The critical state of graduate medical education funding

Medicare provides insurance coverage to elderly and disabled Americans and it also supports graduate medical education (GME). In 2009, Medicare paid $9.5 billion to teaching hospitals for resident training—$3 billion to cover direct costs of approximately 100,000 residency positions and $6.5 billion for the indirect costs of patient care associated with resident training.1


GME – Graduate Medical Education
AAMC – American Association of Medical Colleges
ACA – Affordable Care Act
IME – Indirect Medical Education
DGME – Direct Graduate Medical Education
BBA – Balanced Budget Act
COGME – Council on Graduate Medical Education
DHHS – Department of Health and Human Services
MedPAC – Medicare Payment Advisory Group
NCFRR – National Commission on Fiscal Responsibility and Reform
CBO – Congressional Budget Office
IOM – Institutes of Medicine
RAND – Research and Development Corporation

Given the current budget constraints and economic recession, federal financial support for GME is under greater scrutiny, and in the past five years, legislators have sought to reduce GME funding.2 Last year, the Joint Select Committee on Deficit Reduction proposed GME budget cuts of 50 percent in early versions of the Budget Control Act of 2011. Although these drastic cuts were omitted from the final bill, the automatic funding reductions scheduled to occur will result in across-the-board cuts. Medicare, biomedical research, and other health care expenditures, including GME, are expected to see a 2 percent cut due to the sequestration.3 Deficit reduction is still a high-acuity political goal, and the potential for further cuts to GME remains a concern.

Compounding the tension about GME funding is the growing shortage of physicians, particularly in primary care and general surgery. The American Association of Medical Colleges (AAMC) projects a physician shortage of 62,900 physicians in the U.S. by 2015 due to the increased medical care needs of an aging population and a growing number of people who will be insured under the Affordable Care Act of 2010 (ACA).4 Without an increase in the number of training positions and funding for GME, many medical graduates will be unable to complete the training required to practice independently and therefore will not be in a position to meet the expanding health care demands of the U.S. population.4

Current funding streams

Before Medicare, hospitals funded GME. During Medicare’s implementation, legislators believed that society at-large would eventually find other means to bear the costs of GME.5 Despite attempts to establish long-term alternative sources of support for more than a decade, no policy has significantly addressed Medicare funding for GME, and Medicare remains the primary formal financier of these programs, contributing 72 percent of all tax-financed support. Other federal payors include Medicaid (11 percent), the U.S. Department of Veterans Affairs (10 percent), the U.S. Department of Defense (3 percent), and the Bureau of Health Professions (3 percent).6 State and local governments also finance GME programs, but specific amounts vary widely.

Teaching institutions may fund GME activities and infrastructure through various informal sources as well. Tracking GME financiers is difficult because educational infrastructure is often paid through the teaching hospital’s general revenues or grants for research, not funds specifically designated for education.7 It is through this general revenue stream that private insurers provide unofficial, indirect support of GME through individually negotiated payments to teaching hospitals.


The federal government, primarily through Medicare, subsidizes training programs through direct and indirect payment methods.8 In 2009, Medicare allocated approximately $3 billion in direct graduate medical education (DGME) payments and $6.5 billion in indirect medical education (IME) payments, averaging out to more than $100,000 per resident per year.7 Both DGME and IME payments are hospital-specific, based on the institution’s share of Medicare patients and the resident-to-bed ratio as a measure of teaching intensity.8 See Table 1 for a summary of DGME and IME funding streams.

Table 1. Explanation of Medicare’s direct and indirect funding streams for GME8

Recipient Residents and residency programs Teaching hospitals
Description Covers resident stipends and fringe benefitsPays salaries of supervising facultySubsidizes educational overhead costs Adds on to Medicare’s prospective paymentsPaid directly into hospital general revenuesSubsidizes the capital costs and inefficiencies of running educational programs
Calculation A hospital-specific per resident payment applied to Medicare’s share of inpatient days A logarithmic formula which results in a 5.5% higher Medicare payment for every 0.1% increase in the resident-to-bed ratio
Amount $3 billion in 2009 $6.5 billion in 2009

