SPEAKERS       CONTENTS       INSERTS    Tables

 Page 1       TOP OF DOC
43–165 CC
1997

TEACHING HOSPITALS AND MEDICARE DISPROPORTIONATE SHARE HOSPITAL PAYMENTS

HEARING

before the

SUBCOMMITTEE ON HEALTH

of the

COMMITTEE ON WAYS AND MEANS

HOUSE OF REPRESENTATIVES

ONE HUNDRED FIFTH CONGRESS

FIRST SESSION

MARCH 11, 1997

Serial 105–7

 Page 2       PREV PAGE       TOP OF DOC
Printed for the use of the Committee on Ways and Means

COMMITTEE ON WAYS AND MEANS
BILL ARCHER, Texas, Chairman

PHILIP M. CRANE, Illinois
BILL THOMAS, California
E. CLAY SHAW, Jr., Florida
NANCY L. JOHNSON, Connecticut
JIM BUNNING, Kentucky
AMO HOUGHTON, New York
WALLY HERGER, California
JIM McCRERY, Louisiana
DAVE CAMP, Michigan
JIM RAMSTAD, Minnesota
JIM NUSSLE, Iowa
SAM JOHNSON, Texas
JENNIFER DUNN, Washington
MAC COLLINS, Georgia
ROB PORTMAN, Ohio
PHILIP S. ENGLISH, Pennsylvania
JOHN ENSIGN, Nevada
JON CHRISTENSEN, Nebraska
WES WATKINS, Oklahoma
J.D. HAYWORTH, Arizona
 Page 3       PREV PAGE       TOP OF DOC
JERRY WELLER, Illinois
KENNY HULSHOF, Missouri
CHARLES B. RANGEL, New York
FORTNEY PETE STARK, California
ROBERT T. MATSUI, California
BARBARA B. KENNELLY, Connecticut
WILLIAM J. COYNE, Pennsylvania
SANDER M. LEVIN, Michigan
BENJAMIN L. CARDIN, Maryland
JIM McDERMOTT, Washington
GERALD D. KLECZKA, Wisconsin
JOHN LEWIS, Georgia
RICHARD E. NEAL, Massachusetts
MICHAEL R. McNULTY, New York
WILLIAM J. JEFFERSON, Louisiana
JOHN S. TANNER, Tennessee
XAVIER BECERRA, California
KAREN L. THURMAN, Florida

A.L. Singleton, Chief of Staff

Janice Mays, Minority Chief Counsel

Subcommittee on Health
BILL THOMAS, California, Chairman
 Page 4       PREV PAGE       TOP OF DOC
NANCY L. JOHNSON, Connecticut
JIM McCRERY, Louisiana
JOHN ENSIGN, Nevada
JON CHRISTENSEN, Nebraska
PHILIP M. CRANE, Illinois
AMO HOUGHTON, New York
SAM JOHNSON, Texas

FORTNEY PETE STARK, California
BENJAMIN L. CARDIN, Maryland
GERALD D. KLECZKA, Wisconsin
JOHN LEWIS, Georgia
XAVIER BECERRA, California

Pursuant to clause 2(e)(4) of Rule XI of the Rules of the House, public hearing records of the Committee on Ways and Means are published in electronic form. The printed hearing record remains the official version. Because electronic submissions are used to prepare both printed and electronic versions of the hearing record, the process of converting between various electronic formats may introduce unintentional errors or omissions. Such occurrences are inherent in the current publication process and should diminish as the process is further refined. The electronic version of the hearing record does not include materials which were not submitted in an electronic format. These materials are kept on file in the official Committee records.

C O N T E N T S

 Page 5       PREV PAGE       TOP OF DOC
    Advisories announcing the hearing

WITNESSES

    Prospective Payment Assessment Commission, Joseph P. Newhouse, Ph.D., Chairman; accompanied by Donald A. Young, M.D., Executive Director

    Physician Payment Review Commission, Gail R. Wilensky, Ph.D., Chair; accompanied by Anne Schwartz, Ph.D., Special Assistant to the Executive Director

    Harborview Medical Center, Seattle, Washington, and American Hospital Association, David Jaffe

    Hopper, Cornelius L., M.D., University of California

    Kern Medical Center, Bakersfield, California, and National Association of Public Hospitals & Health Systems, Gerald Starr

    Montefiore Medical Center, Bronx, New York, and Greater New York Hospital Association, Spencer Foreman, M.D

    Saint Francis Hospital and Medical Center, Hartford, Connecticut, and Association of American Medical Colleges, David D'Eramo

SUBMISSIONS FOR THE RECORD
 Page 6       PREV PAGE       TOP OF DOC

    American College of Preventive Medicine, statement

    Integrated Healthcare Association, Pleasanton, CA, statement and attachment

TEACHING HOSPITALS AND MEDICARE DISPROPORTIONATE SHARE HOSPITAL PAYMENTS

TUESDAY, MARCH 11, 1997
House of Representatives,
Committee on Ways and Means,
Subcommittee on Health,
Washington, DC.

    The Subcommittee met, pursuant to notice, at 12:35 p.m., in room 1310, Longworth House Office Building, Hon. Bill Thomas (Chairman of the Subcommittee) presiding.
    [The advisories announcing the hearing follow:]

    ADVISORY

FROM THE COMMITTEE ON WAYS AND MEANS

SUBCOMMITTEE ON HEALTH

CONTACT: (202) 225–3943
 Page 7       PREV PAGE       TOP OF DOC

FOR IMMEDIATE RELEASE

February 27, 1997

No. HL–5

Thomas Announces Hearing on Teaching

Hospitals and Medicare Disproportionate Share

Hospital Payments

    Congressman Bill Thomas (R–CA), Chairman, Subcommittee on Health of the Committee on Ways and Means, today announced that the Subcommittee will hold a hearing on teaching hospitals and Medicare disproportionate share hospital payments. The hearing will take place on Tuesday, March 11, 1997, in 1310 Longworth House Office Building, beginning at 1:00 p.m.
    
    In view of the limited time available to hear witnesses, oral testimony at this hearing will be heard from invited witnesses only. However, any individual or organization not scheduled for an oral appearance may submit a written statement for consideration by the Committee and for inclusion in the printed record of the hearing.
    
BACKGROUND:
    
 Page 8       PREV PAGE       TOP OF DOC
    Teaching hospitals provide critical services to Medicare beneficiaries and others as well as training health care providers and producing state of the art medical research. Costs of operating teaching hospitals tend to be higher than those in other hospitals, in large part due to their multiple missions related of patient care, education, and research. Medicare recognizes these higher costs through graduate medical education (GME) payments: direct GME and an indirect medical education (IME) adjustment. For fiscal year 1996, Medicare payments to teaching hospitals for direct GME were $2.2 billion and IME payments were $4.3 billion.
    
    The Balanced Budget Act of 1995 would have reformed Medicare payment for teaching hospitals and GME through the establishment of a trust fund to support these activities. Despite President Clinton's veto of the Balanced Budget Act of 1995, the Subcommittee remains committed to reforming Federal support for teaching hospitals and will examine the President's fiscal year 1998 budget proposals regarding GME in this context.
    
    In addition to GME payments, Medicare makes payments to hospitals that tend to serve a larger proportion of indigent Medicare beneficiaries. Referred to as the Medicare disproportionate share hospital (DSH) adjustment, payment is made through a percentage add-on to the hospital rate using Medicaid and Supplemental Security Income as a proxy. Because of changes in both Medicaid and welfare, the use of these factors to compute the disproportionate share payment requires reexamination. In recent years, Medicare DSH payments have increased nearly 400 percent, from $1.1 billion in fiscal year 1989 to $4.3 billion in 1996.
    
    In announcing the hearing, Chairman Thomas stated: ''Across the country, teaching hospitals have assumed an important role in the communities they serve. These institutions are currently being challenged by a changing health care system. They are rightly concerned over Medicare spending proposals. It is time to take a close look at the Federal contribution to these hospitals, and decide how that contribution can be better focused to maintain the best of what these institutions have to offer.''
 Page 9       PREV PAGE       TOP OF DOC
    
FOCUS OF THE HEARING:
    
    This hearing will focus on special payments to teaching hospitals for GME and Medicare DSH payments.
    
DETAILS FOR SUBMISSION OF WRITTEN COMMENTS:
    
    Any person or organization wishing to submit a written statement for the printed record of the hearing should submit at least six (6) copies of their statement and a 3.5-inch diskette in WordPerfect or ASCII format, with their address and date of hearing noted, by the close of business, Tuesday, March 25, 1997, to A.L. Singleton, Chief of Staff, Committee on Ways and Means, U.S. House of Representatives, 1102 Longworth House Office Building, Washington, D.C. 20515. If those filing written statements wish to have their statements distributed to the press and interested public at the hearing, they may deliver 200 additional copies for this purpose to the Subcommittee on Health office, room 1136 Longworth House Office Building, at least one hour before the hearing begins.
    
FORMATTING REQUIREMENTS:
    
    Each statement presented for printing to the Committee by a witness, any written statement or exhibit submitted for the printed record or any written comments in response to a request for written comments must conform to the guidelines listed below. Any statement or exhibit not in compliance with these guidelines will not be printed, but will be maintained in the Committee files for review and use by the Committee.
 Page 10       PREV PAGE       TOP OF DOC
    
    1. All statements and any accompanying exhibits for printing must be typed in single space on legal-size paper and may not exceed a total of 10 pages including attachments. At the same time written statements are submitted to the Committee, witnesses are now requested to submit their statements on a 3.5-inch diskette in WordPerfect or ASCII format.
    
    2. Copies of whole documents submitted as exhibit material will not be accepted for printing. Instead, exhibit material should be referenced and quoted or paraphrased. All exhibit material not meeting these specifications will be maintained in the Committee files for review and use by the Committee.
    
    3. A witness appearing at a public hearing, or submitting a statement for the record of a public hearing, or submitting written comments in response to a published request for comments by the Committee, must include on his statement or submission a list of all clients, persons, or organizations on whose behalf the witness appears.
    
    4. A supplemental sheet must accompany each statement listing the name, full address, a telephone number where the witness or the designated representative may be reached and a topical outline or summary of the comments and recommendations in the full statement. This supplemental sheet will not be included in the printed record.
    
    The above restrictions and limitations apply only to material being submitted for printing. Statements and exhibits or supplementary material submitted solely for distribution to the Members, the press and the public during the course of a public hearing may be submitted in other forms.
 Page 11       PREV PAGE       TOP OF DOC
    

    Note: All Committee advisories and news releases are available on the World Wide Web at 'HTTP://WWW.HOUSE.GOV/WAYS_MEANS/'.
    

    The Committee seeks to make its facilities accessible to persons with disabilities. If you are in need of special accommodations, please call 202–225–1721 or 202–225–1904 TTD/TTY in advance of the event (four business days notice is requested). Questions with regard to special accommodation needs in general (including availability of Committee materials in alternative formats) may be directed to the Committee as noted above.

—————


NOTICE—CHANGE IN TIME

    ADVISORY

FROM THE COMMITTEE ON WAYS AND MEANS

SUBCOMMITTEE ON HEALTH

CONTACT: (202) 225–3943

FOR IMMEDIATE RELEASE
 Page 12       PREV PAGE       TOP OF DOC

March 6, 1997

No. HL–5-Revised

Time Change for Subcommittee Hearing on

Tuesday, March 11, 1997,

on Teaching Hospitals and Medicare

Disproportionate Share Hospital Payments

    Congressman Bill Thomas (R–CA), Chairman of the Subcommittee on Health, Committee on Ways and Means, today announced that the Subcommittee hearing on Teaching Hospitals and Medicare Disproportionate Share Hospital Payments previously scheduled for Tuesday, March 11, 1997, at 1:00 p.m., in 1310 Longworth House Office Building, will begin instead at 12:30 p.m.
    
    All other details for the hearing remain the same. (See Subcommittee press release No. HL–5, dated February 27, 1997.)

—————


NOTICE—CHANGE IN ROOM
 Page 13       PREV PAGE       TOP OF DOC

    ADVISORY

FROM THE COMMITTEE ON WAYS AND MEANS

SUBCOMMITTEE ON HEALTH

CONTACT: (202) 225–3943

FOR IMMEDIATE RELEASE

March 6, 1997

No. HL–5-Revised

Room Change for Subcommittee Hearing on

Tuesday, March 11, 1997,

on Teaching Hospitals and Medicare

Disproportionate Share Hospital Payments

    Congressman Bill Thomas, (R–CA), Chairman of the Subcommittee on Health, Committee on Ways and Means, today announced that the Subcommittee hearing on Teaching Hospitals and Medicare Disproportionate Share Hospital Payments previously scheduled for Tuesday, March 11, 1997, at 12:30 p.m., in 1310 Longworth House Office Building, will be held instead in B–318 Rayburn House Office Building.
 Page 14       PREV PAGE       TOP OF DOC
    

    All other details for the hearing remain the same. (See Subcommittee press release No. HL–5, dated February 27, 1997.)

—————


    Chairman THOMAS. The Subcommittee will come to order. I want to welcome you to the Health Subcommittee's fifth hearing this year.
    Today we'll hear from witnesses about Medicare special payments to teaching hospitals and to those hospitals that treat a large share of poorer Americans. As you heard during the Subcommittee's two hearings on this topic last year, teaching hospitals obviously serve an important role in our health care delivery system in terms of developing technological innovations, pioneering medical research, and delivering care to Medicare beneficiaries and the less fortunate, besides the final training of our Nation's doctors and other health care providers.
    Since its inception, Medicare has assumed a substantial role in subsidizing teaching hospitals and, in turn, over the years these institutions have become quite dependent on the Medicare Program.
    Recently, ProPAC, the Prospective Payment Assessment Commission, reported that in 1995, Medicare payments to major teaching hospitals for inpatient services were nearly 20 percent higher than costs, the largest Medicare margin in nearly a decade. However, during the same period, total major teaching hospital revenues for all services and all payers were only 3.7 percent higher than hospital expenses.
    Today the Subcommittee will examine to what extent Medicare is subsidizing costs which benefit all of society, not simply those who participate in the Medicare Program.
 Page 15       PREV PAGE       TOP OF DOC
    The Congressional Budget Office's analysis indicates that Medicare is projected to subsidize hospitals that treat a larger share of indigent care by $26.4 billion over the next 5 years. The Subcommittee will explore recommended changes to the Medicare disproportionate share hospital payment and the impact of implementing those recommendations on hospitals in both urban and rural areas.
    The BBA, Balanced Budget Act of 1995, included what we consider to be bold changes in teaching hospital payment policies. It established a new financing structure which, in addition to Medicare funding, started at $17.5 billion and, leaving conference, was at $13.5 billion from general revenues.
    This year, Congress is reexamining Medicare's payment to teaching hospitals in the context of the BBA and the growing market pressures on both these institutions and the Medicare Program. I might add, parenthetically, a number of our colleagues have examined other methods of payment, including a so-called all-payers system, as well.
    While our discussions today will focus on the President's fiscal year 1998 budget proposal to reduce payments to teaching hospitals, I would ask the witnesses to be as creative as possible in their testimony so that the Subcommittee benefits from a broad policy discussion. And I believe you should lead, for example, ''Therefore.''
    The President has proposed a shift in home health services from part A to general revenues to extend the life of the trust fund. I won't go into the characterizations by prominent economists and others, but it was an attempt to extend the life of the trust fund by shifting specific costs from part A to the general fund.
    If we were to entertain such a shift, instead of specific Medicare benefits, wouldn't it make more sense to transfer the portion of Medicare part A that actually subsidizes a broad-based societal benefit, such as teaching hospitals and disproportionate share payments, from part A to general revenues so that we could not only extend the life of the trust fund with the transfer, but transfer broad-based societal benefit payments rather than specific benefit payments.
 Page 16       PREV PAGE       TOP OF DOC
    I look forward to an open discussion in this important area. I would particularly like to welcome one of our second panel members—I didn't see him earlier; Jerry Starr? There he is. Only because he represents a 270-bed county hospital in my hometown and he's going to get back home before I do—yes, if they still are in business when we get back. That's a comment from the gentleman from California, but you ain't far off the mark, unfortunately.
    Thank you for being here. I understand the size of this hearing room—I still believe it's better than B–318. We may do a rotation on the seating periodically or something that will allow everybody to sit for a portion of the hearing.
    And with that, I'll yield to my colleague from California, Mr. Stark, for his opening statement.
    Mr. STARK. Well, thank you, Mr. Chairman. Like so many of the issues we're facing, I think your proposed solutions and the conundrums you raise are ones that this Subcommittee will have to address.
    My interest in graduate medical education is more, I guess, because we're trying to keep hospitals that provide high quality care for Medicare and Medicaid uninsured than the idea that we have to use taxpayer dollars to encourage people to join the highest 1 percent of salaried elite in our society. We do need good teaching hospitals as safety net hospitals, but there is a lot of evidence we're paying too much for this education, certainly for too many specialists.
    New York's North Shore University Hospital is getting more than $147,000 a year per resident, and that might explain why their program is growing rapidly. But cutting teaching hospitals just puts pressure on the public hospitals who do a good job treating the poor and the uninsured.
    And, to compound this issue, we've got to do something with the disproportionate share formula through nobody's fault, really, but through a change in the way much medical care is being delivered. For example, I think I'm correct in assuming that in Tennessee, where you suddenly have Tenn Plan, every hospital qualifies as a disproportionate share hospital. Arguably, we need a new formula to figure out which hospitals really are under pressure from the uninsured and the underinsured and what kind of a proxy to use to recalculate that.
 Page 17       PREV PAGE       TOP OF DOC
    So we have these two programs that need coordination and some decision by this Subcommittee as to how we'll allocate scarce resources among them.
    I share your admonition to the witnesses that we need some new ideas because we're dealing with really a different way to calculate this. I think everybody agrees we know who needs help. How do we identify them in simple law so that we don't have to go on a hospital-specific basis?
    Thanks very much.
    Chairman THOMAS. Thank you. Not only identify them but we need to figure out the most appropriate way to supply that subsidy.
    The first panelists are not new to us, painfully not new to us. Dr. Newhouse with his Executive Director, Dr. Young, and Dr. Wilensky. Dr. Newhouse is from the Prospective Payment Assessment Commission and Dr. Wilensky is from the Physician Payment Review Commission, accompanied by Dr. Schwartz.
    So Dr. Newhouse, Dr. Wilensky, your written testimony will be made a part of the record and you can address us in any way you see fit, as long as you're creative in your solution options.

STATEMENT OF JOSEPH P. NEWHOUSE, PH.D., CHAIRMAN, PROSPECTIVE PAYMENT ASSESSMENT COMMISSION; ACCOMPANIED BY DONALD A. YOUNG, M.D., EXECUTIVE DIRECTOR

    Mr. NEWHOUSE. Thank you very much, Mr. Chairman. It's a pleasure to be back with you this afternoon to talk about teaching and disproportionate share payments.
    There are two types of teaching payments: Indirect medical education payments, which are designed to compensate teaching hospitals for the sicker patients they treat, for the broader range of technologies they offer; and direct medical education payments, which are designed to pay Medicare's share of costs of training residents and the faculty that teach.
 Page 18       PREV PAGE       TOP OF DOC
    The Disproportionate Share Program is designed to ensure access to care by the Medicare population to those hospitals that treat a large number of low-income patients and are under severe and growing financial pressure.
    The amount of moneys here are substantial. In 1996 the indirect and direct medical education payments were just under $7 billion and disproportionate share was just over $4 billion.
    As you see in my first chart, there are about 1,000 teaching hospitals that receive the teaching payments, about 2,000 hospitals that receive the disproportionate share payments, and a little under 700 that receive both. These tend to be the larger hospitals, and they account for a disproportionate share of the payments.
    Hospitals generally are in an environment that's increasingly price competitive. It's harder for the hospitals to subsidize low-income patients and teaching missions to the degree they had before, and there is no mechanism in the at-risk program for getting the teaching and disproportionate share payments to the hospitals that they're designed for.
    The teaching and disproportionate share payments, as you mentioned in your opening statement, have led to very high PPS, prospective payment system, margins. My chart 2 shows that major teaching hospitals have a PPS margin that's just under 19 percent, and large urban disproportionate share hospitals have a PPS margin just over 16 percent on their Medicare patients. But if you look at my chart 3, the story is just the opposite on total margins. The large major teaching hospitals have the lowest margin of any teaching hospitals or nonteaching, and the large disproportionate share, large urban, also have the lowest margins, 3.5 percent against 6, 7, and 8 percent for other groups.
    Now, the teaching hospital payments, if you look at chart 4, have been growing in the last 6 years. The indirect payment grew from $2.5 to $4 billion and the direct or GME payment grew from $1.5 to over $2 billion.
 Page 19       PREV PAGE       TOP OF DOC
    This has, in part, reflected an increase in the total number of residents because both of these programs increased their payments approximately in proportion to the number of residents. As you see in my chart 5, the number of residents has been pretty steadily increasing since the beginning of the prospective payment system. From 1990 on, it went up 18 percent and that led to approximately an 18-percent increase in these payments.
    Now, we have a number of recommendations for you. On the teaching side, I think our most important recommendation is that the link between the number of residents at a hospital and the amount of money it gets from Medicare be severed. And how those moneys would be distributed could be done in a variety of ways that we can talk about in the questions and answers, but if a hospital hired another resident we would not give it more money.
    Similarly, on the indirect side, we would reduce the amount of the markup that Medicare pays slowly toward its so-called empirical level; that is, the rate at which hospital costs empirically go up with a measure of teaching. We recommend starting out with a reduction from the current 7.7 to 7 percent. The empirical level is around 4 percent.
    Finally, under the current methodology, only full-time equivalent residents in the hospital are included in the payment calculations, and we would like to be flexible enough to support training outside the hospital and in freestanding facilities.
    Finally, we recommend, as you alluded to in your opening statement, a broader based financing mechanism that could be financed in a number of ways, and maybe we can talk about that in the questions and answers.
    On the disproportionate share side of the ledger, we have several recommendations. The current program pays hospitals according to the number of low-income patients it treats, but the number of low-income patients is merely reflected in the number of Medicare SSI patients and Medicaid patients. We would recommend you broaden that to include a much more comprehensive measure of low-income patients, which we believe can be done with a fairly minimal addition to hospitals' administrative burden.
 Page 20       PREV PAGE       TOP OF DOC
    Another major change we would make in disproportionate share payments is to set a threshold where the hospital would receive payments at the same level for urban, rural, and various size hospitals. You see in my chart 8 that hospitals with similar low-income patients percentages now get quite different payments because the thresholds are set at quite different levels.
    Finally, as I mentioned, we would pay hospitals for treating low-income patients that are currently in the at-risk program, and we would pay the teaching payments also if those continue to be tied to Medicare patients or patients that are in the at-risk program. This would require hospitals know that a Medicare patient is in the at-risk program. We believe that can be done.
    I'll look forward to your questions, Mr. Chair.
    [The prepared statement follows:]

Statement of Joseph P. Newhouse, Ph.D., Chairman, Prospective Payment Assessment Commission

    Good afternoon, Mr. Chairman. I am Joseph Newhouse, Ph.D., Chairman of the Prospective Payment Assessment Commission (ProPAC). I am accompanied by Donald Young, M.D., the Commission's Executive Director. We are pleased to be here to discuss Medicare's teaching and disproportionate share (DSH) policies. During my testimony, I will refer to several charts. These charts are appended to the end of my written testimony.
    As you know, Medicare makes additional payments to teaching hospitals and hospitals that serve a disproportionate share of low-income patients to recognize their special missions. Teaching payments are intended to recognize the roles teaching hospitals play in conducting medical research, developing technological innovations, and training the future health care work force. There are two distinct types of teaching payments, each with a different purpose and payment methodology. The indirect medical education (IME) adjustment to payments under Medicare's Prospective Payment System (PPS) is intended to compensate teaching hospitals for the additional patient care costs they incur as a result of treating a more complex patient population, developing and utilizing new technologies, and offering a broad range of sophisticated services. The second teaching payment, the direct graduate medical education (GME) payment, is intended to pay for Medicare's share of costs associated with operating a residency program. It covers costs such as residents' salaries and benefits and faculty supervision, as well as the portion of the residency program's overhead allocated to the hospital.
 Page 21       PREV PAGE       TOP OF DOC
    Medicare pays hospitals that serve a disproportionate share of low-income patients an additional amount to ensure access to care for the Medicare population. These hospitals frequently are the only source of care for the patients that use them, and they are under intense financial pressures because of their willingness to serve the poor. The DSH adjustment to PPS payments helps to alleviate these pressures and preserves the availability of these hospitals to both poor and non-poor Medicare beneficiaries.
    Medicare spending though these special payments is not inconsequential. In 1996, IME and GME payments to teaching hospitals were about $6.8 billion, and DSH payments were about $4.3 billion. About 45 percent of all hospitals paid under Medicare's prospective payment system received teaching or DSH payments in 1996, and these facilities accounted for about two-thirds of PPS discharges.
    Mr. Chairman, it is important to keep in mind that there is substantial overlap in the hospitals that receive teaching and DSH payments. Hospitals that have both teaching and disproportionate share classifications receive about 75 percent of total IME payments and 65 percent of total DSH payments. These hospitals benefit financially from receiving both types of special payments, thus changes to these policies will also have the greatest impact on them.
    These hospitals also are facing a financial environment that is increasingly price-competitive, which threatens their ability to fulfill their teaching and low-income service missions. Medicare makes explicit payments to support these hospitals under its fee-for-service program but, as I will discuss later in my testimony, there is no mechanism to distribute teaching and DSH payments under the risk contracting program. These enrollees currently represent 11 percent of the Medicare population, and this share is growing rapidly. As it grows, and the fee-for-service population necessarily declines, Medicare support to these hospitals will erode if no change is made.
    This afternoon, I would like to first describe Medicare payments to and use of teaching and DSH hospitals, as well as their financial performance. Then, I will discuss Medicare's policies specific to each of these payment types and the Commission's recommendations to improve these policies. These recommendations are discussed in more detail in our Report and Recommendations to the Congress, which we released on March 1.
 Page 22       PREV PAGE       TOP OF DOC

PPS Payments and Financial Conditions of Teaching and DSH Hospitals

    More than 1,000 teaching hospitals receive Medicare payments under PPS. These hospitals accounted for an estimated 43 percent of PPS discharges and 53 percent of PPS operating payments in fiscal year 1996 (see Chart 1). Facilities with the most intensive teaching activity—at least one resident for every four hospital beds—are referred to as major teaching hospitals. While these institutions represented only 5 percent of all PPS hospitals in 1996, they accounted for 11 percent of discharges and 19 percent of payments in 1996.
    Almost 2,000 hospitals receive Medicare DSH payments, accounting for an estimated 51 percent of discharges and 58 percent of PPS payments in fiscal year 1996. DSH hospitals located in large urban areas accounted for a quarter of all discharges and received one-third of all PPS payments, even though they represent only 15 percent of all hospitals.
    As I noted, almost 700 hospitals receive both teaching and DSH payments. These institutions comprise about two-thirds of all teaching hospitals and one-third of all DSH hospitals. They accounted for 27 percent of all PPS discharges and 36 percent of all PPS payments, as well as 65 percent of total DSH payments and 75 percent of IME payments.
    With the additional monies provided through the special payments, Medicare has more than adequately compensated teaching and DSH hospitals for the costs of treating Medicare patients. ProPAC analyses show that, since the first year of prospective payment in 1984, teaching and DSH hospitals' PPS margins—which compares Medicare's inpatient operating and capital payments to the corresponding costs—have exceeded those of other hospitals, and these differences have widened over the years (see Chart 2). In 1995, the average hospital had an estimated PPS margin of 7.9 percent. Major teaching hospitals had the highest average PPS margin, 18.6 percent; large urban DSH hospitals followed closely with a margin of 16.2 percent. As I mentioned earlier, there is a large overlap between these two groups.
 Page 23       PREV PAGE       TOP OF DOC
"The Official Committee record contains additional material here."

