Share this article on:

GME Reform: How Graduate Medical Education Falls Short in Training Neurologists

Hiscott, Rebecca

doi: 10.1097/01.NT.0000455668.36507.f9
Back to Top | Article Outline




A recent report from the Institute of Medicine has proposed sweeping changes to the funding and structure of graduate medical education. Neurologists weigh in on what those changes could mean for the specialty.

A recent report from the Institute of Medicine (IOM) has proposed sweeping changes to the funding and structure of graduate medical education, or GME. Across five overarching recommendations, the IOM authors proposed restructuring GME funding in a way that would prioritize residency training across specialties and geographic regions that are currently underrepresented in medicine. They also called for greater transparency from medical institutions in the way GME funds are administered.

Neurologists involved with GME spoke to Neurology Today about the impact the IOM report could have on the specialty, and weighed in on how the current GME system fails to train the number of neurologists needed to treat an aging US population, increasingly affected by chronic diseases of the aging brain.

Currently, Medicare and Medicaid provide more than 90 percent of all federal funding for GME, and Medicare alone contributes $9.7 billion per year of a more than $15 billion budget, the IOM report noted. The authors recommended freezing federal funding at its current level, updated for inflation, for the next 10 years, at which time it would be re-assessed.

At the same time, the report suggested a major restructuring of GME funding, splitting funds into two streams—one to support current GME programs, and another for a “transformation fund” that would support structurally innovative GME programs.

The authors also recommended that Medicare-funded training positions be allocated to “priority specialty and geographic areas,” moving some residency training programs from academic hospitals to clinics and ambulatory, community-based settings, where Americans are increasingly seeking most of their care.

Commenting on the report, AAN President Tim Pedley, MD, FAAN, Henry and Lucy Moses professor of neurology at Columbia University's College of Physicians and attending neurologist at the Neurological Institute of New York, told Neurology Today, “I don't think anybody could claim that our graduate medical education is an entirely rational process that meets the nation's needs. There is no mechanism within the current GME system for producing the kind of physicians that our health care system needs. We don't even have a mechanism for determining what kind of physicians we need, and where we need them most, in any kind of rational way.”

But despite many salient (if controversial) suggestions, the IOM report doesn't entirely address one of the most significant problems plaguing the medical community, including neurology, he said: a growing shortage of physicians across a variety of much-needed specialties. The report's authors refer to data pointing to a looming physician shortage as “variable” and “historically unreliable.”

In reality, a number of specialties, including neurology, are seeing dwindling numbers of practitioners, Dr. Pedley said. And in brushing this issue aside, the IOM report has not adequately addressed the fact that under the current GME system, medical school graduates tend to overlook specialties that treat an aging population.

Back to Top | Article Outline


Several associations, including the American Hospital Association, the American Medical Association, and the Association of American Medical Colleges (AAMC), have released statements calling for an end to the freeze on Medicare spending for residency training, which has been in effect since 1997. According to the AAMC, there will be a shortage of more than 90,000 physicians within the next decade, consisting of a shortfall of roughly 45,000 primary care physicians and 46,000 surgeons and specialists. The AAMC has said that the only solution is to increase federal funding for residency programs.

The IOM's interest in siphoning funds for new training programs out of academic hospitals and into clinics and community-based settings is troubling, Christiane Mitchell, director of federal affairs and government relations for the AAMC, told Neurology Today.

“Academic medical centers have this great, dynamic learning environment, especially for specialties like neurology,” she said. Residents “have to learn to use the equipment that's available in the hospital [and] do inpatient as well as outpatient care.”

“The IOM implies in their comments that right now, a resident trains in an academic center and it's an ivory tower [where] they have no community exposure whatsoever,” she added. “That's not true. Residents train at private practice, they train at a VA ambulatory clinic, they train at a community health center—they absolutely are in the communities, learning to deliver care in a non-hospital setting.”

However, the AAMC does support the IOM's mandate to ensure that new GME programs train physicians with the skills and experience necessary for better addressing the needs of the current and future patient population, Mitchell said.

