| | Great expectations: the 21st century health workforceReceived 4 September 2002; accepted 7 September 2002. Abstract Health workforce studies have mostly predicted an oversupply of physicians, a shortage of primary care doctors, and an excess of specialists. As the target date of many of these studies is now passed, it is clear that we are evolving into a shortage of physicians, especially specialists, and that primary care will increasingly be done by nonphysician clinicians. The “knowledge society” requires a different workforce than that predicted by most health planners.
The late 19th and early 20th centuries are a period of history that the economic futurist, Peter Drucker, has labeled as one of “epochal transformation.” He defines epochal transformation as a time when the changes are so great that no one living at the time can imagine the world in which their grandparents lived or their parents were born. Drucker characterizes our epochal era as the Knowledge Society, where the coin of commerce is knowledge rather than industry [1].
In a similar vein, Ray Kurzweil, named in 1988 by the Massachusetts Institute of Technology as Inventor of the Year noted: “The exponential growth of technology in the first two decades of the 20th century matches that of the entire 19th century. The exponential growth of technology in the first five years of the 21st century will inexorably match that of the entire 20th century” [2].
Population demographics will undergo radical alteration in the 21st century. The world will be divided between countries with low birth rates and an aging population, and emerging nations with high birth rates and a young citizenry. Europe will have less population in 2020 than in 2000. The United States, which has an aging population and a low birth rate, would have had a population profile similar to Europe were it not for population growth from immigration. Projected immigration will be primarily from Latin America and Asia, altering the population ethnic distribution. In California, groups that are minorities already outnumber the conventional majority.
Another demographic change of enormous consequence is that 78 million baby boomers, the largest number of people within a narrow age range ever to occupy the earth, and their offspring, the baby boomlet, will alter the current population age distribution. These evolving population demographics have profound implications for health care.
Medicine, for the first time in history, is simultaneously undergoing both a social and a scientific revolution. While some reports, such as the one of the World Health Organization, are critical of the American system of health care, our system at its best is the best.
Gary S. Becker, the 1992 Nobel Laureate in Economics at the University of Chicago, wrote in the January 31, 2000, issue of Business Week, “Longer life was the 20th century’s greatest gift.” Consider that 175,000 years ago average life expectancy of primitive man was 25 to 30 years. Even by 1900, when tuberculosis was the leading cause of death, life expectancy was only 45 years (Table 1). Life expectancy in the Western world grew from 45 years at the beginning of the 20th century to nearly 80 at the dawn of the 21st century. Death rates from our leading cause of death, heart disease, are less than half of what they were in 1950. The healthiest elderly people in the world are the aged in the United States.
While much is heard of the cost of health care, vast improvements in the health of the public have had positive consequences for the economy and shaped modern society. The health economy alone in the year 2000 is in excess of $4 trillion, four times larger than the entire world economy in 1900. Improved health and life expectancy have a significant effect on the economic and political well being of societies. Demographers at the Rand Corporation in the 1980s predicted a significant change in the political structure of the Union of Soviet Socialist Republics. Using the assumption than an unhealthy society is a politically unstable one, they concluded that the USSR was vulnerable to significant political instability. The conclusions were based, in part, on data showing that the USSR was the world’s only industrialized society in which life expectancy was falling. The life expectancy in the USSR declined 33% between 1990 and 1994, or 63.8 to 57.7 years for men [3].
The opposite proposition, ie, that a healthy society is an economically and politically stable society, is an evolving concept. Funding First, an initiative of the Mary Woodward Lasker Charitable Trust, has on its Web site a report, “Exceptional returns, and the economic value of American’s investment in medical research.” Kevin Murphy and Robert Topel, of the University of Chicago School of Economics, and economics professor William Nordham of Yale University, have estimated the economic value of increased life expectancy to units of income. They calculated a huge effect, in that national incomes rapidly grew about 2% per year after adjusting for mortality. Increases in life expectancy during the decades of the 1970s and 1980s were worth $57 trillion to America, a figure six times larger than the entire output of tangible goods and services. These calculations by the economists indicate that reducing deaths from cancer by just one fifth would be worth $10 trillion to Americans, or double the national debt. Since mortality rates could fall faster in poorer countries with the potential for large immediate gain, similar calculations would show an even greater rate of change than in countries with existing low mortality rates.
