In early 2002, the Moran Company was engaged by the Greater New York Hospital Association and the Association of American Medical Colleges to analyze current issues in Medicare indirect medical education (IME) payment policy in order to identify the areas policymakers will need to evaluate in deciding the merits of possible changes to current law. As an independent health care research and consulting firm, the Moran Company does not take positions on policy proposals or pending legislation. Instead, the two organizations asked the Moran Company to prepare a White Paper to assist Federal policymakers in evaluating possible proposals to amend Medicare reimbursement policies directed at hospitals engaged in the post-graduate training of physicians. The following is an excerpt from that paper.

Medicare-Focused Arguments for Reducing IME Payments
Policymakers seeking reductions to Medicare IME reimbursement have put forth a number of arguments. They have argued that the Medicare IME formula has unintended incentives, that Medicare pays teaching hospitals disproportionately for IME, and that current levels of support are inappropriate in light of Medicare inpatient margins in teaching hospitals relative to other hospitals. In addition, some question whether Medicare should be supporting the social missions of hospitals generally.

It has been argued that the IME formula provides unintended incentives, such as to decrease hospital beds and increase residents trained. Research on the extent to which hospitals act on these various incentives is mixed.¹ However, in an effort to reduce these incentive effects, the Balanced Budget Act capped the number of residents that could be counted for IME payment purposes at each teaching hospital's 1996 level, and otherwise restrained the growth in the resident-to-bed ratio.

Others have suggested that private sector payers should share more of the burden of extra costs associated with teaching hospitals. According to the Medicare Payment Advisory Commission [MedPAC], the private payer payment-to-cost ratio has been much lower for major teaching hospitals than for non-teaching hospitals. . . .² [P]rivate payers may be reducing payments even further than in the past, in some cases refusing to offer teaching hospitals any increase over normal rates.

A main argument for reducing IME payments is supported by data demonstrating that teaching hospitals have substantially higher margins on their Medicare business than other hospitals. In 1999, inpatient Medicare margins for all hospitals were, on average, 11.9 percent. For major teaching hospitals this figure was 22.3 percent versus 11.6 percent for other teaching hospitals, and 6.5 percent for non-teaching hospitals.³

Finally, some point out that the logic of providing support for the social mission of teaching hospitals under Medicare IME is not rational, and that it is a "disguised" payment that should be more transparent. This argument suggests that these types of payment mechanisms provide little accountability for the funds provided for the social missions of teaching hospitals.⁴

A Broader Context for Considering IME Reductions
While it is reasonable to debate these aspects of Medicare IME financing policy, it is important to recognize that these issues all approach the decisions about IME funding from the perspective of whether it is a rational expenditure of Medicare funds. Our analysis suggests that proceeding in this fashion without thinking several moves ahead runs the risk of triggering policy problems that are avoidable if forethought is given to the impact of major payment cuts to teaching hospitals in key markets. A broader perspective may lead one to conclude that further consideration is warranted prior to implementing further reductions in IME support. This section presents aspects of the contextual environment in which the cuts will occur, and explores potential unintended and perhaps undesirable consequences from these cuts.

Hospital Total Margins, 1990-99: Comparison of Major Teaching, Other Teaching, and Non-teaching Hospitals

The Role of IME Payments to Teaching Hospitals' Overall Financial Viability
Policymakers considering IME payment policy should do so in the context of the significant role that Medicare reimbursement plays in teaching hospitals' overall revenues. According to MedPAC's analysis of 1999 data (the most recent data available), the total margin for major teaching hospitals was 2.4 percent, compared to 4.0 percent for non-teaching hospitals; MedPAC's analysis shows that when hospitals are grouped according to teaching status, major teaching hospitals have consistently had the lowest total margins over time⁵ (see figure at right). Total margins generally reflect the best-case scenarios for hospitals because they contain revenues associated with non-patient care activities, most notably investment income.

Managed Care Market Forces
Policymakers may also wish to consider that a large proportion of teaching hospitals are located in highly competitive markets and are under substantial financial pressure. Prior to the era of managed care, private payers largely paid teaching hospitals charges, which covered the payers' share of the incremental cost of the academic medical enterprise. The era of managed care put pressure on these premiums: while teaching hospitals could once command fees from managed care organizations (MCOs) that were five to 10 percent higher than other hospitals because of their prestige and high-technology equipment,⁶ this is no longer necessarily the case. In highly competitive markets, some private payers are refusing to reimburse teaching hospitals at higher rates than other hospitals or are charging patients higher copays and deductibles if they seek care at teaching hospitals.⁷

