Printed from:

Comprehensive GME Reform—The Time Is Now

Samuel Cullison, MD

(Fam Med 2014;46(9):731-3.)

One of the hallmarks of family medicine residency training is the intentional harmonizing of such training with the character of future practice. Since more than 90% of primary care services are delivered in outpatient ambulatory care settings, family medicine training requirements have emphasized, in fact required, a substantial emphasis on ambulatory, community-based training through our ACGME accreditation requirements. However, nearly all Medicare Graduate Medical Education (GME) training dollars are provided to hospitals, not training programs. This payment system creates incentives for hospitals to focus residents’ training within the hospitals themselves, to utilize residents to support the hospitals’ patient care workforce needs, and in some instances retain the Medicare GME payments to support non-GME and non-family medicine relevant programs often at the expense of providing resources for the educational needs of the residency. This “square peg in a round hole” effect results in a conflict between the training goals of family medicine and the financial and staffing needs of hospitals.

STFM, along with other family medicine organizations, has promoted the reform of Medicare GME for more than a decade. With the recognition that graduate medical education is on every Washington, DC, target list for deficit reduction and the rising support for countervailing proposals to increase residency positions without safeguards to produce the right number and mix of physicians for the country needs, family medicine organizations reexamined their GME reform policies last fall. Our goal was to prevent large-scale cuts to Medicare GME solely for deficit reduction purposes, while keeping intact our vision for the development of a more primary care-centric physician workforce training system. With the July release of the IOM report “Graduate Medical Education That Meets the Nation’s Health Needs” raising the consciousness of policymakers regarding the accountability of Medicare GME, the family medicine organizations felt this was an advantageous time to begin a national conversation on reforming Medicare GME.

New GME policy recommendations were released at a Summit on GME Reform in Washington, DC, on September 15, 2014 by the American Academy of Family Physicians (AAFP). These recommendations were finalized by the Academic Family Medicine Advocacy Committee and then adopted by each member organization, which includes STFM, AAFP, American Board of Family Medicine, the Association of Departments of Family Medicine, Association of Family Medicine Residency Directors, and the North American Primary Care Research Group. We expect these recommendations will be the foundation of our federal advocacy on Medicare GME for the foreseeable future.  

Taken together, these recommendations allow for a budget-neutral method of increasing innovation, right-sizing the physician workforce including emphasizing primary care workforce expansion, promoting more ambulatory/community-based training, and rationalizing the use of federal dollars to meet the nation’s true health care needs.

We anticipate that there may be individuals and institutions in academic medicine that will question or actively oppose these policies. Therefore, I will highlight each recommendation and offer some context about why they are needed.

Recommendation: Expand the number of residency positions for first-certificate residency positions, particularly in needed priority specialties, by limiting direct graduate medical education (DGME) and indirect medical education (IME) payments to the training for first-certificate residency programs, utilizing the savings for expansion.

This recommendation ensures that GME would fund basic training for all physicians and produce a physician workforce ready to meet the health care needs of the nation as a whole.

It also meets a prime goal of the recent Institute of Medicine Report that encourages production of a physician workforce that is better prepared to work in, help lead, and continually improve an evolving health care delivery system that can provide better individual care, better population health, and lower cost.1

“There is compelling evidence that health care outcomes and costs in the United States are strongly linked to the availability of primary care physicians. For each incremental primary care physician, there is [sic] 1.44 fewer deaths per 10,000 persons. Patients with a regular primary care physician have lower overall health care costs than those without one.”2

Merely increasing the total number of physicians regardless of specialty won’t resolve the important workforce issues faced by our current health care system. For example, resident physician slots rose 17.5% from 2003–2012 despite a cap on Medicare funded slots, and there was a disproportionate increase in physicians being trained as subspecialists despite a greater demand for generalists.1

During the time period mentioned, primary care training positions shrank in number, while subspecialty training positions expanded. That is because the clinical revenue for a Medicare GME unfunded subspecialty fellowship position can more nearly support the total training costs to the sponsor while a primary care training position cannot. Thus, since the cap on GME funding occurred in 1996 with federal legislation, the cap did restrain primary care training expansion but failed to limit comparatively higher reimbursed subspecialty training positions. This has contributed to our unbalanced health system with a preponderance of subspecialists and a shortage of primary care.

Recommendation: Align financial resources with population health care needs through a 0.25% reduction in IME payments—from the current 5.5% to 5.25%—and allocate these resources to support innovation in graduate medical education, with an emphasis on expansion to more states and communities, particularly for primary care and community-based education.

