The Association of American Medical Colleges (AAMC) predicts an estimated shortage of more than 91,500 physicians in the United States by 2020, including 45,400 primary care physicians and 46,100 medical and surgical specialists.1 Up to 32 million previously uninsured Americans will gain insurance coverage in 2014, many who need specialty and surgical care to remedy previously untreated conditions.2 In addition, every day for the next 19 years, 10,000 Americans will turn 65, doubling the size of the population that uses the most clinical services. The disease burden of cardiovascular disease, diabetes, and cancer will increase. The number of cancer survivors will surge from 14 million today to 18 million by 2022. Once “cured,” these individuals will need ongoing care.3
Currently, physician workforce projections are based on a future health care system that is either a version of one’s “ideal” system or a modified version of the existing one. Even if researchers, policy makers, and stakeholders could agree on the ideal system of care, workforce planning cannot neglect the growing demand for health care services in the current system. The urgency of patients’ increasing health care needs combined with the long training periods required for medical professionals call for prudent action, even while attempting a complete makeover of the U.S. health care system.
In this article, we argue that workforce planning should begin with the current system and make adjustments based on empirical data that accurately reflect emerging trends. Through a historical review, we show that an approach that concentrates on creating only a “needs based” or “ideal system” causes long-term workforce challenges that take years to overcome.
The Cyclical Nature of Workforce Assumptions and Projections
Historically, projections of physician oversupply have been predicated on assumptions that the health care delivery system would change significantly and rapidly, and each of these projections has been followed by warnings of shortages. In reality, the number of physicians educated and trained (and other adaptations in the marketplace) has changed with the projections, which tends to render any projection useful only as a short-term assessment of what will occur without changes to the system. Each projection cycle has lasted approximately 20 years, which is the amount of time it can take both to train a generation of physicians and to see future delivery system implementations become a reality. In short, past needs-based projections have failed.
In 1910, Abraham Flexner4 suggested that the United States had too many physicians, too much “specialism,” and low-quality care. In 1932, the Rappleye Report by the AAMC’s Commission on Medical Education found that the United States faced a surplus of physicians, based on the population needs assessed at the time.5,6 A drastic reduction in the number of medical schools ensued, and, consequently, the per capita supply of physicians fell from 175 to 125 per 100,000 persons in the span of three decades.7
By 1958 and 1959, the Bayne-Jones and Bane Reports, respectively, included the findings of their needs-based assessments. These reports predicted a growing shortage of physicians (40,000 by 1975) and called for medical schools to increase enrollment.8,9 As a result, the federal government subsidized the expansion of medical school enrollment and allowed more international medical graduates to immigrate to the United States.10 Several other reports reinforced this notion that a physician shortage loomed, including the Coggeshall Report to the AAMC in 1965 and the president’s National Advisory Committee on Health Manpower in 1967. By 1968, both the AAMC and American Medical Association (AMA) had committed to increasing the number of medical schools. By 1982, the number of schools had grown to 127 (from 89 in 1972), and the number of graduates had doubled.