Neurology and other non-surgical specialties may in fact benefit from training in an outpatient setting, if a community-based program can be designed and administered in a way that would provide high-quality training, said Dr. Pedley.

“The current training of almost all physicians emphasizes acute hospital-based care, whereas most of the care, even in academic medical centers, is delivered in the outpatient arena,” he said. “Especially with an aging population, we're dealing with chronic conditions—and that is particularly germane to neurology, because many of the conditions that the elderly have are diseases of the aging brain. Increasingly, the challenge to our health system [is] providing the best care to patients in outpatient and home settings.”

The IOM may be correct in recommending that GME funds be allocated more judiciously and transparently, added John W. Engstrom, MD, FAAN, chair of the AAN's Graduate Education Subcommittee and director of Neurology Outpatient Services at the University of California, San Francisco Spine Center.

“The dollars allocated for training are paid to hospitals, not to training programs, not to departments, and there's only a fraction of those dollars that ultimately find their way to training programs,” he said. “I think there are a number of perverse incentives that marry the financing of training with training in a manner that makes it hard to innovate, because there are stakeholders that have vested interests in keeping things the same.”

Back to Top | Article Outline


The current GME system provides training of the highest caliber, Dr. Pedley said. However, “we are not training a physician pool that reflects the health needs of our country, whether that's rural versus urban care, or chronic versus acute, or hospital-based versus outpatient, and so on. The missing piece is that [residency] training is not tied in some way to the actual health needs of the country.”

The IOM report noted, for example, that the number of US residency positions has increased by 17.5 percent in the past decade—medical school enrollment, meanwhile, has increased 28 percent—despite the 1997 cap on Medicare-funded slots. Indeed, some hospitals have chosen to absorb the costs of extra residency programs because of the crucial services they provide. Still, the greater number of residency slots does not produce the number of physicians needed to adequately meet the health needs of the population.

“The idea that you would increase the medical school class because you think there is a physician shortage, but then you wouldn't increase residency slots, is almost unimaginably flawed reasoning,” said Dr. Engstrom. “What's been happening is that as the number of US medical graduates goes up, the number of foreign medical graduates landing residency positions is going down, and the effect is no net increase in physicians.”

For neurology in particular, exposure to training and education may simply come too late in the course of medical education, said Imran Ali, MD, FAAN, chair of the AAN's Undergraduate Education Subcommittee, professor of neurology and senior associate dean for academic affairs at the University of Toledo. “Students' exposure to neurology training in medical school is limited, so sometimes the opportunity for them to decide to choose neurology [as a residency] is not there or comes in late,” he said.

Another problem still, which the IOM report noted, is that the physician workforce is becoming increasingly specialized in a way that is at odds with the population's needs.

Although the National Resident Match Program (NRMP) maintains that there are more open residency positions than there are US. Graduates — a 2014 NRMP report estimates that there are 1.54 available positions per graduating allopathic US senior — those estimates fail to take into account the number of medical school graduates from foreign institutions applying for American residencies, in addition to a small number of graduates who failed to match the previous year who have re-entered the match. And as US medical school debt continues to skyrocket, graduates are increasingly basing their choice of residency on lifestyle factors, which puts neurology at a competitive disadvantage just when more Americans are beginning to need neurologic care.

“At the present time, applicants choose their specialties based on personal priorities,” said Dr. Pedley. “For many, the two most important ones are income and quality of life, and that is why, for example, dermatology and anesthesiology residencies have a great oversubscription.” Neurology does not compare especially favorably to surgical specialties in terms of income and workload, he said.

With the current structure of GME, “we don't have a method to say, we need this many neurologists, we need this many dermatologists, we need this many orthopedic surgeons, and then relate that to the payment that Medicare is providing for training of physicians,” he added.

Back to Top | Article Outline


The IOM report is not an approved piece of legislation, and its recommendations are still far from being implemented. Therefore, the neurology community should avoid formulating a knee-jerk response to it, Dr. Engstrom said.



