OBJECTIVES: To review the evidence of the clinical and cost-effectiveness of anakinra, an interleukin-1 receptor antagonist (IL-1Ra), for the treatment of rheumatoid arthritis (RA) in adults. DATA SOURCES: Electronic bibliographic databases. Scrip, Food and Drug Administration (FDA) submissions for new drug applications, European Agency for the Evaluation of Medicinal Products (EMEA) reports and the pharmaceutical company submission to the National Institute for Clinical Excellence. REVIEW METHODS: Studies were identified that included randomised controlled trials (RCTs) or economic evaluations of anakinra in adult patients with RA. Existing health economic reviews were also assessed. Data were extracted and quality assessed using a structured approach. The Birmingham Rheumatoid Arthritis Model (BRAM) was used to compare disease-modifying antirheumatic drug (DMARD) sequences, chosen to reflect current clinical practice, with and without anakinra, at different points in the DMARD sequence. RESULTS: Five high-quality RCTs of anakinra in adult patients with RA, involving a total of 2905 patients, of whom 2146 received anakinra, were identified. The results of the clinical trials were consistent with clinical benefit (compared with placebo) as measured by American College of Rheumatology (ACR) composite response rate at 6 months. Variation in response rate was seen across the trials, which is likely to be a reflection of the size of the trials and the wide range of doses evaluated. Consistent benefit was seen at the higher dose evaluated. Benefit was evident both with monotherapy and when used in combination with methotrexate. Data on the efficacy end-points evaluated in a large pragmatic safety study have not been made available, which is of concern. Anakinra treatment was associated with a high incidence of injection-site reactions. Serious adverse events were infrequent, but longer term follow-up is required. No fully published economic evaluations of anakinra in patients with RA were identified. The BRAM gives a base-case estimate of the incremental cost-effectiveness ratio (ICER) of anakinra of 106,000 pounds to 604,000 pounds/quality-adjusted life-year (QALY). In the sensitivity analyses substantial variations were made in key parameters and ICERs were shown to be responsive. However, ICERs did not drop below 50,000 pounds/QALY in any univariate sensitivity analysis. CONCLUSIONS: Anakinra can be considered modestly effective in the treatment of RA based on ACR response, although no conclusion can currently be made on the effect of treatment on disease progression. Adjusted indirect comparison suggests that anakinra may be significantly less effective at relieving the clinical signs and symptoms of RA, as measured by the ACR response criteria, than tumour necrosis factor (TNF) inhibitors all used in combination with methotrexate, although these results should be interpreted with caution. The BRAM produces an ICER for anakinra substantially higher than those for infliximab and etanercept. However, patients may respond to anakinra when they have not responded to other TNF inhibitors, as these agents have a different mechanism of action. Thus, anakinra may produce a clinically significant and important improvement in some patients that they could not otherwise have achieved. Further research would be valuable in the following areas: RCTs to evaluate the efficacy, safety and cost of anakinra over the longer term; comparative trials of anakinra with other DMARDs and biological modifiers; assessment of the role of anakinra in the treatment of patients who have failed to achieve a benefit while taking infliximab or etanercept; assessment on the impact of DMARDs and anakinra on joint replacement, mortality and quality of life; controlled clinical trials of combination therapy with two anticytokines; investigations into newer biological therapies; and the utility of radiographic outcomes in clinical trials of RA.