applications for research but also establish and improve both the standards of care and the cost-effectiveness of care through prospective outcomes research. Focusing upon which treatments work best, outcomes research seeks to determine the effectiveness and efficiency of various alternative treatment strategies for a particular medical disorder.

In this critically important new arena of research, Duke University Medical Center has already established an enviable record of leadership, particularly in assessing various approaches to heart disease. The Duke Databank for Cardiovascular Diseases, for example, is a national resource for assessing outcomes for patients with cardiovascular disease. Established in 1969, the Databank is the world's largest cardiovascular disease database and has played an instrumental role in numerous studies determining which treatments provide maximum benefit to patients. Using such detailed historical information from the Databank, Duke researchers recently reported that the treadmill exercise test, which costs from $200 to $300, is just as effective in diagnosing heart disease in most patients as the more invasive angiogram, which costs several thousand dollars to administer.

As the nation's attention focuses on the problems of health care delivery, we believe it critical that we continue our strong commitment to excellence throughout the spectrum of research. While numerous examples can be cited illustrating the link between basic research and patient care, few are more powerful than the work being conducted at Duke University Medical Center's Joseph and Kathleen Bryan Alzheimer's Disease Research Center. Beginning almost a decade ago with the most basic research into the brain chemistry of deceased Alzheimer's patients, a team of Duke researchers is rapidly unraveling the genetic influences on Alzheimer's disease. This group has already identified the genetic markers for Alzheimer's and now believes that possible therapies are on the horizon.

As the need for primary care has increased, we have developed and are putting into practice the Duke Health Network which will consist of close affiliations between Duke, other hospitals and a large number of primary care practitioners. As students and residents require more and more training in non-hospital settings, the Duke Health Network will provide an optimal setting for this training as well as comprehensive, cost-effective and accessible health care for our region..

We at Duke are proud of the efforts we began several years ago, not in anticipation of legislation but because it was the right thing to do. We plan to continue these efforts and to redouble them. However, there are two issues that bring us at Duke major concern that I would like to share with you. The first is the funding of the academic health center.

Funding for the academic health center is a public-private partnership that subsidizes not only the educational and clinical research efforts of the academic health center but more importantly pays for the disproportionate share of critically ill patients and those who have unusual diseases. I like to think of one component of the academic health center as a fire or police station that is always open, 24 hours a day to handle any medical need that comes along, such as severe, complex illnesses or unusual diseases. For example, our coronary artery bypass team, trauma team and burn team all are available 24-hours a day, 365-days a year whether needed or not. Our bone marrow transplantation unit provides cures for breast cancer not previously available. This level of care cannot be obtained in a community hospital. This care is not only expensive, it is also inefficient (because of the firehouse phenomenon); however, it is absolutely necessary. Therefore, even with the streamlining of our processes and our teaching of cost effectiveness, we will not be able to provide such levels of care at a competitive price.

If we now add the cost of educating our students, our residents, and communicating this knowledge to our community as well as the costs for the broad range of research conducted to improve the care we deliver, it is obvious that the academic health center is more expensive and requires appropriate subsidies for its societally needed services. These subsidies, at present, are provided by increased charges to private insurers and by payments from Medicare under the rubric of GME, direct graduate medical education payments and indirect medical education adjustments. These payments help to fund the core missions of

the academic health center. Managed competition (the premise upon which the Health Security Act [HSA] is based) could unravel academic health centers' entire financing system of cross-subsidization but would make accommodation for only a portion of the needed compensation by replacing it with two smaller funds.

I am pleased that the HSA recognizes that academic health centers are unique national resources and that they fulfill special societal needs in the health care system. I strongly support the need to fund separately the spectrum of costs associated with an academic mission including the costs of graduate medical education and other health professionals, and the special and unique patient care costs that make it difficult for these institutions to compete in the current environment. I agree that all payers should contribute to the financing of both accounts.

However, I am concerned about the level of the financing of the two accounts and how the funds are distributed. Overall, the amount available to fund these costs is insufficient. Proponents of the HSA have argued that, if enacted, teaching hospitals would be better protected and more adequately financed than if the current situation were maintained. They compare the current level of Medicare payments for direct graduate medical education and indirect medical education costs--nearly $6 billion in FY 1994 to the $9.6 billion total amount that teaching hospitals would receive in the year 2000 under the HSA. In addition, many HSA-proponents believe that teaching hospitals will be able to "make up the difference" by commanding premium prices in the delivery system based on their service offerings and reputations.

