Sleep disturbance is negatively associated with subjective well-being in older people, but the potential underlying mechanisms of this association remain unclear. This study aimed to disentangle the pathways linking subjective versus objective sleep disturbance to subjective well-being through the serial mediation effect of loneliness and depression among community- dwelling older people.

This cross-sectional study was conducted in Sakai city of Japan. A total of 212 aged 65 and over participated in this study. The Athens Insomnia Scale, UCLA Loneliness Scale, Geriatric Depression Scale, and Self-perceived well-being were used to assess subjective sleep quality, loneliness, depression, and subjective well-being, respectively. A non-wearable actigraphy device was used to evaluate the objective sleep quality. Total sleep time, sleep latency, sleep efficiency, wake after sleep onset, number of awakenings, and average activity during sleep were recorded. Serial multiple mediation analysis was performed using SPSS PROCESS Version 4.1 macro. This study was approved by the Institutional Review Board of Osaka University.

Subjective sleep disturbance was found to impact subjective well-being through three significant mediation pathways: (1) loneliness (B=-0.024, 95% CI=-0.055, -0.004), which accounted for 25.72% of the total effect, (2) depression (B=-0.020, 95% CI=-0.044, -0.001), which accounted for 20.94% of the total effect, and (3) loneliness and depression (B=-0.008, 95% CI=-0.019, -0.001), accounting for 8.93% of the total effect. The total mediating effect was 55.58%. As for the objective sleep disturbance, the wake after sleep onset can indirectly impact subjective well-being through loneliness (B=0.005, 95% CI=0.001, 0.010), depression (B=-0.005, 95% CI=-0.011, -0.001), and both (B=0.002, 95% CI=0.001, 0.004); the number of awakenings can indirectly impact subjective well-being through loneliness (B=0.041, 95% CI= 0.012, 0.085), depression (B=-0.034, 95% CI=-0.076, -0.002), and both (B=0.018, 95% CI=0.005, 0.036); the average activity during sleep can also indirectly impact subjective well- being through loneliness (B=0.137, 95% CI=0.034, 0.275), depression (B=-0.128, 95% CI=- 0.282, -0.010), and both (B=0.055, 95% CI=0.011, 0.118).

These findings provided new insights into possible avenues for improving subjective well- being among older people through sleep-based interventions with a multi-faceted approach to mental health.