Etanercept and methotrexate are effective in the treatment of rheumatoid arthritis but no data exist on concurrent initiation or use of the combination compared with either drug alone. We aimed to assess combination treatment with etanercept and methotrexate versus the monotherapies in patients with rheumatoid arthritis. In a double-blind, randomised, clinical efficacy, safety, and radiographic study, 686 patients with active rheumatoid arthritis were randomly allocated to treatment with etanercept 25 mg (subcutaneously twice a week), oral methotrexate (up to 20 mg every week), or the combination. Clinical response was assessed by criteria of the American College of Rheumatology (ACR). The primary efficacy endpoint was the numeric index of the ACR response (ACR-N) area under the curve (AUC) over the first 24 weeks. The primary radiographic endpoint was change from baseline to week 52 in total joint damage and was assessed with the modified Sharp score. Analysis was by intention to treat. Four patients did not receive any drug; thus 682 were studied. ACR-N AUC at 24 weeks was greater for the combination group compared with etanercept alone and methotrexate alone (18.3%-years [95% CI 17.1-19.6] vs 14.7%-years [13.5-16.0], p<0.0001, and 12.2%-years [11.0-13.4], p<0.0001; respectively). The mean difference in ACR-N AUC between combination and methotrexate alone was 6.1 (95% CI 4.5-7.8, p<0.0001) and between etanercept and methotrexate was 2.5 (0.8-4.2, p=0.0034). The combination was more efficacious than methotrexate or etanercept alone in retardation of joint damage (mean total Sharp score -0.54 [95% CI -1.00 to -0.07] vs 2.80 [1.08 to 4.51], p<0.0001, and 0.52 [-0.10 to 1.15], p=0.0006; respectively). The mean difference in total Sharp score between combination and methotrexate alone was -3.34 (95% CI -4.86 to -1.81, p<0.0001) and between etanercept and methotrexate was -27 (-3.81 to -0.74, p=0.0469). The number of patients reporting infections or adverse events was similar in all groups. The combination of etanercept and methotrexate was significantly better in reduction of disease activity, improvement of functional disability, and retardation of radiographic progression compared with methotrexate or etanercept alone. These findings bring us closer to achievement of remission and repair of structural damage in rheumatoid arthritis.