STRIP OFFSET FOLIO 1 HERE
"The Official Committee record contains additional material here."

STRIP OFFSET FOLIO 2 HERE
    Another indication of the impact of teaching and DSH payments on hospitals' performance under Medicare is the PPS margins for hospitals that receive neither of these payments. About 55 percent of hospitals meet this criteria, and in 1995 their average PPS margin was 0.5 percent. These hospitals, however, account for only 33 percent of discharges.
    Our analyses also indicate, however, that the overall financial condition of teaching and DSH hospitals is much less robust than their PPS margins might suggest. In 1995, the average total margin for all hospitals—which measures the difference between hospitals' total revenues and total expenses—was 5.6 percent. Major teaching hospitals had total margins of 3.7 percent. DSH hospitals in large urban areas had the lowest total margins of any group we examined, 3.6 percent (see Chart 3). Even though these margins are relatively low, they are still the highest total margins for these hospitals in the past 10 years, and higher than during the period before PPS began.
    Teaching and DSH hospitals have total margins that are lower than those of other hospitals for several reasons. These include the provision of large amounts of uncompensated care and their difficulties in obtaining additional revenues from private payers to support the extra costs associated with the special roles they serve.
"The Official Committee record contains additional material here."

STRIP OFFSET FOLIO 3 HERE
 Page 24       PREV PAGE       TOP OF DOC

Medicare Teaching Payment Policies

    As I mentioned, Mr. Chairman, Medicare's teaching payment policies are intended to recognize the multiple missions of teaching hospitals. In addition to providing patient care, these missions include conducting medical research, developing technological innovations, and helping to ensure there is a well-trained physician work force for the future. Medicare spending to support the special roles served by teaching hospitals has grown rapidly over the past several years, increasing 63 percent between 1990 and 1995 (see Chart 4).
    One of the primary factors driving both IME and GME spending growth is increases in the number of residents. Currently, both IME and GME payments rise approximately in proportion to the number of residents at each hospital. Hospitals, therefore, have an incentive to train more physicians. In fact, since 1990, the number of residents has increased by 18 percent. Historically, the increase in the number of specialty residencies has far exceeded the rise in primary care physicians. Over the past few years, however, the growth in primary care has been greater. The number of residents who are graduates of international medical schools also has risen rapidly over the past several years (see Chart 5).
    The Commission recommends modifying the IME and GME polices so that Medicare's payment policies do not influence hospitals' decisions concerning the number of residents they train. In addition, we also believe the teaching payments should be available to those organizations that incur the costs of training, but that policies be flexible enough to allow training in other settings when appropriate. The current payment methods are biased towards training in the hospital setting, even though training in other settings may be increasingly necessary for future practicing physicians.
    I would like to take a moment to briefly summarize the views of the Commission related to each of the teaching adjustments.
 Page 25       PREV PAGE       TOP OF DOC
"The Official Committee record contains additional material here."

STRIP OFFSET FOLIO 4 HERE
"The Official Committee record contains additional material here."

STRIP OFFSET FOLIO 5 HERE

IME Payment Policy

    As I mentioned earlier, teaching hospitals generally have higher patient care costs than other hospitals. The Commission agrees that teaching hospitals should receive additional payments that are associated with these higher costs, but Medicare's payments for these costs are higher than can be justified by their teaching status.
    The IME adjustment is a per discharge percentage add-on that is applied to Medicare PPS payments. Since 1989, PPS payments for Medicare fee-for-service discharges have been adjusted upwards by about 7.7 percent for every 10 percent increase in teaching intensity, measured using a ratio of residents to beds. ProPAC analyses indicate, however, that Medicare operating costs per discharge are only 4.1 percent higher for each 10 percent increase in teaching intensity. The Commission recommends that for fiscal year 1998, the IME adjustment should be lowered to 7.0 percent; this would mean a 9.1 percent decrease in each hospital's IME payments. We believe that teaching hospitals should be able to adjust to this reduction without compromising their service functions. Ultimately, we think the adjustment should more closely correspond to the actual relationship between teaching intensity and costs.
    The Commission also recommends that the IME payment method be changed so that payments do not fluctuate with variations in the number of residents or beds. Currently, IME payments are based on the number and mix of Medicare patients, as well as teaching intensity. We believe payments should continue to vary based on patient volume and case-mix, but that the relationship between payments and changes in the number of hospital residents should be modified. This could be accomplished in a number of ways, including maintaining each hospital's teaching intensity at its level for the current year level or some previous year. Over time, however, it may be necessary to revise the level for each hospital to reflect substantial changes in the size of residency programs.
 Page 26       PREV PAGE       TOP OF DOC
    Alternatively, only a portion of any change in the number of residents could be recognized for payment purposes. Under this policy, hospitals that increased their number of residents would not be paid in full for the additional residents; but, at the same time, hospitals that reduced their residency positions would not lose the entire amount associated with that reduction.
    Finally, under the current IME methodology, only full-time equivalent residents that train in the inpatient or outpatient areas of the hospital are included in the payment calculations. We believe the payment method should be flexible enough to support training in settings outside of the hospital, including free-standing ambulatory facilities, and managed care organizations.

GME Payment Policy

    Direct graduate medical education (GME) payments cover Medicare's share of residents' salaries and benefits, the services of supervising physicians, and the general operating costs of running hospital residency programs. The amounts hospitals receive are based on their costs per resident in a base year (for most hospitals this is 1984) updated to the current year using the Consumer Price Index. These amounts vary widely. In 1993, per resident payment amounts ranged from around $20,000 to over $100,000 (see Chart 6). Medicare's share of these amounts is determined according to the proportion of the hospital's total inpatient days accounted for by Medicare fee-for-service beneficiaries.
    As with the IME adjustment, the Commission recommends that GME payments not change based on variations in the number of residents a hospital trains. The current system discourages hospitals from reducing the size of their residency programs when it is appropriate to do so because they stand to lose a relatively large payment for each resident. Removing the direct link between payments and the number of residents a hospital trains would weaken the adverse incentives associated with the current method.
 Page 27       PREV PAGE       TOP OF DOC
    To accomplish this, Medicare could consider replacing the current per resident payment amount with a lump-sum payment for resident costs that would be based on the hospital's historical share of GME spending. Alternatively, the current method could be retained but revised so that a change in the number of residents would result in a smaller than proportionate change in payments. In this way, hospitals would still receive partial payments for residency slots they eliminated and not receive full payment if they decide to take on additional residents. At some point in the future, however, Medicare GME payments would need to be adjusted to reflect changes in the number and distribution of residency positions.
    The Commission also recommends that GME policies be modified to provide financial support for training residents in sites outside the hospital, such as ambulatory care clinics and health maintenance organizations. Hospitals currently have some leeway in this regard, but only if they own the ambulatory facility and incur the resident costs. There are no provisions to pay other types of providers directly for resident training. The Commission believes GME payments should be as neutral as possible concerning the site where residents receive training.
"The Official Committee record contains additional material here."

STRIP OFFSET FOLIO 6 HERE

Establishing a Broader-Based Financing Mechanism for Teaching Hospitals

    Mr. Chairman, teaching hospitals are the flagships of the most widely respected health care system in the world, yet the Medicare program is the only payer to make explicit payments nationwide to support their teaching and research missions. While other payers have implicitly helped fund these costs though higher prices for inpatient care services, these purchasers are becoming less willing to pay more to these facilities in an increasingly price-conscious environment. The Commission believes separate funding mechanisms should be developed that would include contributions from a broader range of sources. While there would be a number of technical issues to decide, explicit financial support for medical education and teaching hospital costs would enable these facilities to compete with other hospitals for patients and enhance efficiency, while supporting their multiple missions.
 Page 28       PREV PAGE       TOP OF DOC

Disproportionate Share Payment Policy

    Since 1986, Medicare has made special payments to PPS hospitals that treat a disproportionate share of low-income patients. DSH payments increased almost four-fold between fiscal years 1989 and 1996, from $1.1 billion to $4.3 billion (see Chart 7). Almost 2,000 hospitals receive DSH payments, but the top 250 hospitals account for about half of the total, and about 95 percent of all DSH payments go to urban hospitals.
    Similar to the IME adjustment, DSH payments are distributed through a percentage add-on to Medicare fee-for-service discharge payment rates. The DSH adjustment is determined by a complex set of formulas that incorporate information about a hospital's location, size, and the proportion of low-income patients it treats. The low-income share is calculated by summing two ratios: Medicaid patient days as a share of total patient days, and patient days for Medicare patients eligible to receive Supplemental Security Income (SSI) cash payments as a share of total Medicare patient days. Hospitals do not receive DSH payments unless the low-income share meets a threshold. This threshold varies for different hospital groups, ranging from 15 percent for large urban hospitals to 45 percent for small rural hospitals.
"The Official Committee record contains additional material here."

STRIP OFFSET FOLIO 7 HERE
    Mr. Chairman, as Medicare seeks to reduce the overall growth in payments, it is especially important that available DSH funds be targeted to those hospitals most in need of financial support because of their role in providing access to Medicare beneficiaries. The current method is increasingly inadequate to achieve this goal. A primary reason is because the formula relies, in part, on Medicaid utilization to identify low-income patients. The Medicaid utilization measure has never been a good overall indicator of service to the poor because the portion of the low-income population covered by Medicaid varies markedly from state to state. As states implement Medicaid and welfare reforms, this disparity is likely to be exacerbated. The recently enacted welfare law also changes eligibility for SSI payments, which will affect the distribution of DSH payments under the current formula.
 Page 29       PREV PAGE       TOP OF DOC
    The DSH formula is flawed also in that it emphasizes hospitals' location and size and excludes other measures of identifying care to the poor, primarily uncompensated care. Under the current method, hospitals that have similar low-income patient shares have vastly different payment adjustments depending upon whether they are located in urban or rural areas and how many beds they have (see Chart 8). For example, urban hospitals with 100 or more beds and rural hospitals with 500 or more beds that have low-income percentages of 45 percent receive a 26 percent payment add-on per discharge, while smaller urban and rural hospitals with similar low-income burdens receive an adjustment that is one-fifth that much.
    In addition, many of these hospitals provide similar levels of uncompensated care, but this is not accounted for in the current formula (see Chart 9). Public major teaching hospitals provide the greatest amount of this care and generally receive the highest Medicare DSH adjustments. But, on average, hospitals in rural areas provide a similar proportion of uncompensated care as a share of total costs as hospitals in large urban areas, yet the rural hospitals receive much lower DSH add-ons from Medicare.
"The Official Committee record contains additional material here."

STRIP OFFSET FOLIO 8 HERE
"The Official Committee record contains additional material here."

STRIP OFFSET FOLIO 9 HERE
    The Commission believes that the current DSH payment method should be changed. Our March report sets forth a number of details to guide the development of a new formula. Briefly, the Commission recommends that DSH payments be determined using a measure based on providers' costs of treating poor patients regardless of geographic location. In the Commission's view, such a measure would be the best way to determine the amount of low-income care hospitals furnish. Under our proposal, the DSH measure would include the costs associated with Medicaid patients and Medicare patients who receive SSI payments, but it also would include the costs of care furnished to patients in other indigent care programs and hospitals' uncompensated care costs. These costs as a share of the hospital's total patient care costs would reflect the proportion of resources the hospital devotes to caring for the poor.
 Page 30       PREV PAGE       TOP OF DOC
    Revising the DSH formula would target payments more precisely, thus providing the most financial help to those hospitals that serve the greatest share of poor patients. It also would avoid the current inconsistencies arising from differences in Medicaid eligibility requirements across states. To implement the Commission's proposal would require collecting some additional data from hospitals, but this requirement could be met without substantially increasing hospitals' current reporting burdens.

Teaching and DSH Payments for Hospitals That Treat Risk Enrollees

    Mr. Chairman, teaching and DSH special payments currently apply only within Medicare's fee-for-service program. Over 4 million beneficiaries, however, are enrolled in Medicare's risk contracting program, and this number is expected to increase rapidly over the next few years.
    When risk plan enrollees are treated in a teaching or DSH hospital, the risk plan pays the hospital a mutually agreed upon payment; there is no requirement that risk plans pay teaching or DSH hospitals additional amounts to support the missions associated with these institutions. Often, hospitals must accept lower prices from risk plans or jeopardize their Medicare patient base. Even in those cases where a risk plan may pay higher prices to certain hospitals, these payments may not explicitly correspond to the IME, GME, and DSH payments received under the fee-for-service program.
    The Commission believes that hospitals that receive teaching and DSH payments under the fee-for-service program should receive comparable payments for the risk enrollees they treat. Several approaches could be taken to develop a mechanism to distribute these payments. One option, for example, would have hospitals submit their billing information for risk patients to Medicare. Medicare could then compute teaching and DSH payments based on a calculation of payments the hospital would have received if the patients had been in the fee-for-service program. Regardless of the approach, however, it would be important that hospitals be able to identify patients that are enrolled in a risk contracting plan.
 Page 31       PREV PAGE       TOP OF DOC

Conclusion

    Mr. Chairman, this nation's teaching and disproportionate share hospitals fulfill unique roles in our health care system. As Medicare payments are necessarily constrained to preserve the Hospital Insurance Trust Fund, it is imperative that Medicare's teaching and DSH policies are structured to ensure that Medicare payments are appropriate to support the missions served by these hospitals.
    This concludes my formal statement, Mr. Chairman. I would be pleased to answer any questions from you or other members of the Subcommittee.

—————


    Chairman THOMAS. Thank you, Doctor.
    Dr. Wilensky.

STATEMENT OF GAIL R. WILENSKY, PH.D., CHAIR, PHYSICIAN PAYMENT REVIEW COMMISSION; ACCOMPANIED BY ANNE SCHWARTZ, PH.D., SPECIAL ASSISTANT TO THE EXECUTIVE DIRECTOR

    Ms. WILENSKY. Thank you, Mr. Chairman, for inviting PPRC, the Physician Payment Review Commission, to share its views.
    There will be three chapters in PPRC's 1997 report that are relevant to today's hearings: The changes in the health care market and its effect on academic medical centers, how the changes in the market have affected physician supply and specialty mix, and the implications of financing graduate medical education and teaching hospitals from a trust fund.
 Page 32       PREV PAGE       TOP OF DOC
    As you know, since 1990 PPRC has looked at issues relating to the supply and specialty distribution of physicians and the Medicare financing of graduate medical education; that is, the direct portion of graduate medical education.
    What I'd like to do in the brief time I have in this oral testimony is to highlight some of the issues as the Commission sees them and then to go back and spend just a few minutes talking about the purpose of Medicare funding. Some of the changes that are being discussed or you may be considering or were being considered in the last session of Congress goes to the heart of why it is that Medicare funds graduate medical education. We think before you make changes that issue needs to be revisited.
    First, some of the issues. One is that payments are based on the number of residents and this, of course, creates an incentive to increase the number of residents being trained. You've already heard mention of this. In figure 1 of PPRC's testimony, we show the very substantial increase that has occurred in the number of residents being trained. It is partly because of an increase in the length of training that has occurred. It is also partly because of the increase in the number of international medical graduates.
    The second point is that the payments under direct medical education are based on institution-specific historic costs; more specifically, based on what a hospital was showing as its costs in 1984–85, updated for inflation. That has produced very variable payments, from a low of $11,000 per resident to more than $100,000 per resident.
    In 1993 the Commission recommended that there be a standardized amount paid. We have not changed from that recommendation.
    The next issue is that the payments are made primarily for training in inpatient settings. I think you all know there has been substantial movement to take health care outside of inpatient settings. Medicare policy doesn't reflect these changes that have gone on in medicine. And furthermore, hospitals are not accountable for how they spend that money. There's an issue of fungibility and that is one that, again, gets raised once you look at the notion of paying only for inpatient settings.
 Page 33       PREV PAGE       TOP OF DOC
    A fourth issue is the one that Dr. Newhouse has already mentioned concerning managed care, which is that the structure of the payment occurs so that the teaching hospital does not receive payment directly, at least, when a managed care enrollee is admitted to a teaching hospital. PPRC has recommended in the past and continues to recommend that the dollars be removed and another mechanism be found to distribute those moneys so that you can use the teaching facility or also so that you can sponsor training in managed care settings, but to do so in a more directed way.
    And a final issue we think needs to be raised has to do with the way Medicare supports nursing education. Medicare is usually discussed in terms of medical education with regard to its impact on academic health centers and physician training, but of course Medicare is a very important source of money for the training of nurses. Some $250 million went to nurse training in the last year and some other $83 million to other allied professionals. But it only goes for those who are trained at the diploma level and not at the graduate level.
    When Medicare was enacted in 1965, that made some sense. About 77 percent of the nurses were in diploma programs. It no longer makes a great deal of sense. In fact, as of 1990, less than 8 percent of the nurses were in diploma programs and that is projected to go down even further.
    Let me spend a few minutes talking about the broader issue, which is the purpose for Medicare's financing graduate medical education. These are very big issues and I obviously can only touch upon a few of them. There are at least three reasons why Medicare might wish to support direct medical education payments. This is a philosophical and practical discussion that we had at one of our recent Commission meetings.
    One is to make sure seniors have access to teaching hospitals. Although it does not appear that that is a problem, that was initially one of the driving forces behind having Medicare payments go for teaching.
 Page 34       PREV PAGE       TOP OF DOC
    A second is to ensure the viability of teaching hospitals. As Dr. Newhouse has indicated, the margins for the major teaching hospitals, because of Medicare, has, in fact, been very high, although their overall margins are not high. And it is clear that without this very substantial subsidy by the Medicare Program, these major teaching hospitals would have had some financial difficulties.
    And the third justification is to support the production of physicians to meet the needs for our seniors. And I just want to summarize a couple of the findings that are included in some detail in this year's report.
    The bottom line, as I think you know, is that it doesn't appear the production of physicians is exactly a problem in the United States. In fact, it looks like we have, if anything, a substantial excess supply, particularly in terms of some of the specialists.
    PPRC has done some analysis of what is going on in the physician market and in the market for residents, as an economist looks at these issues. It does look like the physician labor market is changing but the changes are more modest than the anecdotes would suggest.
    There was a decline in income 2 years ago. There was an increase in terms of overall specialties but overall, in the last 2 years, there's still been some decline in physician income.
    It also appears there's a mixed picture with regard to the choice between generalism and more specialized fields. It is easier for people who are in primary care residencies to get jobs when they finish. But it also appears there's still a very strong demand for specialists, and residents in specialty fields are finding jobs, although with some difficulty.
    We have some trouble looking at specific markets. Our data are usually available only at the national level. We know that some of the changes are very specific in terms of their geographic impact, and we think it's important to look there, as well. But in terms of being worried about whether there will be an adequate physician supply to take care of the physician population, that does not seem to be an issue.
 Page 35       PREV PAGE       TOP OF DOC
    We think it is important Congress think through in a very clearly articulated way why it is that Medicare is supporting graduate medical education and disproportionate share funding. That may lead you to some different policy conclusions than what was built in the program in 1965.
    Thank you. I'd be glad to answer any questions.
    [The prepared statement and attachments follow:]

Statement of Gail R. Wilensky, Ph.D., Chair, Physician Payment Review Commission

    Mr. Chairman and members of the Subcommittee, I appreciate the opportunity to be here this afternoon to present the Physician Payment Review Commission's views concerning Medicare payment to teaching hospitals. As you know, the Omnibus Budget Reconciliation Act of 1990 expanded the Commission's mandate to include consideration of the supply and specialty distribution of physicians, and Medicare financing of graduate medical education (GME). We have been working on these issues ever since.
    The Commission's 1997 annual report to Congress will include three chapters relevant to today's hearing:
    •  how changes in the health care market have affected academic medical centers,
    •  how changes in the market have affected physician supply and specialty mix, and
    •  the implications of financing graduate medical education and teaching hospitals from a trust fund.
    My testimony today primarily focuses on the last of these but I will touch on the others where relevant. My remarks focus on three key areas:
 Page 36       PREV PAGE       TOP OF DOC
    •  problems associated with current policy that the Congress may wish to address,
    •  the importance of clarifying policy objectives, and
    •  the implications of financing GME and teaching hospitals from a trust fund.
    Although the Commission's 1997 report will not make recommendations on GME financing, there is much that we can offer as you begin to deliberate this issue. First, some of the recommendations we made in 1993 are still relevant (for example, those concerning financing of training in ambulatory settings). Second, in this report, we provide the Congress with an framework for discussing changes in GME financing. We also have some analytic activities planned for later this spring that could assist this subcommittee as it debates changes in Medicare policy.
    Before I begin, let me clarify how the PPRC and the Prospective Payment Assessment Commission have divided responsibilities in these areas where our mandates overlap. PPRC has primarily concerned itself with Medicare payments for the direct costs of graduate medical education (residents' compensation, faculty supervision, and overhead), leaving consideration of the indirect medical education adjustment (IME) to ProPAC. This is because the indirect adjustment was not designed to support teaching per se. Rather it was meant to compensate teaching institutions for their relatively higher costs attributable to teaching and patient mix. The PPRC has not considered issues related to disproportionate share hospital (DSH) payments.

