related indices, but these associations were not interpreted as evidence of wearable measurement validity. These findings demonstrate the potential utility of using wearable sensor data to evaluate longitudinal changes in health-related behavior and physiological signals in community-based DX initiatives, while also revealing important feasibility limitations, including substantial attrition (only 222 of 355 participants met the 14-day criterion; 90-day dropout 58.3%) and indicating that observed within-participant changes should not be interpreted as demonstrated intervention effects. Community context may also be relevant to sustained participation in such programs.