Objective: The aim of this study was to verify the role of spiritual well-being as a moderator of the effect of living arrangement on depressive state in the elderly. Methods: We conducted a cross-sectional study using semi-structured interviews in Okinawa on 2010. Four hundreds thirteen (165 men, 248 women) participants were asked about 5-item Geriatric Depression Scale (GDS-5) as dependent variable, living arrangements (living alone or living with others) as independent variable, spiritual rating scale for elderly as mediating variable, and demographic and socioeconomic variables as control variables. The main and interaction effects were assessed using hierarchical multiple regression and simple slope analyses. Results: The proportion of elderly who living alone was 26.6% (n 110), and they were significantly higher scores of GDS-5 than the elderly who living with others (P 0.001). Significant negative correlation was found between depressive state and spiritual well-being (r 0.27, P 0.001), and the result indicated that the elderly with higher spiritual well-being were lower depressive state. Results of hierarchical regression analysis showed that the correlation between living arrangement and spiritual well-being was positively associated with increased risk for depressive state (β 0.10, P 0.037). Simple slope analyses revealed that the elderly who living alone had significantly higher depressive state scores than those who living with others only in low spiritual well-being (P 0.002), but not in high spiritual well-being (P 0.445). Conclusions: Our findings suggest that spiritual well-being might play an important role in preventing the elderly who living alone from depression, and interventions aimed at promoting spiritual well-being in the elderly who living alone may lead to improvements in mental health.