Current Medicare Policy

    Graduate medical education is largely financed through patient care revenues generated by hospitals. The federal government is the largest single explicit financing source through the Medicare program. Other payers have less explicit mechanisms. Teaching hospital charges to private payers reflect the direct costs of GME (for example, residents' stipends) although these payers do not identify and separately pay for them.
 Page 37       PREV PAGE       TOP OF DOC
    In Medicare's early years, GME was funded like other hospital services on a retrospective, reasonable cost basis. With the adoption of prospective payment, new policies were needed to ensure equitable payment for teaching hospitals. The costs of graduate medical education are now recognized under two mechanisms:
    •  direct medical education payments to hospitals for residents' stipends, faculty salaries, administration, and overhead allocated to residency programs; and
    •  an indirect medical education adjustment.
    In the early years of prospective payment, direct medical education payments were essentially a pass-through of costs related to training. Since the passage of the Consolidated Omnibus Budget Reconciliation Act of 1985 (COBRA), direct payments have been based on three factors:
    •  hospital-specific per resident costs from the 1984 or 1985 cost-reporting years updated for inflation,
    •  the number of full-time equivalent residents, and
    •  Medicare's share of inpatient days.
    Under OBRA93, the inflation adjustment for residents in nonprimary care fields was eliminated for fiscal years 1994 and 1995. As a result, the per resident amount differs for primary care and other residents.
    The hospital's per resident base amount is multiplied by the number of full-time equivalent (FTE) residents during the payment period times Medicare's share of inpatient days. Residents beyond the initial period of residency (up to the minimum for board eligibility, not to exceed five years) are weighted as 0.5 FTEs. There is also a special exception that allows residents in accredited geriatrics training programs to be considered as 1.0 FTE for an additional 2 years.
    The indirect medical adjustment, by contrast, is a hospital-specific percentage amount (based on the ratio of interns and residents per bed) added to each DRG payment. As I mentioned just a moment ago, the IME adjustment was developed to compensate teaching institutions for costs associated with teaching such as the involvement of residents in patient care, and the severity of illness of patients who require the specialized services available only in teaching hospitals.
 Page 38       PREV PAGE       TOP OF DOC

Problems With Current Medicare Policy

    The Commission has identified five critical problems with Medicare's current methods of financing graduate medical education. Policymakers should be concerned that Medicare payments for the direct costs of graduate medical education are:
    •  based on the number of residents in a teaching hospital,
    •  based on institution-specific historical costs,
    •  made primarily for training in inpatient settings,
    •  structured so that teaching hospitals do not receive payment when enrollees of Medicare managed-care plans are admitted to these hospitals, and
    •  structured to support nursing education in diploma schools, rather than at the graduate level.
    Let me describe in a little more detail why the Congress might choose to address these issues.

Payment Based on the Number of Residents

    Medicare payment for graduate medical education is proportional to the number of residents training in each institution. For direct medical education payments, there is a payment for each full-time equivalent resident (FTE). For the indirect adjustment, the add-on to per case payments increases as the ratio of interns and residents to beds in an institution increases.
    Under both mechanisms, more residents result in more Medicare dollars flowing to the institution. Although the number of residents is not strictly a function of Medicare policies, these policies do create an incentive for hospitals to organize patient care services around residents rather than other caregivers (for example, nurse practitioners, physician assistants or staff physicians). In recent years, the number of residents has increased dramatically (Figure 1). This growth reflects both increases in the length of training and the growing number of positions filled by international medical graduates.
 Page 39       PREV PAGE       TOP OF DOC
    Medicare's rules could be changed so that payment is no longer proportional to the number of residents. Under the Medicare restructuring legislation proposed in the last Congress, payments to hospitals would have been based on their historical share of direct and indirect payments. The implication is that hospitals that increase the number of trainees do so at their own expense; those that decrease the number of trainees are held harmless for this action. Implementing such changes will require methods of periodically rebasing payments so that hospitals do not receive payments into perpetuity based on the number of residents they had at one point in time.

Direct Medical Education Payments Are Highly Variable

    Because Medicare payments for the direct costs of graduate medical education are based on institution-specific historical costs, per resident payments vary substantially across hospitals, from a low of $11,000 to more than $100,000 per resident. This variation may reflect differences in the true costs of training. But it also reflects differences in accounting practices (for example, how overhead costs are allocated across departments) and inaccuracies in measuring the number of full-time equivalent residents. Other differences, for example, whether or not faculty physicians are employees of the hospital, also affect training costs but there are questions about how much of this variation Medicare should support.
    Variation in direct payments could be addressed by paying hospitals a standardized amount per resident as recommended by the Commission in 1993. Important implementation questions concern whether this payment should be simply the national average payment, the national average adjusted by differences in local wage rates, specialty or other factors, or whether it should be derived by other means.

 Page 40       PREV PAGE       TOP OF DOC
Payment to Hospitals

    A third problem with current policy is that payments may only be made to hospitals, despite the fact that much of physician practice now takes place in ambulatory settings. This emphasis on inpatient care also does not provide residents with the competencies they will need to practice in managed care. Although residents do spend time in ambulatory settings, time spent in ambulatory sites and compensation for outpatient faculty is recognized only for the purposes of Medicare direct medical education payment if the hospital ''incurs all or substantially all'' of the costs of training.
    This policy does not reflect the changes that have occurred in medicine since Medicare was enacted. Medical care no longer takes place primarily in hospitals. While educators have changed curricula to provide more training experiences outside the hospital, Medicare's financing mechanisms pose an obstacle to broader changes. Paying primarily for time spent in inpatient settings creates a disincentive to move training to other sites such as outpatient clinics, nursing homes, physician group practices, and managed-care plans.
    It is also worth noting that hospitals are not held accountable for the use of the Medicare payments they receive. These dollars are fungible; once they flow into the hospital, they can be used for any purpose, not just the training of physicians. No empirical data exist to examine how hospitals are using these funds.
    There are several ways to address these concerns. The disincentive to moving training outside the hospital could be removed by permitting time in other settings to be counted for the purposes of either direct or indirect payments. For example, the Administration's fiscal year 1998 budget proposal would count work in nonhospital settings for the purposes of indirect payments. Or payment could be extended to entities other than hospitals for the time residents spend in those settings. This is the approach recommended by the Commission back in 1993. It is also included in the Administration's budget proposal which would extend GME payments to federally qualified health centers for primary care residents when a hospital is not paying for the residents' salary in that setting. A complication in setting such payments is the lack of information about the costs of training in ambulatory sites.
 Page 41       PREV PAGE       TOP OF DOC
    A third alternative, that would both permit training to be moved out of the hospital and introduce more accountability, would be to make payments for the full direct cost of medical education to training programs or consortia rather than hospitals. This approach raises a number of implementation issues including how much should be paid and the measure of Medicare activity for determining Medicare's share of costs.

Lack of Medicare Payment for Medicare Managed-Care Enrollees

    Growth of Medicare's risk contracting program has raised two additional concerns about current policy for financing GME and teaching hospitals. First, as I described in testimony before this subcommittee two weeks ago, risk-plan payments are based on fee-for-service expenditures at the county level which include both GME and DSH payments. The effect of including these payments is shown in Figures 2 and 3. Overall GME and DSH payments account for about 5.5 percent of Medicare capitation rates although the share varies across the country. In effect, managed-care plans are receiving payment for activities they are not obligated to support.
    Second, Medicare does not make additional payments to teaching institutions when they serve Medicare managed-care enrollees. Payments for direct costs are currently based on the notion that Medicare should pay its share of GME costs with the share determined by the percentage of Medicare discharges. But the count of discharges does not include Medicare managed-care enrollees.
    INSERT OFFSET FOLIOS 10 TO 11 HERE
    [The official Committee record contains additional material here.]

—————


 Page 42       PREV PAGE       TOP OF DOC
    As I mentioned at the previous hearing on Medicare managed care, the Commission recommends that medical education payments be removed from calculation of capitation payments, and that other methods be developed to pay for managed-care enrollees' use of teaching facilities or for training in managed-care sites. This approach was taken by the Senate in the last session of Congress although it was dropped in the conference agreement on Medicare restructuring. The final bill provided for payment to teaching institutions admitting Medicare managed-care enrollees but funded these payments from general revenues, rather than extracting them from capitation payments.

Nursing Education

    In addition to supporting the costs of GME, Medicare also provides cost-based reimbursement to hospitals for operating training programs in nursing and a variety of allied health fields including cytotechnology, dietetics, hospital administration, inhalation therapy, medical records administration, medical technology, nurse anesthesia, occupational therapy, physical therapy, pharmacy, and X-ray technology. In 1994, Medicare paid an estimated $248 million for nursing education and $83 million for allied health education under this provision.
    Medicare support for nursing education primarily supports diploma programs because of its origins as a mechanism to ensure equitable payment to hospitals for the costs they incur in treating Medicare patients. The Congress may wish to reconsider, however, as to whether the program should be supporting these programs given changes in nursing education and practice. When Medicare was enacted, 77 percent of nurses were trained in hospital-operated diploma programs. By 1990, however, less than 8 percent of graduates were from diploma schools. Moreover, the annual percentage of diploma-prepared graduates is expected to decline further, dropping to about 2.5 percent by the year 2020.
 Page 43       PREV PAGE       TOP OF DOC
    Nursing education at the graduate level receives no Medicare support despite the potentially greater need for nurses trained as nurse practitioners and clinical nurse specialists. These disciplines did not even exist when the Medicare program was first implemented and lack of support cannot be interpreted as a conscious policy decision.

The Purpose of Medicare Financing

    Policymakers are now focusing on how to make Medicare's policies more rational, equitable, and less of an impediment to change as educators seek to adapt GME to changes in the nation's health and its health system. In the last Congress, this committee and others proposed placing Medicare revenues in a trust fund from which payments could be made to hospitals, training programs, consortia, or medical schools based on principles other than hospital costs and admissions. Since these proposals did not become law, the Commission assumes that they form the starting point for debate in this Congress.
    There are many ways to reallocate Medicare dollars, none of which is inherently correct. What matters is what Congress considers the purpose of such support. All other decisions (for example, who can get the money and how much they are eligible to receive) should flow from a common understanding about these purposes.
    There are at least three different reasons why it may be appropriate for the Medicare program to provide special support for teaching hospitals (Table 1):
    •  to ensure beneficiary access to teaching hospitals,
    •  to ensure the viability of teaching hospitals, and
    •  to support the production of physicians to meet Medicare beneficiaries' needs.
    A case can also be made that there is no compelling rationale for continuing Medicare support.
 Page 44       PREV PAGE       TOP OF DOC

Table 1


Maintaining Beneficiary Access to Services in Teaching Institutions

    The report accompanying the 1965 Social Security Amendments notes that the original intent of Medicare's support of the special costs incurred by teaching institutions was to ensure that beneficiaries could receive care in the best hospitals, those where training also occurred. Medical research costs, by contrast, were not to recognized by Medicare because these were not closely related to normal patient care, and there was ample funding from other sources.
    This rationale of tying Medicare dollars to activities benefiting Medicare patients remains evident in current policy despite the changes that occurred with the adoption of prospective payment. Under both direct and indirect formulas, Medicare only pays hospitals that admit Medicare beneficiaries, and determines its share based on the number of bed days in each institution.
    Does beneficiary access to teaching hospitals remain a compelling policy concern? Despite the many changes in the health care system, public perceptions still often tend to equate teaching status with high quality care. Many complex services, however, such as open-heart surgery and organ transplants, that were previously available exclusively in academic settings, are now provided in nonteaching hospitals where costs are typically lower and outcomes are often comparable.
    It is important to note that under managed care, decisions regarding whether to use a teaching hospital are made primarily by plans, not patients. Although there are no good data on the extent to which Medicare risk contractors use teaching hospitals, the assumption is that most managed-care plans will use teaching hospitals only if comparable services are unavailable elsewhere or if including these providers in their networks makes the plan more marketable. It may be difficult to argue, therefore, that Medicare support is critical to ensuring beneficiaries' access to teaching institutions when there are no such assurances of access for its managed-care enrollees.
 Page 45       PREV PAGE       TOP OF DOC

Ensuring the Viability of Teaching Hospitals

    A second rationale for Medicare support of teaching hospitals is that the nation's academic medical centers are unique resources which have laid the groundwork for important medical advances, ensured the quality of medical education and the competence of practicing physicians, and continue to provide a wealth of specialized health services.
    Beginning with recognition of the costs of clinical education as reasonable costs, Medicare policies have been designed to ''level the playing field'' so that teaching hospitals are not disadvantaged in a competitive environment by a payment system that does not account for all the relevant factors that explain their higher costs.
    This rationale may have greater relevance now as academic centers face competitive pressures arising from growth in managed care. Given the increasingly cost conscious health environment and declining use of hospital services, at issue is how much federal support for academic medical centers is warranted and whether recognition of these costs by Medicare in the form of direct and indirect medical education payments is an appropriate vehicle. Those arguing for cutbacks in federal support point out that since the prospective payment system was implemented, teaching hospitals' Medicare inpatient margins have exceeded those of other hospitals. In fiscal year 1995, major teaching hospitals (those with at least 0.25 residents per bed) had the highest Medicare inpatient margin at 18.6 percent (Table 2). Margins for other teaching hospitals were 7.6 percent and those of nonteaching hospitals were 3.7 percent.

Table 2


    Moreover, teaching hospitals are taking steps to respond to the changing health care environment. They are developing new sources of revenue; creating new organizations to negotiate managed-care contracts; building networks of providers through mergers, acquisitions, and contract arrangements; and taking steps to differentiate their products. They are also working to reduce costs and improve operational efficiency.
 Page 46       PREV PAGE       TOP OF DOC
    Total margins, as opposed to Medicare margins, however, suggest the pressures now being faced by teaching institutions (Table 2). Major teaching hospitals have the lowest total margins at 3.7 percent (compared to 6.5 percent for nonteaching hospitals), reflecting a combination of heavy uncompensated care burdens, inefficiencies, and payments from many payers that do not recognize the full incremental costs of education. Medicare's special payments to teaching institutions therefore are viewed as a thin buffer from powerful market forces that do not value missions beyond patient care.

Supporting the Production of Physicians to Meet Beneficiary Needs

    A third rationale for Medicare funding of GME is that the program should support the production of physicians sufficient to ensure that Medicare beneficiaries have access to appropriate medical care. Since there are no other identifiable financing sources for physician training, public financing may be needed to ensure that Medicare beneficiaries have access to physicians. If these costs were not subsidized, potential candidates might decide to enter other professions. Public support of GME thus ensures that there is a sufficient supply and mix of physicians to meet beneficiary health care needs.
    This rationale is consistent with concerns at the time of Medicare's enactment about an impending physician shortage. Policymakers responded to these concerns by offering construction grants to health professions schools that agreed to increase enrollment, providing direct funding to schools based on enrollment, and extending loans and scholarships to students. Residency programs also expanded rapidly during this period. Between 1970 and 1989, the number of training programs increased by nearly 40 percent; the number of positions offered almost doubled.
    Today, however, the concerns are different. One school of thought is that whatever the problems in physician supply and specialty mix, these are better left to the market. In its 1995 and 1996 annual reports, and once again in this year's report, the Commission has examined how changes in the market for health services are actually affecting the types of employment opportunities available to physicians. Despite the anecdotes we have all heard, the empirical data do not suggest that the market is leading to substantial changes in the work force. While there are some signs that specialty mix may be changing in response to market demand for primary care physicians, there are also signs of continued strong demand for physicians in highly specialized fields. In addition, despite common beliefs to the contrary, many indicators do not reflect an oversupply of physicians. For example, physician incomes continue to grow (although at lower rates than in the past) (Table 3). Interest in medicine as a career is at an all-time high.
 Page 47       PREV PAGE       TOP OF DOC

Table 3


    A second school of thought is that Medicare funding could be used to address work force concerns if subsidies were targeted, rather than open-ended. Rather than ending Medicare support for clinical education because it has led to the overproduction of physicians, Medicare could attach conditions to its dollars to take some of the steps on which there is broad agreement—for example, training fewer physicians overall, providing training experiences in outpatient settings, and training physicians to locate in underserved areas. Based on this rationale, appropriate policy responses might include paying for fewer trainees (as proposed by the Administration), basing direct payments on a national average, and permitting use of funds in settings other than hospitals and hospital-owned outpatient sites.
    Some have also argued that Medicare subsidies are necessary to ensure that disadvantaged students are able to select medicine as a career. That is, if residents were required to pick up the costs of training themselves, only the privileged could become doctors. Broad support for GME is not an effective strategy for encouraging diversity in the physician work force, however. A more effective approach would be to fund individuals, rather than institutions, either through Medicare or through the Public Health Service's grants and scholarship programs.

Implications of Trust Fund Financing

    Finally, let me describe some of the implications of financing GME and teaching hospitals from a trust fund (Table 4). First, the existence of a trust fund creates the opportunity to correct all the problems I discussed earlier. Spending can be oriented for specific purposes, such as training in ambulatory settings or for training in specific specialties. Or it can be structured to be neutral with respect to the type and setting of training.

Table 4


 Page 48       PREV PAGE       TOP OF DOC

    A second implication is that the trust fund shifts financing from an open-ended entitlement to a fixed amount of funding. Under current law, the amount paid per resident is fixed but total funding increases with increases in the number of residents. A trust fund could be structured so that the total amount of funding is fixed, and the payment per resident is variable (for example, decreasing as the number of residents grows).
    Third, financing from general revenues creates a new political dynamic for stakeholders. Given concerns about the solvency of Part A trust fund (from which direct and indirect payments are made), infusion of general revenues would insulate teaching institutions from stringent Medicare cuts. For teaching hospitals, however, reliance on general revenues is a risky strategy. Currently, Medicare payments are an entitlement; there is always enough money for those eligible. General revenues, however, are subject to the annual appropriations process and could be whittled away over time as teaching institutions are forced to compete with others for scarce discretionary funds.
    Finally, the existence of a trust fund creates an administrative structure that could be used for accepting and distributing revenues from other payers. Although I am aware that the current Congress does not appear inclined to compel other payers to contribute to GME financing, leaders in academic medicine see creation of a federal trust fund as an important first step in securing broader support.
    At least some of the policy goals envisioned under previous trust fund proposals could be accomplished by more modest changes in Medicare policy than creation of a trust fund. For example, a trust fund is not necessary for: paying based on factors other than the number of residents; recognizing the costs of training in ambulatory settings; paying standardized per resident amounts, paying teaching hospitals when a Medicare managed-care enrollee receives care in their institutions, or paying for graduate clinical education of nurses. It is also possible to reduce the number of trainees receiving public support without creating a trust fund. This may more easily accomplished (administratively and politically) with a trust fund, however, because explicit decisions do not have to be made about who not to fund. Instead if the number of trainees increases above the desired level, amounts paid per resident drop.
 Page 49       PREV PAGE       TOP OF DOC

Conclusions

    In the months ahead, the Commission will continue exploring the implications of making payments from a GME and teaching hospital trust fund and will be providing you with guidance about different policy alternatives. Issues to be considered include which entities should be eligible to receive trust fund payments and methods to ensure their accountability, the implications of paying hospitals based on their historical share of Medicare dollars, other options for allocating trust fund revenues, and how to set allocations to accommodate change over time in where training occurs. The implications of some alternatives, for example, the distributional consequences at the hospital level of changing Medicare payments, will be modelled using data from Medicare cost reports and the Health Care Financing Administration's Intern and Resident Information System.

