In this article, a rationale for routine use of self-report questionnaires in rheumatology clinical care is presented. Studies performed according to structured clinical research methodologies, including population-based studies, inception cohort studies, randomized controlled clinical trials, and meta-analyses of these trials, have significant limitations in describing accurately the long-term natural history and results of treatment of rheumatoid arthritis; the most accurate data have been derived from clinical observations of consecutive, nonselected patients in routine clinical settings. Self-report questionnaires provide accurate and representative data concerning clinical status and traditional laboratory and variables in large numbers of patients can be developed using microcomputer hardware and software available only over the last decade. Further collection of self-report data in rheumatology clinical care should result in more informative descriptions of the long-term natural history and results of therapy in rheumatoid arthritis.