Modernizing the Paths to Certification in Internal Medicine and Its Subspecialties Lee Goldman, MD
T
he training requirements for internal medicine and its subspecialties have remained fundamentally the same over the past several generations. Physicians must first complete core training in internal medicine, and board certification in internal medicine must precede board certification in the medical subspecialties or eligibility for added qualifications (Table). The initial length of training required for board certification in internal medicine has generally been 3 years, especially recently as the number of competencies to be obtained during training has increased. Exceptions to this 3 year expectation were widely granted a generation ago, when core internal medicine training was overwhelming inpatient, and “short-tracking” (i.e., requiring only 2 years of internal medicine rather than 3 years before subspecialty training) was more common. Now, however, the rare exceptions to the 3-year requirement are reserved for trainees in a physician-scientist track, in which an additional year of research training would counterbalance, at least in part, 1 less year of internal medicine training. This current approach to training and certification in internal medicine can be compared metaphorically with cell maturation. Trainees remain pluripotent stem cells for 3 years before either failing to differentiate, and thereby remaining as internists only, or differentiating into subspecialists. The duration of stem cell training is in part historic and in part driven by the increasing variety of skills and competencies that fall under the broad umbrella of internal medicine. Unlike true stem cells, however, general internists typically forfeit their differentiation potential soon after the completion of their 3 years of training.
IS CERTIFICATION STILL CONSISTENT WITH THE REALITIES OF TRAINING AND PRACTICE? To the extent to which the current approach to board certification is historic, it is worthwhile to remember that internists originally were prepared to be broad-based consultants for the large number of general practitioners whose training was limited to an internship and who would need help for the diagnosis and management of a
Am J Med. 2004;117:133–136. Requests for reprints should be addressed to Lee Goldman, MD, University of California, San Francisco, Department of Medicine, 505 Parnassus Avenue, Box 0120, San Francisco, California 94143-0120. © 2004 by Elsevier Inc. All rights reserved.
Table. Board Certification and Added Qualifications in Internal Medicine Required Time
Subspecialty Board Certification Cardiovascular Endocrinology, diabetes, and metabolism Gastroenterology Hematology Medical oncology Nephrology Pulmonary disease Rheumatology Added Qualifications, Prior Medical Subspecialty not Required Adolescent medicine Critical care medicine Geriatric medicine Sports medicine Added Qualifications after Cardiovascular Certification Clinical cardiac electrophysiology Interventional cardiology
Total Training Months
Clinical Training Months
36 24
24 12
36 24 24 24 24 24
18 12 12 12 12 12
24 24 12 12
— 12 12 12
12 12
12 12
substantial proportion of the serious acute and chronic conditions of adults. This historic role of internists was squeezed from two directions. Consultative expertise became the domain of the increasing number and spectrum of medical subspecialists, who could provide not only cognitive consultation but also the procedures that were often recommended. Simultaneously, 3 years of training provided a new generation of family physicians with skill sets that were very different from those of general practitioners but no longer so different from those of general internists, especially in the primary care setting. The result of both of these forces was for many general internists to become primary care physicians rather than consultants. With these changes, the amount of time that general internists spent in the hospital declined. Other trends, including a higher threshold for hospitalization, the marked reduction in elective medical admissions for diagnostic evaluations, pressures to shorten length of stay, and the forces of managed care further diminished the inpatient presence of the typical general internist and led to the rise of hospitalists. Although some office-based general internists continue to have vibrant inpatient 0002-9343/04/$–see front matter 133 doi:10.1016/j.amjmed.2004.05.001
Modernizing Certification in Internal Medicine/Goldman
practices, a very large proportion of newly trained and even experienced internists are now declaring themselves either predominantly as office-based generalists or as inpatient hospitalists. In less than a decade, the fledgling field of hospitalists is on track to grow to be larger in number than any of the medical specialties, even cardiology (2). As the range of skills and competencies needed to manage the broad spectrum of problems in ambulatory and inpatient medicine expanded, the American Board of Internal Medicine required a full 3 years of training for all but a rather small number of physician-scientists. Recently, the Society of General Internal Medicine even suggested that internal medicine training might be expanded from 3 years to 4 years (3), a change that is unlikely to increase the attractiveness of internal medicine. At the same time, the length of some subspecialty training programs has increased, driven by the need to obtain outpatient as well as inpatient and procedural skills, as well as by requirements for research.
