Certification in internal medicine: 1989–1992

Objective: To determine whether changes in the demographic/educational mix of those entering internal medicine from 1986 to 1989 were associated with differences among them at the time of certification. Participants: Included in the study were all candidates for the 1989 to 1992 American Board of Internal Medicine certifying examinations in internal medicine. Measurements: Demographic information and medical school, residency training, and examination experience were available for each candidate. Data defining quality, size, and number of subspecialties were available for internal medicine training programs. Results: From 1990 to 1992, the total number of men and women candidates increased as did the numbers of foreign-citizen non-U.S. medical school graduates and osteopathic medical school graduates; the number of U.S. medical school graduates remained nearly constant and the number of U.S.-citizen graduates of non-U.S. medical schools declined. The pass rates for all groups of first-time examination takers decreased, while the ratings of program directors remained relatively constant. Program quality, size, and number of subspecialty programs had modest positive relationships with examination performance. Conclusions: Changes in the characteristics of those entering internal medicine from 1986 to 1989 were associated with declines in performance at the time of certification. These declines occurred in all content areas of the test and were apparent regardless of program quality. These data identify some of the challenges internal medicine faces in the years ahead. Received from the American Board of Internal Medicine, Philadelphia, Pennsylvania. This research was supported by the American Board of Internal Medicine but does not necessarily reflect its opinions or policies.