Objective
This study aimed to investigate the relationship between the frequency and duration of exercise and symptoms of depression and anxiety. 

Methods: This is a household survey conducted with a sample of 2,297 adults living in the city of Rio de Janeiro. The frequency and duration of exercise were measured using two closed-ended questions. Symptoms of depression and anxiety were assessed using the PHQ-9 and the GAD-7, respectively. Additionally, sociodemographic and anthropometric information (weight and height) was obtained. Bivariate and multivariate regression models were used to estimate the odds ratios (OR) of the associations between the variables.

Results
A total of 2,201 (age= 38.3 years (SE = 0.4), 62% women) individuals were evaluated. Compared to the non-exercisers, individuals who exercised at least once a week showed significantly lower odds of presenting depressive symptoms, even after adjusting for sex and BMI (OR = 0.7 [95% CI: 0.50–0.96], p = 0.03). Similar results were observed for frequency (2-4 times per week) and duration (≥1 hour and <2 hours). However, after adjusting for sex and BMI, these associations were no longer significant. Finally, no association was found between exercise and anxiety symptoms. 

Conclusion
Physical exercise was associated with a lower likelihood of presenting depressive symptoms. Additionally, a moderate frequency and duration of exercise, initially associated with reduced odds of depressive symptoms, was no longer significant after adjusting for sex and BMI. Thus, body weight and sex appear to mediate the relationship between physical exercise and depressive symptoms in our sample.