In 1980, just as federal subsidies elapsed, the Graduate Medical Education National Advisory Committee (GMENAC) calculated the number of physicians needed to provide predetermined “necessary and appropriate” services and reported a needs-based physician surplus of 70,000 by 2000.11 GMENAC recommended that the United States limit medical school enrollment and severely restrict the number of international medical graduates entering the United States.11
By the 1990s, with the promotion of managed care, many groups predicted a physician surplus, including the national Council on Graduate Medical Education (COGME), the National Academy of Science’s Institute of Medicine, the Pew Health Professions Commission, the AAMC, the AMA, and other national physician associations. These organizations based their projections on the assumption that health maintenance organizations (HMOs) would lead to an even greater surplus of physicians than GMENAC predicted in the 1980s,12 especially of specialists.13 COGME published several reports between 1992 and 1998 to reaffirm its concern about this predicted physician surplus.14–19
Additionally, in agreement with the prevailing belief that the U.S. health care system would drastically shift towards tightly managed care, Congress passed the Budget Control Act of 1997. This legislation capped the number of residency slots supported by Medicare at the then-current level, which had begun to plateau since 1994 at the height of the managed care era.20 The 1997 Consensus Statement on Physician Workforce, signed by the AAMC and five other leading medical societies, committed to reducing the number of entry-level residency positions to 110% of the number of U.S. medical school graduates, while keeping the proportion of generalists and specialists the same.21 Though many of those organizations have reversed their projections, the United States now finds itself perilously close to filling the 110% residency positions—2,706 U.S. citizens matched into a residency position after attending a foreign medical school, leaving only 3,601 of the 26,392 entry-level residency positions in 2013 available to non-U.S. citizens or permanent residents.20
Despite this decrease in the number of physicians trained in the 1990s, the number of physicians in the United States grew by more than 320,000 between 1980 and 2000; yet, no significant surplus of physicians was reported.12 Some researchers attributed this absorption of capacity to supply-induced demand, but less than 2% of the health care services provided have been shown to be attributable to physician-induced demand.22 By the late 2000s, only a decade after anticipating a physician surplus, a growing number of reports projected physician shortages in both primary care and specialty practices.23,24 In fact, the Organization for Economic Cooperation and Development (OECD) reported the U.S. per capita physician and hospital bed supply to be lower than most countries belonging to OECD.25
In health care, demand typically refers to the population’s desire for services that leads to use. Demand or “use-based” workforce projections estimate the number of health professionals required to provide the level of health services that will be used by a given population. This approach focuses on reality at a given point in time; however, it assumes equilibrium of physician supply and demand at that point, ignoring any existing shortages or surpluses of providers. It also assumes that current use patterns will continue into the foreseeable future.6
On the other hand, the needs-based, or the “ideal system,” projections are predicated on subjective determinations of a population’s need for physician services by panels of experts, who delineate “the medically modifiable morbidity (illness) burden of a defined population.”6 The problem with this approach is threefold: (1) the “need” is defined subjectively (often while counting unnecessary services but not errors of omission), (2) the “need” does not reflect “demand” or use, and (3) the mechanism for implementing a needs-based workforce projection is often a subjective version of the ideal health care system that generally does not exist.6
From past experiences with projections, we have learned that prudent workforce planning requires a realistic view of the future, including fewer expectations of radical system change. Systemic changes are complex, difficult to achieve, and take time. Planning solely for an ideal system that could take years to implement—or not—may cause long-term workforce challenges to meeting health care demand.
Current Changes to Our Health Care System
Under the Patient Protection and Affordable Care Act (ACA), prevention, access and care coordination, and cost control are top health policy priorities.26 The transformation underway will be necessary to help offset the predicted physician shortages,6 including the ongoing efforts of the Center for Medicare and Medicaid Innovation. These efforts include an initiative by 27 states to link Medicaid payments with patient-centered medical homes (PCMHs), and the move by the private insurance industry to support new delivery models.27
How well these initiatives will succeed in achieving long-term care delivery change is unclear. For example, recently, 9 of 32 accountable care organizations (ACOs) left the pioneer program.28 In addition, we have seen that Americans often reject the efforts of entities such as HMOs to decide or define their health care needs. Finally, patients in the United States want to see their health care provider when they are ill, rather than for preventive care or health maintenance, irrespective of cost.29
Considerations for Current Workforce Projections
Today, workforce projections that do not assume radical changes in the U.S. health care system continue to suggest a physician shortage.1 The demand for both primary and subspecialty care, including pediatric subspecialties, surgery, oncology, and neurology, is on the rise.30–34 In addition, emergency departments have had difficulty finding surgical staff, and wait times to see endocrinologists have been increasing, ranging from days to months.35,36 Major drivers of this specialty care use in the United States are our aging population and our increasingly unhealthy behaviors.