Under Medicare, GME has been viewed as a “public good” deserving billions of dollars from state and federal public funds.5 However, some analysts claim that Medicare distributes this funding with insufficient accountability for the proportion and quality of medical specialists produced.1,9 Relatively little information is available about what it truly costs participating hospitals to train residents or where the funds are specifically directed in their organization.2 Without accountability, teaching hospitals have been primarily focused on their individual workforce needs and more profitable specialties.9 Structured as it is, the current system of funding does not incentivize programs to train physicians for broader public interests or evaluate meaningful outcomes of their graduates.9

The number of GME-funded positions has been stable since 1997, when it was capped by the Balanced Budget Act (BBA).10 The cap was included because organizations at the time predicted an oversupply of physicians, so they wanted to limit spending and align the number of GME positions to the number of U.S. medical graduates. Further funding modifications were made in 1999, 2000, and 2003 to reduce the IME payment to its current factor (see Table 1).


Although there are no federal requirements that Medicaid programs contribute to GME, it remains the second largest funder of these programs.6 Most Medicaid programs have appropriated funding for GME with direct and indirect payments structured similarly to Medicare.11 Medicaid explicitly paid an estimated $3.78 billion for GME programs in 2009.12

With no requirement for states to provide for GME, recent economic instability and budget constraints have led to a significant reduction in the number of states making Medicaid payments to GME programs. In 2005, a total of 47 states provided GME support of $3.18 billion through Medicaid, representing 6.6 percent of the program’s inpatient hospital expenditures.12 By 2011, Arizona, Massachusetts, Montana, Rhode Island, Vermont, and Wyoming stopped making payments to GME.2 Nine additional states—Michigan, Missouri, Nebraska, Nevada, New Hampshire, New Mexico, Oklahoma, Oregon, and Pennsylvania—have considered ending Medicaid payments to GME. Many others, including Florida and Washington, have decreased funding in the last few years. Based on a 50-state survey, the AAMC expects these cuts in GME funding to continue as states face ongoing fiscal pressures.12

Private insurers

Private insurers support GME through higher payments negotiated with teaching hospitals; however, the actual amount is difficult to calculate, as the proportion of these payments that is attributed to education is not specifically identified.7 Due to numerous private contracts and the respective bargaining power of providers and private insurers, these contributions are highly variable. Private insurers are expected to cover the proportion of GME for their own patients; however, no policy has mandated funding from the private sector. The financing from private insurers has no connection with the amount of work residents do for insurers’ beneficiaries because residents do not charge for services. A study at one teaching hospital estimated that the amount of services residents provide to privately insured patients would have yielded $232,726 of revenue annually.13

Predicted shortages

In the past decade, 62 reports have identified physician shortages in underserved areas and in many specialties.2 The AAMC estimates that the overall deficit of physicians will reach 62,900 by 2015, of which 29,800 will be in primary care.1 The predicted shortage will result from the GME system’s inability to train enough residents to keep up with the rate of retiring physicians or meet the growing demand for health care access. It is important to keep in mind that not only is the U.S. population continuing to expand, but the ACA now guarantees insurance for every American, and some experts doubt that the current system can cope with the influx of newly insured patients.

Shortfalls in the workforce have nothing to do with waning interest in meeting the nation’s growing health care needs. The number of U.S. medical school graduates has continued to rise due to increasing enrollment in medical schools, expanding satellite programs, and newly founded medical schools. However, if Medicare’s cap on GME funding remains locked at approximately 100,000 positions, graduating physicians will be unable to attain the training needed to practice independently. The influx of approximately 7,000 international medical graduates each year intensifies the competition for limited residency positions. It has been estimated that domestic production of medical school graduates alone will functionally surpass the number of GME positions by 2015.2

In the recent debate over health care reform, Democrats, with the support of many medical organizations, proposed to lift the cap and increase the number of Medicare-funded GME positions by 15 percent. However, the American Academy of Family Physicians opposed the amendment because they believed that any additional GME spots should be used to train primary care physicians. As a result, Congress decided not to increase Medicare support but instead redistribute about 900 unused spots to programs reserved for primary care.5 The decision to retain the cap was based, in part, on the fact that teaching hospitals had been able to create about 8,000 new training positions since the limit was imposed. However, the growth of residency spots has been slow—approximately 0.9 percent per year over the past decade—and most of these positions were subspecialty training fellowships rather than in primary care.2 In the current deficit-focused political environment, the $15 billion price tag for the 15,000 proposed positions contributed significantly to the political opposition to lifting the cap. Experts believe that maintaining current GME support from Medicare and seeking alternative funding sources are likely the best outcome GME could achieve at this time.2

Recommendations for reform

Several independent governmental and nongovernmental organizations have issued recommendations about how the federal government, through Congress, should restructure Medicare’s contribution to GME to lower costs and encourage rational strategies in training the physician workforce.

