A critical question for the management of patients with rheumatoid arthritis (RA) is the timing and extent of pharmacological intervention. There is good evidence that early disease is important, and data showing early intervention have particular advantages. However, in the past, several difficulties have arisen with this approach. The availability of patients seen in early arthritis clinics, the ability to predict outcome and the development of new methods of monitoring have made a large difference to the possibilities for therapy. This article describes why it is now reasonable to treat patients who have 'standard' RA with a disease-modifying antirheumatic drug (DMARD) at presentation. In those patients with a worse prognosis, a more aggressive regime is justified. For those who do not fulfil criteria laid down by the American College of Rheumatologists, treatment with a DMARD may still be appropriate on a cost-benefit analysis. Studies addressing these issues are currently being undertaken.