Some researchers, however, have suggested that variations in physician supply are not correlated with population health and are indicative of areas with an oversupply of physicians.37 Yet, simple ratios fail to capture the health status or health needs of the population in different regions of the country.38 The number of active physicians may represent drastically different clinical full-time equivalent (FTE) counts; for example, active physicians in Massachusetts do not represent the same number of clinical FTEs as those in states with fewer academic medical centers, such as South Dakota or Iowa.
In 2010, Americans over the age of 65 accounted for a disproportionate share of ambulatory care visits, hospitalizations, procedures, and high-intensity services (see Figure 1).30 The variation in age-specific cancer care also illustrates the increasing complexity of ailments among the elderly—146 new cancers per year per 100,000 Americans between the ages of 40 and 44; 2,806 new cancers per year per 100,000 Americans between the ages of 70 and 74.39 As the overall number of elderly Americans has grown, and medical research has improved survival, the number of U.S. cancer survivors has quadrupled to nearly 14 million people since the 1970s and is expected to reach 18 million by 2022.40
This good news speaks to the medical accomplishments of the U.S. health care system and biomedical research, but it also indicates that more care for this population will be required. For instance, cancer treatments may produce long-term physical and psychological side effects as patients live longer and continue to use services.41 In addition, although only 2% of children today are affected by childhood cancer—which is generally among the most treatable with an overall survival rate exceeding 80%42—those who transition to adult hood face a wide array of “late” side effects after chemotherapy and radia tion: heart problems, osteoporosis, nerve damage, early menopause, inferti lity, leukemia, and the recurrence of cancer.43,44 Treatments for these and other as yet unidentified long-term side effects require the use of additional high-level health services, and survivors need ongoing care.
Moreover, behaviors such as sedentary lifestyles, poor diets, and smoking have increased the rate of chronic disease. Although prevention, primary care, and better habits may eventually improve the health status of Americans, current projections suggest that today’s U.S. population will be more obese and unhealthy than any prior generation.45
Finally, under the ACA, up to 32 million more Americans will have access to health care starting in 2014. These previously uninsured individuals will have unmet health care needs, and primary care visits will identify requisite specialty care treatments.2,30 Together, all of these considerations contribute to demand-based projections indicating a continued increase in the use of health services as the population increases, ages, and is given more access to health care insurance coverage.
Considerations for Future Workforce Projections
Overall, the “ideal system” workforce planning assumptions about specific major changes to health care in the United States have been wrong, with the resulting projections in stark contrast to reality. Today, the increasing success of medicine has allowed people to live longer, although with more chronic comorbidities. In addition, the high burden of disease in the United States is increasingly related to individuals’ behaviors including poor diet and sedentary lifestyles. Despite attempts to encourage healthy living, it would be unrealistic to expect immediate results. Efforts to change the health care delivery system are necessary, but changes will take decades to implement.
In 2010, the United States had approximately 215.1 active patient care physicians per 100,000 people. Yet, even as individual health needs grow, the number of physicians per capita is decreasing.46 Four in 10 practicing physicians are over 55 years old and expected to retire in the next 10 to 15 years.47 Younger physicians seek a better life–work balance, further reducing the number of FTEs available to care for patients.48 In fact, most of today’s physician workforce projections likely overestimate the effective supply of future physicians.
Many more Americans will be insured under the ACA beginning in 2014. With data collected by PCMHs and ACOs, workforce planning will be better informed about actual health care use patterns and better able to accommodate actual delivery system changes. We argue, therefore, that training 4,000 more physi cians per year than the current trend will prepare us to meet the anticipated increased demand for services.49 At the same time, we must continue to assess population growth, regional and state-specific needs, and the evolving changes to health care delivery systems, including scopes of practice.
Prudent workforce planning requires a balance between preparing for the ideal system and accommodating current health care use patterns, particularly because of the protracted physician training. No single approach will solve the looming physician shortage entirely. From our experience, though, the danger of planning for an ideal health care system alone is being unprepared for the actual needs of the population.