The U.S. Department of Health and Human Services’ (HHS) Council on Graduate Medical Education (COGME) is responsible for “an ongoing assessment of physician workforce trends, training issues and financing policies,” and recommends “appropriate federal and private sector efforts to address identified needs.”14 COGME is composed of 17 representatives from primary and specialty practices, student associations, teaching hospitals, health insurers, and businesses. In December 2010, COGME focused on primary care and recommended that GME payment and accreditation policies be restructured to produce “a physician workforce that is at least 40 percent primary care” by encouraging outpatient training, specifying GME positions for primary care, and maintaining Medicare funding for primary care residency programs.15 It also briefly stated that funding for GME should involve both governmental and non-governmental sources, but did not specify how this goal should be achieved.15

The Medicare Payment Advisory Commission (MedPAC) advises Congress on overall Medicare spending. From 2007 to 2009, the 17 members of MedPAC recommended that Congress reduce Medicare’s IME payments after an analysis found that only 45 percent of the IME payments can be analytically justified to cover the higher costs of Medicare inpatient care.8 In June 2010, MedPAC unanimously voted to recommend cutting $3.5 billion in annual IME payments by reducing the IME payment calculation from 5.5 to 2.2 percent. The commission explicitly expressed concern about the physician workforce mix and recommended that education and training programs focus on incorporation of evidence-based medicine, team-based care, and shared decision making.1 To encourage action toward these goals, MedPAC recommended: (1) increasing accountability and pay for performance; (2) public disclosure of Medicare payment and teaching costs; and (3) analysis of workforce data.

In response, the House Energy and Commerce Subcommittee on Health convened a hearing and questioned the feasibility of redirecting IME payments when hospital budgets already are stretched thin and demands for quality of care are on the rise. The AAMC determined that the MedPAC reductions in IME would have resulted in a loss of 72,600 jobs and $653 million in state and local revenues, costing the U.S. economy a total of $10.9 billion.16 Congress did not pursue the reduction in IME payments.

The bipartisan National Commission on Fiscal Responsibility and Reform (NCFRR), created in 2010, issued a report that called for bringing Medicare’s GME payments in line with the costs of medical education by limiting hospitals’ DGME payments to 120 percent of the national average salary paid to residents in 2010. In the report, the commission also concurred with MedPAC that the IME payments should be reduced to 2.2 percent. The proposal fell short of the 14 votes needed for formal endorsement and House and Senate consideration.17 According to MedPAC, the NCFRR proposal would have cut federal expenditures in GME by $6 billion by 2015 and $60 billion by 2020.18

In March 2011, the U.S. Congressional Budget Office (CBO) analyzed a proposal that consolidated all GME funding streams into one direct payment to teaching hospitals and reduced the indirect portion of funding by more than half.19 The CBO projected that this move would save $69 billion over 10 years. However, the CBO also noted that this proposal would result in the following: lower compensation for residents, IME payments growing more slowly than inflation, fewer education-directed activities, and less care for the uninsured. States would also lose discretion over the portion of GME that previously came from Medicaid.19 Ultimately, Congress did not pursue this proposal, to the relief of many in the health care community.

