BACKGROUND: The costs of interferon beta are high ( 10000/patient year). The impact on health status and quality of life is uncertain. These factors, plus limited funding, restrictive indications and guidance for use mean prescribing in the UK is low and variable. To address these issues, a trial of interferon beta was proposed. The research costs of the trial were substantial ( 10.3 million). The treatment and support costs of the trial could be in the region of 75 million. OBJECTIVES: To assess the potential impact of the trial on the costs and benefits of treatment for people with multiple sclerosis (MS). METHODS: The evaluation compared the potential use, costs and QALY s of interferon beta, azathioprine and usual care with and without the trial. Decision analysis was used to estimate the expected costs and quality adjusted life years (QALY s) in two situations with the trial: (i) the treatment with the highest QALY is used; (ii) treatment decisions are made on efficiency criteria (expected costs and QALY s). RESULTS: If treatment decisions were made to maximise QALY s, the expected cost and QALY s of the trial were - 3026,000,000-+ 3904,000,000 and -5551-14012 QALY s. The expected cost/QALY gained was 0- 369000. Treatment decisions made on efficiency criteria were associated with expected costs and QALY s of - 8292,000,000-+ 167,000,000, and -5355-+14969 QALY s. The expected cost/QALY lost was 0- 662728. CONCLUSIONS: The expected value of the trial depends upon the decision making criteria used and impact of the trial on treatment decisions. If the trial leads to adoption of treatments with a net cost/QALY less than 100,000, it would lead to long term savings in the costs of care, but losses of 251-5355 QALY s in the MS population.