This retrospective study evaluated treatment with sulfasalazine (SAS) in a mean dosage of 2.1 g/day in 95 patients with rheumatoid arthritis (RA) who were followed-up for 3 months to 4 years. Mean disease duration was 7 years; 79 patients had previously received at least one disease-modifying drug. Four per cent of patients were lost to follow-up. Mean duration of treatment was 15 months (3 weeks-50 months). Treatment continuation rates were 57% at one year, 40% at two years, and 26% at three years. Reasons for discontinuation of SAS included adverse effects (n = 24), inefficacy (n = 33), and death unrelated to SAS therapy (n = 2). In four patients, SAS was discontinued within three months of the first dose because of a severe adverse effect (diffuse erythematous rash, diffuse bullous rash, hepatitis with jaundice, agranulocytosis). SAS-induced biologic markers for lupus were seen in one patient. Furthermore, 12% of evaluable patients developed antinuclear antibodies during SAS therapy. The SAS treatment continuation rate was higher (p = 0.05) among patients under 40 years of age (n = 18) than among older patients. This difference was due to a correlation between age and tolerance with less SAS-induced side effects in patients under 40 years of age (p = 0.03). The SAS treatment continuation rate was unrelated to the duration of rheumatoid arthritis or number of previous maintenance treatments. This study suggests that rheumatoid arthritis patients under 40 years of age exhibit better tolerance to SAS therapy.