OBJECTIVES: Estimate the effect of socioeconomic factors on the prevalence of depression in the Brazilian adult population from 1998 to 2008, emphasizing the impact of variables such as income, gender, education level and the comorbity of other chronic diseases. METHODS: A health supplement was added to the National Household Sample Survey (PNAD) for the years of 1998, 2003 and 2008. In addition to information regarding income, education, household structure among other socioeconomic data, the surveys show information about the health condition of individuals, including chronic diseases. With these data, nonlinear methods were applied to estimate the marginal effects of each of the independent variables. The two main methods used were probit and pooled probit. RESULTS: Brazil is a very large and culturally diverse country with many socioeconomic contrasts. This characteristic helps a more comprehensive analysis of the effects of income, education, regional and the other socioeconomic factors on de prevalence of depression. This study found that income and education has a negative correlation with the probability of an individual suffer from depression. In addition, Brazilian women are twice as likely to suffer from depression as their male counterparts. Other variables like unemployment, being divorced or widowed or not having children have shown a positive correlation with the incidence of depression. Finally, cancer, kidney disease and cirrhosis were the chronic illnesses that had the highest rates of comorbity with depression. All the results found where coherent with the results found in the references researched. CONCLUSIONS: Depression is a disease that permeates modern society, and factors such as poverty and income inequality can have an important effect on an individual´s probability of going through a depressive episode during his life, especially individuals with chronic diseases.