—————


    Mrs. JOHNSON [presiding]. Thank you.
    Dr. Schwartz.
    Ms. WILENSKY. She's just here to accompany me.
    Mrs. JOHNSON. Thank you very much for your testimony this morning. Everyone has testified? OK.
    I'd like to elicit the panel's comments on any work or thinking any of you have done on how we would distribute payments from a trust fund for medical education. If we reroute dollars into a trust fund, how do we distribute them? In the Republican bill last year, we used a very crude mechanism. We did it on the basis of the preceding year's Medicare patient census, but you can't keep doing that.
 Page 50       PREV PAGE       TOP OF DOC
    And particularly in a world of mergers and changes and where your long-term goal is to reduce the number of residencies you are subsidizing, how would you distribute that money from the trust fund?
    Ms. WILENSKY. We have some ideas about how you would not distribute it that I would suggest first and then, again, some general principles. PPRC does not have specific recommendations, although I know the Institute of Medicine is going to be delivering a report soon that will look at those issues.
    Our concerns have been that you not link the payments to the number of residents being produced, an issue that Dr. Newhouse also raised; and that while initially looking at historical shares is a way that you can begin, it will clearly not make sense to continue doing so in the long term. Finally, when you make the payments, you should allow for payments to go outside just the inpatient setting, either to institutions more broadly defined or to institutions and consortia.
    Again, let me emphasize the general point, which is to the extent that payments were made, in part, to encourage physicians in general to have graduate training and particularly to make sure that opportunities were available to low-income or disadvantaged physicians to have graduate training, the payment to institutions is a crude or clumsy way to go about it. Some of the funds might be redirected, to the extent that that was the concern, to make sure physicians in certain specialties or physicians from certain disadvantaged backgrounds are, in fact, able to continue their specialty training. The institutional basis of the payment should be rethought.
    Mr. NEWHOUSE. Our Commission did not take a position on how to do this. I would distinguish for you between the indirect medical education payments and the direct medical education payments.
    On the indirect side, I would suggest some historical basis of the kind you used, with some intent to revisit that some years down the line. If there is a broader or so-called all-payer mechanism established, then I would use some measure of all patients, not just Medicare patients.
 Page 51       PREV PAGE       TOP OF DOC
    On the direct medical education side, as Dr. Wilensky noted, there's a tremendous amount of variation across the country in these payments. Our Commission had some discussion of that. Some of that variation, of course, is due to the fact that different institutions have different proportions of Medicare patients. Some of it is because the cost allocations for the faculty side are sometimes contributed and sometimes salaries are paid. If salaries were paid, then Medicare picked up its share of those salaries.
    But this goes back to 1984, trended forward, as Dr. Wilensky noted, and what was right in 1984 may not be right now.
    So my personal bias is to support some slow move toward uniformity, but that is my personal view, not the view of the Commission.
    Mrs. JOHNSON. Thank you.
    I'd like to conclude my questioning by asking if you have any comments on the administration's proposal for the New York demonstration project.
    Ms. WILENSKY. These are comments of myself personally and not a PPRC recommendation. It is likely, in my view, that ultimately when a full package of reform is put together that providing some support at the same time you try to encourage a downsizing of the number of residents might be a part of the final package.
    My concern about the New York demonstration is twofold. The first is I'm not sure we'll ever know what happened because it's not set up in the normal way that research design is set up with a control, an experimental group, and a clear assessment strategy.
    But even more than that, I'm afraid you're giving away everything and not getting enough in return. The problems with graduate medical education are far greater than simply the number of residents being trained and the mix between generalists and specialists. You also need to address the whole issue of the payments and their variation, as well as the major philosophy about why you're doing it and what you want to get in return.
 Page 52       PREV PAGE       TOP OF DOC
    I think providing this kind of payment at this point without getting enough in return is a mistake, but do not question so much the idea of providing some support in a transition period. I think that probably will be a part of an ultimate package of policy changes.
    Mr. NEWHOUSE. Our Commission did not take any position on the New York demonstration, so these are my own views.
    When or if the link between the number of residents and Medicare payments is severed or partially severed, I think it's important that that ultimately lead to some savings to Medicare or to the government, and the New York demonstration provides that over time by phasing down the payments that would be made if residency slots are cut.
    So while I don't want to comment on the specifics of the phasing down, that general notion, I think, is consistent with the way to go.
    The other thing I would point out about the New York demonstration is that to the degree that it allows hospitals to save some money, not have their payments cut completely, by reducing residents, it's consistent with the spirit of our recommendation, but the New York demonstration is entirely voluntary.
    So the spirit of our recommendations would be to go something like the spirit of the New York demonstration but do it for everybody. As it is, the New York hospitals, the ones that want to reduce residents anyway, will gain by this and the ones that don't want to reduce residents or want to increase can just stay under the old system.
    Mrs. JOHNSON. Thank you very much.
    Thank you, Mr. Chairman.
    Chairman THOMAS. The other problem I have with a ''demonstration project'' is when it includes two-thirds of the universe, that is the problem. When you're looking at it, it makes it fairly difficult to call it a demo project. I think a relief project is more appropriate, and I don't know whether you do that without running through us, in terms of a policy change.
 Page 53       PREV PAGE       TOP OF DOC
    Does the gentleman from California wish to inquire?
    Mr. STARK. Thank you, Mr. Chairman.
    In this residency stuff, it just seems to me we have a lot of approximations out there. I think it's probably fairly certain the residents don't get much money. They get maybe $20,000 or $30,000.
    Ms. WILENSKY. $40,000.
    Mr. STARK. $40,000?
    Ms. WILENSKY. But not $120,000.
    Mr. STARK. It runs $20,000 to $40,000.
    Now, the hospitals want to sign them up. Some of them have more residents than they have beds, right? So you can just put a resident in every bed and let them hug the patient.
    If residents would walk into hospitals and say, ''Here I am; I'll do this free; use me; train me and use me; I'll volunteer,'' leaving the subsidies out of this for 1 minute, does anybody know what it would cost a hospital, on average, hard costs, out-of-pocket costs, to train them? Assuming they're getting something back, I presume there's some benefit which the resident adds, some productivity; is that correct? Some service or resource that——
    Ms. WILENSKY. That the residents provide?
    Mr. STARK. Is that correct?
    Ms. WILENSKY. Oh, absolutely.
    Mr. STARK. OK. So they're contributing $20,000 to $40,000 to the hospital. Now, how much does it cost them to supervise them? On the hospital's side, what is it? Does anybody know that or have any studies?
    Ms. WILENSKY. There have been studies that have been done but they have varied widely.
 Page 54       PREV PAGE       TOP OF DOC
    Mr. STARK. This is not with the hospital accountant loading in the limousine for the hospital administrator and all that sort of thing, but just what——
    Ms. WILENSKY. I'd be glad to provide what we know about it.
    Mr. STARK. Make a guess.
    Mr. NEWHOUSE. We can provide some information on that.
    Mr. STARK. Double?
    Dr. YOUNG. We have some information from the Medicare Cost Report. Now, this is as was reported by hospitals. What we did was break their costs that they report into three categories. The first category is resident salaries and fringe benefits——
    Mr. STARK. Yeah, but these guys are contributing, now.
    Dr. YOUNG. That's right. The second category is physician supervision and other costs related to that, and the third category is allocated overhead—the heating bill and all of that.
    Mr. STARK. Let me ask you this. How far off would I be if I said the allocated overhead is marginal?
    Dr. YOUNG. It is a legacy of cost-based reimbursement. It is marginal.
    Mr. STARK. I understand that but in reality, if you sent a cost accountant in——
    Dr. YOUNG. Under today's payment systems it is marginal.
    Mr. STARK. OK. Supervision makes sense. What I'm trying to get a handle on is, What does the resident contribute? What is the value of their service? If there were some way to calculate all this we might get to—Gail, you suggest a universal national fee?
 Page 55       PREV PAGE       TOP OF DOC
    Ms. WILENSKY. I think we ought to regard this as a training service delivery effort. We ought to basically be netting out the difference between what the physicians are contributing and what they receive in teaching. And for those that we think can't or wouldn't be able to receive additional training, because they are too poor or don't have access to basically tuition kinds of moneys, that a program be used to target for those individuals.
    That is a clear function. It's not clear to me, in the long term, what Medicare's continuing role is to the institutions. In the interim, there are a lot of reasons to want to support——
    Mr. STARK. No, I understand. It's a proxy for paying extra money to these institutions which are very expensive. Whether they're worth it or not is——
    Ms. WILENSKY. Well, there's no historical precedent that I'm aware of in other areas besides graduate medicine for the Federal Government to be paying institutions directly of this nature.
    Mr. STARK. You guys are going to come up with a formula, right? No? To do this?
    Mr. NEWHOUSE. We'll try to respond to what you asked.
    Mr. STARK. The other side of this coin is the DSH stuff. I want to come at this just a little bit. I think that Gail, you said you want to give the rural hospitals a little more DSH, right? A little bit? Did you say that in your testimony?
    Ms. WILENSKY. Today?
    Mr. NEWHOUSE. I said that.
    Mr. STARK. You said that—Joe, OK. You wouldn't pay that to the cost basis rural hospitals, I believe. Is that correct?
    Mr. NEWHOUSE. We view the DSH payment as a way of helping hospitals that have large low-income patients, and we would set the threshold at——
 Page 56       PREV PAGE       TOP OF DOC
    Mr. STARK. Then you're telling me that this is going to cause a minimal reduction in payments to urban and a little bit higher reduction in suburban areas, correct?
    Mr. NEWHOUSE. Depending on the details of the scheme, yes.
    Mr. STARK. All right. So basically you could say that the DSH payments are flat. Fair?
    Mr. NEWHOUSE. I'm not sure I follow you.
    Mr. STARK. A little bit more for the rural, a little bit less for the urban, and a little bit less for the suburban.
    Mr. NEWHOUSE. Yes.
    Mr. STARK. Now, the problem is that the uninsured people are going up pretty rapidly. There are probably 40 to 41 million uninsured. I don't know; it depends on how they count them. That's one-fifth of the people under age 65 uninsured. Twenty-five percent of New York residents, they tell us, are uninsured.
    If we keep the DSH payment constant, how do we take care of the uninsured?
    Mr. NEWHOUSE. We did not make a recommendation on the total size of the pot.
    Mr. STARK. But I'm just saying, as a practical matter——
    Mr. NEWHOUSE. Our proposed targeting method would help your problem because we would include the uninsured in the low-income count that triggers the DSH payment, whereas the current formula does not. It only——
    Mr. STARK. How are you going to count that in Tennessee?
    Mr. NEWHOUSE. I'm going to count Medicaid plus uninsured plus indigent care. Basically count all except privately insured and regular Medicare——
 Page 57       PREV PAGE       TOP OF DOC
    Mr. STARK. But you won't know in Tennessee. They're all on Tenn Plan.
    Mr. NEWHOUSE. Oh, I'm sorry. Medicaid will also count. It's not that I'm going to get rid of Medicaid. I'm just going to——
    Mr. STARK. How are you going to know?
    Mr. NEWHOUSE. It won't matter because they'll either be Medicaid or they'll be uncompensated care.
    Mr. STARK. Or low income.
    Mr. NEWHOUSE. Or low income. I'll count all of them.
    Mr. STARK. But then every hospital that has Tenn Care people is—it's just those hospitals that have the most will get DSH?
    Mr. NEWHOUSE. Yes, correct.
    Mr. STARK. OK.
    Ms. WILENSKY. One of the problems, of course, is trying to use part A Medicare wage tax money that's supposed to go for certain functions of inpatient——
    Mr. STARK. If we treated the Defense Department the way we're talking about treating Medicare, we'd have to close down the Defense Department and privatize it to Wackenhutt & Pinkerton.
    Now, we are never going to have enough tax revenue to pay for the Defense Department and the gap and the budget deficit that's being created there—this trust fund figment—we're obligated to pay for the medical care of seniors and we aren't going to stop that, in spite of any Chicken Little ideas that the sky is falling, and people are trying to address the problem.
    I would rather have it be dedicated, obviously, but I don't much care if it comes out of general revenues or not. I'd sure as hell rather do that than the ignition facility at Lawrence Lab for $20 billion or something that we know won't work. At least we have some idea that medical care will cure people who are sick.
 Page 58       PREV PAGE       TOP OF DOC
    I don't know. I'm more concerned about how we figure out how to fairly reimburse the hospitals to which the uninsured gravitate because they can get into an emergency room and they can't get care someplace else.
    And this proxy that we've used, we've all accepted. It was probably very inefficient when we weren't being so chintzy or forced to be so tight with a dollar, so maybe that's good for us, but then we've got to find a more accurate way to determine who's providing uncompensated care and try and, I think, reimburse them so that these people truly do get care.
    You could say we could block grant the States but are we sure they'll do it? I don't think the States can afford it. I think we have to continue to help. It's up to you guys. You're the experts here, the wonks in this world. Come up with some formulas.
    Ms. WILENSKY. Well, in part, the frustration is the split in this case, in disproportionate share payments, between Medicare and Medicaid makes it very difficult to have a rational policy for taking care of uninsured people.
    Mr. STARK. I hear you but that's what we've been doing, right, since—well, when did DSH come in?
    Mr. NEWHOUSE. In 1986.
    Ms. WILENSKY. It became a serious Medicaid issue in 1989.
    Mr. STARK. I don't know. We can argue about how much it'll be here, because that's a thing relative to how much we're going to spend on B–2 bombers, but once we decide, you guys should be able to help us up our comfort level that we're going to get it to the people we intend to get it to. That's what we're asking you for.
    Thank you, Mr. Chairman.
    Chairman THOMAS. I thank the gentleman.
    I apologize. I'm going to be in and out. I've got another Subcommittee that I have to deal with, as well.
 Page 59       PREV PAGE       TOP OF DOC
    This whole direction bothers me and bothers me a lot and I think you know it. I'm not someone who's been in this field either as a chosen profession or for years and years. So there are a number of disadvantages to that but frankly, the primary advantage is I look at what people have been doing and I say, ''This is ridiculous.'' After a while I'm beginning to hear back from other folk and the one answer I don't want to hear is, ''We do it this way because we do it this way.''
    I happen to believe there were very real political reasons why we began funding the way we funded. There may continue to be very real political reasons based upon which groups are receiving a subsidy.
    I think that's a legitimate debate issue that should be carried out and not simply locked in and argued that we are ''funding'' medical education, and I'll use disproportionate share first as an example.
    What it does, in my opinion, is inhibit our ability to fund graduate medical education, especially in today's marketplace, with the innovative delivery of medicine in different locales and different structures.
    The fact that we have very large, very important hospitals in urban centers that provide medical services to particular profiles of individuals who otherwise might not be able to pay for them is a different question, and I think it's clearly becoming more a different question every year, as we look at the size of the commitment in terms of the subsidy.
    In addition to that, the more traditional ways of funding graduate medical education, as we've seen, don't make a lot of sense.
    But what bothers me most, and in my opening remarks—my colleagues were not here when I said it—is our inability or refusal to get out of the box. All we do is create miniadjustments—the demonstration project is one of those—as we continue the current structure.
 Page 60       PREV PAGE       TOP OF DOC
    What is your reaction to the idea I mentioned in my opening remarks that if the President is bound and determined to buy some years on the part A trust fund by transferring costs that are currently in part A to part B, it makes more sense, to me, to transfer a broad-based societal funding benefit, like graduate medical education, than it does home health care. Especially if he's going to transfer it without attachment to the premium or to the copay which existed prior to 1980, it's simply a 100-percent drain on the general fund.
    I think a higher calling on the general fund, rather than a specific benefit to beneficiaries, would be the funding of graduate medical education.
    Now, we proposed in the Balanced Budget Act a separate general revenue fund and asked you folk and others, within 3 years by creating a commission, to give us a better way of funding education. The question of whether we should continue to subsidize any particular group at an appropriate amount, increase it or decrease it, is a separate question.
    By moving it to the general fund, we can begin to engage, in my opinion, in the discussion about who gets what, when and how, which will lead us to changes that, frankly, we need to discuss—regarding subsidizing low income. Do we subsidize it through the Medicaid structure or through graduate medical education? Obviously, I think the choice is defining the group and assisting it.
    But when we talk about general fund drain and who's getting it, a subsidy to particular groups, it might lead us into examination of those retired millionaires who are getting 75 cents on every dollar.
    It may also lead us to the argument that we have a group of people who are receiving no health insurance and we have another group of people who are receiving an open-ended opportunity for medical insurance because of the Tax Code, and that we need to reexamine those who are enabled by the Tax Code to get open-ended benefits and those who, through no fault of their own but by accident of where they work, get no benefits.
 Page 61       PREV PAGE       TOP OF DOC
    And it just seems to me we can't forget about those other groups getting open-ended and only concentrate on how we are able, then, to find additional resources to provide for those who don't benefit enough to get an open-ended benefit, try to move them to that position, but to take all of our resources, put it together, figure out a rational reallocation so that everyone gets something, and then figure out how to move everyone up, rather than an open-ended benefit for one group and no benefit whatsoever for another.
    And the first step, to me, in getting there, in our jurisdiction, is to take what we're currently spending out of a dedicated payroll trust fund to aid hospitals who deliver medical care to certain patients and reimburse this crazy relationship between beds and residents and get it into the general fund with a commitment, over a very short period of time—3 years, let's say—to take the dollar amounts, because we're going to have to continue to redistribute them under the system that we now have in place, and get it so that you get a distinct and discrete funding of medical education and a separate funding of medical services, subsidizing particular groups.
    What's wrong with that?
    Ms. WILENSKY. As an individual policy wonk, I find your logic compelling. As Chair of PPRC, PPRC has no official position. I, in fact, do think you've raised some very important issues.
    The one hesitation or qualification would be the one that you yourself referenced, which is if the switch of home care into part B is done without being subjected to the normal constraints of part B, then there really isn't any good reason for not swapping one for the other. If it were to be subjected to the normal constraints, then you would be taking a part and putting it into general revenue without that.
    But given that that is not what the President has proposed, then I think the rationale you've just offered is compelling.
 Page 62       PREV PAGE       TOP OF DOC
    Chairman THOMAS. And the amount that's currently coming out of part A, in the area that I described, is close to $70 billion.
    Ms. WILENSKY. It's greater. I thought it was closer to $80 billion.
    Chairman THOMAS. I'm trying to get the exact numbers. It is close enough to say how ironic that if you want to transfer something, you can accomplish what the President said he wanted to by moving these broader-based societal funds and get the same length on the trust fund but, frankly, a higher calling on the general fund to fund medical education than to pay for particular benefits to particular beneficiaries.
    It just seems to me that's a far smarter thing to do if we're looking for something to transfer from part A to part B and, being honest about it, that we're trying to buy years' extension on the trust fund, because if anything has ceased its useful life, it's the way in which we fund graduate medical education.
    Mr. NEWHOUSE. Mr. Chairman, I have essentially the same thing as Dr. Wilensky. ProPAC doesn't have an official position but personally, I think insofar as these programs, both the teaching program and the Disproportionate Share Program serve a broader social purpose than buying medical care for the elderly and the disabled, that there's a very strong argument for financing them out of general revenues.
    Chairman THOMAS. I've got a bunch of other questions but I'll move on.
    The gentleman from Wisconsin.
    Mr. KLECZKA. Thank you, Mr. Chairman.
    Last session we talked about the trust fund and I don't think we fully developed the idea. However, I think more than the taxpayers and Medicare should be contributing to that type of medical education.
 Page 63       PREV PAGE       TOP OF DOC
    Let me understand this more fully. Do third-party payers also pay not a segregated fee for graduate medical training but as part of their reimbursement to a medical facility, a hospital?
    Ms. WILENSKY. They used to, frequently because they didn't know it. It wasn't there. They were being charged amounts that included funding that was probably including some of the graduate medical education, but it was not——
    Mr. KLECZKA. But today's billings are devoid of that segment?
    Ms. WILENSKY. Today's health care market is much more competitive and hospitals have less ability to include charges for services that don't have to do with taking care of patients. It's much more limited, not uniformly across the country but more so than it used to be.
    Mr. KLECZKA. So it is, in fact, true that the major reimburser for medical education is the Medicare Program.
    Dr. Wilensky, I listened to your three rationales why Medicare should pay and it comes from 1984, and I don't think any of them apply today.
    Ms. WILENSKY. I agree.
    Mr. KLECZKA. Oh, OK. We're in synch on that.
    Ms. WILENSKY. I absolutely agree. We raised them because the 13 Commissioners were puzzling through why are we doing this and we raised these as issues. I think there actually was a sense, although no recommendation, that the logic for these three is not obvious.
    Mr. KLECZKA. OK. Well, I think all payers of medical benefits, including premiums from individuals, should contain a piece for medical education, and not only the Medicare Program. Hopefully, we'll get back to the trust fund and an all-payer-type system will be incorporated.
 Page 64       PREV PAGE       TOP OF DOC
    I really have a question and a problem on the $400 million payment to New York not to train doctors. When that was publicly broadcast, I just got a very negative reaction from my constituents that we're going to be in the business now, with all the fiscal problems here on the Federal level, that we're going to be paying people not to do something or paying hospitals not to do something.
    I think even though it's an administration proposal, I think it's goony, to say the least, and will not pass the public smell test.
    Both of you have referred to that portion. Do you want to expand on the inadvisability of doing that?
    Ms. WILENSKY. Well, I think there are two kinds of reasons you might do it and I don't see either present. The first is you really want to know whether something's going to work and so you do a demonstration and you do it in a controlled environment with an evaluation. That's not present here. And, as Dr. Newhouse indicated, the self-selection would really make it a problem, aside from all the other problems.
    The second reason is you have a total package of policy changes and as part of the package change, you want to provide funds to academic centers as you transition to whatever new system you put in place, so you might want to make it less painful going from here to there and therefore, giving them a way to maintain in a sliding-down fashion, the moneys they had been getting for a 5- or 6-year period would be part of an overall reform. But this is not part of an overall reform.
    Mr. NEWHOUSE. Yes, this has been characterized as paying them for not training residents, with the implication that they were getting more, but the historical pattern has been that residents have been going up. Therefore, in some sense, a baseline here would continue to go up absent these payments.
    As it is, if this were done uniformly for everybody, as I suggested, rather than just for the hospitals in New York that wanted to do it, this would end the notion that if you hired another resident, you got more; if you got rid of a resident, you got less. Therefore, that factor would not operate on hospitals' decisions on how many residents they wanted. And if a hospital chose to have fewer residents without the subsidy, that should be a good thing and it would generate—if set up the way it is in New York and could be done for everybody—would generate savings for both Medicare and leave the hospitals with some incentive to downsize.
 Page 65       PREV PAGE       TOP OF DOC
    Mr. KLECZKA. I think it's ironic. The program with the most fiscal problems, i.e., Medicare, is shouldering the whole burden for graduate medical education. I think that's ludicrous and it has to be changed.
    Let me just finish by responding to the Chairman's thoughts on using general purpose revenue, general taxpayer revenue, for training doctors as a higher calling than for paying bills for elderly in a home health care setting. I think if you put that to a referendum, Bill, you'd lose, the question being, ''Would you rather have your tax dollars training doctors or paying for poor people in home health care settings?'' I think it would probably be 80 : 20 in favor of home health care and the transfer to part B.
    That doesn't alleviate the need for correction there because of the vastly increasing expenditures and the inconsistency in reimbursement for home health care, but I'll challenge you on the higher calling of taxpayer funds. I was going to do it when you were out of the room so you couldn't respond.
    Chairman THOMAS. I appreciate the gentleman's comment. He's also fully cognizant of the fact that if we submitted the Bill of Rights to the popular vote it would go down, as well. [Laughter.]
    The gentleman from Louisiana.
    Mr. KLECZKA. Holy moly. [Laughter.]
    Mr. MCCRERY. Thank you, Mr. Chairman.
    I think it's appropriate that this hearing has become somewhat of a freewheeling discussion of the Federal Government's responsibility to provide health care to the elderly. And my good friend from California, as usual, brought up some very important points, although not all of us are as convinced of his conclusions as he is.
    For example, his statement that we have an obligation to provide health care to the elderly and we're not going back on that, I think is correct. However, a question that this fiction that he talked about of the payroll tax and the trust fund that it funds is, How much health care do we have an obligation to provide the elderly? Do we have an obligation to provide all health care that's available at any cost to every elderly citizen?
 Page 66       PREV PAGE       TOP OF DOC
    I don't think we have satisfactorily answered that question and that's part of what this hearing is about today, as we talk about the fiction of a trust fund, and it is a fiction. We know all the money that comes in to the Federal Government is pretty much fungible and we spend it freely, for all kinds of purposes.
    What's not a fiction, though, is the payroll tax that funds the fictional trust fund. People can see that coming out of their paycheck every month, and that serves a purpose, because it does tend to let us know what the cost is of providing that health care to the elderly, at least insofar as hospital payments are concerned.
    Mr. STARK. Would the gentleman yield?
    Mr. MCCRERY. Surely.
    Mr. STARK. How about if we do a consumption tax? If we could hide it in, say, a Federal sales tax, and we'll dedicate it only to health care?
    Mr. MCCRERY. Well, we can discuss that at another hearing.
    Mr. STARK. In the Full Committee.
    Mr. MCCRERY. Yes, in the Full Committee when Chairman Archer wants to talk about a consumption tax. But right now I think payroll tax does serve a very valid function of reminding workers in our society what they are paying to provide health care to the elderly.
    Now, obviously the part B of Medicare is not so readily obtainable in terms of its costs because the work force doesn't pay that directly, although it pays it very much indirectly through general taxes.
    So I would hope we would not conclude or at least not accept the conclusion that we are obligated, as the Federal Government, to give all health care to all elderly, because that's going to become a more and more important question.
    One thing I think we have concluded today is that no formula for allocating funds for medical education is perfect.
 Page 67       PREV PAGE       TOP OF DOC
    Dr. Wilensky talked about her Commission debating the question of why Medicare funds teaching hospitals, and I think that's a legitimate question to ask. One thing you might have talked about before you talked about that question is, Why is it a Federal responsibility at all to fund teaching hospitals? I think that question deserves some debate.
    If we conclude it is a national purpose, along with home health care for the elderly, then I guess it's a Federal responsibility to see that our graduate medical education, medical education in this country is funded, at least partly, by the Federal Government, but we haven't talked about that enough, in my opinion.
    But if we conclude it is a Federal responsibility, then we get to the question of should Medicare fund it. It seems to me to be a rather odd way to fund teaching hospitals. It's convenient because we already make payments, so it causes us a lot fewer headaches than trying to think of a formula. But I think we can all reach the logical conclusion that it's not a very good way to fund teaching hospitals and certainly if we're talking about a health care program that's designed for the elderly, if we take revenues that are collected for that specific purpose and give to teaching hospitals, not everybody, I think, would agree that that should be done.
    So I'm glad we're talking about all these questions. I hope your two panels, your two Commissions, will give those questions some more debate and share your conclusions with us.
    And, Mr. Chairman, I would like for you to encourage more of this discussion among the policymakers who actually have the power to decide. Even though, as Mr. Stark said, we have the wonks with us, we're the policymakers. We have to actually debate these questions and come up with decisions. So I would think maybe one day we'll have a hearing with just us debating these questions. Who knows.
    Chairman THOMAS. I just want to thank the gentleman from Louisiana. While I was complimenting myself for thinking outside the box, the gentleman humbles me because he's thinking outside the room. [Laughter.]
 Page 68       PREV PAGE       TOP OF DOC
    The idea of going to full private funding for graduate medical education is, in fact, a discussion that ought to take place before we then move into the other discussions because it would change significantly the examination of who gets what, when, and how.
    Anyone want to respond?
    Mr. NEWHOUSE. Mr. McCrery, I'd like to distinguish between the payments for indirect medical education and direct medical education.
    Mr. MCCRERY. You did already and I accept that distinction.
    Ms. WILENSKY. But I had meant, when I had used the phrase of questioning the responsibility of Medicare for graduate medical education, to include the broader issue of to what extent or why is it a Federal Government function or a government function at all.
    It is a very unusual role that the Federal Government has taken in terms of Medicare. It is not normally a significant role that the Federal Government takes to pay institutions, as opposed to individuals, for training at the graduate level. I think if there's going to be a redesign, it would be a shame to not address that first issue before you get on to any decisions about reallocations or trust funds.
    Mr. MCCRERY. If you reach a conclusion, come back and let us know.
    Dr. Newhouse, do you want to expound on your distinction between IME and——
    Mr. NEWHOUSE. Yes, that's exactly what I wanted to do.
    Thank you.
    The prospective payment system right now pays close to a flat amount for every hospital, for the standardized rate, anyway. One can, I think, believe with some degree of credibility that teaching hospitals provide a different range of services that's more expensive and treat a sicker class of patients. And one might have a separate peer group for teaching hospitals that would amount to a higher rate. One would compute an average rate for teaching hospitals and an average rate for nonteaching hospitals.
 Page 69       PREV PAGE       TOP OF DOC
    That is, more or less, what the indirect payment is trying to do. It uses a graduated measure of teaching and then, therefore, gets mixed up with how many residents do you have and so forth and so on. But the basic rationale for the payment is that teaching hospitals are really different from nonteaching hospitals in several respects for the services they produce, and therefore should get a higher payment. It's not the same product, if you will.
    Now, the direct medical education side gets to the range of issues you were really discussing and what should be the Federal Government's role in supporting training of residents. I think you brought out the relevant points there, but I would distinguish between those two purposes.
    Mr. MCCRERY. That's a valid distinction. That doesn't mean we ought to broach the question of whether the Federal Government has that responsibility, as well, but it is two different questions.
    Chairman THOMAS. My former colleague from New York, Mr. Kemp, always used to argue that if you want more of something you should subsidize it. Apparently, someone decided we wanted more of them, so we're subsidizing them.
    Does the current gentleman from New York wish to inquire?
    Mr. HOUGHTON. Thank you, Mr. Chairman.
    Mr. Newhouse, I'm going to pick on you for a moment because I have three questions. I want to take a pencil out.
    First of all, I'm interested in the New York hospital situation. I come from upstate. There are no teaching hospitals. But I wonder whether the current program for New York really is inhibiting other States in putting together their own plans, and also whether this is interfering with any other reforms of Congress. So that's number one.
    Number two, you say in your testimony that the payment method should be flexible—the payment method to support medical training should be flexible enough to support training in settings outside the hospital—freestanding facilities, managed care organizations. Maybe you could bring us up to date on how you think a formula like that ought to be established.
 Page 70       PREV PAGE       TOP OF DOC
    Then the third question really talks about the DSH payments. The President's budget, if I understand, suggests relatively modest increases in the Medicare DSH adjustment, but then there's a freezing of these adjustment rates and then the Secretary of HHS is going to develop a new formula within 18 months to determine methodologies and things like that.
    It seems as if we're there, and I just wonder what's wrong with the current formula and why can't we just go ahead and do what we're doing?
    Mr. NEWHOUSE. Let me start with the DSH formula because I'm particularly pleased with what our Commission came up with on that score in response to your request to take a look at the DSH formula.
    We think it's clear enough to proceed now on making some changes in the formula, and the most important changes I would emphasize here are that we would include all low-income patients in a measure of what would trigger payments, not just Medicaid patients. And we would, in effect, set the bar at the same level for urban and rural hospitals of varying sizes, rather than have the bar higher for rural hospitals, as it is now.
    We think both of those changes could be made. There are some other changes we're recommending to you be made and that we don't need to wait 18 months.
    Mr. HOUGHTON. So they could be made within a matter of weeks?
    Mr. NEWHOUSE. We believe there would have to be some additional data collected, but we don't believe that's a major bar. We believe you could proceed to implement them rather promptly, within months, at least, not years.
    I don't know if Dr. Wilensky wants to comment here. I might come back and talk about the other two points you raised.
    Ms. WILENSKY. Let me quickly reference those other two points. I agree with the comments Dr. Newhouse has made with regard to the disproportionate share.
    With regard to why don't other States come in and make their own proposals, it has to do with, at least as I understand it, the ability of HCFA to have demonstrations and what the purposes of those demonstrations are, and one of them is as a research demonstration. You can make a good case as to why, before you adopt a plan for the whole country, you might want to try it and assess the effect. But I think you could look at the New York proposal and say whatever else it will do, it will not allow you to assess the effects of doing this type of a payment because it wasn't set up with the kinds of design or an evaluation plan to let you do this.
 Page 71       PREV PAGE       TOP OF DOC
    So if it's supposed to be a demonstration, then let's be serious about it. HCFA has a very long history of demanding important control information and/or good empirical evaluation strategies. That was clearly not present.
    It's also not clear how you would be treating all these States coming in, the timeline you would allow them, and it is a rather unusual way to try to continue an ability to affect decisions on residents without affecting payment. If you're not going to use it as a learning demonstration, then I think there really is an equity question as to why the hospitals in California or Massachusetts or Texas or places that have heavy concentrations of academic health centers ought not to have the same right to be included if they so choose, as New York.
    So I think it's either a demonstration or there's a very serious equity issue that's being raised. I think there's also a question about whether you're giving away too much for what you get in return, but that's a judgment that you, as Congress or the administration, would have to make.
    With regard to the flexibility of paying outside the inpatient setting, this is a big issue. There are a lot of implementation issues. If you decide you want to make a payment that isn't just tied to the inpatient, are you going to include direct and indirect or are you only going to use the direct payment? As Dr. Newhouse indicated, the indirect has a different justification; a different product is being produced.
    Will it be by the numbers? And who's eligible? Should it be a consortium? Should it be just traditional players?
    So I think that if the Congress wanted to move that way, the Commission would certainly be glad to give you a host of ideas about how to implement that, but there are some very fundamental calls that really are the decisions of the Congress about, if they want to go that way, the kinds of institutions they think should be allowed to make claim.
    I don't want to sound like a broken record but before you can really make those decisions, what purpose you're looking to the Federal Government to accomplish for the direct payment is one of the first questions, before you make these changes.
 Page 72       PREV PAGE       TOP OF DOC
    Mr. NEWHOUSE. Mr. Houghton, I might say in terms of implementing the DSH changes, that would require legislation. So the administration can't do that without your moving.
    Mr. HOUGHTON. Thank you very much.
    Chairman THOMAS. Does the gentleman from California wish to inquire?
    Mr. BECERRA. Thank you, Mr. Chairman.
    Let me ask that we move back for a moment and deal with the question of whether we have an oversupply of physicians. If I'm correct, there are some 700,000 physicians in the country and some 265 million residents in the United States, somewhere around one physician for every 400 residents. And, of course, we're lumping specialists with primary practitioners.
    I know areas where you won't find 1 physician for every 4,000 residents, and certainly they would not say there's an oversupply of physicians. If we're going to move to provide less incentive to increase the number of residents in our teaching hospitals, how do we deal with the problem for those areas that don't have enough physicians to begin with?
    Ms. WILENSKY. I think it goes to the question of whether making payments to institutions, as opposed to making payments to individuals, might be a more effective way to affect the distribution of physicians.
    There was an attempt in the late sixties to have loan forgiveness programs. It didn't work particularly well, but it was at a time when the tuition payments for physicians was around $2,500 and not the $20,000 that it frequently is now.
    I personally believe that a targeted loan forgiveness, targeted subsidies, would be an effective way to impact some of these distributional decisions of physicians and that it would represent a better use of some of the funds that now go out of graduate medical education to institutions.
 Page 73       PREV PAGE       TOP OF DOC
    Mr. BECERRA. So, Dr. Wilensky, you're saying we should go ahead and use some of the funds that we currently have going into this pot for teaching institutions to actually encourage graduating physicians to go into the areas that are most underserved?
    Ms. WILENSKY. It's my personal strong preference. It is not a recommendation of PPRC.
    The second issue is that especially because, in the aggregate, there does appear to be an oversupply, but there certainly are distributional issues, both by specialty and certainly by location. This reflects the fact that some areas have high concentrations of uninsured and low-income populations and that again, one of the problems that we're dealing with is how to try to have financing and demand for services by people who don't have the resources to pay, and that is impeding some of the distribution issues.
    But if you look since the eighties, there has been some change in location of physicians, with more physicians going into more rural areas than had occurred and certainly more physicians going into primary care and general medicine than had occurred earlier. But I think you could have a much bigger impact if you targeted some of that money to change the behavior of physicians in ways that were socially desirable.
    Mr. BECERRA. And Dr. Newhouse, before you answer, could I just ask one followup question of Dr. Wilensky?
    Do we have enough money in the pot right now to do those things, to take some of the money out and use it for incentives to have people locate in certain areas?
    Ms. WILENSKY. I think you do.
    Mr. BECERRA. OK.
    Dr. Newhouse.
    Mr. NEWHOUSE. First, I think I'm right that even if hospitals cut all the way back on the number of residents—well, let me back up. Let me say that I think your objective here ought to be neutrality toward hospitals and incentives to hire residents. That's the thrust of our recommendation.
 Page 74       PREV PAGE       TOP OF DOC
    Now, right now, hospitals' resident counts are around 140 percent of U.S. medical graduates. Some people have advocated trying to cut that back to 110 percent. Even if that were done, and it's not clear that if you were neutral, hospitals would cut back that far, but even if they did, I'm almost certain it is true that physician-to-population ratios are going to continue to increase.
    So what I saw lurking there was the premise that if we step on the brake, will physicians go down? I don't think they will. I think they're going to continue to go up.
    Mr. BECERRA. And that's not my concern. My concern is still the distribution.
    Mr. NEWHOUSE. Right. So I was going to come to the distribution. On the distribution, and again I'm speaking for myself, not for the Commission, my preference would be you do something about the demand side for care or people's ability to pay. I don't know that it does a lot of good to put a physician out there if nobody comes through the door because they don't feel they can afford it.
    Mr. BECERRA. Good point. And so your solution would be?
    Ms. WILENSKY. There's not enough money in this pot to fix that problem. [Laughter.]
    Mr. BECERRA. I think, Dr. Wilensky, in your testimony you mentioned that we should try to base our payments on some—have them be generalized so that——
    Ms. WILENSKY. Standardized.
    Mr. BECERRA. Standardized. Thank you. That would concern me some if you're talking about a high-cost area versus a low-cost area. You do mention perhaps gauging it on labor costs and so forth.
    Ms. WILENSKY. Yes.
 Page 75       PREV PAGE       TOP OF DOC
    Mr. BECERRA. Can you expand on that a bit?
    Ms. WILENSKY. Well, we have now enormously variable costs and Dr. Newhouse indicated that in 1984 and 1985 when the formulas were put together, there were very different accounting rules and also very different historical divisions in terms of whether some of the practice costs were part of the practice plans or whether they were part of the hospital. You had incentives. You had some hospitals much more savvy about what the allocation would do in future years or much luckier about where they made their allocation.
    So that while there are differences in cost of living and there are appropriate differences that you would like to build into a pattern, what we see here is just vastly in excess of that. So it's standardized in the history, really, or tradition of Medicare, which is to make adjustments for some specific cost factors that are appropriate, but only those.
    Mr. BECERRA. And so that's really still the same thing. You're still talking about standardized——
    Ms. WILENSKY. Yes. When we used the term ''standardized,'' it was not meant to exclude appropriate adjustment factors like cost of living.
    Mr. BECERRA. Dr. Newhouse, if I may ask a question regarding the IME adjustment, I know you're proposing we go down to 7 percent per 10 percent of increases in teaching intensity. The administration is proposing to go down to 5.5 percent over 5 years?
    Mr. NEWHOUSE. They're proposing going to 5.5 percent by 2002, I believe, yes.
    Mr. BECERRA. About 5 years. Your proposal just covers the one year——
    Mr. NEWHOUSE. Correct.
    Mr. BECERRA [continuing]. But I was wondering if you could comment on what you think should happen over the next several years after that, given that I think your estimate was that costs are being overestimated, that we really are only incurring costs of 4.1 percent.
 Page 76       PREV PAGE       TOP OF DOC
    Mr. NEWHOUSE. Well, the costs, I think, have been properly estimated. We're just paying more than that. So the Commission would go down to ultimately the cost level. We're a little concerned about the transition and we were taking it a year at a time.
    I think there's no doubt that I speak for the majority of the Commission, anyway, that we would try to get to the empirical level ultimately. That's been a longstanding recommendation of the Commission.
    Mr. BECERRA. So although you don't have it in your recommendation, you would ultimately urge us to go toward a system that gets us to the empirical level?
    Mr. NEWHOUSE. Yes. You have to decide, as the policymakers, how much money to spend here, but we would, for the indirect side, I think there's a very good case for getting to the empirical level.
    Chairman THOMAS. The gentleman's time has expired.
    Does the gentleman from Nebraska wish to inquire?
    Mr. CHRISTENSEN. Thank you, Mr. Chairman.
    I want to follow up on Mr. Becerra's question, Dr. Wilensky, talking about the targeted loan forgiveness. Could you explain to me what the program looked like in the early sixties?
    Ms. WILENSKY. As best I can remember it, I think it was around 1966 or 1968 and it allowed for physicians who would choose to go into shortage areas to have the loans that they took out covering their tuition forgiven or, alternatively, the physician could pay off the loan, pay out. I assume there may have been some additional charge or some interest payment, but there was either a payout or payback in terms of service.
    What tended to happen in the late sixties is that these well-paid professionals frequently decided they'd rather pay back than make the location or specialty decision choices that they had received the loans for. And given the relative tuition payments and the relative incomes, it's not surprising that that was not enough to have an incentive for a physician to choose where they otherwise might not choose.
 Page 77       PREV PAGE       TOP OF DOC
    I think now you are seeing very different circumstances. You have tuition levels that are frequently in the $20,000, sometimes $25,000 range, not counting living expenses. You have medical students who are frequently graduating very heavily in debt as a result, being pressed to go or feeling pressed to go into urban areas or specialties where they can try to repay some of that. It seems to me much more likely, in that type of environment, where you could get someone's attention by offering a loan forgiveness program which had good strings attached in terms of this is what you have to get in order to have your loans paid off.
    Mr. CHRISTENSEN. So you think the parameters have changed since——
    Ms. WILENSKY. I think they've changed substantially. I think, frankly, it also may be possible because there are far more couples that are in the same field. That was also my understanding of one of the problems, that it was the spouse as well as the physician who had to be satisfied with the location decision. I think in the late sixties the income level was really very different.
    Mr. NEWHOUSE. Let me elaborate a little on that. My recollection is that the loan forgiveness was primarily part of the National Health Service Corps legislation, which was the early seventies.
    Ms. WILENSKY. This is in addition.
    Mr. NEWHOUSE. In addition? All right.
    Ms. WILENSKY. But the same point was there, too.
    Mr. NEWHOUSE. This was at a time when rural areas were predominately staffed by general practitioners who were retiring. The wave of specialists that were to come had not gotten there yet. As a result, there was a sense that there were large geographic areas that were very understaffed.
    As time passed and we produced more and more specialists, they gradually diffused out toward smaller and smaller towns. It was found that these large areas were very few and far between, and the definition of an underserved area moved from one that was just defined on the basis of how many doctors per person were there to one that also included measures of health status, such as the infant mortality rate. There was a change in terminology toward medically underserved areas, so that the whole concept became broader.
 Page 78       PREV PAGE       TOP OF DOC
    In fact, my recollection, at least at one time and maybe it's still true, is that Chinatown in San Francisco was defined as a medically underserved area on this criteria. Normally, one would not think of the city of San Francisco as a medically underserved area.
    Chairman THOMAS. Would the gentleman yield? On holidays, however, that was not the case.
    Mr. CHRISTENSEN. I would like to explore this area more to see if there's a possibility we could look at this again and fashion a program, if we can afford it, and you said that, in your own opinion, you thought the funds may be there.
    One other quick question before my time runs out. Concerning nurses, I visited with the local head of the nurses association last month in my office. Do they have legitimate reasoning on obviously not having the funding of their GME and should we look at that? And if we can afford to help greater areas of Nebraska have a doctor in their area, should we look at the funding mechanism for GME with nurses, as well?
    Ms. WILENSKY. The Commission has looked at the issue with regard to nurses and nurses' education for a number of years and the overall conclusion is that some of the same issues that got raised with the physician payments are relevant here; that is, they ought to be more comparable, they ought to be more fair than what now goes on.
    Again, you have this historical basis that is used and, in the case of nurses, you have this peculiar tie to a diploma education that was relevant in 1965 that is not particularly relevant in 1997.
    But again, before you make a final decision, you really need to question what is the role of the Federal Government, as opposed to State governments or government in general, in this training. And when you have answered that to your satisfaction, then I think you have—to the extent you think there is a role for physicians, that you can make a case that there's a role in the training of nurses.
 Page 79       PREV PAGE       TOP OF DOC
    Again, you'll have to ask whether it should be the institution or the individual. In both cases, in the aggregate there appears to be if not oversupply, a lot of abundance.
    Mr. CHRISTENSEN. Thank you. Thank you, Mr. Chairman.
    Chairman THOMAS. The Chair does not wish to engage in a second round of questioning but I believe the gentlewoman from Connecticut has a question for clarification.
    Mrs. JOHNSON. Just very briefly. I notice in your testimony that 674 of the hospitals—of the 5,153 hospitals, 674 are both teaching and DSH and that they receive 75 percent of the total IME and 65 percent of the total DSH payments.
    Have you run the impact on this group of hospitals of a cut in IME and a cut in DSH? They would clearly be impacted in a way that none of the other hospitals would be impacted.
    Mr. NEWHOUSE. They would certainly be impacted by an IME cut. To be clear, we did not have a recommendation on the size of the pot for DSH. Our recommendations went to how DSH would be allocated among hospitals. And in the sample run we did, one could choose a lot of different formulas and they would come up with different allocations, as you can appreciate, but in one specimen run we did, the major public teaching hospitals would come out ahead relative to where they are now in the DSH payments, and all other hospitals would lose slightly.
    Dr. YOUNG. As your deliberations continue, we will be able to follow through with followup kinds of analysis. We've done it so far in the aggregate. We've not done it for small groups of hospitals, in part related to data problems, which our data set will be increasing in size.
    So as your deliberations continue, we will provide you with that information.
 Page 80       PREV PAGE       TOP OF DOC
    Mrs. JOHNSON. Thank you. Thank you very much for your testimony today.
    Chairman THOMAS. Thank you. We may be getting some written questions to you, in part, to run numbers based upon some of the suggestions that have been made by Members.
    Thank you very much.
    Ms. WILENSKY. Thank you.
    Mr. NEWHOUSE. Thank you.
    Chairman THOMAS. I'd ask the next panel to come forward. Dr. David D'Eramo, who is the president and chief executive officer, Saint Francis Hospital in Hartford, Connecticut. However, he's here, I believe, for the Association of American Medical Colleges.
    Dr. Cornelius L. Hopper, who is vice president for health affairs, University of California.
    Dr. Spike Foreman. In the absence of Rick Lazio, Dr. Foreman, I want to welcome you, on behalf of the Greater New York Hospital Association.
    I already introduced Jerry Starr, chief executive officer of the Kern Medical Center, here on behalf of the National Association of Public Hospitals & Health Systems.
    And David Jaffe is executive director and chief executive officer of Harborview Medical Center in Seattle, Washington, on behalf of the American Hospital Association.
    I want to welcome all of you. I will tell you that any written statement that you have for the record will be made a part of the record, without objection, and that beginning with Dr. D'Eramo and moving across the table, you can inform this Subcommittee in any way you see fit about your concerns, hopes, and fears for the graduate medical education.
    Mrs. JOHNSON. Mr. Chairman, before Dr. D'Eramo begins, I'd like to welcome him to the table of this Subcommittee. He has been a leader in health care reform initiatives in Connecticut. And, as you know, while we got started late, the pace has been fast.
 Page 81       PREV PAGE       TOP OF DOC
    Thank you. Nice to have you here, Dr. D'Eramo.

