This study examines long-term outcomes of total shoulder arthroplasty (TSA) via survivorship analysis, patient questionnaires, and minimum 10-year physical examinations. The study group consisted of 320 consecutive TSAs performed in 267 patients between 1974 and 1988. Diagnoses included rheumatoid arthritis (69%), osteoarthritis (22%), and juvenile rheumatoid arthritis (4.7%). Minimum 10-year physical examination follow-up was obtained on a subset of 72 TSAs at a mean (+/- SD) of 14.0 +/- 2.7 years. A Disabilities of the Arm, Shoulder and Hand (DASH) questionnaire was obtained from 80 patients with 103 TSAs at a mean of 15.4 +/- 3.4 years after the index procedure (range, 10.4-23.2 years). Kaplan-Meier survivorship rates with revision as the endpoint at 5, 10, 15, and 20 years were 98%, 93%, 88%, and 85%, respectively. Of the shoulders, 22 (6.9%) required a revision, most commonly for loosening of one or both components (15 shoulders). Dislocation occurred earlier than other causes of revision or complication (P < .05, analysis of variance). Minimum 10-year physical examination follow-up revealed lasting, significant improvements in range of motion and strength. The patients' subjective assessments of TSA were favorable in that 92% felt that their shoulder was "much better" or "better" after TSA. The mean DASH score was 49 +/- 25; no significant differences were found among diagnoses. Long-term analysis of the Neer-type TSA revealed survivorship rates comparable to other joint replacements. The significant improvements in relief of pain, shoulder range of motion, and strength are associated with a high degree of patient satisfaction.