(1) Multiple sclerosis is a neurological disorder characterised by disseminated inflammatory lesions of the myelin sheath, the white fatty layer that covers the neurons in the brain and spinal column. The clinical repercussions are highly variable, ranging from minor deficits to severe disability and even death. Some patients have exacerbations separated by remissions of varying length, while others experience fairly steady progression of disability. This review article, based on Prescrire's standard in-house methodology, examines treatments proposed to slow the progression of multiple sclerosis. (2) For exacerbations (4 trials, a total of 140 patients), an improvement was noted in 64% of patients treated with intravenous methylprednisolone for 5 days, versus 40% of patients on placebo. (3) Interferon beta is the best-assessed of the disease-modifying drugs. Various marketed interferon beta products have been compared with one another in trials involving more than 1000 patients in total. There are no major differences in effectiveness or adverse effects. The latter are frequent, especially a flu-like syndrome and depression. (4) After a first episode suggestive of multiple sclerosis (3 trials, a total of 1160 patients), 35% of patients treated with interferon beta subsequently developed the disease, versus 50% of patients in the placebo groups. This advantage seemed to persist for at least 8 years. (5) In relapsing-remitting multiple sclerosis (4 trials, a total of 1426 patients), interferon beta prevented about one exacerbation every 2.5 to 3 years but had no clear preventive impact on progression of disability. (6) In secondary progressive multiple sclerosis (5 trials, a total of 3082 patients), interferon beta's effectiveness, if present, was limited to patients who continued to have exacerbations. (7) Natalizumab does not appear to be more effective than interferon beta. It is associated with serious adverse effects, about which only limited information is available. (8) Glatiramer has not been shown to prevent exacerbations or progression of disability. Intravenous immunoglobulin is still being tested in multiple sclerosis. In initially aggressive forms of multiple sclerosis, mitoxantrone has a negative risk-benefit balance (2 trials, 236 patients); it prevents less than one exacerbation during two years of treatment but causes heart failure in 1% to 2% of patients after 10 months of treatment. (9) Patients' anxiety as their disability worsens must not overshadow the fact that there are no treatments available with strong clinical effectiveness, and that all existing treatments carry a risk of serious adverse effects. Pending the arrival of something better, only interferon beta has a favourable risk-benefit balance, and only in selected patients.