STATEMENT OF DAVID D'ERAMO, PH.D., PRESIDENT AND CHIEF EXECUTIVE OFFICER, SAINT FRANCIS HOSPITAL AND MEDICAL CENTER, HARTFORD, CONNECTICUT; AND CHAIRMAN, COUNCIL OF TEACHING HOSPITALS AND HEALTH SYSTEMS, ASSOCIATION OF AMERICAN MEDICAL COLLEGES

    Mr. D'ERAMO. Thank you all and good afternoon, Mr. Chairman, and distinguished Members of the Subcommittee. I am Dave D'Eramo, president and chief executive officer of Saint Francis Hospital and Medical Center in Hartford, Connecticut. I also have the privilege of serving as chairman of AAMC, the American Association of Medical Colleges Council of Teaching Hospitals and Health Systems.
    I'm pleased to be here and for the insights gained through the discussion preceding. I believe the thoughts I wish to posit are largely complimentary, at least philosophically.
    This afternoon I'd like to suggest two steps that would help assure the future of American medicine. The first is establishing a shared responsibility approach for financing the special missions of teaching hospitals and medical schools. The second is removing the special payments from the calculation of Medicare managed care rates. Finally, I'll address reducing Medicare payments for graduate medical education.
    First, on behalf of the AAMC, let me express appreciation to the Subcommittee and, in particular, to Chairman Thomas and Representative Johnson from my own district for recognizing that teaching hospitals face unique and potentially devastating problems in the current price-competitive environment.
    As was pointed out earlier, during the last session, with the leadership of Chairman Archer and the Ways and Means Committee, the Congress devised, what I believe was, an innovative approach to funding the costs related to graduate medical education. That is the Balanced Budget Act of 1995, which included a trust fund, a trust fund with general revenue complemented with contributions from the Medicare Program.
 Page 82       PREV PAGE       TOP OF DOC
    The AAMC supported this proposal and saw it as an important first step in establishing a new framework for shared responsibility financing of clinical education.
    The AAMC endorses the trust fund concept. It does so as a means of assisting both teaching hospitals and medical schools.
    Like teaching hospitals, the fact is, medical schools rely on clinical revenue. They do so to support their education and research missions. Clinical practice is one of the few sources of flexible revenue for underwriting innovative teaching programs and providing important seed money for new research activities. And, in a price-competitive environment, reductions in clinical revenue threaten medical schools' financial stability and their ability to foster the education and research missions from which all Americans benefit.
    While we should further engage in the exercise of thinking outside the room or the box for the sake of achieving and realizing constructive change in real time without risking the potential implosion of our institutions' current medical education infrastructure, I'd suggest there are at least two pragmatic rationales for the creation of a trust fund.
    One is that it serves as replacement funding to ameliorate the impact, and there will be an impact, of reductions in Medicare IME and DGME payments. And the other rationale, which the AAMC hopes to emphasize, is the restoration of dollars that private payers, in the past, have contributed, along with Medicaid.
    So that's why we believe, as part of the shared responsibility, that Medicare should contribute to a trust fund on behalf of its beneficiaries, including those enrolled in risk plans.
    The association strongly supports removing the IME, the direct graduate education and DSH components, from payments from the calculation of the adjusted AAPCC and paying them directly to teaching or DSH hospitals when they serve risk plan enrollees.
    Congress designed these hospital payments for specific policy objectives. Additionally, both ProPAC and PPRC have said that carving them out of the rate would improve the HMO formula to reflect more closely the actual costs of providing care to risk enrollees. And, as the centerpiece of GME financing, Medicare sets the standard for participation. It should continue its vital role on behalf of both fee-for-service and managed care enrollees.
 Page 83       PREV PAGE       TOP OF DOC
    Clearly, the AAMC understands the fiscal difficulty Congress faces in extending the solvency of the part A trust fund and hopes that any reduction in teaching-related payments would be proportionate to other decreases in Medicare spending, would be implemented gradually, and the impact monitored closely.
    While many have pointed to the high PPS margins as justification for reducing indirect medical education and direct graduate medical education payments, the fact is, teaching hospitals are fragile. And, as was pointed out earlier today, the total margins continue to be lower than those of other hospitals, and they do lose money on Medicare patients when all of the revenues and costs are considered.
    Finally, the association supports changes in GME payments to encourage primary care and ambulatory training, allowing DGME funding to flow to the entity that incurs the cost and counting the time that residents spend in nonhospital settings for IME payment purposes. We think both of those are positive steps to GME reform.
    I thank you for the opportunity to appear before the Subcommittee.
    [The prepared statement and attachment follow:]

Statement of David D'Eramo, Ph.D., President and Chief Executive Officer, Saint Francis Hospital and Medical Center, Hartford, Connecticut; and Chairman, Council of Teaching Hospitals and Health Systems, Association of American Medical Colleges

    Mr. Chairman, and distinguished members of the Subcommittee, I am David D'Eramo, Ph.D., President and CEO of St. Francis Hospital and Medical Center in Hartford, Connecticut. I also am Chairman of the AAMC's Council of Teaching Hospitals and Health Systems (COTH). The AAMC welcomes the opportunity to testify on the future financing of graduate medical education (GME). The Association represents all of the nation's 125 accredited medical schools, approximately 400 major teaching hospitals, including 75 Veterans Affairs medical centers, the faculty of these institutions through 89 constituent academic society members, and the more than 160,000 men and women in medical education as students and residents.
 Page 84       PREV PAGE       TOP OF DOC
    I come before the Subcommittee today to express the Association's concern about the future financing of the special missions of teaching hospitals and medical schools in a fiercely price competitive environment. I will offer possible financing approaches so that these institutions, which constitute the backbone of the American health care system, can continue to meet their societal responsibilities. I will also address the role of Medicare in financing GME and its related costs under the fee-for-service and managed care portions of the program.
    The AAMC expresses its appreciation to the Subcommittee for recognizing that teaching hospitals face unique and potentially devastating problems in maintaining their crucial missions in the current environment. We would like to give special thanks to Chairman Bill Archer and Representatives Thomas and Johnson for their continuing leadership and concern. The Subcommittee's broad understanding of the functions and diversity of teaching hospitals obviate the need for me to explain their special role in society.(see footnote 1)

    The AAMC understands the difficulty Congress faces in reducing the rate of growth in the Medicare program to extend the solvency of the Part A Trust Fund. The Administration's proposed reductions for Medicare and Medicaid would have a significant impact on the health care system. Moreover, the changes contemplated would have especially profound effects on the nation's teaching hospitals and medical schools. Teaching hospitals serve large numbers of the poor and the elderly and depend heavily on Medicare payments for direct graduate medical education (DGME), indirect medical education (IME), and disproportionate share (DSH). Many teaching hospitals also serve large segments of the Medicaid population. In 1995, Medicare and Medicaid payments constituted one-half of all patient revenue for the average COTH member.
    Before addressing specific proposals, however, I would like to comment on recent reports of near record prospective payment system (PPS) inpatient margins among teaching hospitals. New data on hospital financial performance show that the average teaching hospital, like all other hospitals, has improved its financial position relative to Medicare PPS and all other patients. At its January meeting the Prospective Payment Assessment Commission (ProPAC) presented PPS inpatient and total margins. The average PPS inpatient margin for major teaching hospitals increased to 18.6 percent in 1995, while their average total margin was 3.7 percent. The general trend for most hospital groups shows increasing PPS and total margins, but these 1995 data are preliminary and subject to change. Only 60 percent of hospitals had reported data, and large teaching hospitals and those with late-year Medicare Cost Report closings were under-represented in the sample. Additionally, the aggregate margins are ''average'' financial indicators and mask the wide variation among teaching and all other hospitals' performance.
 Page 85       PREV PAGE       TOP OF DOC
    Dramatic declines in hospital costs are driving improvements in margins. Teaching hospitals have made great strides in increasing labor productivity and reducing length of stay. They are successfully adapting to a rapidly changing environment. For all hospitals, payments per discharge have increased steadily, while costs per discharge have been decreasing since 1993–94. For major teaching hospitals, the annual change in Medicare costs per discharge during the period 1993–96, after adjusting for inflation, was minus 3.3 percent, the same percentage as was true for non-teaching hospitals. Whether hospitals can achieve these striking results over the long-term without diminishing quality and access to health care is doubtful. Even after reducing their costs at the same impressive rate as non-teaching hospitals, teaching hospitals are still at a disadvantage when competing on price due to their added roles in the delivery system.
    Gains in operating efficiency notwithstanding, the AAMC notes that the payment policies adopted by Congress have been a resounding success in assuring the vitality of these institutions. Over the past several years, Congress has pursued a deliberate course of acknowledging the broader mission and overall financial viability of teaching hospitals, as measured by equity in total margins across hospital groups, to assure access and quality of care for Medicare beneficiaries and other patients. Major teaching hospitals depend heavily on IME and DSH payments; they receive two-thirds of all DSH payments. In FY 1996, one-third of teaching hospitals' PPS payments came from the IME and DSH payment adjustments because they so often care for severely ill patients who require specialized services and large numbers of low-income poor. Some policy makers have pointed to a high average PPS inpatient margin as reason for decreasing Medicare payments to teaching hospitals, specifically the level of the IME and DSH payment adjustments. While these mission payments have contributed to their higher average PPS inpatient margins compared to non-teaching hospitals, the total margins of major teaching hospitals remain consistently lower than those of any other hospital group.
 Page 86       PREV PAGE       TOP OF DOC
    Other factors, such as teaching hospitals' overall Medicare activity and the ability to cost shift, should be considered in measuring financial viability. Teaching hospitals have negative margins on average when all Medicare-related payments and costs are compared. According to ProPAC's June 1996 report, major teaching hospitals experienced losses in 1994 from treating Medicare patients when all inpatient and outpatient costs were considered. Teaching hospitals also registered higher losses from uncompensated care and had less ability to cost shift compared to other hospitals, which led to lower total margins. Public major teaching hospitals cared for few privately insured patients who generally pay higher rates. Private major teaching hospitals were able to attract privately insured patients, but made pricing concessions to maintain their patient volume. Their private payer payments were not nearly as much above costs as other types of hospitals.

