BackgroundBright light therapy and exercise interventions are effective methods for treating seasonal and non-seasonal affective disorders. Synchronization of internal circadian rhythms with the external environment by light therapy and physical activity may partly explain its efficacy. In the present study, we objectively measured daytime light exposure and physical activity in real life situations with elderly participants, and investigated the association between farming habits and the prevalence of depressive symptoms. MethodsThis cross-sectional was conducted among 1005 participants (mean age: 71.5) of a community-based cohort study. Depressive symptoms were assessed by the Geriatric Depression Scale (GDS score ≥ 6) and administration of antidepressant. ResultsFarming habit with long duration (> 7.0 h/week) showed significantly lower odds ratios (OR) for depressive symptoms (adjusted OR 0.63, 95% confidential interval,0.41 to 0.96) compared with participants without farming habit independent of confounders such as age, gender, body mass index smoking, drinking, daytime ambulatory systolic blood pressure, diabetes, living alone, education, income, and daylength. Even in farming with short duration (≤ 7.0 h/week), we found significant association with lower OR for depressive symptoms (adjusted OR 0.64, 95%CI, 0.42 to 0.97). Light exposure and daytime physical activity measured by wrist actigraphy were significantly higher among participants with longer farming habits (p for trend < 0.01). Physical activity mediated 12.0% of association between farming habit and depressive symptoms. LimitationsA cross-sectional association may be found because the participants with depressive symptoms tended to avoid farming. A longitudinal study is warranted to determine the direction of causality. ConclusionsParticipants with farming habit showed significantly lower OR for depressive symptoms than those without farming habit, and it was partly mediated by physical activity.