Design-Driven Smart Wearable System for Personalized Intervention to Improve Sleep Quality in Older Adults

Abstract

Sleep disturbances are common among older adults and contribute to declines in health, cognition, and overall well-being. To address the limitations of passive, monitoring-centered digital sleep solutions, this study proposes a design-driven smart wearable system that integrates multimodal physiological sensing, adaptive personalization, and non-pharmacological behavioral and sensory-based interventions. A randomized crossover trial involving 40 older adults demonstrated that activation of the personalized intervention module — including sound-based relaxation guidance, light-based circadian support, and vibration-assisted behavioral cues — led to significant improvements in both subjective and objective sleep outcomes. Participants experienced a mean reduction of 2.2 points in Pittsburgh Sleep Quality Index (PSQI) scores, alongside improvements in sleep efficiency, wake after sleep onset, and sleep fragmentation. A subsample undergoing home-based polysomnography (PSG) showed moderate to strong correlations between wearable-derived and PSG-derived sleep parameters (r = 0.68–0.79), supporting the physiological validity of the system. High adherence and usability ratings further indicated that design-driven personalization effectively enhanced engagement, a key barrier in conventional wearable-based sleep interventions. These findings suggest that adaptive, non-pharmacological wearable interventions can provide a scalable and accessible approach to precision sleep health management in aging populations.