Why Teaching Hospitals and Medical Schools Are Worried About the Future

    Teaching hospitals and medical schools are apprehensive about their ability to maintain their missions. Providing an environment in which health professional education, clinical research and innovation can flourish adds to the cost of patient care services at teaching hospitals and medical schools. Both entities have traditionally relied on a complex and delicate web of support from clinical revenue to finance their additional missions.

The Current Financing of Teaching Hospitals.

    Teaching hospitals have long depended on revenues from patient care to cover most of the costs of the many services they provide for society. Patient care dollars enable teaching hospitals to support specialized services that are particularly expensive but vital community and regional resources. Patient care revenues also help cover the costs of treating those who cannot pay for their care, training health professionals, and providing an environment for research.
 Page 87       PREV PAGE       TOP OF DOC
    Teaching hospitals are experiencing increasing difficulty in maintaining their educational and other social missions, which add to their cost structures, because they must meet the price competition from nonteaching hospitals when negotiating with nongovernment managed care contractors. Private insurance companies, businesses, and some government purchasers of health care want to pay the lowest possible price for only those services that their enrollees receive. In addition, state Medicaid programs are increasingly entering into risk-based contracts under which teaching hospitals must compete with nonteaching hospitals.
    At present, only two purchasers of services—Medicare and, in many states, Medicaid—specifically recognize the additional costs of teaching hospitals. Since the focus of this hearing is the Medicare program, I will not address Medicaid's participation in GME financing, except to note that most state Medicaid programs, but by no means all, include payments under their fee-for-service systems for the equivalents of Medicare DGME and IME payments. Under Medicaid managed care, however, these payments often are not made to teaching hospitals. The AAMC is concerned that state Medicaid programs are retreating from making special payments to teaching hospitals for their education and other societal missions.

The Current Financing of Medical Schools.

    Medical schools play a pivotal role in anchoring academic medical centers and producing the public goods of education, research and clinical innovation. The schools educate our future physicians and biomedical research scientists, conduct 50 percent of the total extramural research supported by the National Institutes of Health (NIH), and provide the faculty physicians who staff teaching hospitals, direct the residency programs, generate clinical discoveries, and provide outstanding medical care to a disproportionate share of the nation's indigent and uninsured. Although the schools appear to be robust enterprises, their financial structures are fragile, and their financial stability is highly dependent on clinical revenues.
 Page 88       PREV PAGE       TOP OF DOC
    Medical schools support their education and research activities from diverse sources of income, but on average, only 10–20 percent of that income is derived from secure sources such as tuition and fees, endowment earnings and gifts, or (mainly for the public schools) state support. About 20 percent of their income comes from the NIH and 10 percent comes from a variety of other sources. Nearly 35 percent comes from fees generated by the faculty's practice of medicine, and about 15 percent is from direct payments from teaching hospital partners. In other words, nearly 50 percent of the schools' aggregate annual revenue is derived from the practice of medicine, and the surpluses generated from these revenues have been a vital source of flexible funds for new academic investments.
    Revenue comes into the medical school with varying degrees of restriction in use (mild to extreme) and must be maintained in discrete accounts that may be accessed only for designated, specific purposes and not commingled. Grants and contracts provide the clearest example of highly restricted revenues. Research grants and contracts are awarded for the express purpose of supporting the proposed research. Gifts to medical schools are characteristically restricted, sometimes to a broad area of application, but more often to work on specific diseases or to support the scholarship of specific faculty.
    The financial health and stability of medical schools are best reflected, however, in the discretionary revenues and unrestricted fund balances that traditionally have been available to schools and departments for support of core academic objectives, for which designated, or restricted, funds are either insufficient or non-existent. Such core purposes would include support of faculty salaries, stimulation of new academic programs in education and research, investment in equipment and facilities, initiating new research programs, and providing the substantial funds required for cost-sharing on sponsored research. Discretionary revenues are also called upon to support new educational initiatives and curriculum changes, for example, those required to provide medical students with expanded clinical training in ambulatory and community sites.
 Page 89       PREV PAGE       TOP OF DOC
    Tuition and fees and state and local appropriations are ostensibly major sources of unrestricted funds. However, tuition and fees are relatively modest, accounting for only 4 percent of total medical school revenues in 1995. Although state and local appropriations may be quite substantial, much of this revenue is committed to existing needs, often for support of tenured faculty.
    Facilities and administrative cost recovery on research grants and clinical revenues, primarily from faculty practice plans, are both important sources of medical school discretionary revenues. A portion of clinical revenues is converted to unrestricted funds through the excess that remains after all of the direct and indirect costs of the clinical practice have been met, including the costs of uncompensated care.
    Faculty practice plan revenues support academic programs in several ways. First, some revenue is directly transferred to the medical school, its departments, and other research institutes and centers. This is discretionary money available to deans and department chairs to underwrite important teaching programs and a range of scholarly and research activities. An AAMC study estimated that in 1992–93, a total of $2.4 billion, or 28 percent, of faculty practice revenue, was used to support clinical research, scholarship and teaching programs. Second, the revenues are used to compensate clinical faculty for time spent in teaching and research. If clinical faculty were not engaged in these academic activities, their clinical productivity and the generation of patient care revenue would be far higher. Finally, some faculty practice plan revenues provide direct support for residents and fellows, and indirect support for academic programs by paying the operating expenses of clinical practices in which these programs are intermingled. However, research also is supported by a variety of public, private, and institutional sources, including faculty practice plan revenue.
    The federal government, especially the NIH, has been and remains the cornerstone of research support at medical schools and teaching hospitals. Institutional cost-sharing has been a growing factor in federally-sponsored research, and research sponsored by philanthropic and industrial sources rarely pays the full costs borne by the institution. The shortfalls caused by uncompensated research costs are offset significantly by funds from clinical practice income. These funds also provide seed money for innovative faculty research projects prior to demonstration of their competitiveness for external funding, support for investigators during temporary periods in which funding is not available, information technology, renovation of research facilities, major research equipment, faculty recruitment, administrative compliance with state and federal regulations, and assorted overhead expenses not reimbursed through other mechanisms.
 Page 90       PREV PAGE       TOP OF DOC
    Medical schools are concerned about meeting their responsibilities of education and research in a changing competitive health care system. Purchasers of health care services show a reluctance to share responsibility for the added costs associated with teaching and research. The impact of these changes is being felt particularly by medical schools in areas of high managed care penetration, where revenues generated per faculty member and faculty practice margins have been declining since 1991. As managed care penetration accelerates and price competition intensifies, the impact on medical schools will become more generalized, and the unique and delicate financing structure of medical schools will unravel. The ability of medical schools to sustain their academic missions is increasingly at risk.

A Shared Responsibility Approach

    As complex institutions with multiple and varied funding streams, medical schools and teaching hospitals are subject to many different environmental pressures, but their dependence on clinical revenue makes them extremely vulnerable to changes in the delivery system. Teaching hospitals and clinical faculty are adapting to a market-driven health care system, but no matter how successful they are at reducing their costs and managing efficiently, they will always be at a competitive disadvantage because of their additional responsibilities of education, research and specialized patient care.

The AAMC believes that a ''shared responsibility'' approach to financing the special missions of teaching hospitals and medical schools is necessary to assure the future of American medicine.

    The AAMC has consistently supported a policy that graduate medical education and other societal missions are the shared responsibility of all entities that pay for hospital and health-related services on behalf of their enrollees. Two Congressional advisory bodies agree. ProPAC, in its March 1997 report to the Congress, recommends that explicit payments for GME and teaching hospital costs should not be limited to the Medicare program. The Physician Payment Review Commission (PPRC) has called for the participation of all purchasers of health care services in financing the costs of GME. The federal Council on Graduate Medical Education (COGME) has recommended all-payer funding as well.
 Page 91       PREV PAGE       TOP OF DOC
    Two innovative approaches to funding the costs of medical education emerged in the 104th Congress. Both would have established a trust fund from which payments could be made to support the special costs associated with the education mission. The AAMC endorses the concept of a trust fund as a means of assisting teaching hospitals and medical schools in meeting their mission-related costs.
    With the leadership of the Ways and Means Committee the trust fund concept was included in H.R. 2491, the Balanced Budget Act of 1995, as adopted by the U.S. House of Representatives on October 19, 1995 and by the U.S. Senate on October 28, 1995. This legislation would have created a Teaching Hospital and Graduate Medical Education (THGME) Trust Fund consisting of five separate and distinct accounts. Three of the five accounts would have been funded by appropriated general revenue, and the Medicare program would have contributed funds to the two other accounts. Payments would have been made to teaching hospitals or qualified graduate medical education consortia.
    Another approach to a trust fund was taken in the Medical Education Trust Fund Act of 1996 (S. 1870) by Senator Daniel Patrick Moynihan (D–NY). Senator Moynihan has again introduced the bill, the Medical Education Trust Fund Act of 1997 (S. 21) in the 105th Congress. The bill would establish a trust fund for graduate and undergraduate medical education and provide three sources of funding: an assessment on private sector health insurance premiums, Medicare, and Medicaid. Five accounts would be established, including accounts for teaching hospitals and one for medical schools. To our knowledge, this is the only bill that would provide funds for medical schools to partially offset their loss of revenue from clinical practice in a competitive environment. Companion legislation (H.R. 881) was introduced in the House of Representatives on February 27, 1997 by Representatives Nita Lowey (D–NY) and Louise Slaughter (D–NY).

 Page 92       PREV PAGE       TOP OF DOC
The AAMC strongly supports the creation of a trust fund for medical schools and teaching hospitals.

    Shared responsibility for clinical education is an essential ingredient in a competitive delivery environment. Purchasers of health care services are now largely unwilling to pay a premium for education and research. Competitive pressures preclude them from doing so. The federal government alone is in a position to make formal and explicit what has always been informal and implicit by facilitating financial support for medical school academic programs and teaching hospitals through contributions to a trust fund. In so doing, it assumes an appropriate and necessary role: to ensure that all beneficiaries of the public goods created by medical schools, their faculties, and teaching hospitals continue to participate in their support.
    A trust fund would provide a framework for all segments of the health care delivery system and society to share in the financing of clinical education. Once a financing mechanism is created, several difficult design issues must be addressed. One pivotal issue is how to finance the trust fund. Financing could be provided by general revenues, as proposed in the Balanced Budget Act of 1995, or through an assessment, or tax, on purchasers of health care services, as in the Medical Education Trust Fund Act of 1997.
    The Balanced Budget Act of 1995 devised a novel approach to assist teaching hospitals in meeting the special costs associated with their education mission. The Association endorses the trust fund concept, but recognizes that there are at least two different rationales for the creation of the trust fund. One rationale is that the trust fund serves as replacement funding to ameliorate the impact of reductions in Medicare DGME and IME payments. However, the AAMC hopes that, to a significant degree, the creation of a trust fund, which includes non-Medicare revenue, could be viewed as a transition toward establishing a fund based on shared responsibility. Designated general revenue funds could be viewed as replacing dollars that Medicaid programs have contributed and that private payers traditionally have provided through higher payments to teaching hospitals.
 Page 93       PREV PAGE       TOP OF DOC
    Another important design issue is the structure of the payment mechanism. The AAMC supports retaining the structure and methodology of the current Medicare payment system for GME, namely the continued formula-driven contributions of separate DGME and IME payments. The AAMC has long held that these two payments with an education label serve separate and distinct purposes and should continue to be paid as individual contributions to a trust fund. The AAMC also believes that the funds should be distributed using a methodology that is related to teaching intensity and to the complexity of the patient, not to payer status. If payments are made based on Medicare utilization, for example, some institutions may not receive adequate funding for their involvement in education. We understand that new methodologies for distributing funds are being studied by the Institute of Medicine (IOM) at the request of the Ways and Means Committee. The AAMC is eager to work with the members of this Subcommittee to enact a trust fund for medical schools and teaching hospitals.
    I would now like to focus the rest of this testimony on the Administration's specific Medicare budget proposals for FY 1998. The President has made proposals in both the managed care and fee-for-service portions of the program that would have a profound impact on teaching hospitals and teaching physicians. They include eliminating the mission-related payments from the calculation of the Adjusted Average Per Capita Cost (AAPCC), reducing the IME adjustment, changing rules for making DGME and IME payments, and creating incentives to control high-volume inpatient physician services.

Eliminating IME, DGME and DSH Payments from Medicare Managed Care Rates

    As Medicare beneficiaries transfer from fee-for-service to capitated arrangements, teaching hospitals and DSH hospitals (often one and the same) are being placed at an ever greater competitive disadvantage in the marketplace. The reason for this is a direct consequence of the manner in which Medicare calculates the rate it pays to risk contracting plans for each enrolled Medicare beneficiary. This rate, called the Adjusted Average Per Capita Cost (AAPCC), is based on aggregate Medicare fee-for-service expenditures in a given county, including program spending for IME, DGME and DSH payments to hospitals. As a result, payments intended by Congress to support the societal missions of GME and indigent care are going instead to managed care organizations who refuse to pass the payments on to teaching and DSH hospitals. At the same time, the expenses that these payments are designed to offset remain on the teaching and DSH hospitals' books and render them more costly than their competitors.
 Page 94       PREV PAGE       TOP OF DOC

The AAMC strongly supports proposals to exclude IME, DGME and DSH payments from the AAPCC rates and pay these mission-related dollars directly to teaching and DSH hospitals whenever they provide care to Medicare beneficiaries who are enrolled in risk-based plans.

    The Medicare program should contribute to the mission-related activities of teaching and DSH hospitals on behalf of its beneficiaries in the traditional fee-for-service program and in risk-based plans. As the historical centerpiece of GME financing, Medicare sets the standard for participation.

The Administration's proposal would not adversely affect the savings needed to extend the solvency of the Medicare Part A Trust Fund.

    The Administration, in its FY 1998 budget, proposes to eliminate the three mission-related payments from the calculation of the AAPCC rates. This proposal would result in a shift of current program expenditures, reducing payments to Medicare HMOs and increasing payments to teaching and DSH hospitals. The carved-out payments would be transferred to a funding pool and then paid directly to teaching and DSH hospitals or to managed care plans that operate residency programs when they provide services to risk plan enrollees. A carve-out also is featured in The Coalition's Common Sense Balanced Budget for FY 1998 and in bills introduced by Representative Ken Bentsen (D–TX) and Senator Judd Gregg (R–NH).

Removing IME, DGME and DSH payments from the calculation of the AAPCC rates would partially reduce geographic variation in the rates.

 Page 95       PREV PAGE       TOP OF DOC
    For technical reasons, a number of groups, ranging from ProPAC and PPRC to COGME to the American Academy of Actuaries, have called for modifying the AAPCC methodology. The methodology has come under heavy criticism because it results in rates that vary widely among counties. One reason for the geographic variation is the inclusion of the special teaching-related and DSH payments in the rates. Recently the American Academy of Actuaries noted that removing the special payments would improve the HMO formula by reflecting more closely the actual costs of providing care to risk enrollees.
    In addition to the technical aspect of reducing variation, both ProPAC and PPRC have recommended removing the special payments from the rates and paying them directly to teaching and DSH hospitals to achieve the policy objectives of the Medicare program. They have questioned whether all risk plans should receive the special payments in their rates, since the funds are earmarked to compensate specific types of hospitals for distinct functions and risk plans are unlikely to pay the extra costs that these hospitals incur.

The AAMC believes that the inclusion of the mission-related payments in the AAPCC rates leads to inequities in the benefit package that beneficiaries receive, creating a two-tiered system of benefits for the elderly.

    Many risk enrollees receive more generous benefits because plans use payments in excess of their costs to fund extra benefits to attract enrollees. Medicare AAPCC rates are not based on the plans' costs of providing services, but on historic fee-for-service expenditures. If a plan's payments exceed its expected costs—including allowances for administrative expenses and a reasonable profit—as reported to the program, Medicare requires the plan to return the excess either to the federal government or to provide additional benefits to risk enrollees. Plans can charge additional premiums, but many choose to waive some or all the premiums associated with these additional benefits, perhaps related to local market pressures. As a result, there is variation in the benefits available to beneficiaries across the country. A recent ProPAC analysis confirms that HMOs are more likely to participate in urban areas that have higher AAPCC rates and they tend to offer more generous benefit packages than plans in areas with lower payment rates. The AAMC believes that, given the relationship between risk plan participation and payment rates, managed care plans have used the mission-related payments embedded in the AAPCC rates for other than their intended purposes, namely to provide extra benefits to enrollees and to increase operating profits beyond the normal rate of return that Medicare incorporated in the rate setting process.
 Page 96       PREV PAGE       TOP OF DOC

The theory that risk contractors pass these mission-related dollars back to teaching and DSH hospitals is simply not borne out by the facts.

    Since HMOs receive the same AAPCC amount regardless of the type of provider with which they contract, they have a great incentive to systematically avoid teaching and DSH hospitals when the patient care services they provide can be obtained more cheaply at hospitals without these responsibilities. Even when teaching and DSH hospitals have contracts with HMOs, contract prices need not, and most often do not, recognize the special costs incurred in providing their special missions. Teaching hospitals simply want a level playing field to begin negotiations. They are willing and able to compete for patients in the marketplace, but without access to the special payments embedded in the AAPCC rates, they can never hope to enter into fairly priced contracts.
    The AAMC understands that removing the mission-related payments from the AAPCC rates is only one issue in improving risk payment methodology. How to remove these dollars, whether at the national or individual county level, is a crucial issue and will have interactive effects with other proposed changes. Excluding the mission-related payments at the national level, essentially an across-the-board reduction, would decrease unfairly the AAPCC rates of those counties where no or little teaching or service to the low-income poor occurs. On the other hand, abruptly removing these payments at the individual county level could have significant consequences in some areas.
    Despite broad agreement on the policy appropriateness of excluding mission-related payments from risk plan rates, many observers have questioned whether the newly determined rates would be sufficient to pay health plans without stunting the growth of Medicare managed care enrollment. A ProPAC analysis of 1995 data showed that AAPCC rates would decline on average 5.3 percent and in half of all counties, rates would be lowered by 2.5 percent or less. However, teaching and disproportionate share hospitals are concentrated in certain geographic areas, and in those counties, the impact on rates would be more severe. In these counties, the mission-related payments could be phased out over a transition period of a few years. Included with other proposed modifications to the methodology, such as a minimum annual update, a payment floor, a blended rate uncoupled from fee-for-service spending, and improved risk adjustment methods, payment rates to risk plans could be made more accurate, equitable and predictable. The AAMC is working jointly with the American Hospital Association to analyze options for removing the special payments from the AAPCC rates and their interactions with other proposed changes to the methodology and would be happy to share the findings with the Subcommittee as they become available.
 Page 97       PREV PAGE       TOP OF DOC
    The AAMC recommends both near-term actions to address the immediate issue at hand, as well as longer-term actions to resolve the current Medicare payment methodology for the AAPCC. In the long term, initiatives should be undertaken to identify and study potential alternative contracting mechanisms to the AAPCC methodology. For the near term, once DGME, IME, and DSH payments are removed from the AAPCC, the AAMC recommends that separate payment methodologies, which mirror the current Medicare regulations and are administratively feasible, be applied to each component of the DGME, IME and DSH payments. This approach could be accomplished through direct payments to hospitals by continuing to use the current Medicare payment methodologies and settlement process with some relatively minor technical changes. The AAMC urges the Congress to address this methodological issue in an urgent manner as part of its package of proposals to reform the Medicare program. The Association recognizes that while this problem is more prevalent in some parts of the country than in others, it will be increasingly difficult to resolve as national enrollment in Medicare risk plans grows. As beneficiaries increase their participation in managed care plans, or exercise other options such as Medical Savings Accounts (MSAs), and fee-for-service payments decline, the mission-related dollars lost by teaching and DSH hospitals will increase substantially. The same issues also are arising under proposals to increase enrollment in Medicaid managed care programs. The AAMC believes that modifying the AAPCC calculation would at least partially ameliorate the competitive disadvantage that teaching hospitals bring to the negotiating table, remove barriers to expanding risk-based contracts among Medicare beneficiaries and strengthen the existing, risk-based coordinated care program.

Reducing Medicare Payments to Teaching Hospitals and Teaching Physicians

    The AAMC recognizes that unrestrained growth in Medicare spending threatens the long-term solvency of the Federal Hospital Insurance (HI) Trust Fund, and supports reforms to align trust fund income and outlays. However, the Administration's proposed changes in the Medicare and Medicaid programs would have significant effects on the nation's health care system, and especially on teaching hospitals and medical schools. Under the fee-for-service portion of Medicare, which still accounts for over 85 percent of all beneficiaries and is expected to retain a majority of the elderly for years to come, the program makes two explicit payments, IME and DGME, to teaching hospitals. The Administration has proposed reducing these payments by nearly $8 billion between FY 1998 and FY 2002. It also has proposed three GME reforms. Another proposal that would affect teaching hospitals and teaching physicians is the Administration's proposal to create incentives to control high-volume physician services, which would save $2 billion over 5 years. In the absence of a marketplace where all insurers or sponsors of patient care programs share responsibility for supporting the academic missions, the historical, explicit payments to teaching hospitals, DGME and IME payments, take on critical importance.
 Page 98       PREV PAGE       TOP OF DOC
    Teaching hospitals also rely heavily on DSH payments, receiving about two-thirds of total payments. Changes in both the welfare and Medicaid programs are eroding the measures in the formulae used to calculate Medicare DSH payments. New measures of service to the low-income poor, such as uncompensated care, should be evaluated with an eye toward targeting DSH dollars toward hospitals that serve the indigent.

The Indirect Medical Education (IME) Adjustment.

    Since the inception of the prospective payment system in 1983, Medicare has made payments through the IME adjustment for the higher operating costs of teaching hospitals. While its label has led many to believe that this adjustment compensates hospitals solely for GME, its purpose is much broader. Both the House Ways and Means and the Senate Committees specifically identified the rationale behind the adjustment: This adjustment is provided in light of doubts...about the ability of the DRG case classification system to account fully for factors such as severity of illness of patients requiring the specialized services and treatment programs provided by teaching institutions and the additional costs associated with the teaching of residents...the adjustment for indirect medical education costs is only a proxy to account for a number of factors which may legitimately increase costs in teaching hospitals (House Ways and Means Committee Report, Number 98–25, March 4, 1983 and Senate Finance Committee Report, Number 98–23, March 11, 1983).

The AAMC believes that any reduction in the current level of the IME adjustment should be proportionate to other reductions in the rate of Medicare spending.

    The President has proposed a reduction in the IME adjustment from its current 7.7 percent for each 10 percent increment in a hospital's resident-to-bed ratio to 5.5 percent by FY 2002, a 29 percent reduction over 5 years. Any proposed decrease in IME payments is not a reduction in the rate of increase of program spending, but is a real cut for teaching hospitals. Coupled with private sector losses, reductions in Medicaid spending and other cuts in projected Medicare payments, reductions in IME or DGME payments will force teaching hospitals to bear an unfair burden of Medicare payment reductions, making it more difficult for them to sustain their additional missions.
 Page 99       PREV PAGE       TOP OF DOC
    As the Congress contemplates adjustments to assure the solvency of the Trust Fund, payments for hospital services are principal targets. All hospitals' Medicare payments are affected by changes in the factor for updating the basic PPS prices, but only certain types of hospitals experience the effect of changes in IME and DSH payment policy. While teaching hospitals recognize the need to control Medicare expenditures to protect the long-term solvency of the program, these institutions would be affected not only by IME reductions, but also by reductions in the update factor and the erosion of DSH payments.
    While some policy makers regard the $4.3 billion in current IME funds as a source from which to obtain budget savings, it must be remembered that IME payments, while a fairly small proportion of total Medicare spending, are absolutely vital to a relatively small number teaching hospitals. ProPAC has shown that IME funding is concentrated in relatively few teaching hospitals: about 140 teaching hospitals receive one-half of all IME payments. If the level of the IME adjustment is reduced, it would have a significant negative impact on the hospitals at the very high end of the continuum of teaching intensity.

The AAMC believes that any changes in Medicare payment policy should be implemented gradually with an annual evaluation of their impact on the financial viability of different groups of hospitals.