“I think it's a good time to take a long deep breath, and rather than come up with a gut reaction to the IOM report, try and be a little more deliberate and pick through the pieces,” he said. “When it comes to this [report], both the devil and the angel are in the details. There may be things which will be better for patients and better for our field in the long run, but we need to see the specifics. At the moment, the report is really a policy statement.”

Right now, the most important takeaway from the IOM report is that GME is not likely to continue in its current form. Policymakers, as well as various medical communities, will need to work together to find a rational way to link graduate medical training to the nation's actual health care needs, said Dr. Pedley. Part of that will be to reform incentives and drive prospective residents into the specialties where they are most needed.

“Somehow, we need to improve the quality and distribution of care within the United States, making it more equal across the country, [while] retaining the best of what we do,” he said.

“The debate is not going to go away, and the status quo is not going to continue,” Dr. Engstrom added. Rather than shy away from controversy, or dismiss the IOM report entirely, he said, “I think it makes a lot more sense for neurology to position itself to be firmly embedded in the debate.”

Back to Top | Article Outline


The question of how to reform GME funding has become especially pressing in recent years, as multiple reports have suggested a growing shortage of primary care physicians, as well as a lack of physicians in specialties that will be needed to treat an aging population.

In a study published online in the April 17, 2013 issue of Neurology, researchers conducted a statistical analysis to predict the future shortage of neurologists in the United States through 2025. Taking into account the number of new neurologists trained each year, the shifting demographics of the neurology workforce, the needs of an aging US population, and increased demand associated with healthcare expansion under the Affordable Care Act, the researchers predicted that the demand for neurologists would increase from approximately 18,180 in 2012 to 21,440 by 2025. But in that time, the supply of neurologists would only increase from 16,366 (an 11 percent shortfall) to 18,060 (a 19 percent shortfall). The supply-demand gap would also continue to widen between urban populations and traditionally underserved rural and inner city areas, they predicted.

“As more residents subspecialize (e.g., in sports medicine, as hospitalists, and in neurointensive care), there may be even fewer neurologists to provide care to patients with chronic conditions,” they wrote.

Back to Top | Article Outline




  1. Maintain Medicare GME funding at the same level (adjusted for inflation) for 10 years, at which time funding would be re-assessed.
  2. Build a GME policy and financing infrastructure. Create a GME Policy Council in the Office of the Secretary of the US Department of Health and Human Services, which would oversee a strategic plan for Medicare GME financing and develop future policies for the use of Medicare GME funds. Create a GME Center within the Centers for Medicare & Medicaid Services that would manage the operational aspects of GME Medicare funding and manage the GME Transformation Fund.
  3. Create one Medicare GME fund with two subsidiary funds: An operational fund that would distribute support to currently approved residency training programs, and a transformation fund that would “finance initiatives to develop and evaluate innovative GME programs,” and allocate GME funds in priority specialty and geographic areas.
  4. Modernize GME payment methodology by creating one GME funding stream that would fund organizations with GME programs, based on a national per-resident amount (with a geographic adjustment) and implement performance-based payment to GME sponsoring organizations.
  5. Medicaid GME funding should remain at each state's discretion, but Congress should mandate the same high level of transparency and accountability as the new Medicare GME plan.
Back to Top | Article Outline


•. Institute of Medicine. Graduate medical education that meets the nation's health needs. Washington, DC: The National Academies Press, 2014.
    •. Wilensky GR, Berwick DM. Reforming the financing and governance of GME. New England Journal of Medicine 2014; 371.9–793: 792–793.
      •. Dall TM, Storm MV, Chakrabarti R, et al. Supply and demand analysis of the current and future US neurology workforce. Neurology 2013; 81: 470–478.
        •. National Resident Matching Program. National Resident Matching Program, results and data: 2014 main residency match. Washington, DC: National Resident Matching Program, 2014.
          •. American Association of Medical Colleges. Physician shortages to worsen without increases in residency training. AAMC Center for Workforce Studies, 2010.
            •. More about GME:
              © 2014 American Academy of Neurology