The treatment of rheumatoid arthritis with disease modifying drugs is based on the hypothesis that they exert a more profound effect on the disease process than non-steroidal anti-inflammatory drugs. Short-term studies of disease modifying drugs show significant clinical and laboratory improvements in comparison to placebo. Longer-term studies suggest that this improvement is not maintained. Attempts to assess the effect of disease modifying and immunosuppressive drugs on radiological progression suggest that they may have a week protective effect, but there is no evidence to suggest that this is maintained in the long term. Studies of corticosteroid therapy indicate that in large doses it possesses disease modifying properties and is capable of retarding radiological progression. The benefit-risk ratio of such doses is unacceptable. Lower doses have not yet been adequately investigated. Rheumatologists should regard existing suppressive agents as moderately effective in the short term. Few patients remain on single drug therapy for prolonged periods and available evidence does not suggest that these drugs are capable of altering the long-term outcome of rheumatoid arthritis.