While the total of these set-aside funds would exceed current Medicare spending for direct graduate medical education costs and the indirect medical education (IME) adjustment, this premise indicates an apparent misunderstanding of the current competitive environment and the level of support that the academic mission requires. The Medicare program supports only a portion of the academic mission. Data from hospitals belonging to the AAMC's Council of Teaching Hospitals show that Medicare payments cover only about 20-33 percent of the costs associated with the academic mission. The other 67 to 80 percent must be obtained from public and private payers who provide the balance of funding for these additional costs, primarily through increased charges for services.

Historically, teaching hospitals have financed their multiple functions through crosssubsidization. For example, patient service revenues have supported graduate medical education and other academic activities; routine service revenues have supported tertiary care patients; revenues from high volume services have supported low volume services; and payments from paying patients have supported charity care patients. However, during the past few years, as the overall costs of medical care have risen sharply, private health care payers have adopted payment systems--such as capitation, aggressive contacting and discounting--that restrict their payments to cover only goods and services they believe are necessary and of identifiable benefit to their enrollees. Costs associated with the education and research missions of teaching hospitals are not generally recognized by these payers.

In the newly price competitive environment, there is pressure to identify the crosssubsidized products of teaching hospitals. I believe that teaching hospitals will no longer be able to "make up the shortfall" to fund the costs associated with their academic missions through higher charges to patients. Therefore, the overall financing of the two funds must be adequate to ensure the continued financial viability of these institutions.

Notwithstanding my specific comments on the funding of the workforce and AHC proposals, there is another issue of major concern to the academic community. Managed competition, the fundamental premise on which the HSA is based, would not only unravel medical schools' entire financing system of cross-subsidization, and replace only a portion of it. My colleagues and I are concerned about the ability of medical schools to continue to support physician education, particularly at a time when medical schools and teaching physicians are being called upon to transform the medical education system from one that

focuses on specialist training in hospital inpatient settings to a more inexpensive system of generalist training in ambulatory, non-hospital sites.

For several reasons, medical schools will have difficulty sustaining this elaborate system undergirding the education and research missions. Federal support is increasingly constrained with medical schools expected to accept a greater share of the costs. Pressures brought to bear on medical service costs will likely lead to declining income from the faculty clinical practice, and less money available to support educational and research efforts. In order to preserve the patient base critical for medical education and research, faculty physicians are being drawn into developing networks with affiliated teaching hospitals and are being asked to accept capitated or discounted payments from private payers. As community physicians are forced to align with various health plans in integrated networks, their willingness to "contribute" teaching services are being threatened.

Undergraduate medical education in the clinical setting, directed by the medical schools, is not recognized explicitly by any payment system, but like other academic costs, it has been financed indirectly. The shift to a more explicit financing system threatens the ability of medical schools and teaching hospitals to fund this activity through other sources of support. Funds from physicians' clinical incomes cannot be expected to maintain their current levels. Fundamental forces are causing the traditionally cross-subsidized products to rise to the surface, yet only in two arenas has the HSA provided assistance. My colleagues and I believe that a complete and adequate financing system for academic medicine must be provided, and we would be pleased to work with members of Congress and the Administration to remedy the situation.

My second major concern is with the number and distribution of individuals caring for our patients. There is a general notion that the number of primary care givers should increase, and we agree with that premise. However, the exact magnitude of that need is simply unknown at this time. Additionally, as it becomes clear that certain primary care functions can and should be performed by nurse practitioners and physician assistants, the need for primary care physicians will not be as great.

Similarly, assessing the need for specialty care is extremely complex. The mcthodology for determining exactly how many and what kind of specialists is not perfected. While it is likely that there are too many of some, are we to believe that there are too many of all? This problem becomes further complicated by the welcome addition of 37 million currently uninsured Americans. To assume that we should change our residency programs to achieve the mandate for 55 percent primary care physicians is premature. I urge the Congress to enact legislation that would create a National Health Professional Workforce Advisory Board.

This Board should be empowered to develop the methodology to assess the needs in primary care and specialty care throughout this nation. I urge you to set general goals for an increase in primary care but to remove the mandate of a specific number that has little basis in fact.

In return, I can assure this committee and the American people that their investment in the critical functions of the academic health centers will be rewarded many times over by the contributions these centers make to improve the quality of care and the excellence of our still-evolving American health care system.

Thank you.

Chairman STARK. Dr. Foreman.

STATEMENT OF SPENCER FOREMAN, M.D., IMMEDIATE PAST CHAIR, ASSOCIATION OF AMERICAN MEDICAL COLLEGES, AND PRESIDENT, MONTEFIORE MEDICAL CENTER, BRONX, N.Y.

