OBJECTIVES: To assess clinical response after switching between anti-tumor necrosis factor-alpha (anti-TNF-alpha) agents in patients with rheumatoid arthritis (RA). PATIENTS AND METHODS: This study included 99 patients diagnosed with RA American College of Rheumatology, 1987), on anti-TNF-alpha therapy, to assess the therapeutic response after 24 weeks. Switching was performed if, after 12 to 24 weeks, a severe adverse event was reported (toxicity: T) or if no reduction greater than 0.6 in the initial Disease Activity Score 28 (DAS28) occurred (inadequate response: IR). In case of IR, the patient was considered as primary failure (PF). Secondary failure (SF) was defined as loss of response after initial improvement. Remission (DAS28 < 2.6), low disease activity (between 2.61 and 3.2), and functional improvement [increase in the initial Health Assessment Questionnaire (HAQ) > 0.2] were assessed by use of linear regression analysis. The significance level adopted was P < 0.05. RESULTS: Switching was performed in 39 (39.4%) patients, especially due to PF (24.3%), SF (35.1%) and T (40.5%). The retention rate of the first agent was 60.1%, and the mean time for switching was 14.2 ± 10.9 months. After switching, a tendency towards a decrease in DAS28 was observed (4.7 ± 1.4; P = 0.08), but not in the HAQ (1.2 ± 0.77; P = 0.11). Around 43% of the patients achieved good/moderate EULAR response. The major determinant of switching was a higher initial DAS28, independent of age, duration of disease, and functional capacity. CONCLUSION: Switching between anti-TNF-alpha agents is a valid strategy to control disease activity, despite the low likelihood of remission and no significant improvement in functional capacity.