Current workforce projections1,2 suggest a substantial physician shortage over the next 10 to 15 years. This month’s special workforce issue provides background and analysis of current thinking about how academic health centers (AHCs) and others can best address this shortage. Even more articles on workforce issues and future needs are available this month in Health Affairs; both journals collaborated to make this possible.
Support for the likelihood of a future physician shortage may be found in a 2008 report by Dill and Salsberg,1 which projected future physician workforce needs from a baseline of 2006 through 2025 using a physician supply-and-demand model. Their “most plausible estimate” suggested a deficit of 159,300 full-time equivalents (FTEs) by 2025. Another projection of physician workforce needs, using somewhat different assumptions, from the Health Resources and Services Administration Bureau of Health Professions,2 states that between 2005 and 2020, physician FTE requirements will grow to 976,000 from 802,100 (a 22% increase), and supply will grow from 811,800 to 926,600 (a 14% increase), leaving a deficit of about 49,000 if service models remain relatively constant. Although these estimates are somewhat different, they both project significant deficits under current conditions of care delivery. The authors of the 5 commentaries and 22 articles in this special issue of Academic Medicine generally accept these reports’ assumptions and findings of a future workforce shortage.
A Model of Workforce Dynamics
What can be done to lessen the impending shortage? I think the first step is to have a model of workforce dynamics, to help us understand the various perspectives and pressure points to which interventions can be applied and what the effects and consequences might be.
I have created the following model, in which physician workforce sufficiency is broken down into three components:
Phase 1. Production and flow of physicians and other providers into the system
Phase 2. Population demand, provider supply and efficiency, and delivery system dynamics
Phase 3. Attrition and outflow of providers through death, retirement, work reduction, or disability
In Phase 1, the production of providers would include such elements as the numbers and characteristics of medical students, international medical graduates (IMGs), advanced practice registered nurses (APRNs), physician assistants (PAs), and graduate medical education (GME) training positions, as well as the duration of training for each of the groups just mentioned. Because medical students and IMGs cannot provide medical care without residency training, the number of physicians that can be added to the workforce currently depends on the numbers and types of GME positions. Increases in the numbers of medical students or IMGs will have limited impact on the inflow of providers if there is no expansion of the GME capacity. In fact, increases in the numbers of medical students without an increase in GME positions may lead to the unfortunate paradox of partially trained doctors who cannot complete their training.
Requirements for GME are currently not necessary for APRNs or PAs, who can practice immediately on completion of their basic education. Although the shorter time frame for educating APRNs and PAs compared with educating physicians represents a significant advantage for addressing workforce deficits with these providers, this solution begs a question. Because GME is generally considered to be a critical part of the continuum of physician education, how can a training program that lacks this critical element produce independent providers who can compare favorably to physicians? The lack of such required graduate training for PAs and APRNs suggests either that assumptions about the importance of GME are incorrect or that the training of PAs and APRNs is probably not adequate for them to have sufficient skills to do the work of physicians. A process to provide adequate supervision to ensure that APRNs and PAs are competent for independent practice might be a way to address this dilemma.
Once providers complete their training, they enter Phase 2. In this phase, the demand for services by patients as well as the features of the delivery system and provider efficiency play important roles in workforce sufficiency. A variety of efficiency factors can stretch the supply of providers to meet the demand of the population for services. These efficiency factors include payment incentives for productivity,3 administrative support—including availability of clerical, nursing, and other support for case management—and information systems.4 Provider characteristics such as age and gender can influence the numbers of hours worked and lengths of careers,1 which can affect the number of providers needed over time.
In addition to the characteristics of providers and the dynamics of the delivery system, the demand for services by the population will influence the sufficiency of the workforce. As the population grows, increases its average age, and develops more chronic diseases, the demand for care will grow. Demand for services can also be affected by economic growth and the amount of money allocated to health services. In 2002, Cooper et al5 eloquently demonstrated the impact of economic growth and other efficiency factors in a workforce model that predicted a shortage of specialists. If workforce projections anticipate reductions in demand that do not materialize, the projections will underestimate the workforce needs; this occurred with Weiner’s6 projections during the time of anticipated growth of managed care in the 1990s. Management of complex patients with multiple chronic illnesses to avoid hospitalization is one current example of how more efficient management strategies can reduce population demand for services. In this issue of Academic Medicine, Dow et al7 describe how one health system identified high-risk, chronically ill patients and provided coordinated management to improve quality and reduce costs.