Traditional Medicare GME provides funding to hospitals in order to produce physicians needed to care for the Medicare population and the country. However, this hospital-centric method of paying for training has not produced the quantity or types of primary care physicians the nation needs. Traditional GME financing has rigid payment formulas that don’t allow for the innovation and change required to move medical education to produce physicians with the appropriate training needed to improve the value of our health care delivery system, including evidence-based medicine, population health, team-based care, care coordination, and team-based shared decision making with patients.

By changing from traditional financing that is calculated according to Medicare beneficiary bed-days to new funding methods that reallocate existing GME funding directly to primary care residency programs regardless of sponsor, those educational consortia, or other non-hospital agencies have the opportunity through proper financial incentives to create or expand ambulatory and community-based training programs to meet the current and future health care needs of the nation.

Recommendation: Fund the National Health Care Workforce Commission. This commission would lead a strategic process that moves away from guaranteed financing of GME programs to a system of accountability and planned production goals.

While some health care organizations/entities may disagree about the specifics of physician shortages, we can all agree that having a national body of experts functioning to identify our nation’s needs is critical. Most other developed countries in the world already do this. The outcome could be a coherent national strategy to produce a suitable workforce needed for effective care and outcomes, including specialty production, location, and diversity.

Recommendation: Require all sponsoring institutions and teaching hospitals seeking new Medicare and/or Medicaid-financed GME positions to meet minimum primary care training thresholds as a condition of their expansion.

Of note, some additional nuances to this policy are:

• Measurement of primary care production by training institutions should be based on the status of graduates 5 years after medical school, not their initial certification.

• Over-cap positions should not be eligible for new GME funding.

• 50% of new positions should be in primary care; of those, 50% should be in family medicine.

There is compelling evidence that health care outcomes and costs in the United States are strongly linked to the availability of primary care physicians.2 If any workforce proposal is to restrain the growth in health care spending successfully, it must also support programs that build the family medicine and primary care workforce. This raises the quality of the care provided while restraining overall health system costs.

Data show primary care production leading to a ratio of at least 40% primary care will improve our nation’s financial bottom line. COGME’s 21st Report states that GME funding should be prioritized to accelerate physician workforce alignment with population and health care delivery needs.3

The Census Bureau projects that the US population will increase by 15.2% in the period of 2012–2025, and the population older than 65 will grow by 60%. Based on these projections, a recent study published in the Annals of Family Medicine estimates the number of primary care office visits will increase from 462 million in 2008 to 565 million in 2025, and by 2025 we will need nearly 260,687 practicing primary care physicians—51,880 more than the current workforce. In this time of increasing primary care need, we must recognize the importance of maintaining and expanding the pipeline of primary care production.4

While we acknowledge that family medicine and other primary care physician production is not the only national need, it is the largest need. Therefore, we felt it was important to include in our recommendations stipulations on how new slots will be distributed by specialty.

Recommendation: Demonstrate a commitment to primary care by all institutions who receive new Medicare or Medicaid-financed positions, through the establishment of primary care maintenance-of-effort requirements.

We can’t allow an expansion of subspecialty training positions by institutions in a way that disadvantages primary care positions. According to expert testimony by the Altarum Group before the Council on Graduate Medical Education, “Until primary care preferences rebound, it will be important to avoid excess growth in non-primary care training positions. Annual growth in non-primary care positions in excess of 2% will cause the primary care share of new physicians to decline, regardless of how many new primary care positions there are…” For 40% of US graduates who match in primary care, primary care is their second choice. We need to maintain current primary care positions and expand them further.

As with all of our policy recommendations, our goal is to improve the US health care system by achieving the triple aim: improving the experience of care, improving the health of populations, and reducing per capita costs of health care. To achieve this goal, we still have much work to do. Congressional action is required to make these policy recommendations a reality. I ask you to join me in our advocacy efforts to move this reform agenda forward.

Correspondence: Address correspondence to Dr Cullison, Methodist Health System Dallas, 1441 N. Beckley Avenue, Dallas, TX 75203. 214-947-2356. Fax: 214-947-2358.



  1. Institute of Medicine. Graduate medical education that meets the nation’s health needs. Accessed September 24, 2014.
  2. Starfield B, Shi L, Macinko J. Contribution of primary care to health systems and health. The Milbank Quarterly 2005;83(3):457-502.
  3. Council on Graduate Medical Education. COGME 21st Report: Improving value in graduate medical education. Accessed September 24, 2014.
  4. Petterson SM, Liaw WR, Phillips RL Jr, Rabin DL, Meyers DS, Bazemore AW. Projecting US primary care physician workforce needs 2010–2025. Ann Fam Med 2012;10(6):503-9. Doi: 10.1370/afm.1431.

Copyright 2018 by Society of Teachers of Family Medicine