The aim of this study was to evaluate the efficacy and safety of etanercept (ETA) for treating rheumatoid arthritis. A systematic review was performed to search for randomized clinical trials comparing subcutaneous doses of ETA at 25 mg twice a week or 50 mg weekly to a placebo group, with or without methotrexate. Studies of low quality (less than 3 points on Jadad's scale) were excluded. The efficacy was assessed by using the criteria of the American College of Rheumatology (ACR). Safety data were evaluated based on serious adverse events, serious infections, malignancy and deaths. Withdrawals as a result of adverse events or lack of efficacy were also evaluated. Eight studies met the inclusion criteria, comprising 2385 patients. In the efficacy meta-analysis, a greater number of ETA-treated patients achieved the efficacy criteria within 6 months of treatment, where the relative risk (RR) was 2.94 [2.27, 3.81] for achieving ACR20, 5.28 [3.12, 8.92] for ACR50 and 4.83 [1.74, 13.47] for ACR70. After 1 year, the RR for achieving ACR20, ACR50 and ACR70 were 1.14 [1.07, 1.23], 1.36 [1.21, 1.53] and 1.56 [1.30, 1.88], respectively. This response rates were higher for ETA-treated patients in comparison with control group patients. For safety, there were no statistically significant differences between treated patients and controls. This was also confirmed by withdrawals as a result of adverse events, which were not statistically different between the two groups. However, more patients withdrew from control groups because of a lack of efficacy as compared with ETA groups (RR = 0.48 [0.30, 0.78]).