The Institute of Medicine (IOM) played an important role in GME policy debates with its influential 2001 report, Crossing the Quality Chasm, which provided a vision for GME by addressing workforce, compensation schemes, quality, safety, and responsiveness of the health care system.20

In December 2011, seven senators requested that the IOM study the governance and financing of GME to address the significant concern of health care.21 The letter notes the inadequacy of medical training to meet the nation’s medical needs, and the need for high-quality, low-cost health care. It calls explicit attention to the following issues: accreditation, reimbursement, workforce supply, geographic distribution of physicians, care of the underserved, access, and maintenance of an appropriately skilled workforce. The Macy Foundation recently awarded the IOM $750,000 to research these problems.22

Unsuccessful proposals for GME reform

Two bills were proposed in 2001 that attempted to solicit more consistent and equitable contributions to GME programs from private payors. The All-Payer Graduate Medical Education Act was introduced by Rep. Ben Cardin (D-MD) and would have established a trust fund for private payors to contribute a 1 percent assessment of private insurance premiums.23 These contributions, estimated to total approximately $4 billion, would then be used to make DGME and IME payments. Medicare would continue its DGME payments; however, the IME payment add-on ratio would be determined by the proportion of Medicare revenues to total revenues instead of the proportion of Medicare inpatient days. The inclusion of private payor GME payments would have reduced Medicare’s IME payment factor from 5.5 to 4.8 percent. A similar bill, the Medical Education Trust Fund Act, was introduced by Sens. Jack Reed (D-RI) and Hillary Clinton (D-NY). Their proposal had all private payors contribute a 1.5 percent assessment on premiums to a trust fund. Medical schools and teaching hospitals would apply for these funds through the Secretary of HHS.23 Health insurers strongly opposed both of these bills, which ultimately failed in Congress.

The original ACA legislation called for an additional $230 million in funding to support GME training of primary care physicians at community health centers. However, on May 25, 2010, the House voted 234 to 185 to eliminate this additional funding despite numerous reports projecting an increased demand for more health care providers.24 Ultimately, the ACA succeeded in establishing the Primary Care Residency Expansion Program, which provides $80,000 for resident positions designated for primary care, even if the number of total positions exceeds the training program’s GME cap.3 The ACA also increased funding for the National Health Service Corps to $1.15 billion and allowed residency programs to count outpatient training experiences toward GME payments.15

In response to growing national debt and political disagreement about whether to extend the U.S. debt ceiling, the Budget Control Act of 2011 established a Joint Select “Super” Committee on Deficit Reduction, which was intended to devise a bipartisan solution to balance the federal budget. Cuts to Medicare GME funding were in early versions of the Budget Control Act, but were ultimately dropped. Due to the inability of the Joint Select Committee to reach a consensus on budget cuts by November 23, 2011, an automatic sequestration will, by default, result in a $1.2 trillion cut in federal discretionary spending over 10 years.4 Medicare, research, medical education, and other health care expenditures, including GME, are expected to see a 2 percent cut as a result of the sequestration.3 However, with previously proposed GME budget cuts of 50 percent receiving serious consideration from federal legislators, the potential for further cuts to GME remains a real concern.25

In an effort to reduce federal expenditures, the Obama Administration’s 2012 budget proposal eliminated annual funding of $317 million, which had been earmarked to support pediatric GME training.26 The justification for this proposal centered on the view that dedicated children’s hospitals should not receive Medicare funding because children do not qualify for Medicare coverage. However, a bipartisan vote by the House Energy and Commerce Committee extended the pediatric GME program for five more years.2 In the 2013 budget proposal, the Obama Administration again pushed to remove support for pediatric GME training, this time proposing to cut $88 million. Furthermore, the 2013 budget proposal called for cuts to Medicare’s IME payments at all hospitals, advocating a reduction of $9.7 billion over 10 years and encouraging more HHS oversight.22

A way forward for GME reform

The IME component of GME funding is based on the assumption that the education process creates inefficiencies at teaching hospitals—inefficiencies that make them less competitive in the marketplace relative to other hospitals.1 However, the research does not support this assumption, and the extent to which uncompensated educational activities outweigh the work residents do for hospitals is unclear. MedPAC estimated that only 45 percent of the IME payments can be analytically justified.8 Medicare and Medicaid should fund studies to estimate the degree to which the IME payments are required and whether they vary by institution and specialty. It is essential that teaching hospitals be fairly compensated for the public good of training future physicians.