Good health is valued by the public, and all political parties appreciate the contribution of medical research to the betterment of our society. The National Institute of Health budget continues to be well supported by our lawmakers and the public.
In the past year, the Institute of Medicine of the National Academy of Science published two reports: the first being To err is human, and the second, Bridging the quality chasm. In Bridging the quality chasm, 10 expectations for the health care system are listed as goals. Among the goals is state-of-the-art health care available 24 hours a day. At the present, that goal will be difficult to obtain, as we lack the financial resources, universal health insurance, and the number of physicians with the proper skills, to implement the emerging scientific breakthroughs.
The question before medical educators is what kind of a workforce is capable of providing the 21st century needs of the public, while also translating the marvelous technologic advances into clinical applications, all at a reasonable cost. The public not only expects us to develop the science of new medicine, they also expect provision of the site of its application, our Council of Teaching Hospitals and Organizations (COTH) and the workforce to implement it [4].
The cover of the 10 September 2001 issue of US News and World Report depicts a busy emergency room, and is titled, “Crisis in the emergency room.” It describes the increasing pressure on emergency services. Ironically, September 11 is, of course, the day of the tragedy of the terrorist’s attacks on the World Trade Center and the Pentagon. The rescue and health care workers’ efforts reflected great credit on our profession. They accomplished their work, as is usually the case, by local and regional disaster plans, and contact among professionals that happen to be personally acquainted. We still need, as has been advocated for many years, a good coordinated hospital specialized bed system to work with the Federal Emergency Management Administration (FEMA) and the military.
The need for a civilian military hospital system to work with FEMA has been under discussion for years. Now, with the threat of continued terrorist acts in the United States that could include biologic and chemical agents, we need to act swiftly. Our emergency system needs enhanced, central coordination. Specialized trauma, burn, dialysis, and critical care units need to be coordinated in a systematic way with military hospitals to provide a program for national emergency medical services to serve the public.
Health care in the 21st century  First, the implementation of managed care in the United States has failed, having pleased neither the providers nor the public, and only transiently reducing costs. It is clear, however, that a legacy of managed care will be a more organized health system. Secondly, it is equally certain that we will not have a primary care gatekeeper, staff model health maintenance organization (HMO) driven healthcare system. The new science requires a specialized workforce. It is our challenge to provide a workforce with the expertise to implement the scientific advances. History of workforce predictions The first workforce study in 1819 was actually a survey by Dr. Samuel Bard, student of John Hunter and Dean of Columbia School of Medicine. He was the only faculty member of King’s College, the forerunner of Columbia, to remain on the faculty in the post-Revolutionary War period. Many of the rest of the faculty were Tories and fled to Canada, Nova Scotia, or England. Bard compared the product, curriculum, and quality of Columbia, Pennsylvania, Harvard, and Maryland with their common antecedent, the University of Edinburgh. In more recent times, the Flexner Report in 1910 dealt with the health workforce of some 3,000 graduates of medical schools, as part of the recommendation for restructuring medical education in the United States. During the 1920s and 1930s when medicine was predominantly solo general practice, it was understood that a doctor shortage existed. Even with accelerated medical education (B-12 Program) during World War II, a physician shortage still existed in the post war period. During the post-World War II period, specialization blossomed as physicians returning from military service were supported during residency, in those pre-Medicare GME days, by the G.I. Bill. The physician shortage was addressed by an increase in numbers of medical schools from 88 to 126 (now 125 medical schools). The increased number of schools and expansion of class size in existing schools was responsible for the increase of physician graduates from about 7,800 to the current 17,000 annually. Subsequently, for purposes of establishing more precise workforce planning and methodology, the Graduate Medical Education National Advisory Committee, (The GMENAC Report) in the 1970s developed estimates of required numbers of physicians by specialty per population at decade benchmarks [5]. Many of the target dates have passed, thus allowing the accuracy of the GMENAC predictions to be evaluated, which include the prediction of a 145,000 physician surplus by 2000 (Table 2). It is clear that many GMENAC predictions, quite reasonable at the time they were made, have not come to pass. The GMENAC study and others subsequently