Objective. To critically appraise evidence of the effectiveness of continuing medical education (CME) in rheumatic diseases for primary care physicians.Methods. Three physicians independently applied preset criteria to evaluated CME interventions published between January 1966 and August 1993.Results. Eight of 166 articles identified were critically appraised, 7 of which had positive results. Marked heterogeneity in educational interventions, evaluative methods, and outcomes was noted.Conclusion. Despite generally positive results, weak methodology precludes drawing firm conclusions about the effectiveness of CME in rheumatic diseases.Rheumatic diseases are among the most common causes of morbidity, disability, and health care utilization in the general adult population. Their impact on primary care is extremely high. In the United States, rheumatic diseases have been found to be the main reason for 9% of all physician visits (1) and for 11-13% of visits to office-based general and family physicians (2). Similar percentages have been found in primary care settings in a number of countries, irrespective of the health care system (3-7). Osteoarthritis and regional joint pain are the disorders most frequently encountered in the ambulatory setting (2)(3)(4)(5)8 while in hospital settings, inflammatory disorders predominate.In contrast to the importance of rheumatic diseases in the general population and in primary care settings, the training of primary care physicians in the examination of joints and in the diagnosis and treatment of common rheumatic diseases has been noted to be deficient. Inadequate training at the undergraduate level has been found in the US, the UK, and Australia (9-14). Postgraduate training has also been found to suffer from a paucity of appropriate training experiences in rheumatology (15,16). Although core educational guidelines have been developed for family practice residents (17), few residency training programs include mandatory experiences in rheumatology.Inadequate training in rheumatic diseases may lead to suboptimal patient management. For example, two Swedish studies documented poor agreement between primary care and rheumatology diagnoses (18,19), while late referral was found in Britain for spondylarthritic diseases and for rheumatoid arthritis in London and Dublin rheumatology clinics (20,21). Suboptimal primary care practices were found in Michigan for rheumatoid arthritis (22) and in Ontario for gout (23).Given the historical lack of adequate training in rheumatic diseases and the evidence suggesting suboptimal patient management practices, the role of continuing medical education (CME) takes on particular importance. Most states in the US require CME for renewal of licenses (24), and a number of specialty boards and colleges in Canada and the US require CME, either alone or as part of a recertification procedure.Despite the move toward mandatory CME, a debate continues about the effectiveness of CME in changing physicians' behavior and in influencing patient outcomes (24-32). ...