Acknowledgments: The authors wish to thank Tannaz Rasouli for her insight and review.
2. Hanchate AD, Lasser KE, Kapoor A, et al. Massachusetts reform and disparities in inpatient care utilization. Med Care. 2012;50:569–577
3. de Moor JS, Mariotto AB, Parry C, et al. Cancer survivors in the United States: Prevalence across the survivorship trajectory and implications for care. Cancer Epidemiol Biomarkers Prev. 2013;22:561–570
5. Rappleye WC Medical Education: Final Report of the Commission on Medical Education. 1932 New York, NY Association of American Medical Colleges
6. Dill MJ, Salsberg ES The Complexities of Physician Supply and Demand: Projections Through 2025. 2008 Washington, DC Association of American Medical Colleges Center for Workforce Studies
7. Grover A. Critical care workforce: A policy perspective. Crit Care Med. 2006;34(3 suppl):S7–S11
8. U.S. Department of Health, Education, and Welfare, Office of the Secretary. The Advancement of Medical Research and Education Through the Department of Health, Education, and Welfare; Final Report of the Secretary’s Consultants on Medical Research and Education. 1958 Washington, DC US Government Printing Office
9. Bane F. chair; Surgeon General’s Consultant Group on Medical Education. Physicians for a Growing America. 1959 Washington, DC U.S. Government Printing Office
10. Health Professions Educational Assistance Act of 1963. Pub. L. 88-129, Sept. 24, 1963, 77 Stat. 164.
11. Summary Report of the Graduate Medical Education National Advisory Committee to the Secretary, Department of Health and Human Services. 1981 Washington, DC U.S Department of Health and Human Services, Public Health Service, Health Resources Administration, Office of Graduate Medical Education . Pub. no. 81-651
12. Weiner JP. Forecasting the effects of health reform on US physician workforce requirement. Evidence from HMO staffing patterns. JAMA. 1994;272:222–230
13. Salsberg ES, Forte GJ. Trends in the physician workforce, 1980–2000. Health Aff (Millwood). 2002;21:165–173
14. Council on Graduate Medical Education. Improving Access to Health Care Through Physician Workforce Reform: Directions for the Twenty-First Century, Third Report. 1992 Rockville, Md Council on Graduate Medical Education/Health Resources and Services Administration
15. Council on Graduate Medical Education. Recommendations to Improve Access to Health Care Through Physician Workforce Reform, Fourth Report. 1994 Rockville, Md Council on Graduate Medical Education/Health Resources and Services Administration
16. Council on Graduate Medical Education. Managed Health Care: Implications for the Physician Workforce and Medical Education, Sixth Report. 1995 Rockville, Md Council on Graduate Medical Education/Health Resources and Services Administration
17. Council on Graduate Medical Education. Physician Workforce Funding Recommendations for Department of Health and Human Services’ Programs, Seventh Report. 1995 Rockville, Md Council on Graduate Medical Education/Health Resources and Services Administration
18. Council on Graduate Medical Education. Patient Care Physician Supply and Requirements: Testing COGME Recommendations, Eighth Report. 1996 Rockville, Md Council on Graduate Medical Education/Health Resources and Services Administration
19. Council on Graduate Medical Education. International Medical Graduates, the Physician Workforce, and GME Payment Reform, Eleventh Report. 1998 Rockville, Md Council on Graduate Medical Education/Health Resources and Services Administration
21. American Association of Colleges of Osteopathic Medicine; American Medical Association; American Osteopathic Association; Association of Academic Health Centers; Association of American Medical Colleges; National Medical Association. Consensus Statement on Physician Workforce. Advisory #97-9, February 28, 1997. 1997 Washington, DC Association of American Medical Colleges
