RA is a chronic disease that affects 0.5 to 1% of the population. The economic impact of RA on individuals and society is enormous and the costs of RA rise steeply with disease severity. A therapy that reduces disease progression could be expected to lead to reductions in resource use as well as maintaining quality of life. OBJECTIVE: To estimate the costs and consequences of adding infliximab to the care of patients with severe rheumatoid arthritis (RA) already being treated with methotrexate. METHODS: Estimates of the impact of infliximab on disease progression were obtained from the ATTRACT trial where 428 RA patients were randomly assigned to methotrexate or methotrexate plus infliximab. Since patients in the ATTRACT trial were followed for only 54 weeks, we developed a Markov model in order to estimate the long-term consequences of RA. The model was based on a cohort (ARAMIS) involving 4258 consecutively enrolled RA patients followed in nine centres in USA and Canada. Markov health states were based on the Health Assessment Questionnaire and on drug treatment. For the first year, costs were calculated using the resource utilisation by UK patients in the ATTRACT trial and applying UK unit costs. Long-term costs were obtained from the Norfolk Arthritis Register (NOAR) cohort. Utilities were based on visual analogue scale assessments in ATTRACT (first year) and ARAMIS (long-term). RESULTS: In the base-case analysis, the incremental cost per QALY of infliximab was £33618. Assuming radiographic stabilization of joint disease for patients treated with infliximab after the first year of treatment (as suggested in the long-term data from ATTRACT trial) the cost-effectiveness ratio falls to £5111 per QALY. Sensitivity analyses were performed to allow for uncertainty in some of the estimates. CONCLUSION: Infliximab is likely to be a cost-effective treatment for patients suffering from severe RA.