The Flexner Report in 1910 was the first workforce study of the modern era. In 1900 there were 160 medical schools graduating 5214 students. They were the survivors of 457 schools that dotted the landscape with medical schools in the 19th century, especially in the midwestern region of the United States.1 In the late 19th century, the United States was producing nearly half of the world’s medical graduates. By the time of the Flexner report in 1910, the number of schools had decreased to 131, graduating 4440 students. Flexner felt that quality of education in at least half of the schools was unacceptable and that half should close. The impact of his report was immediate; by 1924 only 79 schools, with 3562 students, were in existence.2
During the Great Depression, the number of medical schools decreased further. By World War II a doctor shortage existed. The War further sapped the domestic workforce by drafting many of the younger physicians for service in the military. In some states insufficient physicians remained on the home front and some medical societies closed for the duration. World War II initiated the first federal subsidy of medical education through the V-12 program. It abbreviated premedical requirements to 2 years and mobilized the newly minted graduates for military service immediately. Following World War II, the GI Bill provided an educational opportunity for returning veterans. Peter Drucker called the GI Bill one of the most important pieces of legislation ever passed in the United States as it developed an educated workforce for the Knowledge Society.3 The GI Bill provided undifferentiated, general medical officers, or general practitioners, an opportunity to specialize on return to civilian life. An important dimension of the GI Bill was that for the first time a stipend was provided so that individuals could afford to take a residency and have a family. The residency systems rapidly expanded to accommodate the returning GIs. Many residencies were hastily formed to accommodate the large number of returning veterans. In Surgery, more than 600 surgery training programs were created. Most of those hastily-founded residencies have consolidated and marginal programs have closed so that only 253 remain. Over the past 50 years, however, the number of graduates of general surgery programs has remained remarkably constant with approximately 1000 finishing chief residents each year.4
The population growth and improved economy of the late 1940s and early 1950s resulted in need for more health professionals. Accordingly, the number of schools of medicine increased from 79 in 1950 to 127 in 1988. The number of students graduated increased from 5553 in 1959 to 15,919 in 1988.2
The doubling of medical schools, and students graduating, spawned a number of health workforce studies in the 1970s because of concern that too much expansion had occurred. The most extensive study, the Graduate Medical Education National Advisory Committee (GMENAC), was the most influential and controversial study. The study authors anticipated and advocated a primary care, gatekeeper workforce to replace the aging, solo-practice, general practitioner. The most debated projection was a prediction of a surplus of 145,000 physicians by the year 2000.5
In the same time period, the American Surgical Association Council and the American College of Surgeons’ Board of Regents were on parallel tracts to study a spectrum of future needs of surgical workforce: education, certification, and other issues. The 2 organizations joined forces and with external funding produced over 5 years a report in 1974 called the Study of Surgical Services in the United States (SOSSUS). The Chair of the Executive Committee was Dr. George D. Zuidema, and the Vice-Chairman was Dr. Francis D. Moore who was also Chair of the Subcommittee on Surgical Manpower. The report concluded that a sufficient number of surgeons was probably available but recommended enhancing academic medicine by reducing the number of residencies and other changes.6p17–90 With the perspective of 50 years, the study seems to have accurately predicted the reduction of residencies. The closure of residencies, however, usually resulted in consolidation without reduction in graduation numbers. In several areas the authors’ predictions were based on estimates by the Census Bureau that the US population would be 250 million by the year 2000 when it actually grew to over 300 million. Another miscalculation was based on estimates for the amount of time surgeons spend in nonoperative patient care and office practice.
The next big change occurred in 1965 when the Medicare law was enacted. It provided direct and indirect payment for graduate medical education. Because Medicare funds about 30% more resident positions than there are graduates of US medical schools, the determinant educational experience for producing physicians for the US workforce becomes residencies. To fill the postgraduate year (PGY) 1 residency positions, over 15,000 US medical graduates are joined by 5000 to 7000 new medical graduates who did not attend US medical schools. Most are international medical graduates (IMGs) from India, with the second largest number being US citizens who train in Caribbean medical schools (USMGs).
In 1974 SOSSUS reported that the US health workforce of surgeons was 6.93/100,000. In 1994 Jonasson et al estimated that in the 20 years since SOSSUS was published, the number of general surgeons had increased slightly to 7.1 per 100,000.7 In part as an outgrowth of GMENAC, the Health Services Research Agency, which collects health professional data, began observing additional dimensions of health workforce. The Council on Graduate Medical Education (COGME) was founded, as part of the Consolidated Omnibus Budget Reconciliation Act of 1986, and continued much of the methodology, as well as the point of view of the GMENAC.
