Rheumatoid arthritis (RA) is a chronic condition that affects about 1% of the adult population. In a historical cohort of Minas Gerais State, 11,573 RA patients registered in the Outpatient Information System (SIA) between 2008 and 2013 were identified. For this study we adopted the public funding body's perspective and the values were adjusted by the national inflation index (IPCA) of December 2015. Etanercept was the most expensive treatment. The mean cohort age was 52 years old and most of the patients were women. Multiple regression analysis indicated a negative association between higher expenditure and age, female sex, and diagnosis at entry in the cohort and positive association between high expenditure and the Human Development Index (HDI) of the municipality and use of tumor necrosis factor agents. This study identified the factors that have an impact on RA drug treatment expenditure. Also, we showed that methods that enable extracting demographic and expenditure data of administrative information systems may represent important tools in the construction of economic studies to subsidize economic health evaluations, especially from the standpoint of the managers.