Under the current GME system, Medicare and Medicaid’s contributions are transparent because funding is provided using a formulaic approach based on public data. However, it is unclear how recipient programs use these funds because the internal system for distributing DGME and IME funding varies by institution. Because many of the funding streams that support GME flow into the general revenues of teaching hospitals, it’s difficult to ascertain what amounts are used for education. No mechanisms exist to hold residency programs accountable for their GME spending, so Medicare and Medicaid have no way to influence how the funding is used. Tracking performance measures and outcomes for each program, such as the number of graduates who enter undersupplied specialties and practice in rural settings, is a good first step toward ensuring the maximum public benefit.

Stakeholders invested in GME

Stakeholders invested in GME

However, to encourage priority programs, Medicare and Medicaid must take a second step and divorce GME funding mechanisms from patient payments. Under the current system, GME payments to teaching hospitals depend solely on the number of residents and the percentage of patients with Medicare or Medicaid. Hospitals are rewarded for the quantity of residents and publically funded patients, not for the quality of their educators or their graduates’ ability to meet future workforce needs. Training programs must be held accountable for their performance and GME funding needs to be specifically designated for training purposes.

In light of recent legislative proposals to cut GME funding and the reductions proposed by the Obama Administration, the GME system cannot continue to rely solely on federal funding; alternatives must be considered. To encourage accountability and create incentives to align the current training system with public demands, all stakeholders invested in GME must be actively involved in its financing (see figure on left). Every stakeholder has a different perspective and agenda, but all are invested in having a functioning health care system staffed by well-trained physicians and should support GME in some form.

Table 2. Conceptual framework for creating and evaluating GME funding alternatives*

Equity Those that bear the cost of the activities should receive benefits that are proportional to their contributionsFunding should be distributed to meet current and future needs of the entire population
Adequacy An adequate system must provide funding to support the training needs of a high-quality physician workforceStable funding should be available to allow teaching programs to invest in high-quality training programs
Efficiency An efficient system must encourage effective educational programs at an economical priceFunding must adequately subsidize educational activities so that teaching institutions remain fiscally solvent
Accountability An accountable system must directly demonstrate the efficacy of resource allocation to achieve desired goalsFunding recipients should be held accountable for producing a workforce to meet the needs of the public with respect to the supply, specialty mix, and geographic distribution
Administrative feasibility A feasible system must ensure that its administrative burdens and costs do not outweigh its associated benefits
*Adapted from RAND Working Paper: Alternative Ways to Finance Graduate Medical Education.7

An effective conceptual framework must drive the creation and evaluation of GME funding alternatives. In 2006, HHS commissioned the Research and Development (RAND) Corporation to investigate the current GME system and establish one such framework. The mechanism RAND developed uses five critical measures: equity, adequacy, efficiency, accountability, and administrative feasibility (see Table 2).7 Overall, the RAND analysis revealed that Medicare’s most appropriate way forward was to continue IME support of teaching hospitals to compensate for higher patient care costs, but to shift responsibility for DGME payments to a separate federal organization dedicated to funding residency activities so that payment would not be tied to service use. RAND team members also stated that funding residency programs directly would foster accountability and provide a method for federal funders to align their payments with public health priorities.


The role of Medicare and Medicaid in GME has been a longstanding source of debate. The BBA caps, a significant reduction in state support, and recent legislative cuts to GME have made the state of GME funding more precarious than ever. Further reductions may severely impair the ability of the GME system to continue to train future physicians. Training more physicians to serve an increasing elderly population and the millions that will be newly insured by 2014 will require more support from all stakeholders, including federal and state governments. The rapid expansion of medical school capacity to meet health care demands, without proportionally increasing GME spots, adds further pressure to an already strained physician workforce and would disregard educational investments U.S. medical school graduates have made.

With further cuts looming and the mission of increasing training spots to meet current and future demands, GME faces a very real crisis. These competing agendas should serve as a signal to the medical community that the current GME funding policy is unsustainable. Previous proposals have attempted to address this problem by distributing the burden of GME financing to private stakeholders through the all-payor trust fund. However, reducing federal contributions without adequately assessing the costs of training and ability of other stakeholders to contribute has been unsuccessful. The use of a conceptual framework will allow for a structured analysis as various solutions for GME reform are proposed. With a clear understanding of costs, effective allocation of resources may create a more efficient GME system. With the input and consideration of various groups, relationships between payors and beneficiaries can help guide the creation of a fair, equitable system. To ensure the sustainability of U.S. health care, the GME system must continue to produce a high-quality physician workforce trained to address the growing needs of the public in a way that does not overburden any stakeholder.


