Epidemiological aspects of studying outcome in rheumatoid arthritis.

Knowledge of the outcome of rheumatoid arthritis (RA) is important both to the patient and to the rheumatologist. However, little data exist on the long-term outcome and effects of treatment on the disease. Consideration is given to the problems that arise in the choice of methods and measures to be used in an outcome study. Patient selection is an underestimated factor, which often produces biases and problems of comparability. Useful lessons can be learnt from examining outcome studies in other chronic diseases.