concluded that the post-World War II response to the physician shortage had overcompensated and resulted in an evolving physician surplus. With the GMENAC study and others predicting an excess of 145,000 physicians by year 2000, consensus has been that an overproduction of physicians, a shortage of generalists, and a surfeit of specialists were evolving. In response, the number of medical school graduates has not increased in the United States since 1980, and many specialties have had little or no growth. In addition to the number of physicians, attention has focused on the types of physicians being trained rather than on the specific services they provide. The various formulas of proportioning generalists to specialists were based on the pre-1980 impression of European health care systems, which at the time were composed of a workforce in which approximately half of the physician pool was in generalist specialties [6]. However, the evolution of the European workforce specialty ratios, functioning somewhat different than in the United States, is fairly similar to our current generalist:specialist distribution. Nevertheless, emphasis on producing a primary care workforce became a goal of medical educators. Many state legislatures established quotas of primary care requirements for their medical schools. The Council on Graduate Medical Education (COGME) founded in 1985 by the Consolidated Omnibus Budget Reconciliation Act (COBRA) has recommended that graduate medical education (GME) positions should be 110% of the United States Medical Graduates (USMG), and proportioned 50:50 generalists to specialists. Graduate Medical Education positions currently are 130% of the USMG and our historical ratio of 35% generalists to 65% specialists still exists [7]. The COGME recommendations have several flaws. Specialists are never defined. The conclusions are based on the assumption universally held over the past 20 years that a surplus of physicians and too many international medical graduates (IMG, FMG) filling our residency position exist. Moreover, The COGME recommendations were partially based on prediction of generalists to specialists requirements for a health care system expected to evolve along the organizational pattern of staff model health maintenance organizations (HMO). One of the problems of focusing on ratios of generalists to specialists is that approach only partially addresses the different skills and services provided. The approach lumps generalists and specialists into service stereotypes. While a general internist and a family practice physician provide many services in common, a urologist, pulmonologist, and a neurosurgeon have few interchangeable or overlapping skills. In terms of output from residency, a static annual workforce output of about 60 pediatric surgeons is not comparable with the average output of 7,500 graduates of internal medicine residencies. The COGME recommendations have had an impact. Efforts to increase the ratio of generalists graduating from medical schools has occurred. A solid generalist workforce has been created allowing us to claim credit for that accomplishment. Efforts to educate our legislatures that we have accomplished one part of health workforce goals, a sufficient number of generalists should be advanced. For the 21st century workforce, our legislators should be encouraged to help in developing the needed number of some specialists, and clinical scientists. While medical educators were focusing attention on the number of US medical school graduates, the number of residency positions increased to 130% of USMG. The graduates of offshore medical schools and foreign medical schools matched into the positions not filled by USMG which contributed the equivalent of about 40% more practicing physicians annually finishing GME than produced by US schools of medicine. As such, GME became more of a determinant of the practicing health workforce than USMG. Today, only 65% of the participants of the National Residency Matching Program (NRMP) annual GME match of about 20,000 positions are filled by graduates of US medical schools. International medical graduates (IMG, FMG) and US graduates of foreign medical (USIMG, USFMG) fill most of the remaining positions. As such, the number of practicing physicians in the United States who are graduates of foreign medical schools remains at about 25%. Osteopathic physicians, about 6% of the practicing physicians workforce, are increasing four times faster than MD graduates. There are even plans to open nine more Schools of Osteopathic Medicine—an increase from the current 19, recently increased from 14. Nonphysician practitioners While the medical community has focused on the balance between specialists and generalists, a quiet revolution has occurred. There has been a large increase (twice that of physicians) in the nonphysician providers (NPP), alternative medicine practitioners, and other nontraditional health care workers. Some estimate than only 10% of health care is actually provided by physicians today. The implications of a spectrum of health care providers are substantial. Blue Cross Blue Shield of Illinois considers services provided by chiropractors to be primary care. In 37 states, nonphysician practitioners can write prescriptions. Everyone is aware of the overlap between optometrists