    Congressional decisions on Medicare payment policies should be made in the context of their impact on the entire health care system. Non-federal COTH members account for 6 percent of the nation's hospitals, but nearly 2 million, or almost 20 percent, of all Medicare discharges. For many COTH member hospitals, Medicare payments comprise from one-quarter to one-third of all their revenue. Clearly, changes in Medicare payments will have a profound impact on these institutions.
 Page 100       PREV PAGE       TOP OF DOC
    While some policy makers have pointed to teaching hospitals' current PPS inpatient margins as reason to reduce the level of the IME adjustment, the AAMC believes, as explained earlier, that PPS inpatient margins are simply one narrow measure of financial performance. Teaching hospitals' total margins remain below those of other hospital groups. Congress has indicated that the level of the IME adjustment should reflect the broader mission and overall financial viability of teaching hospitals to assure access and quality of care for Medicare beneficiaries and other patients. ProPAC, in its March 1997 report, notes that a ''large and immediate reduction in IME payments might make it difficult for teaching hospitals to support their unique missions'' and has recommended that changes in the level of the adjustment should be ''made gradually and monitored closely to ensure that access... is not adversely affected.''

Three Proposals to Change IME and DGME Payments.

    Proposals to change the rules for making DGME and IME payments have been stimulated by both the need to limit the growth in Medicare expenditures and the need for an appropriately-sized and -trained physician work force. For FY 1998, the Administration has offered three separate proposals to reform GME: capping the total number of residents and the total number of non-primary care residents for IME and DGME payments; allowing GME payments to non-hospital settings; and counting resident' time spent in non-hospital settings for IME payments. Taken together, they would save the program $3.4 billion over 5 years. These proposals seek to achieve a more appropriately configured physician work force by placing limits on the total number of residents for which the program will pay, shifting the balance of primary care and specialist physicians, and encouraging training in non-hospital settings. These proposals are generally consistent with current AAMC physician work force policies.

 Page 101       PREV PAGE       TOP OF DOC
Placing a Cap on the Number of Residents.

    The Administration proposes to cap both the total number of residents and the number of non-primary care residents that may be counted for IME and DGME payments. The Medicare program currently imposes no limit on the total number of residents it will support for DGME and IME payments. While the details of this proposal are not yet available, it attempts to limit the total number of residents the program will pay for and to influence the mix of residents toward primary care specialties.

The AAMC agrees with the Administration that the size of the residency training system needs adjusting, but we believe other approaches also should be considered.

    Last month, the AAMC, in concert with five other national associations, issued a Consensus Statement on the Physician Work Force. It recommends a closer alignment between the number of entry level residency positions and the number of graduates of U.S. medical schools, to be achieved primarily by limiting federal funding for residency positions. Training opportunities for foreign born physicians would continue to be available, but their training would not be financed by the Medicare program.
    At the same time, the AAMC believes that changes in Medicare funding must make provisions for appropriate transitional mechanisms to address the impact on crucial patient care functions of hospitals that are adversely affected by a substantial reduction in the number of residents, many of whom are graduates of foreign medical schools. For some hospitals, where residents provide a large proportion of patient services, the immediate elimination of Medicare support for residents would cause substantial access and service problems for Medicare beneficiaries. The AAMC notes that the Medicare demonstration project in New York, where hospitals that have chosen to participate in the experiment are reducing the number of residency positions by meeting annual targets in exchange for transition funding, includes a process and a time table so that patient access to services is not reduced. The New York experiment offers another approach to adjusting the GME system by uncoupling the level of funding from the number of residents for a limited period.
 Page 102       PREV PAGE       TOP OF DOC
    The AAMC is very aware that, notwithstanding the broad policy view that the nation is headed toward an oversupply of physicians, there remains a significant number of underserved inner city and rural areas. To ensure that physicians set up practices in underserved communities, we recommend that federal incentives be made available to students who pursue careers as generalist physicians and establish practices in inner city and rural areas. To the degree that decreases in the level of GME funding and residency downsizing results in program savings, the AAMC believes that savings could be used to encourage physician migration to underserved communities. The AAMC also supports increased funding for the National Health Service Corps which provides inducements to practice in underserved areas. With respect to the Administration's proposed cap on the number of residents, the AAMC strongly recommends a process that would allow for new training programs and expansion of programs in certain areas. This is particularly important for the development of training sites in rural areas.
    The AAMC believes that reductions in the number of residency positions, or ''right-sizing'' the GME enterprise, should begin with constraining or eliminating support for residents who are not graduates of accredited U.S. medical schools. This process will require difficult decisions. Hospitals that cannot reduce positions due to service demands by the populations they serve must get adequate financial assistance to maintain their patient care mission. However, after the process is complete, the quality of graduate medical education training and the size and compositions of the physician work force will be better aligned.
    Some policy makers have proposed a voucher system to both control the number of residency training positions and encourage ambulatory training. According to this plan, placing the payment in the hands of the resident would enable the trainees to choose the training site. The AAMC believes that the current system has many characteristics of a voucher system. While the resident does not possess an actual voucher certificate, the Medicare program makes a payment to the hospital only if the resident has chosen to train in the hospital's program. While a voucher system appeals to some in terms of achieving national work force objectives, many operational questions about how a voucher system would work, such as the payment amount of the voucher, the number of training years it would cover, and on what basis it would be distributed, are yet to be addressed by policy analysts.
 Page 103       PREV PAGE       TOP OF DOC

Allowing GME Payments to Non-Hospitals.

    Current law regarding Medicare DGME payments explicitly states that DGME payments may be made only to hospitals. The Administration proposes to allow non-hospital settings, such as Federally Qualified Health Centers (FQHCs), to receive GME payments for primary care residents when a hospital is not paying the resident's salary. The AAMC supports changes in Medicare DGME funding to encourage residency training for all specialties, when appropriate, in non-hospital, ambulatory sites, and believes that Medicare DGME payments should be made to the entity that incurs the cost. Recipients of payments could be teaching hospitals, medical schools, multi-specialty group practices or organizations, such as GME consortia, that incur training costs. However, the AAMC does not support payments being awarded directly to training programs, since ultimately the organization in which the program functions must determine the institutional commitment to graduate medical education. The AAMC also does not support allowing non-hospital settings to receive IME payments. The IME adjustment recognizes differences in inpatient costs between teaching and non-teaching hospitals.

Counting Work in Non-Hospital Settings for IME.

    Current law does not allow hospitals to count residents for IME payment purposes in settings other than inpatient hospital units or the hospital outpatient department. Many policy makers believe that this rule should be changed to remove barriers to training physicians in non-hospital, ambulatory settings. The AAMC supports the proposal to count residents' work in non-hospital settings for IME payments. In making this change, hospitals will be able to count residents in non-hospital ambulatory training sites for their resident-to-bed ratios. The total number of residents counted by the hospital would be capped under the Administration's plan at some historical level.
 Page 104       PREV PAGE       TOP OF DOC

Creating Incentives to Control High-Volume Inpatient Physician Services.

    The Administration proposes to limit payments to groups of physicians practicing in hospitals where the volume and intensity of services per admission exceed the national median to encourage physicians to become more efficient. Medicare would withhold 15 percent of each payment for each physician service delivered by a high-cost medical staff starting on January 1, 2000. If physician groups successfully managed the volume and intensity of services during the year, Medicare would return the withhold. If they were over the limit, part or none of the withheld funds would be returned.

The AAMC joins the PPRC and several physician organizations in opposing this proposal.

    The Association believes that applying a prospective withhold is not only inappropriate, but fundamentally contrary to Congressional intent and every physician's participating agreement with the Medicare program. This proposal would have a significant negative impact on both teaching and non-teaching physicians. It would unfairly focus payment reductions on only a small fraction of providers or services. This proposal incorrectly assumes that all physicians in a hospital belong to a single practice group. It introduces physician profiling as a punitive device, rather than as an educational or quality control tool. Perhaps most important, the proposal could have a serious impact on quality of care if the volume targets were set to encourage medical staffs to skimp on care. Finally, it could result in disruptions in certain hospitals as physicians switch institutions. If regional referral centers or tertiary hospitals were designated as ''high cost'' hospitals, Medicare beneficiaries might face limited access to needed specialized services. As the PPRC found several years ago, there is little evidence to support the notion that a state, regional, or even a local level volume performance standard would provide a better incentive for physicians to reduce costs than the national volume performance standard.
 Page 105       PREV PAGE       TOP OF DOC

Conclusion

    Fifty years ago, a revolution in medical innovation began, turning teaching hospitals and medical schools into the backbone of medical progress. An infusion of funds into the central missions of these institutions—research, education and patient care—has stimulated an era of hope and discovery. The rise of managed care, the consolidation of providers and price competition may now unravel the financial threads that hold teaching hospitals and medical schools together.
    Only the federal government can create an equitable, effective system to share responsibility for the nation's health and medical progress. The AAMC believes that establishing a ''shared responsibility'' fund for the special missions of teaching hospitals and medical schools is a crucial and important step for all Americans. This approach to financing the special missions of academic medicine is an issue that deserves the Subcommittee's close attention.
    As part of the federal responsibility, the Medicare program, since its inception, has supported a portion of the costs related to graduate medical education, setting a standard that has been in place for more than thirty years. As the bulwark of graduate medical education financing, Medicare's continued participation—on behalf of its fee-for-service and its managed care beneficiaries—is essential in a competitive delivery system. The program's policies profoundly affect how other payers view their roles in support of GME. The Congress should remember the important precedent established and maintained by the Medicare program when setting a course for the future of GME financing.

—————

 Page 106       PREV PAGE       TOP OF DOC

Appendix A

The Characteristics and Roles of Teaching Hospitals

    Teaching hospitals, in addition to all hospitals' mission of providing basic health services to community residents, have the responsibilities of clinical education for all types of health professionals, provision of an environment in which clinical research can flourish, and highly specialized patient care. These responsibilities are combined in many different ways in individual teaching hospitals, depending upon a hospital's mission, its role in the community, the resources available to it, its past history and its view of the future. As a result, a vast continuum of diverse teaching hospitals exists.

Graduate Medical Education.

    Participation in graduate medical education programs is the characteristic that, by definition, separates teaching from non-teaching hospitals. Upon completion of medical school, physicians continue their medical education by completing at least three years of training in residency programs. While some residencies are based outside the hospital, most graduate medical education is sponsored by hospitals. Medical schools and teaching hospitals have devised a range of relationships for the conduct of graduate medical education. At one extreme, the ''freestanding'' residency is established, staffed and controlled by the individual hospital. At other end of the continuum, the residency program is offered jointly by the medical school and one or more hospitals. Along the continuum are a variety of relationships tailored to the needs, resources, and opportunities available. In 1995, nearly 1,025 short-term, non-federal hospitals provided the training sites for over 88,000 residents and clinical fellows in graduate medical education programs. The Veterans Health Administration also supports almost 9,000 filled residency positions, or 8.7 percent of filled residency training positions in the United States.
 Page 107       PREV PAGE       TOP OF DOC

Undergraduate Medical Education.

    Prior to residency training, medical students complete four years of medical school. The ''hands-on'' clinical education of medical students consists of clerkships in hospitals and other clinical settings during which medical students spend a fixed amount of time under the supervision of faculty and residents in various specialties. Residents contribute substantially to the education of medical students, and their presence is often critical to the success of undergraduate programs. In 1995, over 33,000 undergraduate medical students received clinical training at teaching hospitals or their affiliated educational sites.

Nursing and Allied Health Education.

    Hospitals also remain the primary sites for the clinical training of nurses and other allied health professionals. While classroom training for nurses is now more likely to take place in a college or university, nurses still receive the major portion of their clinical education in hospitals. More than 25 other training programs in allied health fields are widely supported by teaching hospitals, including programs for physical therapists, respiratory therapists, and emergency medical technicians.

Provision of an Environment for Clinical Research.

    The nation's teaching hospitals and medical schools are the backbone of innovation in American medicine because they provide the environment for the conduct of clinical research and the introduction of new life-saving drugs, devices, and procedures into clinical practice. What is now commonly accepted medical care, such as treatment of infectious disease, came from laboratory and clinical research in academic health centers. Open heart surgery and life-saving organ transplantation were pioneered at teaching hospitals. From the use of ether in performing ''painless'' surgery 150 years ago, to the development of neonatal intensive care units, and the promise of gene therapy in curing inherited genetic disease, medical schools and teaching hospitals serve as locations for experimentation and development of new knowledge that benefits the world. Many of these advances began in basic research laboratories of universities and their affiliated hospitals; most of the advances were transferred to patient care as clinical research programs at teaching hospitals. After rigorous evaluation in major medical centers, many of these innovations are adopted in other provider settings. Teaching hospitals offer a natural setting for the advancement and early application of medical knowledge by bringing together seriously ill patients and research-oriented faculty physicians.
 Page 108       PREV PAGE       TOP OF DOC

Provision of Patient Care Services.

    In addition to their education and research missions, teaching hospitals are, first and foremost, providers of a broad range of health care services. They provide all levels of patient care—from preventive to tertiary services. They are local institutions providing basic hospital care in their neighborhoods and communities. They also are referral institutions providing tertiary care to statewide and regional populations, as well as community service institutions caring for patients from all economic and social backgrounds. Because of their research activities, teaching hospitals house the newest and most advanced services and facilities and with residents and supervising physicians available around-the-clock, teaching hospitals often care for the nation's sickest patients.

Why COTH Member Hospitals are Different

    All teaching hospitals share three common objectives: education, research and patient care. However, while teaching and non-teaching hospitals operate in the same general organizational, social and financial environment, academic medical center hospitals, defined as short-term, nonfederal members of the AAMC's Council of Teaching Hospitals and Health Systems (COTH), have distinctive organizational and service characteristics. Membership in COTH requires hospitals to sponsor or participate in at least four approved residency programs and have a signed agreement with an accredited school of medicine. Thus, COTH member hospitals, which include 75 Veterans Affairs medical centers, are the backbone of graduate medical education, training about 75 percent of all residents in the U.S.
    Comparing the 276 short-term general, non-federal members of COTH that reported data to the American Hospital Association in 1994 with the 828 other teaching hospitals and 3,853 non-teaching hospitals reveals striking differences about the characteristics of COTH members. Nearly two-thirds of COTH hospitals, but less than one-half of other teaching hospitals, are located in metropolitan areas of over one million population. In contrast, over one-half of non-teaching hospitals are located in rural areas. COTH hospitals are significantly larger than other hospitals. Over one-half of COTH hospitals have more than 500 beds; in comparison over one-half of non-teaching hospitals have under 100 beds. More than two-thirds of other teaching hospitals have between 100 and 400 beds. As large organizations with multiple responsibilities, COTH members also employ many personnel, often serving as economic engines in their local communities or regions. COTH members are primarily sponsored by non-profit organizations. About one-quarter are state university hospitals or major inner city municipal hospitals. Only three COTH members, or 1 percent, are investor-owned hospitals, while 7 percent of other teaching hospitals and 15 percent of non-teaching hospitals are owned by for-profit entities.
 Page 109       PREV PAGE       TOP OF DOC
    COTH members are major providers of patient care services and offer a wide range of hospital services. In 1994, while comprising only 6 percent of all hospitals, COTH members accounted for 20 percent of all admissions; 21 percent of all births; 23 percent of outpatient visits; and 18 percent of all surgeries performed in short-term, non-federal hospitals.
    COTH members' unique responsibilities compel them to serve the needs of their communities differently than other teaching and non-teaching hospitals:
    •  Seventy-one percent of COTH members operate certified trauma centers, compared to only 29 percent of other teaching hospitals and 13 percent of non-teaching hospitals;
    •  Sixty-three percent of non-federal COTH members provide organ transplant surgical services compared to only 16 percent of other teaching hospitals and 3 percent of non-teaching hospitals;
    •  Ninety percent of all COTH hospitals provide both inpatient and outpatient AIDS services, while 69 percent of other teaching hospitals and only 33 percent of non-teaching hospitals provide similar services;
    •  Ninety-four percent of COTH members provide cardiac catheterization services compared to 68 percent of other teaching hospitals and 33 percent of non-teaching hospitals, and COTH hospitals provide similarly disproportionate amounts of open heart surgery and angioplasty services.
    Teaching hospitals provide a disproportionate share of health care services to the most disadvantaged members of our society. Non-federal COTH members have 18 percent of the nation's beds, but 24 percent of all Medicaid inpatient days. In addition, COTH members provide a disproportionate share of uncompensated care. In 1993, COTH members wrote off 45 percent of the charity care ($4.9 billion) incurred by non-federal hospitals, and 27 percent of all bad debt expense.
 Page 110       PREV PAGE       TOP OF DOC

—————


    Mrs. JOHNSON. Thank you, Dr. D'Eramo.
    Dr. Hopper.

STATEMENT OF CORNELIUS L. HOPPER, M.D., VICE PRESIDENT HEALTH AFFAIRS, UNIVERSITY OF CALIFORNIA

    Dr. HOPPER. Mr. Chairman, distinguished Members of the Subcommittee, I'm Dr. Con Hopper, University of California vice president for health affairs, based in the office of the president in Oakland, California.
    I welcome this opportunity to provide very briefly testimony on behalf of the university with respect to Medicare's financing of graduate medical education and disproportionate share payments to teaching hospitals.
    By way of background, the University of California system has five academic medical centers. Moving from south to north geographically, these are located in San Diego, Irvine, Los Angeles, San Francisco, and in the Davis Sacramento complex. Each of these centers includes, at minimum, a school of medicine and one or more university teaching hospitals, but at UCLA and San Francisco there are also a number of other health profession schools.
    With over 12,000 students enrolled in medicine, dentistry, nursing, public health, pharmacy, and optometry, the University of California operates the Nation's largest health professions education program. Our 2,600 medical students represents nearly two-thirds of California's total medical school enrollment, and the 150 residency programs sponsored by the university provide training opportunities for over 4,400 residents, or nearly one-half of the total in California.
 Page 111       PREV PAGE       TOP OF DOC
    UC's medical centers are key components of the university's health sciences teaching and research mission. As you know, they're organizations where important scientific discoveries are made and translated into state-of-the-art health care services.
    But I would point out that the centers also represent a critical part of California's safety net health care system for its poor and underinsured populations. Three of the five centers are located in counties that have no county hospital and they basically subsume that county hospital function. Collectively, our five centers rank second only to the Los Angeles County hospital system in the total volume of Medicaid patients served.
    The dramatic changes in the organization and financing of health care nationally over the past few years, of which this Subcommittee is well aware, is nowhere more evident than in California. UC's medical centers are located in five of the most competitive markets in this Nation. The California market penetration of managed care, with its associated competition and price-driven reimbursement, is more than twice the national average for the private sector and more than four times the national average for Medicare.
    About 70 percent of California's commercially insured population and approximately 40 percent of California's seniors are currently enrolled in capitated plans and by the end of this year, we expect that 50 percent of California's Medicaid patients are expected to be enrolled in similar plans.
    The move into managed care in the highly competitive, and price-conscious California health care market has had major consequences for our academic medical centers. It has meant reduced access to patients, reduced reimbursement rates for services and, in our contracts with capitated health plans, it has meant our direct assumption of financial risk.
    As you've heard on numerous occasions in the past and today, our legitimate health care costs linked to our educational and research missions are not acknowledged and not reimbursed by the health plans with which we contract, and we're no longer able to shift to private insurers a significant portion of the unreimbursed costs that we incur in caring for large numbers of poor patients.
 Page 112       PREV PAGE       TOP OF DOC
    At the same time, State funds for support of the clinical teaching program have remained static and now represent only 2.1 percent of net revenues at our medical centers. These managed care-driven circumstances have led to a currently precarious financial situation at several of our centers and an uncertain future for all of them.
    I think it's important to ask how we've responded to these challenges. First, from the perspective of our clinical services, our hospitals have collectively stripped more than $275 million out of their cost base over the past 4 years. We've made ourselves more efficient by selective staff reductions, consolidation of administrative functions, and streamlining of operations.
    Our medical school faculty are increasingly reorganizing themselves into functional group practices for their clinical activities, and what this is doing is enabling our hospitals and medical faculty groups to negotiate with managed care health plans in their regional markets as single provider entities.
    Our clinical faculty are likewise, of necessity, becoming increasingly efficient and cost-effective deliverers of managed care. Since access to an adequate number and variety of patients is critical to the academic mission of these centers, as well as to their financial stability, our medical centers are also reaching out and establishing linkages with a variety of community-based health organizations and systems.
    These efforts in our five centers are tailored to the specific opportunities and demands of their five separate regional markets. One unique example of this, of which I'm sure some of you have read, is the recent strategic merger of the hospital and clinic operations of Stanford University and UC-San Francisco, with the respective schools of medicine remaining separate and independent.
    Our second response, and I want to emphasize this, is on the educational side. Very importantly, we are taking concrete steps to address future physician work force needs in response to the changed health care environment. In addition to shifting more of our educational activities into community-based clinical settings, the university is now in its fourth year of a formal memorandum of understanding with the California State government that provides for a substantial increase in both the number and relative proportion of generalists that we train.
 Page 113       PREV PAGE       TOP OF DOC
    We've committed to a system-wide 7-year plan to shift the university's ratio of primary care to nonprimary care residency slots from approximately 45 percent in primary care during the 1992–93 base year up to 55 percent by the year 2000. We're currently on track with this plan. By the end of this fiscal year, that is, July 1, 1997, 50 percent of our physicians will be in primary care.
    I might point out that the president of the University of California, a few months ago, appointed a special commission, a national Commission on the Future of Medical Education at the University of California. That commission will report in June, and I fully anticipate that one of their recommendations will be a major acceleration in the downsizing of our residency program in California.
    Notwithstanding these efforts to control our costs and meet the demands of our competitive market and be educationally responsive, there has been a progressive erosion in the operating financial margins of our academic medical centers. A margin of 5 to 7 percent is required to sustain the operation of these centers and to meet their obligations for the support of medical education and research.
    However, from a high of 7.1 percent in 1991–92, our centers, by last year, had dropped to a margin of 2.8 percent, and projections for the future are not at all encouraging. This problem, of course, is not unique to California, as you've heard today.
    At the present time, the competitive market in California has only one significant source of support left for graduate medical education, and that is the Medicaid Program. Yet even here, there are significant problems because, again as you've heard, the current Medicare and managed care reimbursement formula, known as the AAPCC, diverts educational funding from those institutions which train physicians. Yet there is no requirement that HMOs or managed care organizations either contract with teaching hospitals or pass along the payment adjustments intended for education.
 Page 114       PREV PAGE       TOP OF DOC
    The California Health Care Association estimated that nationally, these managed care organizations received about $568 million in 1995 that would otherwise have gone to teaching and disproportionate share hospitals under a fee-for-service methodology. The loss for California hospitals alone, where managed care is so much greater, was about $253 million or 45 percent of the national total. As more and more seniors enroll in managed care and if this inequity is not addressed, our losses will be even greater.
    This issue is critical for teaching hospitals in California and we join with the AAMC in recommending that IME, DME, and DSH payments should be carved out from the Medicare managed care payments and directed to those institutions that incur the actual educational costs.
    We understand that this step may not be the long-term solution this Subcommittee seeks for the financing of graduate medical education and neither do we, but we will address two important short-term goals that we would urge you to consider.
    First and foremost, carving out the mission-related payments and directing them to the teaching hospitals would have the practical results of leveling the playingfield and allowing us to compete more effectively.
    Second, this approach would allow Congress to take the first step toward reestablishing graduate medical education as an important ingredient in the Nation's health and well-being or, in other words, as a public good.
    I think our arguments about DME and IME payments for graduate medical education you have heard on a number of occasions today. Again, we support the position that the AAMC has taken on this.
    Mr. Chairman, as you search for new and innovative approaches to the financing of graduate medical education, I hope you will draw on the experience of California academic medical centers in the Nation's most advanced managed care market. We have some significant successes to share with you, as you know from your recent visit to UCLA, but we also face some extremely difficult challenges.
 Page 115       PREV PAGE       TOP OF DOC
    Continuing to meet our important safety net provider responsibility while sustaining our outstanding educational and research programs require that our academic medical centers remain, first and foremost, financially viable. We seek your understanding and assistance in achieving that goal, and thank you for the opportunity of providing this testimony.
    [The prepared statement follows:]

Statement of Cornelius L. Hopper, M.D., Vice President, Health Affairs, University of California