One of the miscalculations of COGME was that a desirable national policy should be a health workforce reflecting a 50:50-ratio of primary care to specialist. The optimal number of doctors of different specialties became an increasing focus of health policy. The government founded 6 workforce centers located throughout the country; northeastern, Albany, NY; southeastern, Chapel Hill, NC; central, Chicago, IL; southwestern, San Antonio, TX; western, San Francisco, CA; and northwestern, Seattle, WA.
During 2006 to 2007 the Southeastern Regional Health Workforce Center, located at the Cecil G. Sheps Center at The University of North Carolina, assisted the North Carolina Institute of Medicine in addressing the health workforce shortage of the State.8 The Sheps Center, using data from the American Medical Association (AMA), estimates a national ratio of 6 surgeons per 100,000, which is fewer than the SOSSUS study of 1974 (6.93) and the Jonasson, Sheldon paper of 1994 (7.1).7 Because the AMA data overestimates numbers, the more accurate estimate is likely to be closer to 5/100,000. In any case we have fewer surgeons per population than 50 years ago.
Some of these shortages are products of the assessments and recommendations of the studies of the 1970s and 1980s. COGME recommendations of 1993 were based on the assessment that a physician surplus existed. COGME and most of the medical community opined that too few students were entering primary care. Medical organizations, including the Association of American Medical Colleges (AAMC), recommended that half of all US medical graduates should enter primary care.9 In addition the overproduction of physicians was to be regulated by decreasing the number of residents and limiting the total number of residency positions to the number of US medical graduates plus 10%. By 1990 these policy targets had begun to show fruition. Of the 188,207 certificates awarded by the American Board of Medical Specialists between 1980 and 1990, 49% were in primary care fields. In that same decade General Surgery certificates were awarded to 9445 surgeons.10 The American Board of Surgery (ABS) continues to certify approximately 1000 surgeons annually.
Health care in the United States has undergone an enormous transformation in the past 40 years. The changes include a spectrum of providers from different educational backgrounds. Nursing, especially, has developed a number of advanced practice specialties. All nonphysician providers have broadened their scope of practice. For example, nurses can write prescriptions in all 50 states; in many states there are opportunities for some physician groups, such as naturopaths, to do some minor surgery. Licensing boards have great control over scope of practice by different providers. Professional societies of nonphysician clinicians are effective lobbies and frequently go directly to state legislatures to pass legislation allowing increased scope of practice.
About 30% more positions in graduate medical education exist than graduates of US medical schools. For that reason, the determinant of physician numbers is the number of completing residents each year. In 2003, 23,680 physicians entered graduate medical education in all fields; 64% were graduates of the 125 US medical schools; 11% were from osteopathic schools: and 20% were non-US citizen IMGs. So, the United States is a recruiter of physicians from other countries.11p3,5 There is increasing concern that the developed western countries are recruiting physicians from undeveloped countries, which raises issues of international distributive justice.
The World Health Organization recently addressed the problem of an international shortage of health workers by declaring the decade of the health worker in 2007.12 World Health Organization estimates that there is a shortage of approximately 5 million health workers, mainly physicians, nurses, and midwives. The Albany Health Workforce Center estimates that the United States is in need of 4 million health workers.13 The appetite for health workers is great in the western countries and the need is great in the underdeveloped countries; both are competing for a finite pool of workers.
Another contributing problem is that US medical schools and residencies are predominantly located in the eastern US whereas the population is moving to the southeast and southwest. For example, there are 292 resident physicians per 100,000 in the District of Columbia and 26 per 100,000 in the most populous state in the union, California. Because medical students and residents often practice in the region or state where they train, a strategy for avoiding or correcting a doctor shortage is training sites.
Dr. Darrell Kirch has discussed recommendations and efforts of the AAMC to achieve a target of a 30% increase in medical school graduates. That increase, even if it occurs expeditiously, will not have impact for another 15 years because of the long educational track before entering practice. Moreover, if the Balanced Budget Act of 1997 remains in force, the increase in medical school graduates will be lessened. About 15,000 physicians graduate from US medical schools each year and another 5000 to 7000 IMGs join them at the PGY-1 level. If US medical school graduates increase but residency positions remain fixed, many of the new residency positions will be filled by USMGs which usually have priority in the residency selection. The Balanced Budget Act needs to be repealed to reestablish the flexibility for graduate medical education to expand as population increases. In any event it is likely that US dependence and attractiveness for IMGs to supplement USMGs will continue.