and ophthalmologists, and anesthesiologists and nurse anesthetists, but the overlap of services between general physicians and nonphysician providers has been less obvious. The focus on generalism, usually identified with the specialties of general internal medicine, family medicine, pediatrics, and often obstetrics and gynecology, has successfully provided a strong generalist workforce. That success, ironically, has been paralleled by an increase in nonphysician clinicians, who provide many of the same services. Nonphysician providers, obviously, are more able to provide primary care services, than services of specialties such as neurosurgery. Nevertheless, the overlap of services provided is an essential point to consider as the workforce product of our medical schools is examined. North Carolina, for example, has an generalist MD to population ratio of 85 per 100,000, more than the COGME recommendation of 60 to 80 per 100,000 and that of most industrialized countries. As there will always be a need for a high percentage of our workforce to be generalists, maintaining a satisfactory ratio, distribution and skills is essential. The interface with nonphysician providers is an important area to examine. Specialty-specific workforce issues Fig. 1 shows the supply of active physicians reported from 1950 to 2000 and the projected number based on current output. Internal medicine is the largest of the 24 specialties recognized by the American Board of Medical Specialties. Approximately 7,500 residents graduate from residencies in internal medicine annually, of whom two thirds take fellowships and subspecialize; gastroenterology and cardiology are the most popular subspecialties. Internal medicine continues to grow as a specialty. It has attempted to define its practice ethic as both generalist and specialist. In practice, most patients with complex illness will be referred among many medical specialists. Obstetrics and gynecology is a large, popular specialty. At least two workforce trends are operative in the specialty of obstetrics/gynecology. The first is the popularity of the field to women, who now comprise about half of the graduates of US medical schools, and approximately 75% of the first and second year residents in obstetrics/gynecology. As such, it reflects some of the unique professional issues of women in medicine. The second is that obstetrical services are also provided by nonphysician practitioners, licensed midwives; a growing professional group. In North Carolina between 1984 and 1989, the physician population increased by 18%, but the number of physicians providing obstetrical services declined from 896 to 839 [8]. During this same period, the number of physician assistants doubled, and the number of nurse midwives offering obstetric care quadrupled. In 1998, 6.1% of physicians and 15.5% of nurse practitioners treated pregnant women. More rural counties are without an MD providing obstetrical services than previously. Among the trends illustrated by these data is the growth of the NPC, the nurse midwife, the problem of geographic distribution, and the withdrawal from obstetrical services by family medicine practitioners. As national benchmarks of primary care physicians to population have been met in North Carolina, the deficiency is not of specialty, but of the type of service provided as well as geographic distribution. These data point to the need to evaluate the service, rather than the specialty. General surgery is a specialty that traditionally has been popular among students. General surgery, by contrast to internal medicine, has not grown, but has graduated the same number of trainees each year for 20 years, ie, 1,000. Like internal medicine, approximately two thirds of the general surgery graduates take fellowships, of which cardiothoracic and plastic surgery are the most popular. Some of the fellowships lead to Certificates of Added Qualification by examination of the American Board of Surgery (hand, surgical critical care, vascular, and pediatrics). Popular specialties that require or prefer a prerequisite of general surgery are thoracic surgery and plastic surgery. As with general internal medicine in the past, the specialty seems to be losing popularity, although the quality of the USMG remains high, and the number of IMG is low. A variety of explanations address the lessened popularity, including the life style issue. General surgery has not been as successful in recruiting women to the specialty as has obstetrics/gynecology, pediatrics, and so forth. Women, comprising approximately half of our medical students today, are fewer than 20% of the residents in surgery. Anesthesiology is a specialty that has been heavily impacted by managed care, workforce studies, and academic underappreciation. While perhaps the only specialty founded on a uniquely American innovation (Crawford Long in Jefferson, Georgia in 1842), paradoxically, the United States is the only Western industrialized nation where a plurality, if not a majority, of anesthetics are administered by nurses, not physicians. Anesthesiology as an academic specialty experienced growth and achieved departmental status after World War II. The evolution of expertise in pharmacology, pulmonary physiology, intensive care, and pain management became areas of specialty expertise. The