This Article Extract Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when Comments are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to Citation Manager Reprints & Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Muller, R. W. Articles by Snyderman, R. Search for Related Content PubMed PubMed Citation Articles by Muller, R. W. Articles by Snyderman, R. Related Collections Health Professions Education

Graduate Medical Education

PERSPECTIVE

Does Economic Theory Justify Changing Policy That Works?

We appreciate and agree with the authors’ observation that "teaching institutions provide enhanced quality of care and that payments to these institutions need to reflect their higher costs." We also agree that there is confusion about the distinct purposes of Medicare direct graduate medical education (GME) payments and the indirect medical education (IME) adjustment to Medicare per case payments. However, we oppose the authors’ contention, as well as MedPAC’s recommendation, that Medicare should recharacterize the educational costs associated with direct GME payments as patient care costs and combine the two historically separate funding streams—direct GME and IME—into a single adjustment to Medicare’s per case payments.

We believe that a distinct payment that is associated with resident educational costs is a sound and critically important policy decision. It is precisely this explicit recognition by Medicare of its responsibility to fund residency education costs through the direct GME payment mechanism that has helped to sustain the high quality of physician education in this country. This education produces a physician workforce universally regarded as the best in the world. Asserting that the entirety of the higher costs associated with teaching hospitals is "attributable to something other than medical education" is a construct that if adopted as a basis for future policy decisions could have serious, deleterious implications for the future of medical education in this country.

The authors assert that principles of economic theory support their notion that residents, not hospitals or medical schools, are incurring the educational costs associated with their residency period because it is a type of "general training." Because it is not in the interest of any firm to pay for this "general training," since trainees can take what they have learned and use it elsewhere, they argue that economic theory concludes that the trainees—in this case residents—bear the costs of their own education. They further argue that the costs currently characterized as educational costs are better considered as costs associated with the differential level of patient care provided by teaching hospitals. Consequently, they believe that these costs should be combined with the costs underwritten by the IME payment adjustment and the two payment streams combined.

In other words, Newhouse and Wilensky suggest that Medicare should change the way it supports GME because, in their opinion, there is a better theoretical basis for rationalizing the direct GME payments provided to teaching institutions. Given that they are not proposing a change in the aggregate level of Medicare support to teaching hospitals, one must ask whether there are sound policy reasons, beyond economic theory, to justify fundamentally changing an approach that has existed for more than thirty years, fifteen years in its current form?

Unfortunately, the paper is deficient in articulating reasons to make such a dramatic change. Foremost, the authors fail to identify any deficiencies in the current system that could not be, or have not already been, addressed within the current framework. For example, the authors’ concerns about excessive variation in "per resident amounts"—the basis of direct GME payments—was addressed in legislation enacted in 1999 and 2000, which narrowed the range of these amounts.¹

Number of residents.

 
The authors also argue that the current methodology creates an incentive to increase the number of residents because of economic benefits to the hospitals. This argument is unsound on several fronts. But we first point out that even if this were a legitimate concern, the authors suggest no alternative to the current resident-to-bed measure to eliminate this incentive. Moreover, the argument is mostly moot given that the Balanced Budget Act (BBA) of 1997 established a limit on the number of residents associated with Medicare reimbursement, and the total number of residents has remained flat over the past seven years.

As the authors note, the number of residents has grown over time. In large part, these changes are the result of advances in medicine, which have necessitated the development of new specialty residency programs or required an increase in the timespan of residency education. None of us can recall an occasion when we participated in a decision to increase the number of residents because we would get more Medicare funds. Those decisions were prompted by changes occurring in medicine, in our capabilities, and in the changing needs of our communities for physicians. However, we can say that the existence of explicit support from Medicare to partially offset the additional costs that would be incurred was an important factor in facilitating our ability to support residency education programs.²

This does not mean that we believe that Medicare’s current direct GME methodology must remain static. For example, the payment system may need refinement to ensure adequate resources to support educational costs in ambulatory sites, where more and more residency education is occurring.

Impact on residency education.