22. Reschovsky JD, Hadley J, Saiontz-Martinez CB, Boukus ER. Following the money: Factors associated with the cost of treating high-cost Medicare beneficiaries. Health Serv Res. 2011;46:997–1021
24. Center for Workforce Studies. Association of American Medical Colleges. Recent Studies and Reports on Physician Shortages in the US. 2012 https://www.aamc.org/download/100598/data/
. Accessed September 3, 2013
27. Erikson C. Annual address on the state of the physician workforce. 2012 San Francisco, Calif Presented at: 2012 AAMC Annual Meeting
29. Sommers R, Goold SD, McGlynn EA, Pearson SD, Danis M. Focus groups highlight that many patients object to clinicians’ focusing on costs. Health Aff (Millwood). 2013;32:338–346
30. U.S. Department of Health and Human Services, Health Resources and Services Administration, Bureau of Health Professions. . The Physician Workforce: Projections and Research into Current Issues Affecting Supply and Demand. 2008 http://bhpr.hrsa.gov/healthworkforce/reports/physwfissues.pdf
. Accessed August 14, 2013
31. U.S. Department of Health and Human Services, Health Resources and Services Administration. Recommendations for Improving Access to Pediatric Subspecialty Care Through the Medical Home. 2007 Washington, DC Health Resources and Services Administration
32. Lynge DC, Larson EH, Thompson MJ, Rosenblatt RA, Hart LG. A longitudinal analysis of the general surgery workforce in the United States, 1981–2005. Arch Surg. 2008;143:345–350
33. Erikson C, Salsberg E, Forte G, Bruinooge S, Goldstein M. Future supply and demand for oncologists: Challenges to assuring access to oncology services. J Oncol Pract. 2007;3:79–86
34. Gottfried ON, Rovit RL, Popp AJ, Kraus KL, Simon AS, Couldwell WT. Neurosurgical workforce trends in the United States. J Neurosurg. 2005;102:202–208
35. Rao MB, Lerro C, Gross CP. The shortage of on-call surgical specialist coverage: A national survey of emergency department directors. Acad Emerg Med. 2010;17:1374–1382
36. Rizza RA, Vigersky RA, Rodbard HW, et al. A model to determine workforce needs for endocrinologists in the United States until 2020. Endocr Pract. 2003;9:210–219
37. Goodman DC, Grumbach K. Does having more physicians lead to better health system performance? JAMA. 2008;299:335–337
38. Institute of Medicine. Variation in Health Care Spending: Target Decision Making, Not Geography. 2013 Washington, DC National Academies Press
39. Salsberg E, Grover A. Physician workforce shortages: Implications and issues for academic health centers and policymakers. Acad Med. 2006;81:782–787
40. Siegel R, DeSantis C, Virgo K, et al. Cancer treatment and survivorship statistics, 2012. CA Cancer J Clin. 2012;62:220–241
41. Deimling GT, Bowman KF, Sterns S, Wagner LJ, Kahana B. Cancer-related health worries and psychological distress among older adult, long-term cancer survivors. Psychooncology. 2006;15:306–320
43. Casagranda L, Trombert-Paviot B, Faure-Conter C, Bertrand Y, Plantaz D, Berger C. Self-reported and record-collected late effects in long-term survivors of childhood cancer: A population-based cohort study of the childhood cancer registry of the Rhône-Alpes region (ARCERRA). Pediatr Hematol Oncol. 2013;30:195–207
44. Jacobs LA, Pucci DA. Adult survivors of childhood cancer: The medical and psychosocial late effects of cancer treatment and the impact on sexual and reproductive health. J Sex Med. 2013;10(suppl 1):120–126
46. Bodenheimer T, Chen E, Bennett HD. Confronting the growing burden of chronic disease: Can the U.S. health care workforce do the job? Health Aff (Millwood). 2009;28:64–74
48. Staiger DO, Auerbach DI, Buerhaus PI. Trends in the work hours of physicians in the United States. JAMA. 2010;303:747–753