specialty is well suited to large group practices and a more predictable life style than many other specialties. A study by Abt in 1994 predicted an oversupply of anesthesiologists. The Balanced Budget Act impacted anesthesiology severely, and the increase in nurse anesthetists was purported to be a less costly alternative to MD anesthesiologists. As with most specialties when reimbursement for services decreased, the recruiting of new graduates to practices decreased. Senior members of large practices deferred perquisites such as educational time, vacations, and early retirement. As a result, it became grapevine wisdom that anesthesiology was an undesirable career choice for US medical students. The CA-1 positions decreased dramatically from 1,873 in 1994 to 745 in 1996. The number of USMG entering anesthesiology dropped from 87% in 1994 to 43% in 2000. As occurs with program directors of all specialties when unfilled residency positions occur, the positions are filled with international medical graduates [9]. The 2001 NRMP match suggests a rebound of interest in anesthesiology as USMG matching into anesthesiology residency increased 5.8%. While anesthesiology is a specialty with a parallel and competing nonphysician group providing anesthesia care, it is noteworthy that to become a licensed nurse anesthetist, it is necessary first to become a nurse. The nursing profession is under even more workforce stress than medicine, and is unlikely to be a qualitative or quantitative substitute for a physician anesthesiologists. Conclusion and recommendations Although numerous, past medical workforce studies have not provided a basis for planning the number of physicians required for a healthy society. The long educational requirements to become a physician impede effective short term planning in terms of curriculum, physician numbers, and specialty types. Health planners, in the 1970s, when developing workforce criteria for the GMENAC, could not have anticipated a disease called AIDS or that coronary artery bypass would become a common treatment modality for a subset of the population with acquired heart disease. Our generation of planners can only speculate on the educational and workforce requirements for the new medicine based on scientific advances such as applied genomics, stem cells, robotic surgery, epidemiology, and the new national security threat of bioterrorism. Moreover, new diseases, such as AIDS 30 years ago do occur. New methodology of workforce projections, however, have established improved predictive criteria, such as trend analysis, than earlier studies [10]. An additional factor not considered in most workforce studies, is the economy. A prosperous and aging population will invest in health care and research. The National Health Service in the United Kingdom, less generously funded than health care in the United States, is considered by the British to be the most important legislative accomplishment of the last 50 years. Universal health insurance predictably will become a reality in the United States by incremental expansion of public programs. The number of graduate medical education positions determines the physician workforce, not the number of graduates of US medical schools. The COGME recommendation of sizing GME to the US medical school graduates is valid, but should be accomplished by increasing the medical school class positions to approximate available GME positions currently filled by US-FMG, rather than downsizing GME to current USMG. In 1995, 366 USFMG received residency positions through the NRMP. That number had increased in the 2001 match to 1,048. The USFMG now educated offshore or abroad, are usually qualified applicants and US citizens. If class sizes in our 125 medical schools were increased each by 8–10 positions, we could accommodate that group. The US-IMG educated in offshore schools or abroad become part of the US physician workforce anyway by matching into the GME positions not filled by USMG. The public and the students would both be better served if the US students now receiving medical education abroad were accommodated in undergraduate US or Canadian schools of medicine, as they are in graduate medical education. Foreign medical graduates (FMG, IMG) who are not US citizens are a continued source of personnel for residency and for our health workforce. As a generality, the majority match to the large specialties that are less competitive among USMG, or have a plethora of residency positions. In 1995, there were 2,818 matched through the NRMP, representing about one third of the FMG applicants. In the 2001 NRMP match, 2,294 successfully achieved residency positions [11]. Notable in the 2001 match was a substantial increase in successful matching into residency for USFMG and a slight decrease in the success rate for non-US FMG. An issue to examine, is whether undergraduate class size and residency positions could be more closely sized, as has been recommended by COGME, to encompass all physicians practicing in the United States, including the USFMG and non-USFMG. As it seems increasingly unlikely that a physician surplus exists, such a plan would address the need for additional physicians. Qualified graduates of foreign medical schools should still be admitted to GME, but fewer would probably be accommodated if our