Among the unpredictable elements of workforce planning are the career choices of a changing cohort of medical students. Medical students are now almost equally balanced between men and women. The current generation of physicians is less interested in the dominance of professional life over personal lifestyle than previous generations. Between 1998 and 2006 the percentage of USMGs choosing primary care fields declined, especially in Family Medicine (Fig. 1). Recent information from the NRMP reveals USMGs are truly in a buyer’s market. Only 5 specialties have more applicants than positions available: Plastic Surgery, General Surgery, Dermatology, Orthopedic Surgery, and Radiation Oncology.11 In 2001, 68 General Surgery positions were unfilled in the match. General Surgery has rebounded and this year filled all but 2 of its positions; however, this is the first year that the number of positions in General Surgery, filled by USMGs, has dropped below 80%. In the 2007 National Residency Matching Program match, 229 of the 1057 available General Surgery categorical PGY-1 positions were filled by IMGs14 (Fig. 2).
The AMA master file reveals decreasing numbers of general surgeons over the past 5 years, from 27.509 in 1998 to 24,902 in 2002 (Fig. 3). The community of osteopathic physicians continues to increase with 5 new schools in the past few years. In 2005 there were 51,015 osteopathic medicine (DOs) in the AMA master file, which identifies Accreditation Council on Graduate Medical Education graduates.15 There are probably 70,000 osteopaths practicing in the United State today. The American Osteopathic Association lists 4296 self-designated surgeons in all surgical specialties (Personal communication February 19, 2007. Dr. Steven Andes for Dr. Konrad Miskowicz-Retz, American Osteopathic Association.). Of these, 1037 are general surgeons and 1114 are orthopedists. There are 119 doctors of DOs certified by the ABS as general surgeons, of whom 89 have been certified since 1991. (Personal correspondence, Dr. Robert S. Rhoads, ABS April 18, 2007.) The increasing number of DOs, and their access to Accreditation Council on Graduate Medical Education residencies, will be a source of surgeons. However, the rapidly growing number of osteopathic physicians and surgeons will be insufficient to ameliorate the shortage.
Although the focus of workforce studies is usually on healthcare providers, the important focus should be on service provided. All specialties have a breadth of skills, procedures, or services in which some education and training overlaps. However, soon after finishing residency or entering practice, most physicians narrow their scope of practice. In General Surgery nearly 80% of residents, who complete categorical general surgery residencies, pursue further specialization, often in 1 of the 9 principal components of general surgery. Stitzenberg and Sheldon’s, 2005 study reported the most common fellowship chosen by graduates of categorical general surgery residencies is minimally invasive surgery.16 Minimally invasive surgery is not an approved subspecialty of the American Board of Medical Specialties. Thus, with the evolution into early subspecialization, the scope of practice is narrowed almost from the time of fellowship completion.
Fellowships not approved by the American Board of Medical Specialties are, in part, displacing the attractive specialties of the past, such as vascular and cardiac surgery. For most of the past 25 years, cardiovascular surgery and vascular surgery were the most sought fellowships from a general surgery base. Recently the attraction of cardiac surgery to general surgery residents has diminished. In 1997 there were 95 programs and 175 applicants; in 2006 the program number had fallen to 90 with 84 unmatched positions. Less than 100 applicants for thoracic surgery are anticipated from US medical schools in 2007.17 Vascular Surgery, a subspecialty of the ABS since 1986, and recently a designated primary certificate, has been a highly competitive post general surgery fellowship for 40 years. It is now experiencing difficulty filling fellowships. In 2005, 23% of the programs and 21% of the positions went unfilled. Cardiovascular disease is predicted to be the most common disease affecting the increasing elderly population in the 21st century, and a shortage of cardiac and vascular surgeons seems inevitable.
Over the past 25 years, some professional conflict has occurred over the scope of practice of primary care doctors and specialists. There is also overlap in scope of practice among nonphysician clinicians and physicians. The Southeast Regional Center for Health Workforce Studies of The University of North Carolina provides data on licensed healthcare workers in 16 fields, including medical specialties, nurse practitioners, podiatrists, chiropractors, and allied health personnel. With an increased scope of practice, the fastest growing licensed group is nurse practitioners, who have increased 214%. Whereas physician numbers have increased by 25%, the number of physicians delivering babies has decreased. Nurse midwives are providing obstetrical care in some counties that lack physicians.