  1. Medicare Payment Advisory Commission. Report to the Congress: Aligning Incentives in Medicare. Available at: Accessed April 10, 2012.
  2. Iglehart JK. The uncertain future of Medicare and graduate medical education. N Engl J Med. 2011;365(14):1340-1345.
  3. Kaiser Family Foundation Medicare Policy. The Budget Control Act of 2011: Implications for Medicare. Available at: Accessed April 10, 2012.
  4. Association of American Medical Colleges. Don’t Cut Funds That Support Physician Training. Available at: Accessed April 10, 2012.
  5. Iglehart JK. Health reform, primary care, and graduate medical education. N Engl J Med. 2010;363(6):584-590.
  6. Young JQ, Coffman JM. Overview of graduate medical education. Funding streams, policy problems, and options for reform. West J Med. 1998;168(5):428-436.
  7. Wynn B, Guarino C, Morse L, Cho M. RAND Health Working Paper. Alternative Ways of Financing Graduate Medical Education. Available at: Accessed April 10, 2012.
  8. Medicare Payment Advisory Commission. Report to the Congress: Improving Incentives in the Medicare Program. Available at: Accessed April 10, 2012.
  9. Council on Graduate Medical Education. Nineteenth Report: Enhancing Flexibility in Graduate Medical Education. September 2007. Available at: Accessed April 10, 2012.
  10. Vladeck BC. Plenty of nothing—A report from the Medicare commission. N Engl J Med. 1999;340(19):1503-1506.
  11. Centers for Medicare & Medicaid Services. Eligibility. Available at: Accessed October 25, 2012.
  12. Henderson T. Medicaid Direct And Indirect Graduate Medical Education Payments: A 50 state survey 2010. Available at: Accessed April 10, 2012.
  13. Feinstein AJ, Deckelbaum DL, Madan AK, McKenney MG. Unsupervised procedures by surgical trainees: A windfall for private insurance at the expense of graduate medical education. J Trauma. 2011;70(1):136-139.
  14. Council on Graduate Medical Education. Charter. Available at: Accessed April 10, 2012.
  15. Council on Graduate Medical Education. Twentieth Report: Advancing Primary Care. December 2010. Available at: Accessed April 10, 2012.
  16. Association of American Medical Colleges. Proposed Reductions in Medicare IME Payments to AAMC Teaching Hospitals: National and State Economic Impacts. Available at: Accessed April 10, 2012.
  17. The National Commission on Fiscal Responsibility and Reforms. The Moment of Truth. Available at: Accessed April 10, 2012.
  18. Medicare Payment Advisory Commission. Report to the Congress: Reforming the Delivery System. Available at: Accessed April 10, 2012.
  19. Congressional Budget Office. Reducing The Deficit: Spending And Revenue Options. Available at: Accessed April 10, 2012.
  20. Institute of Medicine. Crossing the Quality Chasm: A New Health System For The 21st Century. Available at: Accessed April 10, 2012.
  21. Bingham J, Udall M, Kyl J, Udall T, Grassley C, Bennet M, Crapo M. Letter to the Institutes of Medicine. Available at: Accessed April 10, 2012.
  22. Iglehart JK. Financing graduate medical education—Mounting pressure for reform. N Engl J Med. 2012;366(4):1562-1563.
  23. Association of American Medical Colleges. “All-Payer” Proposals. Available at: Accessed April 10, 2012.
  24. Association of American Medical Colleges. Physician Shortages to Worsen Without Increases in Residency Training. Available at: _factsheet.pdf. Accessed April 10, 2012.
  25. Sells T. Anxiety cloaks medical resident funding; $100M payment cut may be off for now, but Tennessee’s schools still worried. The Commercial Appeal. Available at: 04/anxiety-cloaks-medical-resident-funding/. Accessed April 10, 2012.
  26. Magoon PM, Jaudes PK. CHGME funding. Chicago Tribune. Available at: Accessed April 10, 2012.

Tagged as: , ,


Bulletin of the American College of Surgeons
633 N. Saint Clair St.
Chicago, IL 60611


Download the Bulletin App

Get it on Google Play