US citizen international medical graduates matriculated domestically. A more targeted program for international medical graduates who are foreign nationals committed to return to their country of origin could be developed. Transnational movement of health professionals has always occurred, and is likely to increase as licensure issues are addressed in the European Union. Canada, currently in a severe physician workforce shortage, is experiencing a phenomenon of 15% of their graduating physicians immigrating to the United States. As is the case in the United States, the Canadian workforce, which recently restricted international medical graduates, is now allowing increased immigration of physicians, as well as access to residencies by international medical graduates. Our schools of medicine should prepare students for a focused career and especially with services for which physicians long education uniquely qualifies them to provide. Curriculum alterations could easily be accomplished by better use of the senior year of medical school. The Graduate Professional Education of Physicians, the GPEP report of 1985, made a strong recommendation for better use of the senior year. To prepare for the workforce of the 21st century, students could be tracked in the fourth year into generalists, proceduralists, clinical scientists, health evaluation service, and a miscellaneous category. Schools of medicine and public health educational degrees and curriculum could be merged as an additional desirable alteration. The challenge to produce a medical school graduate able to apply the science and technology of the 21st century is our responsibility and challenge. The knowledge society requires a specialized workforce to implement research and technology advances [12]. While our needed generalist workforce numbers approach being met, insufficient concentration on a specialist workforce has occurred. Even our primary care specialties in the evolving healthcare system are developing more specific rather than general roles. For example, family medicine now has subspecialty certificates in geriatrics and sports medicine. With the growth of nonphysician clinicians, a clearer definition of the 21st century physician’s role is required. While some services are economically and qualitatively better provided by nonphysician clinicians, the complexity of translational modern science requires a highly specialized, scientifically oriented physician as the key health professional for the 21st century. Acknowledgements Presented as the Chair’s Report at the 112th Annual Meeting of the Association of American Medical Colleges, November 2001. See also AAMC Web site, Chair’s Report—George F. Sheldon, MD, Newsroom, press release, November 4, 2001 (available at: http://www.aamc.org). The views expressed are those of the author, not the policy of the Association of American Medical Colleges. References [1].
[1]
Drucker P.
Post capitalist society. New York: Harper Collins; 1994;. [2].
[2]
Kurzweil R.
The age of the spiritual machine. New York: Penguin Books; 1999;. [3].
[3]
Notzon FC, Komarov YM, Ermakov SP, et al.
Causes of declining life expectancy in Russia.
JAMA. 1998;279:793–800. MEDLINE |
CrossRef
[4].
[4]
Levin R, Moy E, Griner PF.
Trends in specialized surgical procedures at teaching and nonteaching hospitals.
Health Affairs. 2000;230–238. [5].
[5]
Graduate Medical Education Advisory Committee
.
Report of the Graduate Medical Education National Advisory Committee (Summary Report). Washington, DC: US Department of Health and Human Services; 1981;
. [6].
[6]
Schroeder SA.
European response to physician oversupply.
JAMA. 1983;251:373–384. [7].
[7]
Council on Graduate Medical Education (COGME)
.
Patient care physician supply and requirements (testing COGME recommendations. Eighth report). Washington, DC: Department of Health and Human Services; 1996;. [8].
[8]
Cecil G.
Sheps Center for Health Services Research and North Caro-lina Area Health Education Centers Program. A twenty year profile of trends in licensed health professions in North Carolina: 1979–1998. Chapel Hill, NC: Cecil G. Sheps Center; 2000;. [9].
[9]
Eckhout G, Schubert A.
Where have all the anesthesiologist gone? Analysis of the national anesthesia worker shortage.
Am Soc Anesthesiol Newslett. 2001;65(4):. [10].
[10]
Cooper RA.
Perspectives on the physician workforce to the year 2020.
JAMA. 1995;274:1534–1543. MEDLINE [11].
[11]
The National Residency Matching Program (NRMP). Table 2, Table 6, Table 10. Association of American Medical Schools (AAMC) website. Available at: http://www.nrmp.org/re_match/data_tables.html. Accessed September 6, 2002. [12].
[12]
Sheldon GF.
The health workforce, generalism, and the social contract.
Ann Surg. 1995;222:215–228. MEDLINE [13].
[13]
Mullan F.
The case for more medical students.
N Engl J Med. 2000;343:213–217. MEDLINE |
CrossRef
a Department of Surgery, 167 Burnett-Womack Clinical Science Bldg., University of North Carolina Medical School, CB 7050, Chapel Hill, NC 27599-7050, USA  Corresponding author. Tel.: +1-919-966-4389; fax: +1-919-966-7841.
PII: S0002-9610(02)01142-X © 2003 Excerpta Medica Inc. All rights reserved. | |
|
FULL TEXT
01142-X/journalimage?img=PIIS000296100201142X.gr1.lrg.gif&fig=FIG1&kwhquery=&issn=0002-9610&ishighres=false&allhighres=false&free=yes','journalimage');)