What are the changes that are needed in the composition of the workforce to lessen the need for more physicians? Ladden et al8 and Garson9 describe changes in the workforce that include nonphysician providers who can work as part of a team, coordinate care between specialists, and pick up some of the activities currently provided by physicians. Pershing and Fuchs10 in their commentary in this issue describe some of the changes that will be needed in medical education to prepare the workforce to deliver care in teams, with multidisciplinary collaboration to meet new delivery models such as those presented by Ladden et al and Garson.
The last part of the workforce model, Phase 3, is about the attrition of providers from their work activities. Providers leave practice, work part-time, migrate to other states and countries, become disabled, or die. The rate at which they leave or reduce their contributions is an important factor in overall workforce sufficiency and one that might be influenced by incentives. Stearns et al11 surveyed family physicians who were 50 years and older; 75% stated that they would like to remain active in teaching, while 55% would like to participate in mentoring. Physicians nearing retirement could be an important resource to address workforce shortages. According to one estimate,1 if the average age of retirement could be delayed by two years, it would reduce the estimated workforce deficit by 24%. Unfortunately, AHCs and health planners have not generally focused on this part of the workforce except to ensure quality and safety of older providers through continuing medical education programs, licensing oversight, and remediation programs.
Based on my review of the workforce articles in this issue, I have three recommendations:
First, current projections of physician workforce deficits appear to be methodologically robust and consistent. Newer trends of resident subspecialization12 may worsen the deficit as physicians narrow the spectrum of medical problems within their purview. The authors of this issue’s articles and commentaries present a variety of possible steps in all three phases of workforce development to avoid the projected workforce deficits and ensure access to quality care for our population. These include expansion of the GME capacity through lifting of current Medicare caps (although this may not be sufficient),13 shortening GME training,13 incorporating new training sites such as community health centers to increase primary care training,14 as well as changes in care delivery systems with increased use of PAs15 and APRNs9,16 and development of incentives to encourage delays in retirement of older physicians.11 The time to begin working on these and other solutions is now rather than when the deficits occur several years from now.
Second, workforce analysis provides an opportunity for collaboration between AHCs and state and regional policy makers to carry out needed research and analysis. For example, although national statistics are valuable, they mask regional variations and cannot answer questions about migration of physicians from state to state, which occurs frequently according to Ricketts.17 Workforce figures also do not tell us whether patients can actually access care regardless of numbers of providers. Only surveys of patients call tell us that. There are currently gaps in state and regional health workforce information. The collection and analysis of statewide data could help guide effective use of resources to meet workforce needs and demonstrate the value of training programs to address shortages. AHCs should consider creation of regional workforce centers with resources and expertise to provide data analysis and recommendation to state and regional leaders. AHCs should also consider the development of a comprehensive approach to workforce planning as part of their mission to include all phases of the workforce cycle, including the attrition phase.
Third, the funding of workforce research, evaluation, and policy activities is critical so that our public funds are spent wisely and effectively in the support of education of our workforce. Thibault18 reviews the contribution of foundations to workforce research in his commentary. However, considering the large current government investment in GME and the great amount of government spending for health care delivery, it would seem prudent for the government to support research on the production, education, and efficiency of the health care workforce.