Dr. FOREMAN. Mr. Chairman, I am Spencer Foreman, the president of Montefiore Medical Center in the Bronx, N.Y., and immediate past chairman of the Association of Medical Colleges; and I am here to speak on their behalf.

The Nation's teaching hospitals and medical schools recognize their great responsibility in the health care system and are confident that, given the necessary tools, they can provide a competent, properly balanced physician workforce. The AAMC appreciates the administration's leadership in proposing comprehensive, high quality, cost-effective coverage for every American, and is particularly pleased that the Health Security Act has as an underlying policy requirement all payer support of the academic mission. The principal portions of my written testimony focus on the workforce and academic health center provisions of the act. My oral comments will, as well.

With respect to manpower development, the AAMC agrees with the need to train more physicians in generalist disciplines and supports an overall national goal of having at least 50 percent of graduating physicians entering generalist careers. However, we are concerned that the act's timetable for achieving that goal may be overly ambitious and that the government will move too quickly to a regulatory approach to accomplish it. We strongly urge allowing time for the new incentives now coming into play in the marketplace to work.

Managed care and revised fee schedules are reducing the demand for specialists and shrinking their income, while the same forces are enhancing the status of generalists and improving their income. We are already seeing significant shifts in specialty preference among graduating medical students. Furthermore, the AAMC is very concerned that a national physician manpower regulatory body would have a great deal of difficulty making the thousands of equitable allocation decisions required to regulate 82 specialties and 7,000 training programs.

I call your attention to tables 3 through 7 in my written statement, which attest to the complexity of the problem and the need for flexibility. Nevertheless, the AAMC is prepared to support a more regulatory approach if there is insufficient progress toward meeting the national goal.

With respect to financing workforce development, the AAMC enthusiastically endorses all-payer support of physician training, but is very concerned that the funding proposed in the Health Security Act is inadequate. The workforce account, which is designed to support the operating costs of graduate medical education, uses as a basis for payment a cost-finding methodology which omits real and presently recognized costs of training including those now covered by Medicare in its direct graduate medical education payments. Furthermore, payments from this fund are to be made based on the national average cost of training.

We are very concerned that moving to an average payment will cause a marked and unwarranted redistribution of support among training programs and will have very serious consequences for those programs which lose substantial amounts of funding.

Chairman STARK. Back up on that 1 minute. Just before the average payment, the preceding paragraph.

Dr. FOREMAN. The workforce account which is designed to support the operating costs of graduate medical education uses as a basis for payment a cost-finding methodology which omits real and presently recognized costs of training, including those now covered by Medicare in its direct medical education payments. Furthermore, payments for this fund are proposed to be made on the basis of national average cost of training.

We are concerned that moving to an average payment will cause a marked and unwarranted redistribution of support among training programs and will have very serious consequences for those programs which lose substantial amounts of funding.

Chairman STARK. In the aggregate institutional support and in programs within the institution?

Dr. FOREMAN. In the aggregate institutional support and in programs within the institution. That is, presently those costs are paid on a cost-finding basis, which is institution specific. If you move to an average, the ones that get high amounts of reimbursement will lose it and those that are low will be unanticipated winners.

Chairman STARK. It is not a popular program in New York?

Dr. FOREMAN. No, sir, it is not. Finally, the AAMC does not support payments being awarded directly to training programs. The Association believes that payments from the workforce account should be made to the entity that incurs the cost.

The second major fund is the academic health center account, and while the AAMC is pleased that the act would create a stream of support for academic health centers, the pool is seriously underfunded at $3.8 billion. Preliminary results from an analysis conducted by Lewin/VHI indicates that the real 1991 inpatient and outpatient cost difference between teaching and nonteaching hospitals, excluding the direct costs of education, was between $9 and $11 billion.

Looking back at that 1991 period, Medicare's $4.2 billion in indirect medical education expenditure served as a proxy payment for those costs. The balance was obtained by cross-subsidization from other payers or in rate regulated States, explicitly as payment adjustments. But with managed competition shrinking payments from other payers, it becomes increasingly difficult to sustain a cross-subsidy system. If the Medicare indirect medical education adjustment is eliminated October 1, 1995, as is proposed in the act, without a substantial enhancement of the academic health center adjustment, there will be a huge hole in the funding required to assure that Medicare beneficiaries and others have access to services provided by teaching hospitals.

The final concern of the academic community is that managed competition is likely to unravel the system of cross-subsidization through which faculty professional services income helps to underwrite medical education. Note on table 2 of my testimony that

83-037 0-94-7