Eighty-six patients with rheumatoid arthritis treated with sulfasalazine were followed for 5 years, or until treatment was discontinued. At the end of 5 years, there was a 22% probability of successfully continuing treatment. Most adverse effects developed in the first 3 months of treatment. In 38 patients treatment was discontinued because of inefficacy. In 18 of these, a brief period of improvement was followed by clinical relapse. Twenty were regarded as having no useful response to sulfasalazine. The treatment continuation rate of 22% at 5 years is in marked contrast to the pessimistic longterm evaluations of second line drugs that have recently been reported.