Objective To estimate the incremental cost-effectiveness of infliximab versus conventional combination treatment over 21 months in patients with methotrexate-refractory early rheumatoid arthritis.

Methods In this multicentre, two-arm, parallel, randomised, active-controlled, open-label trial, rheumatoid arthritis patients with <1 year symptom duration were recruited from 15 rheumatology clinics in Sweden between October 2002 and December 2005. After 3–4 months of methotrexate monotherapy, patients not achieving low disease activity were randomised to addition of infliximab or sulfasalazine+hydroxychloroquine (conventional treatment group). Costs of drugs, healthcare use, and productivity losses were retrieved from nationwide registers, while EuroQol 5-Dimensions utility was collected quarterly.

Results Of 487 patients initially enrolled, 128 and 130 were randomised to infliximab and conventional treatment, respectively. The infliximab group accumulated higher drug and healthcare costs (€27 487 vs €10 364; adjusted mean difference €16 956 (95% CI 14 647 to 19 162)), while productivity losses did not differ (€33 804 vs €29 220; €3961 (95% CI −3986 to 11 850)), resulting in higher societal cost compared to the conventional group (€61 291 vs €39 584; €20 916 (95% CI 12 800 to 28 660)). Mean accumulated quality-adjusted life-years (QALYs) did not differ (1.10 vs 1.12; adjusted mean difference favouring infliximab treatment 0.01 (95% CI −0.07 to 0.08)). The incremental cost-effectiveness ratios for the infliximab versus conventional treatment strategy were €2 404 197/QALY from the societal perspective and €1 948 919/QALY from the healthcare perspective.

Conclusions In early, methotrexate-refractory rheumatoid arthritis, a treatment strategy commencing with addition of infliximab, as compared to sulfasalazine+hydroxychloroquine, was not cost-effective over 21 months at willingness to pay levels generally considered acceptable.

Trial registration number: NCT00764725.