The In-Training Examination in Internal Medicine: an analysis of resident performance over time.

Background: The In-Training Examination in Internal Medicine (IM-ITE) has been offered annually to all trainees in U.S. medical residency programs since 1988. Its purpose is to provide residents and program directors with an objective assessment of each resident's personal performance on a written, multiple-choice examination and the performance of the residency program compared with that of its peers.

Objective: To analyze trends in the demographic characteristics and scores of examinees during the first 12 years of administration of this examination.

Design: Descriptive analysis over time.

Setting: U.S. residency programs in internal medicine, 1988-2000.

Participants: Residents at all levels of training in categorical, primary care, and medicine-pediatrics programs in the United States and Canada. The number of examinees increased from 7500 in 1988 to almost 18 000 in 2000.

Measurements: After calibration of the scores for each examination, test results were compared and analyzed for selected cohorts of residents over 12 years.

Results: More than 80% of residents in medicine training programs participate in the IM-ITE, most on an annual basis throughout their period of training. Test performance improves at a predictable rate with each year of training. Since 1995, international medical school graduates have persistently outperformed graduates of U.S. medical schools. Test results were affected by the timing of the examination, the time that was available to complete the examination, and the actual time that residents spent in internal medicine training before each examination.

Conclusions: The IM-ITE scores generally improve with year of training time spent in internal medicine training before the examination and time permitted to complete the examination. These observations provide evidence that the IM-ITE is a valid measure of knowledge acquired during internal medicine training.