The certification process in adolescent medicine.

The first and second certifying examinations in adolescent medicine were administered jointly by the American Board of Pediatrics (ABP) and the American Board of Internal Medicine (ABIM) on November 15, 1994, to 295 candidates, and on November 18, 1997, to 197 candidates including 170 who were taking the examination for the first time. This report describes the certification process and the characteristics of those taking the first and second examinations in adolescent medicine. The purposes of certification in adolescent medicine, consistent with the purposes of specialty boards, are to improve the quality of patient care during the second decade of life, expand and improve training to include a comprehensive and scientifically based approach to the care of young people as they enter adulthood, and stimulate faculty development and research for the problems of the adolescent population (1). Certification addresses these goals by establishing rigorous standards for training, verifying the knowledge base of candidates through a written examination, and requiring periodic recertification. This certification program recognizes a growing number of physicians with expertise in the medical problems of adolescents and the emergence in the past 30 years of a body of knowledge about the health problems, the effects of rapid biologic and psychologic change, and the interaction of biologic, psychologic, social, and environmental forces that characterize this unique developmental period (2).

Subspecialty internal medicine in the United States: in and outside the hospital.

In the U.S. the subspecialties of internal medicine are well integrated in the overall discipline of internal medicine and each one is also unique. The American Board of Internal Medicine at present certifies 16 unique subdisciplines, which are, in order from the largest to the smallest: cardiology (14,000 certified cardiologists), gastroenterology (10,000 certified), pulmonary disease (9000 certified), critical care medicine (6500 certified), hematology, oncology, nephrology, infectious diseases, allergy and immunology, endocrinology, rheumatology, geriatrics, sport medicine, adolescent medicine. General internal medicine is the first level of certification after three years of training: the above-mentioned subspecialties are the second level (after an additional 2 to 3 years of training). Some subspecialties are at a third level or "third tier". Within cardiology, "third tier" certification recognizes clinical cardiac electrophysiology and interventional cardiology. These disciplines require an additional, fourth year of cardiac training and prior certification in cardiology. Between the departments of internal medicine and the subspecialty divisions there is often a "healthy tension" stemming from unavoidable inequities in the financial contributions of the subspecialties to the department. Currently, there is a decline in specialization rate due to a variety of converging factors, among them restricted direct access to specialists in HMOs, and the developing hospitalist movement: full-time hospitalists are less likely to need subspecialist consultation than other generalists.

Future roles and training of internal medicine subspecialists. American Board of Internal Medicine Task Force on Subspecialty Internal Medicine.

While fellowship training programs are being reduced in size to better conform to societal needs, the training of subspecialist basic scientists and clinical investigators must be protected to ensure continued discovery and the scholarly application of knowledge to patient care. Fewer subspecialist clinicians must be appropriately trained to serve as consultants, as principal care providers, and as scholarly leaders and educators in their subspecialties. This article describes the recommendations of the American Board of Internal Medicine for subspecialty training. To encourage physicians to choose careers as investigators, overlapping but different training paths are delineated for subspecialist clinicians and investigators. More didactic coursework is recommended for both paths. To maximize the contribution of fewer subspecialists, it is essential to provide rigorous training that is appropriately relevant and realistically matched with career opportunities.

The validity of an experimental video examination.

This study investigated the validity of an experimental video examination that had previously been shown to be feasible and have reproducible scores for a sample of candidates taking the 1987 ABIM Cardiovascular Disease examination. Correlations of the three video formats of echocardiograms, ventriculograms, and arteriograms with traditional item formats, similar content multiple choice question scores, experience, and external measures suggest that the video formats could be a worthwhile addition to the examination.

The reproducibility of scores from three video formats.

The purpose of this paper was to assess the reproducibility of three simulations presented on videotape: echocardiograms, ventriculograms, and arteriograms. The results indicate that roughly comparable numbers of cases, but different amounts of testing time, are required to achieve specified levels of reproducibility; score interpretation affected reproducibility in the expected ways.

First American Board of Internal Medicine critical care examination: process and results.

On November 10, 1987, 2,712 diplomates of the American Board of Internal Medicine (ABIM) took the initial examination recognizing added qualifications in critical care medicine (CCM). The 1-day examination, made up of one-best-answer and multiple true/false questions, covered a broad range of topics from the internal medicine domain of CCM. Diplomates were admitted to the examination through four pathways requiring formal training in CCM and two practice pathways; 89% of the diplomates entered through the practice pathways. The absolute standard for the examination resulted in a total group pass rate of 64% (n = 1,725); the pass rate for diplomates with training in CCM was 84%. Examination performance was positively associated with months of critical care training, earlier ABIM examination scores, certification in pulmonary medicine, program directors' ratings of overall clinical competence, and several factors associated with the critical care work environment.

A core component of the certification examination in internal medicine.

OBJECTIVE: To develop and test the psychometric characteristics of an examination of core content in internal medicine. DESIGN: A cross-sectional pilot test comparing the core examination with the 1988 certifying examination and two pretest examinations. SETTING: The 1988 certifying examination of the American Board of Internal Medicine. PARTICIPANTS: A random sample of 2,975 candidates from 8,968 candidates who took the 1988 certifying examination were given the core examination; similarly drawn samples were each given one of two pretests of traditional questions. INTERVENTIONS: A framework for developing an examination of core internal medicine questions was designed and used to develop a 92-question core test with an absolute pass/fail standard. RESULTS: Candidates answered 74% of core internal medicine questions, compared with 64%, 52%, and 53% of traditional questions on the 1988 certifying examination and the two pretests. The discriminating ability of the core internal medicine examination was lower than that of the certifying examination (r-values were 0.28 and 0.34, respectively). The pass rate was 83% for the core internal medicine examination and 57% for the certifying examination; 27% passed the core examination and failed the certifying examination; 1% passed the certifying examination and failed the core examination. CONCLUSION: Core internal medicine questions were easier than but almost as discriminating as traditional questions of the certifying examination. A small percentage of candidates passed the certifying examination but failed the core examination.

The relevance to clinical practice of the certifying examination in internal medicine.

OBJECTIVE: To determine the relevance of the initial certifying examination to the practice of internal medicine and the suitability of items used in initial certification for recertification. DESIGN: Using a matrix-sampling approach, items from the 1991 Certifying Examination were assigned to two sets of judges: directors of the American Board of Internal Medicine (ABIM) and practicing general internists. Each judge rated the relevance of items on a five-point scale. PARTICIPANTS: 54 current or former directors of the ABIM and 72 practicing general internists; practitioners were nominated by directors and their ratings were included if they spent > 80% of their time in direct patient care. RESULTS: The directors' mean rating of all 576 items was 3.98 (SD = 0.62); the practitioners' mean rating was 4.11 (SD = 0.82). The directors assigned to 27 items ratings of less than 3 and the practitioners assigned to 42 items ratings of less than 3; seven of these items received low ratings from both groups. There were differences in the two groups' ratings of the relevance of various medical content categories, but the mean rating of core items was higher than that of noncore items and the mean rating of items testing clinical judgment was higher than that of items testing knowledge or synthesis. CONCLUSIONS: These findings suggest that the initial certifying examination is relevant to clinical practice and that many of the examination items are suitable for use in recertification. Differences in perception appear to exist between practitioners and directors, and the use of practitioner ratings is likely to be a routine part of judging the suitability of items for Board examinations in the future.