The Human Face of the Workforce Dilemma
I wanted to end this editorial by giving a human face to the discussion of workforce. Every July at the hospital where I teach residents, new housestaff arrive from all over the country. On a chaotic early July evening, a new intern presented a patient to me who was short of breath because of large pleural effusions. After visiting with the patient and discussing our options, we decided to remove some of the fluid with a thoracentesis. The intern had never done the procedure before, and so I guided him through it—the informed consent, the sedation, the communications with nurses and the lab, the technical aspects of the anatomy, the equipment we would use, the insertion of needles, the follow-up of results and possible complications, and the documentation. Everything took twice as long as it would have a few weeks before with the old interns, but it went well and I watched the intern’s face as the patient thanked him for relieving her severe shortness of breath. For the first time in days, she could rest and breathe without difficulty. The intern was beaming with delight at having done something he had never done before that had made a difference for someone who was suffering. It was probably the first time he had experienced this feeling of accomplishment in his medical career. I think the moment epitomized in so many ways the rewards and challenges of academic medicine: the opportunity to be a part of the growth of our future doctors, nurses, and other health care providers while safely overseeing the care of our patients. This intern and the thousands like him are the human face of the workforce dilemma that we discuss in this issue of Academic Medicine—the need to educate and maintain enough health care providers to meet the public’s increasing health needs in a resource-constrained environment while maintaining high quality and patient safety—and I hope that we do not lose sight of our interns, our patients, and our faculty as we consider the numbers, assumptions, and formulas that often drive our workforce discussions.
Editor’s Note: The opinions expressed in this editorial do not necessarily reflect the opinions of the AAMC or its members.
David P. Sklar, MD
2. . The Physician Workforce: Projections and Research Into Current Issues Affecting Supply and Demand. December 2008 U.S. Department of Health and Human Services, Health Resources and Services Administration Bureau of Health Professions http://bhpr.hrsa.gov/healthworkforce/reports/physwfissues.pdf
. Accessed August 6, 2013
3. de Brantes FS, D’Andrea BG. Physicians respond to pay-for-performance incentives: Larger incentives yield greater participation. Am J Manag Care. 2009;15:305–310
4. Amarasingham R, Plantinga L, Diener-West M, Gaskin DJ, Powe NR. Clinical information technologies and inpatient outcomes: A multiple hospital study. Arch Intern Med. 2009;169:108–114
5. Cooper RA, Getzen TE, McKee HJ, Laud P. Economic and demographic trends signal an impending physician shortage. Health Aff (Millwood). 2002;21:140–154
6. Weiner JP. Forecasting the effects of health reform on the U.S. physician workforce requirements: Evidence from HMO staffing patterns. JAMA. 1994;272:222–230
7. Dow AW, Bohannon A, Garland S, Mazmanian PE, Retchin SM. The effects of expanding primary care access for the uninsured: Implications for the health care workforce under health reform. Acad Med. 2013;12:1855–1861
8. Ladden MD, Bodenheimer T, Fishman NW, et al. The emerging primary care workforce: Preliminary observations from The Primary Care Team: Learning from Effective Ambulatory Practices Project. Acad Med. 2013;88:1830–1834
9. Garson A. New systems of care can leverage the health care workforce: How many doctors do we really need? Acad Med. 2013;88:1817–1821
10. Pershing S, Fuchs VR. Restructuring medical education to meet current and future health care needs. Acad Med. 2013;88:1798–1801
11. Stearns J, Everard KM, Gjerde CL, Stearns M, Shore W. Understanding the needs and concerns of senior faculty in academic medicine: Building strategies to maintain this critical resource. Acad Med. 2013;88:1927–1933
12. Jolly P, Erikson C, Garrison G. U.S. graduate medical education and physician specialty choice. Acad Med. 2013;88:468–474
13. Whitcomb ME. Decreasing the length of residency training: A public policy perspective. Acad Med. 2013;88:1802–1803
14. Rieselbach RE, Crouse BJ, Neuhausen K, Nasca TJ, Frohna JG. Academic medicine: A key partner in strengthening the primary care infrastructure via teaching health centers. Acad Med. 2013;88:1835–1843
15. Glicken AD, Miller AA. Physician assistants: From pipeline to practice. Acad Med. 2013;88:1883–1889
16. Grover A, Niecko-Najjum LM. Primary care teams: Are we there yet? Implications for workforce planning. Acad Med. 2013;88:1827–1829
17. Ricketts TC. The migration of physicians and the local supply of practitioners: A five-year comparison. Acad Med. 2013;88:1913–1918
18. Thibault GE. The role of private foundations in addressing health care workforce needs. Acad Med. 2013;88:1804–1805