Nonphysician clinicians are capable of providing services similar to that of physicians in many clinical settings. However, few surgical services or specialties can be served by primary care providers or nonphysician clinicians. North Carolina data surveys identify a worrisome trend in the availability of general surgeons; between 1995 and 2005, 47 North Carolina counties experienced a decline in the number of general surgeons. Another 4 counties lost all general surgeons and 18 counties had no surgeons at all.18 In 2006 the Institute of Medicine of the National Academy of Sciences, with leadership provided by Dr. A. Brent Eastman, published a report titled, Future of Emergency Care: Emergency Medical Services at the Crossroads. The report documented diminished involvement and availability for providing emergency care by general surgeons, neurologic surgeons, orthopedists, hand surgeons, plastic surgeons, and others.19 Similarly, in 2006 the American College of Physicians published a white paper, “The Impending Collapse of Primary Care Medicine and Its Implications for the State of the Nation’s Health Care.”20 In addition to the Institute of Medicine, the 2 largest organizations of specialists in the country, the American College of Physicians and the American College of Surgeons, have identified access to their services to be a problem.
Solutions to the health workforce shortage are complicated.21 The first challenge is awareness of the problem and the consequences of a physicians, nurse, and other health workforce shortage. An ongoing need is for regional and statewide data so that decisions about workforce can be based on information of value. As was discussed by Dr. Kirch, efforts are underway to increase physician output by 30%. Dr. Richard A. Cooper and Linda Aiken, RN, at the Wharton School of Business of the University of Pennsylvania, have organized a group called Council on Physician and Nurse Shortage (COPNS) to continue analysis and advocacy for the issue.
The medical community needs to advocate for revocation of the Balanced Budget Act of 1997. This Balanced Budget Act has been problematical because it established downward reduction in Medicare-allowed fees. It also froze Medicare-funded resident positions. A detailed consideration of the medical community should take responsibility for changing the educational environment to accommodate a changing world. A creative approach to team patient care needs to occur. It will hopefully occur within the context of healthcare reform.
1. Bonner TN. Iconoclast: Abraham Flexner and a Life in Learning
. Baltimore, MD: The Johns Hopkins University Press; 2002.
2. Bowles MD, Dawson VP, eds. With One Voice: The Association of American Medical Colleges 1876–2002
. Washington, DC: Association of American Medical Colleges; 2003.
3. Drucker PF. Post-Capitalist Society
. New York: Harper Business; 1993.
5. Government Printing Office. Report of the Graduate Medical Education National Advisory Committee to the Secretary, Department of Health and Human Services. Vol. 1: GMENAC Summary Report
. Washington, DC: Government Printing Office; 1981.
6. Surgery in the United States: A summary report of the study on surgical services for the United States. Sponsored Jointly by the American College of Surgeons and the American Surgical Association; 1975.
7. Jonasson O, Kwakwa F, Sheldon GF. Calculating the general surgery workforce. JAMA
8. North Carolina Institute of Medicine, Durham, NC. Available at: www.nciom.org
. Accessed 30 May, 2007.
9. Council on Graduate Medical Education (COGME). Patient Care Physician Supply and Requirements: Testing Cogme Recommendations
. 8th Report. Washington, DC: Department of Health and Human Services; 1996.
10. American Board of Medical Specialties, Philadelphia, PA. Available at: www.abms.org
. Accessed 30 May, 2007.
11. Jolly P. Charting Outcomes in the Match: Characteristics of applicants who matched to their preferred specialty in the 2005 NRMP main residency march. A collaborative project of the National Resident Matching Program and the Association of American Medical Colleges; 2006. Available at: /www.nrmp.org/matchoutcomes.pdf
. Accessed 29 May, 2007.
13. Center for Health Workforce Studies, University of Albany, School of Public Health, Rensselaer, NY. Available at: http://chws.albany.edu
. Accessed 30 May, 2007.
16. Stitzenberg KB, Sheldon GF. Progressive specialization within general surgery: adding to the complexity of workforce planning. J Am Coll Surg
19. Institute of Medicine of the National Academies Future of Emergency Care: Emergency Medical Services at the Crossroads
. Washington DC: Institute of Medicine of the National Academies; 2007.
© 2007 Lippincott Williams & Wilkins, Inc.
21. Sheldon GF. Workforce issues in general surgery. Am Surg