Background: In western Christian countries, it is generally believed that religiosity is associated with a lower risk for depression which is supported in part by epidemiological evidence. However, the association between religiosity and depression in multireligious countries is unknown. The aim of this study is to evaluate the association between religiosity and subsequent depression in a multireligious population. Methods: Retrospective longitudinal study was conducted in a large international training hospital in Tokyo metropolitan area, Japan, from 2005 to 2018. All adult participants who underwent health check-ups at the center of preventive medicine, without prior history of depression or depression at baseline, were included. Our primary outcome was development of major depression. The outcome was compared by the degree of religiosity, adjusting for potential confounders. Results: Among 67,723 participants (mean age, 46.3±12.3 years; 33,893 males [50.1%]), those who were more religious tended to be older, female, married, had healthier habits, but more overweight/obese and had more medical comorbidities, at baseline. During median follow-up of 2,528 days (interquartile range: 974-4,007), 1,911 (2.8%) developed major depression. Compared to reference group (not religious at all), more religious group tended to have more odds ratios (ORs) for development of major depression in dose-dependent manner. Among them, extremely religious group (OR 1.51, 95% confidence interval (CI): 1.28-1.78 in model 4) and moderately religious group (OR 1.30, 95%CI: 1.14-1.49 in model 4) were statistically associated with increased development of major depression compared to not religious at all group. Those who had increased their religiosity from baseline had statistically lower development of major depression (OR 0.85, 95%CI: 0.75-0.97) compared to those who remained in the same degree of religiosity from baseline. Conclusion: Our large longitudinal study demonstrated that religiosity was associated with future major depression in dose-response manner in a multireligious population which was in the opposite direction from that seen in previous western longitudinal studies. However, those who had increased their religiosity after their initial evaluation had lower risk of major depression. Thus, in contrast with western Christianity culture, Japanese multireligious culture may have different impact on the religiosity-depression association. Our results suggest a new perspective for religiosity-depression association with the notion that cultural and religious diversity must be taken into consideration. Funding Statement: This work was supported in part by Grant-in-Aid for Scientific Research C 18K07604 (YH) from Japan Society for the Promotion of Science and Medical Research Fund (YH) from Takeda Science Foundation. Declaration of Interests: The authors stated: None. Ethics Approval Statement: The Ethics Committee Institutional Review Board at the hospital approved this study (approved number:18-R203).