Background: Relations between diet and mental health continue to be a subject for controversy and an increasing numbers of studies. Recent literature is represented by papers that examine overall diet by way of dietary patterns and its association with depression, replacing previous studies about nutrients. The aim of this cross-sectional study is to evaluate the relation between dietary patterns and depressive episode in the baseline (2008–2010) population of the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil). We analyzed 14,798 participants of ELSA-Brasil. Methods: We constructed dietary patterns based on the Food Frequency Questionnaire using multiple correspondence and cluster analysis; to evaluate depressive episodes, we used the Clinical Interview Schedule-Revised (CIS-R). As an independent variable, we used the patterns: Traditional, Low-Sugar/Low-Fat, Fruit-Vegetables, and Bakery Products. We used multiple logistic regression models to evaluate relations between the dietary patterns and depressive episodes. Results: The Traditional pattern showed the highest percentages of consumption. After adjusting, the Bakery Products (OR = 1.33; 95%CI 1.05–1.70) was associated positively and significantly with depressive episodes only for women. Conclusions: International studies corroborated this finding, suggesting that the Bakery Products pattern could be a marker of a specific population group in which depressive episodes are frequent.