Physical activity domains and incident clinical depression: A 4-year follow-up analysis from the ELSA-Brasil cohort

Aim: To analyze the prospective dose-response relationships between total and domain-specific physical activity (PA) with incident clinical depression.Methods: We used data from two waves (Wave 1: August/2008-December 2010; Wave 2: September/2012-December/2014) of the Brazilian Longitudinal Health Study (ELSA-Brasil) cohort study. Self-reported PA (total, transport, and leisure-time) was the main exposure. Incident clinical depression (new cases of depression between waves) was assessed through the Clinical Interview Schedule-Revised (CIS-R). Poisson regression models, adjusting for potential confounders, were used for data analysis.Results: In 12,709 adults (53.8 % women, mean age: 51.9 +/- 9.0), moderate and high volumes of total PA (1-149 min/week: RR = 0.81, 0.58-1.13, 150-299 min/week: RR = 0.55, 95%CI: 0.40-0.76; >= 300 min/week: RR = 0.64, 95%CI: 0.52-0.80), and any volume of leisure-time PA (1-149 min/week: RR = 0.65, 95%CI: 0.50-0.83; 150-299 min/week: RR = 0.67, 95%CI: 0.52-0.88; RR = 0.61, 95%CI: 0.45-0.82) were associated with a lower risk of incident clinical depression. Transport PA protective only in the lower category (0.1-4.4 mMET-h/wk) (RR = 0.71, 95%CI: 0.54-0.94). Limitations: Other PA domains such as occupational and domestic were not assessed; the use of self-report measures for PA which may be subject to bias and recall issues; lack of assessment of additional potential confounders, such as sedentary behavior and family history of depression.Conclusion: Total and leisure-time PA were associated with lower incidence of clinical depression, even at lower doses. Low, moderate, and high volumes of total and leisure-time PA were associated with lower risk of incident clinical depression. Public health PA interventions aiming to prevent development of clinical depression should consider focusing on leisure-time PA.