The pressures of managed care on teaching hospitals have been well and long documented. Studies demonstrate that teaching hospitals in markets with high managed care penetration are more likely than those in markets with low managed care penetration to have negative total margins.⁸ Analyses of hospital discharge data indicate demand for inpatient services declines as managed care market penetration increases.⁹ In addition, research indicates that patients in HMOs are less likely than other types of patients to receive care in teaching hospitals.¹⁰

Policy Implications of the Geographical Concentration of Teaching Hospitals
In evaluating the policy implications of additional IME payment reductions, it is important to understand that the impact of Medicare reimbursement changes on teaching hospitals may not be homogenous. This is because some teaching hospitals might be able to recoup losses stemming from the IME cut by increasing rates to private payers, while others may not. As suggested in the preceding section, the fiscal impact of IME payment reductions on teaching hospitals in less concentrated markets may be less severe than the impact in major urban markets. Experience over the last few years suggests that, even within major urban markets, IME payment reductions are likely to have disparate effects. In major markets with heavy concentrations of teaching hospitals, we have already seen certain teaching facilities experiencing disproportionately greater fiscal stress than others.¹¹

The general theory of prospective payment in Medicare is that the system should place balanced downward pressure on costs, in order to give all hospitals incentives to improve efficiency. An important corollary of this theory is that hospitals that are unable to continuously improve their cost performance face the risk of being forced to abandon their social missions, the high costs of which may be the source of their financial distress. The problem policymakers face in the present instance is that, in comparison to the era prior to managed care, the fiscal problems of teaching institutions are concentrated in major urban markets with a high clustering of teaching hospitals. The magnitude of these problems, moreover, is often sufficiently large and pervasive that the governance structures of these institutions have no meaningful recourse if revenues fall materially below actual operating costs.

When confronted with these facts, some economists and other policy analysts, whose primary concern is the operating efficiency of the healthcare system, may argue that the insolvency of such institutions is in fact warranted, and that further IME payment reductions are not a cause of instability. The long-term stability of such a policy, however, presupposes that policymakers are sufficiently informed about the consequences of the policy and willing to maintain it in the face of the possible failures of specific national medical institutions. Given that policymakers may find such failures to be problematic, they may wish to consider the context that we have presented when evaluating possible approaches to IME reimbursement policy.

Conclusion
This analysis suggests that policymakers evaluating proposals to reduce IME payments under Medicare will find it desirable to think several moves ahead before fashioning the next major change in Federal financing policy for teaching hospitals. If policymakers proceed with Medicare IME payment reductions without articulating an alternative theory of how Federal support for teaching hospitals should be structured, they risk weakening important parts of the medical infrastructure—a scenario that would be difficult to reverse in the future. Given these prospects, it is clear to us that policymakers will find it desirable to reach outside the context of the Medicare policy debate, and evaluate the overall context of the health care system when considering Federal and Medicare policies towards teaching hospitals. Notes

1. S. Nicholson and D. Song, "The Incentive Effects of the Medicare Indirect Medical Education Policy," Journal of Health Economics 20, no. 6 (November 1, 2001): 909-933.
2. Medicare Payment Advisory Commission, data presented at a public meeting, January 16, 2002.
3. Medicare Payment Advisory Commission, "Report to the Congress" (March 2002): 149.
4. Henry R. Desmarais and Michael M. Hash, "Financing Graduate Medical Education: The Search for New Sources of Support," Health Affairs 16, no. 4 (July/August 1997): 148-163.
5. Medicare Payment Advisory Commission, "Report to the Congress" (March 2002): 157.
6. James Reuter and Darrell Gaskin, "Academic Health Centers in Competitive Markets," Health Affairs 16, no. 4 (July/August 1997): 242-252.
7. Peter D. Fox and Jeff Wasserman, "Academic Medical Centers and Managed Care: Uneasy Partners," Health Affairs (Spring 1993): 85-93. Also "Top Hospitals Fear Insurers' Fee Plan: Patients Would Pay More to Pick Some Teaching Facilities," Chicago Tribune, 8 February 2002; and "HMOs Eyeing Surcharge for High-End Care: Some '02 plans affect visits to medical centers," Boston Globe, 28 August 2001.
8. Gerard F. Anderson, George Greenberg, and Craig K. Lisk, "Academic Health Centers: Exploring a Financial Paradox," Health Affairs 18, no. 2 (March/April 1999): 156-167.
9. Reuter and Gaskin, cited in note 6.
10. Robert Mechanic, Kevin Coleman, and Allen Dobson, "Teaching hospital costs: implications for academic missions in a competitive market," JAMA 280, no. 11 (September 16, 1998): 1015-1019.
11. See, for example, "Losses Mount for Detroit Medical Center," Detroit News, 31 January 2002. See, also, "Boston's CareGroup May Dismantle System," Modern Healthcare 32, no. 4 (January 28, 2002): 30 (including an estimate of a $102 million annual loss for the same teaching hospital).