WHAT ARE THE PROBLEMS WITH THE CURRENT SYSTEM? In the current system, I see several problems. One, as emphasized by Blackwell and Powell (4), is that training simply takes too long. For medical subspecialists, the length of training now is often as long as for their surgical counterparts; for example, it takes about as long to become a board-certified interventional cardiologist as to become a board-certified cardiac surgeon. We are discouraging future clinicians, whose investment in training will have far higher economic yields in alternative specialties (5). We are also discouraging future physician-scientists, because the required training in clinical and research skills means that most will not even seek their first faculty positions until their mid-30s, a time when most of their college classmates have long since matured professionally. A second problem is that our attempts to make everyone pluripotent has reduced the focus of our training and the ability of our trainees to tailor their educational experiences. By shortening the duration of core training and providing a number of appropriate subsequent options, our trainees could better prepare for what they actually want to do.
WHAT TO DO? It is my contention that certification in internal medicine requires more change than the American Board of Internal Medicine has suggested (6). I have proposed that the length of core training, during which the young physician 134
July 15, 2004
THE AMERICAN JOURNAL OF MEDICINE威
Volume 117
Figure 1. Short tracks for certification in internal medicine and its subspecialties.
is a pluripotent stem cell, should be reduced to 2 years (7). This 2-year curriculum would be specifically designed to include the fundamentals of inpatient and outpatient medicine, emphasizing training and learning rather than service. To be eligible for board certification in internal medicine, trainees would then have to spend at least 1 additional year in one or more of a number of options (Figure 1). For those who wish to become office-based primary care physicians, the third year would be predominantly spent in ambulatory internal medicine, analogous to many current primary care residency programs. These trainees would first sit for boards in internal medicine; if they passed, they could then sit for an examination that would provide them with added qualifications in Adult Office Medicine. An alternative would be to spend the third year of training predominantly in the inpatient setting, analogous to many current categorical residency programs, with subsequent certification in Adult Hospital Medicine. In an era of secure, interactive, computerbased examinations, the test for added qualifications could be administered sequentially on the same day after the examinee passed the first test for internal medicine. Some trainees might elect a 4-year pathway, with 1 year of mostly outpatient training and 1 year of mostly inpatient training (Figure 2). This pathway would be a natural option for trainees who currently spend a fourth year as chief residents. Trainees who took the 4-year pathway would be eligible for board certification in internal medicine as well as added qualifications in both Adult Office Medicine and Adult Hospital Medicine. It should be recognized that board certification and added qualifications may or may not correspond directly with privileges granted by medical groups, managed care organizations, or hospitals. Physicians who would become certified as internists with added qualifications in Adult Office Medicine would have as much adult inpatient training as is currently received by residents in primary care internal medicine programs, and more than is currently received by family physicians. Since graduates of these programs routinely receive hospital privileges, it
Modernizing Certification in Internal Medicine/Goldman
HOW COULD THIS PROPOSAL ACCOMMODATE CURRENT PRACTITIONERS?
Figure 2. Long tracks for certification in internal medicine and its subspecialties.
would be expected that graduates of this new pathway would similarly qualify. Trainees could enter subspecialty training programs, including programs in fields such as geriatrics that are jointly sponsored by internal medicine and other boards, after 2, 3, or 4 years of internal medicine training (Figures 1 and 2). Trainees would be eligible to sit for the internal medicine boards after the third year of training, regardless of whether the third year was in Adult Office Medicine, Adult Hospital Medicine, another area of added qualification, or a subspecialty. Board eligibility in the subspecialties of medicine would always require prior board certification in internal medicine. Subspecialty training requirements would generally not change, with one major exception. Many subspecialty boards require dedicated research time for board eligibility. Except in very rare circumstances, previous research time, regardless of whether it led to publications or to an advanced degree, is not counted. As a result, future physician-scientists who have had research training and publications typically are required to duplicate this educational experience during their subspecialty fellowships. I would suggest that the research requirements of subspecialty training routinely be satisfied by prior dedicated research training and research output. Certainly, extra years of research training should qualify. I would also assert that some of the experiences and productive research performed during 4 years of medical school in research-intensive training programs should qualify. My proposal also provides an efficient mechanism for general internists to obtain an added qualification in a discipline that their patients will actually understand: Adult Office Medicine or Adult Hospital Medicine. These added qualifications will address many of the perception problems currently faced by internists (8 –10), including the public’s long-standing difficulty in distinguishing internists from interns.
