OBJECTIVE: To describe the degree of clinical benefit in patients with rheumatoid arthritis (RA) who receive infliximab therapy after lack of efficacy with etanercept. METHODS: In a retrospective study among 6 centers primarily designed to assess the safety of infliximab in combination with leflunomide, a standardized chart review form was used to collect data on 93 patients with RA. During that study, it was noted that some of these patients had switched from etanercept to infliximab. In this study, we compared the response of subjects switching from etanercept to infliximab (n = 20) to that of subjects receiving infliximab with no prior tumor necrosis factor (TNF) therapy (n = 73). RESULTS: The swollen and tender joint count, patient and physician global assessments, morning stiffness, and C-reactive protein all improved substantially in both groups, with no statistical difference in the degree of benefit between the groups. At the time of chart review, switchers had received a statistically higher dose of infliximab than controls (4.4 vs 3.19 mg/kg; p = 0.006) with a total of 5.7 and 5 infusions, respectively. CONCLUSION: In this retrospective study, previous lack of efficacy with etanercept did not predict lack of efficacy with infliximab. Indeed, the degree of clinical improvement was similar in both groups, although switchers were receiving a higher dose of infliximab at the time of chart review. Our findings suggest that clinical response may differ between anti-TNF agents, and lack of response to one agent may not predict a lack of response to another.