To assess the short and longterm efficacy of intensive inpatient treatment of active rheumatoid arthritis (RA), 16 patients, the test group, admitted to a hospital based rheumatic disease unit (mean length of stay 12.4 days) were studied, using clinical and laboratory variables. The comparison group was similarly evaluated, using 10 outpatients with active RA to whom hospitalization was recommended, but refused by the patients. Both groups were studied intensively over a 12-week period. In addition, a 2-year followup was performed on 12 test group inpatients and 8 comparison group outpatients, using the same variables plus a functional status questionnaire. Health care costs were determined for both groups and corrected for a 1985 dollar value. The test group showed significant improvement in morning stiffness, pain, and joint score, whereas the comparison group improved only in pain score during the initial 12-week period. At 2 years, the test group and the comparison group showed significant improvement in morning stiffness, pain, grip strength, and joint score. The comparison group initially had a somewhat lower index of disease activity. The test group maintained their initial improvement and none required rehospitalization. Functional status scores were similar for both groups. Health care costs were initially higher for the test group ($5,065); followup care cost for the test group was $99 less/year than the comparison group over 2 years. Hospitalization on a rheumatic disease unit brought about prompt, sustained improvement in 2 weeks which required nearly 2 years to achieve in the comparison group. Such hospitalization of uncomplicated RA seems warranted to decrease disability and increase the quality of life.