Whenever a new type of certification becomes available, one question is how to address the large number of physicians whose training antedated its availability. I would assert that recertification provides an ideal opportunity for practicing internists to obtain added qualifications in Adult Office Medicine and/or Adult Hospital Medicine. Current diplomates of the American Board of Internal Medicine could be asked to provide evidence of sufficient postresidency experience in either or both of these domains, and then take and pass the appropriate examination(s). This postresidency pathway should be made available to all current and future diplomates in internal medicine, so the decision not to do the training and obtain certification for either of these added qualifications during residency would not permanently limit the options for a board-certified internist.
WILL THIS NEW APPROACH FRACTIONATE MEDICINE, AS HAS HAPPENED IN SURGERY? The fundamental difference between my proposal and what has happened in surgery is that board certification in internal medicine would remain a prerequisite for all subsequent added qualifications and board certification. Unlike surgery, maintenance of this core of internal medicine training and certification should prevent fractionation. My concern is that the alternative, a stubborn attempt to maintain an increasingly fragile status quo, will be more detrimental to internal medicine. If training becomes progressively longer and less pertinent to what our trainees ultimately want to do, there will be increasing pressure to pull internal medicine apart and create independent paths for subspecialists.
IS THIS IDEA JUST TOO CONFUSING, OR CAN IT WORK? The range of options in my proposal carries the potential risk of creating too many options and thereby diluting the credentials of internal medicine. However, my suggestion addresses several key issues. Training will generally be shortened. New, descriptive options for added qualifications will help office-based and/or inpatient-oriented physicians distinguish themselves as doctors for adults. Recertification would become a proactive opportunity to add recognition rather than just a defensive requirement to maintain existing certification. Internal medicine would July 15, 2004
THE AMERICAN JOURNAL OF MEDICINE威
Volume 117 135
Modernizing Certification in Internal Medicine/Goldman
no longer be a discipline in which training must follow a fixed price menu, but rather would provide a more flexible menu to allow its trainees and practitioners to obtain their desired skills in a timely fashion and to receive appropriate credentials that describe what they do. (1)
REFERENCES 1. American Board of Internal Medicine. Policies and Procedures for Certification. Available at: http://www.abim.org/about/p&p_ 2003.pdf. Accessed April 29, 2004. 2. Lurie JD, Miller DP, Lindenauer PK, Wachter RM, Sox HC. The potential size of the hospitalist workforce in the United States. Am J Med. 1999;106:441–445.
136
July 15, 2004
THE AMERICAN JOURNAL OF MEDICINE威
Volume 117
3. Larson EB, Fihn SD, Kirk LM, et al. The future of general internal medicine. J Gen Intern Med. 2004;19:69 –77. 4. Blackwell TA, Powell DW. Internal medicine reformation. Am J Med. 2004;117:107–108. 5. Weeks WB, Wallace AE. Long-term financial implications of specialty training for physicians. Am J Med. 2002;113:393–399. 6. Duffy FD, Zipes DP. The future of certification and recertification. Am J Med. 2004;117:140 –144. 7. Goldman L. Specialized tracks, or one size does not fit all! The ABIM Report. 2000:57–61. 8. Goldman L. Adult (not internal) medicine. Ann Intern Med. 1997; 127:835–836. 9. Salerno SM, Reed W, Landry FJ, et al. Patient perceptions on the capabilities of internists: a multi-center survey. Am J Med. 2001; 110:111–117. 10. Goldman L. The name game: Tradition and change. Am J Med. 2001;110:153–155.