Top Number of residents. Impact on residency education. Appropriate application of... NOTES

 
Another major disappointment is the authors’ failure to explore—even in a conceptual context—either positive or negative impacts of implementing their framework on the quality and availability of residency education and the related educational activities undertaken by teaching hospitals and medical schools. We believe that a decision to rescind a distinct payment associated with education could undercut the vibrancy and strength of GME by signaling to current and future policymakers, and possibly providers, that educating future physicians should not be a primary mission of academic medical institutions. This is a dangerous message that could have serious consequences, especially when teaching hospitals are facing pressures to reduce the level and quality of their multiple missions to be price competitive. At a minimum, it seems imprudent to change a system that is working, without intensive attention to potential negative outcomes.

Appropriate application of economic theory?

Top Number of residents. Impact on residency education. Appropriate application of... NOTES

 
Consequently, we are left with the authors’ desire to satisfy an economic theory to justify overhauling the current system. We are not economists, and, in fact, we have great respect for the authors’ economic expertise. However, we seriously question whether the theoretical model that they espouse is appropriate to apply to GME. Additional discussion is needed to better test the authors’ contention—discussion that should include the characteristics of residency education that are distinct from traditional types of "on-the-job training." We believe that the efforts by teaching hospitals and medical schools, in conjunction with accreditation bodies, ensure that residents remain learners throughout the residency period, as they need to be, not simply apprentices or employees who are receiving on-the-job training.

As an aside, we note that the authors’ economic theory assumes that a firm always acts in its own financial interest. If such were the case, teaching hospitals would eschew the levels of uninsured patients that they treat. Yet these institutions and their affiliated physician faculty members accept and treat these patients because it is a part of their mission. Our academic institutions feel the same way about their teaching mission: It is who they are and what they do.

Despite our strong disagreement with the authors, we appreciate their focus on GME because it gives us an opportunity to raise what we believe is a serious cause for concern: how to ensure dedicated and stable financial support for educating our future physicians and the other valued contributions of teaching hospitals and medical schools. The authors are correct in noting that historically private payers have made payments to teaching hospitals—and, in fact, all hospitals—that are higher than their associated costs. Such payments have helped to finance the costs of missions, such as teaching and treating the poor, that are not covered by Medicare. However, this support is eroding, as evidenced by MedPAC data.³ Consequently, we believe that the fundamental flaw related to GME financing lies not within Medicare, but in the current and potential future failure of many private payers in an increasingly price-competitive marketplace to pay their fair share of the costs associated with the educational and other societal missions of teaching hospitals and medical schools. We believe that the way to fix this flaw is for all payers of health care services to adopt Medicare’s rationale and explicitly contribute to the funding necessary to educate residents as well as to sustain the other critical missions of teaching hospitals that ultimately benefit patients. What is at stake is nothing less than the future quality of health care in this country.

Editor's Notes

 
Ralph Muller, immediate past chair of the Association of American Medical Colleges (AAMC), is president and chief executive officer (CEO), University of Chicago Hospitals and Health System. George Sheldon, AAMC chair, is chairman and Zack D. Owens Distinguished Professor, Department of Surgery, University of North Carolina at Chapel Hill School of Medicine. Ralph Snyderman, AAMC chair-elect, is chancellor for health affairs, Duke University; executive dean, Duke University School of Medicine; and CEO, Duke University Health System. Health Affairs invited their response to the paper by Joseph Newhouse and Gail Wilensky, which precedes this Perspective.

NOTES

Top Number of residents. Impact on residency education. Appropriate application of... NOTES

 

1. See Section 311 of the Balanced Budget Refinement Act of 1999 and Section 511 of the Medicare, Medicaid, and SCHIP Benefits Improvement and Protection Act of 2000.
2. The authors also note that the empirical level of the IME adjustment has fallen over time, while the number of residents has increased. Yet the level of the IME adjustment is attributable to several factors, some of which are unrelated to resident counts. For example, to the extent that the Medicare diagnosis-related group (DRG) system improves with annual changes to DRG classifications or teaching hospitals reduce operating costs that are unrelated to residents, the IME adjustment will fall.
3. Published and unpublished MedPAC data on private payer hospital payment-to-cost ratios.

                      What's this?