    Mr. Chairman, and distinguished Members of the Subcommittee, I am Dr. Cornelius L. Hopper, University of California Vice President for Health Affairs, based in the Office of the President in Oakland. I welcome this opportunity to testify today on behalf of the University with respect to Medicare's financing of graduate medical education and disproportionate share payments to teaching hospitals. Before I begin my testimony, I want to say what a pleasure it is to appear before this Subcommittee.
    The University of California system has five academic medical centers. Moving from south to north geographically, these are located in San Diego, Irvine, Los Angeles, San Francisco, and Davis/Sacramento. Each of the centers includes, at minimum, a school of medicine and one or more university teaching hospitals. The San Francisco and Los Angeles health sciences centers also include a number of other health professions schools. With over 12,000 students enrolled in medicine, dentistry, nursing, public health, pharmacy, and optometry, the University of California operates the nations largest health professions education program. Our 2,600 medical students represent nearly two-thirds of California's total medical school enrollment. The 150 residency programs sponsored by the University provide training opportunities for more than 4,400 residents, or nearly half of California's total.
    UC's medical centers are key components of the University's health sciences teaching and research mission; organizations where important scientific discoveries are translated into state-of-the-art health care services. But the centers also represent a critical part of California's ''safety net'' health care system for its poor and underinsured populations. Three of the five centers are located in counties that have no county hospital. Collectively, the five centers rank second only to the LA county hospital system in the total volume of Medicaid patients served.
 Page 116       PREV PAGE       TOP OF DOC
    The dramatic changes in the organization and financing of health care nationally over the past few years, of which the committee is well aware, is no where more evident than in California. UC's Medical centers are located in five of the most competitive markets in this nation. The California market penetration of managed care, with its associated competition and price-driven reimbursement, is more than twice the national average for the private sector and more than four times the national average for Medicare. About 70 percent of California's commercially insured population and approximately 35–40 percent of California's seniors are in capitated plans. Fifty percent of California's Medicaid population is expected to be in enrolled in managed care plans by the end of this year. Less than 5% of the non-Medicare and non-Medicaid patients seen at our medical centers in 1995–96 were covered under fee-for-service arrangements.
    This move into managed care in the highly competitive, price-conscious California health care market has had major consequences for our academic medical centers. It has meant reduced access to patients, reduced reimbursement rates for services, and, in our contracts with truly capitated health plans, it has meant our direct assumption of financial risk. Legitimate patient care costs linked to our educational and research missions are not acknowledged—and not reimbursed—by the health plans with whom we contract, and we are no longer able to shift to private insurers a significant portion of the unreimbursed costs that we incur in caring for large numbers of poor patients. At the same time, state funds for support of the clinical teaching program have remained static and now represents less than 2.1 percent of net revenues at our medical centers. These managed care-driven circumstances have led to a precarious financial situation at several of our centers.
    How have we responded to these major challenges? First, from the perspective of our Clinical Enterprise, our hospitals collectively have stripped more than $ 275 million out of their cost base over the past four years. We have made ourselves more efficient by selective staff reductions, consolidation of administrative functions, and streamlining of operations. Our medical school faculty are increasingly reorganizing themselves into functional group practices for their clinical activities, and this is enabling the hospitals and medical groups to negotiate with managed care health plans in their regional markets as single provider entities. Clinical faculty are likewise, of necessity, becoming increasingly efficient and cost-effective deliverers of managed care. Since access to an adequate number and variety of patients is critical to the academic mission of these centers as well as to their financial stability, our medical centers are also reaching out and establishing linkages with a variety of community-based health care organizations and systems. These efforts are tailored to the specific opportunities and demands of their five separate regional markets. One unique example of this is the recent strategic merger of the clinical service operations of the still independent and separate UCSF and Stanford medical schools.
 Page 117       PREV PAGE       TOP OF DOC
    On the educational side, we are likewise taking important steps to address future physician work force needs in response to the changed health care environment. In addition to shifting more of our educational activities into community-based clinical settings, the University is now in its fourth year of a formal agreement with the Governor and Legislature that provides for a substantial increase in both the number and relative proportion of generalists trained. We have committed to a system wide 7 year plan to shift the University's ratio of primary care to non-primary care residency slots from approximately 45%—55% during the 1992–93 base year to 55%—45% by July 2001. We are currently on-track with this plan, with approximately 50% of July 1, 1997 positions in primary care.
    Notwithstanding these efforts to control our costs and meet the demands of our competitive market, there has been a progressive erosion in the operating margins of our academic medical centers. A net gain of 5–7% is required to sustain the operations of these centers and to meet their obligations for the support of medical education and research. However, from a high of 7.1% in 1991–92, our centers sustained a drop in net gain by 1995–96 to level of 2.8% and projections for the future are not encouraging. The problem, of course, is not unique to California. There is growing national concern about the loss of clinical revenues and the consequences for research. Dr. Harold Varmus, Director of the National Institutes of Health (NIH) has set up a commission to investigate the issue.
    At the present time, the competitive market in California has only one significant source of support left for Graduate Medical Education: the Medicare program. Yet even here, there are significant problems because the current Medicare managed care reimbursement formula (known as the Adjusted Average Per Capita Cost (AAPCC) diverts educational funding from those institutions which train physicians. Medicare's payment to health maintenance organizations that enroll Medicare beneficiaries includes reimbursement adjustments for direct medical education (DME), indirect medical education (IME), and disproportionate share hospitals (DSHs). Yet there is no requirement that the HMO or managed care organization either contract with teaching hospitals or pass along the payment adjustments intended for education. The California Hospital Association has estimated that nationally these managed care organizations received about $568 million in 1995 that would otherwise have gone to teaching and disproportionate share hospitals under the fee-for-service payment methodology. The loss for teaching hospitals in California, where managed care is so much greater, was about $253 million or 45 percent of the total. As more and more seniors enroll in managed care, our losses will be even greater.
 Page 118       PREV PAGE       TOP OF DOC
    This issue is critical for teaching hospitals in California. We believe that the current policy creates an uneven playing field between teaching and non-teaching hospitals—and represents a windfall for managed care organizations. We believe that the IME, DME and DSH payments should be ''carved-out'' from the Medicare managed care payments and directed to those institutions that incur the actual educational costs. We understand that this step may not be the long-term solution that this Committee seeks for the financing of graduate medical education, and neither do we, but it will address two important short term goals that we would urge you to consider. First and foremost, carving out the mission-related payments and directing them to the teaching hospitals would have the practical result of leveling the playing field and allowing us to compete more effectively. Second, this approach would allow Congress to take the first step toward reestablishing graduate medical education as an important ingredient in the nation's health and well-being; i.e. as a ''public good.''
    The experience of our academic medical centers in California's competitive market place, which will be the experience of much of the rest of the nation in the near future, dramatically underscores the need for a reliable source of funding for graduate medical education. We commend your efforts, Mr. Chairman to establish a separate trust fund for paying for graduate medical education. We believe that such a trust fund should be in addition to, and not a substitute for, Medicare's current contributions to graduate medical education. Our experience in California strongly suggests that such a trust fund should be based on a principle of shared responsibility and participation of all payers.
    Medicare's separate direct (DME) and indirect (IME) payments for graduate medical education and payments to hospitals that treat a disproportionate number of Medicare patients (DSH) together account for almost 10 percent of the total net revenues of our medical centers. These are, by far, the most significant funds that we receive that uniquely address the costs of our special mission. While all hospitals that treat Medicare patients are affected by changes in the Medicare reimbursement update factor, only teaching and DSH hospitals are subject to changes in IME, DME, and DSH funding. We ask that as you look for ways to balance the budget that these special payments be maintained at a level that will allow our teaching hospitals to continue to fulfill their teaching, research, and service missions.
 Page 119       PREV PAGE       TOP OF DOC
    Mr. Chairman, as you search for new and innovative approaches to the financing of graduate medical education, I hope you will draw on the experience of California's academic medical centers in the nation's most advanced managed care marketplace. We have some significant successes to share with you, as you know from your recent visit to the UCLA Medical Center; but we also face some extremely difficult challenges. As I pointed out earlier, our hospitals in San Diego, Orange and Sacramento counties have historically served as the safety net providers for uninsured and underinsured patients. Just last year our medical centers collectively provided about $ 375 million in uncompensated care to such populations. Continuing to meet this important responsibility while sustaining our outstanding educational and research programs requires that our academic medical centers remain financially viable. We seek your understanding and assistance in achieving that goal.
    Thank you for the opportunity to present this testimony. UC stands ready to work with you and the members of your Subcommittee. I will be happy to answer your questions at this time.

—————


    Mrs. JOHNSON [presiding]. Thank you, Dr. Hopper.
    Dr. Foreman.

STATEMENT OF SPENCER FOREMAN, M.D., PRESIDENT, MONTEFIORE MEDICAL CENTER, BRONX, NEW YORK; ON BEHALF OF GREATER NEW YORK HOSPITAL ASSOCIATION

    Dr. FOREMAN. Good afternoon, Madam Chairman, and Members of the Subcommittee.
 Page 120       PREV PAGE       TOP OF DOC
    Though my written testimony offers a broader perspective, I would like to focus my oral comments on the New York Medicare GME demonstration project, which has generated so much recent interest.
    I must say, Madam Chair, I have had the privilege of appearing before this Subcommittee a number of times before. This is the first time I ever felt I ought to be strapped into my chair with a seat belt.
    Here are the facts on the New York demonstration project. Forty-two public and voluntary teaching hospitals and academic medical centers, which collectively train about two-thirds of New York State's 15,000 resident physicians, will participate in a demonstration project which will reduce the number of their residents by about 25 percent.
    The Medicare GME payments for the residency positions that are eliminated will be phased out, but the phaseout will be more gradual than the reduction of residents.
    At the end of the demonstration, all of the reimbursement will have been eliminated, and the Medicare Program will have a permanent savings thereafter.
    Specifically, over the next 6 years, the positions eliminated will generate Medicare savings of about $700 million. The hospitals will retain $400 million of that, and the Medicare Program, $300 million. At the end of the experiment, Medicare will have an ongoing permanent annual savings of $200 million per year.
    Our association proposed the idea for the demonstration in response to a growing concern among policymakers and the profession that there existed a physician surplus which should be addressed by restricting opportunities for residency training, a concern that had focused heavily on New York because it trains 15 percent of the Nation's residents.
    Our chief worry was that the loss of Medicare reimbursement in any downsizing would cause real harm to New York's very fragile health care system because even though 100 percent of the Medicare reimbursement is lost for each resident physician cut, 100 percent of our cost is not. That is because resident physicians are not merely helpers in the care process. They perform an extraordinarily important range of patient care services, most particularly for uninsured patients, or those covered by Medicaid. In fact, in the city of New York, they are the backbone of the care system for these populations.
 Page 121       PREV PAGE       TOP OF DOC
    Although this is generally true of teaching hospitals around the country, nowhere is the problem of the magnitude it is in New York. Forty-eight percent of New York City's population, 3.5 million people, are either uninsured or on Medicaid. These patients, by and large, cannot afford private care either in a doctor's office or in a hospital, and teaching hospitals have for years filled this gap through training programs in which carefully supervised residents provided their care.
    When resident positions are eliminated, the patient care needs formerly met by those residents are left exposed, for there are no other sources of payment to cover them. It was this problem that led us to propose the demonstration because we saw it as a way to model a workable solution and to create crucial time and resources that could allow teaching hospitals on a voluntary basis to phase down their training programs.
    I would like to stress a few important points about the New York demonstration. First, the cuts are real. HCFA will carefully monitor each year's resident count. If we do not meet our yearly reduction targets, we will lose the partial reimbursement we would have received in that year, and if we do not meet the final targets, we will have to pay it all back.
    Second, these reductions cannot be taken at the expense of training programs in primary care, which must either increase or stay the same.
    Third, while the project imposes strict parameters, teaching hospitals and their faculty will be free to make their own decisions about how best to meet the criteria and adjust their academic missions.
    Fourth, although the project does not specifically target reductions in international medical graduates, a policy position which we oppose, we believe American graduates will continue to be more competitive for the smaller number of positions and the practical result of the experiment may well be a reduction in international medical graduates. It will be a topic for evaluation in the study.
 Page 122       PREV PAGE       TOP OF DOC
    Finally, we believe that although the demonstration project is not perfect, we see it as a real opportunity to explore ways to respond to a national concern about physician oversupply and to achieve Medicare savings through a process that permits us to continue to meet the needs of vulnerable communities.
    I would add only one last comment. I would disagree with Commissioner Wilensky's criticism of the lack of an assessment made during the last panel. HCFA is planning a detailed examination of this demonstration project, consistent with appropriate research techniques. Moreover, all demonstration participants will be audited annually on the achievement of their goals.
    I thank you, Madam Chair.
    [The prepared statement and attachment follow:]

Statement of Spencer Foreman, M.D., President, Montefiore Medical Center, Bronx, New York; on Behalf of Greater New York Hospital Association

    Thank you, Mr. Chairman and members of the Subcommittee for inviting the Greater New York Hospital Association (GNYHA) to testify before you today on teaching hospitals and Medicare disproportionate share hospital (DSH) funding.
    My name is Dr. Spencer Foreman. I am the President of the Montefiore Medical Center in the Bronx, the university hospital for the Albert Einstein College of Medicine (AECOM) responsible for the training of 350 medical students and 1,200 resident physicians across several hospital and community-based sites. I am a past Chairman of GNYHA, which represents 175 not-for-profit voluntary and public hospitals and nursing homes in the New York metropolitan area and a current Commissioner of the Prospective Payment Assessment Commission (ProPAC). I have also served on the Committee on the U.S. Physician Supply for the Institute of Medicine and as past Chairman of the Association of American Medical Colleges.
 Page 123       PREV PAGE       TOP OF DOC
    Montefiore is the largest not-for-profit provider of health care and related services to the 1.2 million residents of the Bronx, a community whose socioeconomic and health status indicators rank among the poorest in the United States. Today, Montefiore includes two hospitals that provide more than 47,000 acute care discharges each year; three skilled nursing facilities; one of the nation's largest hospital-based home health agencies; 13 community-based comprehensive primary care centers which provide almost 285,000 ambulatory care visits annually to patients from underserved communities and serve more than 10,000 children in school-based clinics; a 14-site multi-specialty physician group practice that provides about 160,000 annual visits to HMO and fee-for-service patients; and a range of other community-based health services. Through contracts with the City of New York, Montefiore provides physician and other professional services to two public hospitals, North Central Bronx Hospital and Jacobi Medical Center, and health and mental health services at the 16,000 bed Riker's Island Detention Center. With roughly 11,000 employees, Montefiore is the largest private employer in the Bronx.
    You have called this hearing today to discuss subjects that are of critical importance to the members of GNYHA, namely, teaching hospitals and Medicare DSH payments. We urge the Subcommittee to: (1) support the enactment of a critically needed national graduate medical education (GME) trust fund in which all payers equitably support the mission of teaching hospitals; (2) ensure that teaching and DSH hospitals do not bear more than their fair share of reductions in Medicare payments; and (3) correct flaws in current methodologies used to set Medicare managed care payments to HMOs by carving out the GME and DSH components of those rates, paying them according to current Medicare payment rules to the hospitals entitled to receive them, and ensuring that any further adjustments to AAPCC rates be applied in a manner that recognizes differences among regions of the country, such as variations in the cost of living.

New York Medicare GME Demonstration Project
 Page 124       PREV PAGE       TOP OF DOC

    I would like to begin my testimony by describing a voluntary GME demonstration that will begin in New York State this July. The genesis and concept of the New York Medicare GME Demonstration Project in many respects reflects GNYHA's views in the crucial areas that are the subject of this hearing. I will then offer specific comments on the Medicare DSH program and teaching hospital payments.
    The GME demonstration, which was recently approved and announced by the U.S. Health Care Financing Administration (HCFA), is projected to result in a decrease in New York State of 2,228 residency positions and a simultaneous 13% increase in the proportion of primary care and OB/GYN residents in training over the next five years. It will phase out, rather than immediately eliminate, the Medicare GME reimbursement associated with these resident reductions. This will help ensure that educational and patient care quality, as well as the needs of underserved populations, are not compromised in the transition to a less resident-dependent hospital system.

GME Policy Proposals Have Not Met the Needs of Teaching Hospitals in New York

    Past and current proposals to restructure the numbers and types of doctors trained in America have greatly troubled GNYHA. These proposals have focused, among other things, on the elimination of Medicare GME reimbursement for graduates of international medical schools (IMGs);(see footnote 2) reduction in the prices paid for all residents, but especially those in non-primary care specialties; and the establishment of new Federal regulatory bodies to allocate residency positions around the country. The principal goals behind such proposals have been to achieve Medicare program savings and to address perceived problems in the supply and specialty mix of physicians nationwide.
 Page 125       PREV PAGE       TOP OF DOC

    We have always been guided in our evaluation of proposals such as these by concerns about the effects they would have on our precious biomedical complex, as well as their impact on the patients we serve. Our member hospitals serve significant populations of uninsured and low income individuals and families for whom residents are important caregivers. Our evaluations have resulted in deep concern that proposals such as these would wreak havoc on our health care delivery system.

New York's Teaching Hospitals Serve Vulnerable Populations

    Teaching hospitals in New York, which include some of the premier academic medical centers in the nation, have training and patient care missions that are inextricably linked. This is shown in the overlap between hospitals receiving both teaching and DSH payments. As you know, Medicare DSH payments are provided to hospitals that serve a large volume of low income Medicare and other patients.(see footnote 3) Nationally, teaching hospitals receive about 65% of all DSH payments, while 74% of teaching payments go to DSH hospitals.(see footnote 4) This reflects the important role teaching hospitals play in caring for underserved populations. In New York City, the overlap is even more dramatic; in 1996, an estimated 99% of Medicare DSH payments were paid to teaching hospitals, and 85% of teaching reimbursement was paid to hospitals also receiving DSH. Thus, virtually all teaching hospitals in New York City receive DSH payments, and the majority are classified as so-called ''high DSH'' hospitals, a reflection of their low income patient shares.

    It is not surprising that teaching hospitals in New York City are relied upon to this extent to care for medically underserved populations. In 1995, 21.5% of the City's population, or 1.6 million people, were uninsured and another 26%, or 1.9 million people, were Medicaid beneficiaries.(see footnote 5) Unfortunately, the number of persons without insurance is growing, not shrinking; from 1994 to 1995, the proportion of uninsured in New York City rose by a rate of 13%, and this trend is continuing.
 Page 126       PREV PAGE       TOP OF DOC

    Indigent patients lack private physicians. Consequently, residents, who ''learn by doing'' under the close supervision of attending physician faculty, serve as important providers of inpatient and ambulatory care services to such populations. Teaching hospitals, therefore, are at the front line of meeting the health care needs of the 48% of the City's population that lack insurance or are supported by Medicaid.

Past Proposals and Current Policy

    Due to our multiple missions, as well as the poor financial status of our hospitals resulting from years of stringent State regulation of hospital prices, New York's teaching hospitals are particularly vulnerable to cuts in GME and DSH reimbursement. Therefore, we have been adamantly opposed to reductions in these programs.
    For example, proposals to eliminate or curtail reimbursement for IMGs, many of whom are American citizens who attended medical school abroad, would devastate hospitals in underserved urban and rural communities in which IMGs provide essential health care services and draw arbitrary distinctions among residents that are not related to quality, service, or work force goals. IMGs are often concentrated in hospitals located in underserved areas that experience difficulties in recruiting graduates of American medical schools.
    The recognition by some that IMGs would have to be replaced by alternative caregivers supported by a degree of funding is an important first step in recognizing the importance of IMGs to underserved communities. However, issues related to the need for permanent funding, personnel recruitment issues in these communities, and the actual impact of withdrawing support for IMGs have not been adequately explored. An IMG cut is too simple a solution for a complex set of problems.
 Page 127       PREV PAGE       TOP OF DOC
    Other past proposals of concern would have established new residency allocation bodies to assign residency slots to teaching hospitals around the country. This policy would have wrongly relied upon the government to make decisions about the numbers and types of physicians needed in the country. These decisions are best left to educators and health care professionals.
    Finally, GNYHA has noted that current Medicare GME payment rules, which eliminate all GME reimbursement for a resident when that resident is cut, create disincentives to reduce the number of residents in training. This is because a hospital loses 100% of the direct and indirect medical education reimbursement associated with the trainee, yet hospital costs do not decrease in like amounts, particularly if the hospital serves uninsured and underserved populations. Physician residents provide valuable patient care services under close faculty supervision while they participate in GME programs. There is a need to continue providing these services with replacement personnel such as full time physicians, nurse practitioners, and others, as well as by making fundamental changes in the way services are delivered, when a resident is cut. The loss of residents coupled with the elimination of funds to supply these services has made it difficult to implement residency reductions.
    Because of these concerns, GNYHA's leadership last year decided to try to pursue a more proactive course through development of a GME demonstration to test ways to address the issue of resident downsizing while also meeting the needs of New York's medical centers and our communities. The demonstration project represents an important initiative because it strikes a balance between government intervention, market forces, and local initiative. In particular, it allows hospitals to meet the needs of their communities while dramatically reducing the size of residency programs and producing permanent Medicare savings.
    It should be clearly understood that nothing in the design of this project shields demonstration participants from the effects of Congressional action. The demonstration's parameters specifically provide that any legislative or other changes that apply to other hospitals will apply in full force to demonstration participants for the residents they continue to train.
 Page 128       PREV PAGE       TOP OF DOC

Need for the Demonstration Project

    In each of the past two years, more than half of all U.S. medical school graduates (USMGs) have entered residency programs in the generalist fields of family medicine, internal medicine, or pediatrics. This dynamic has not been lost on teaching hospitals, which have begun to take a hard look at their training programs and consider appropriate ways to downsize and rebalance them. In addition, the cost-cutting pressures of managed care have required hospitals to seek new and more streamlined ways of delivering services.
    Recent trends in the number of accredited residency positions nationally shows that growth has slowed and flattened out, likely reflecting these considerations. Thus, while data from the American Medical Association show that the total number of residents in accredited training programs for the past five years increased by 13.7%, from 86,217 in program year 1991–1992 to 98,035 in program year 1995–1996, the last year for which published data are available, the total number increased by only 0.2% from the 1994–1995 to 1995–1996 program year.(see footnote 6)

    However, the challenge in reducing the size of residency programs, particularly at the levels required by the demonstration, is to find a way to pay for the critically important patient care services that would be lost as residency programs shrink or disappear and to make a transition to a delivery system that depends less on residents. Even if some downsizing is occurring on its own in different parts of the country, it is highly doubtful that it would occur at the levels and under the restrictions that will occur under the demonstration. In addition, according to GNYHA staff who helped develop the demonstration and assisted hospitals in evaluating whether or not to participate, many institutions clearly decided to join only because of the special reimbursement features, and others elected not to join because of concerns that the requirements were too stringent or that they could not meet their patient care needs even with the demonstration's special provisions.
 Page 129       PREV PAGE       TOP OF DOC

Process for Securing Demonstration Approval

    Some questions have been raised about the genesis of the demonstration project. It was developed by GNYHA and proposed to HCFA in the summer of 1996. The final terms and conditions developed by HCFA received clearance from the Office of Management and Budget (OMB) in early February, 1997.
    The time frame used to develop the demonstration was driven by the deadline for submission of available residency positions under the National Resident Match Program (NRMP).(see footnote 7) Under the NRMP calendar, hospitals were required to submit a list of positions and desired candidates by February 21, 1997 for the class of residents entering in July 1997. A hospital's submission of a match list constitutes a commitment on its part to accept applicants that are ''matched'' to it through the NRMP. Therefore, February 21, 1997 was the outside date by which to influence the resident class that will enter this July.

Description of Demonstration Features

    Participants in the demonstration have agreed to reduce the full-time equivalent (FTE) number of residents by 20% to 25% from the current level over the next five years, as well as to abide by a number of strict terms and conditions. In return, the GME reimbursement associated with these residents will be phased out over six years rather than eliminated immediately as it would be under current Medicare payment policy. The phase-out of this reimbursement provides crucial time and resources to implement new ways of meeting patient service needs formerly met by residents.
 Page 130       PREV PAGE       TOP OF DOC
    Forty two New York State teaching hospitals that train more than two-thirds of the 15,000 residents in New York State will participate in this voluntary demonstration. They include:
    •  5 voluntary not-for-profit major academic medical centers, including my own;
    •  13 public hospitals, including all of those in the New York City Health and Hospitals Corporation and two county hospitals; and
    •  24 voluntary not-for-profit major and community teaching hospitals.
    Thus, a full spectrum of teaching hospitals is represented.
    Applicability to Medicare managed care. The demonstration is limited to the fee-for-service system and does not affect payments to or from Medicare managed care plans.
    Participation Criteria. Hospitals were able to participate in the demonstration individually; jointly with one or more otherwise eligible hospitals; or through an organized GME consortium recognized by New York State. Applicants had to select one of several participation options, and to meet the requirements of the demonstration in five years. Performance will be measured against the 1996–1997 residency program year.
    The basic participation criteria include reducing the total number of FTE residents by 25% from current levels while maintaining the same ratio of primary care(see footnote 8) and OB/GYN to total residents. There are two alternatives to the basic criteria. In order to increase primary care training, the demonstration allows participants that increase the proportion of primary care and OB/GYN residents by 20% to reduce the total number of residents by 20%. And, in order to support the development of GME consortia, the demonstration allows a consortium organized and recognized pursuant to state law to reduce the total number of residents by 20% while maintaining the proportion of primary care and OB/GYN training at current levels.
 Page 131       PREV PAGE       TOP OF DOC

    The demonstration encourages collaborative efforts by allowing joint and consortium applicants to meet the participation criteria in the aggregate across all sites. Thus, individual sites within a consortium may reduce by more or less than 20% so long as the consortium in the aggregate cuts 20% of its base year residents. This permits an important level of flexibility to ensure that reductions are not made in an overly rigid way. Eighteen of the 42 participants belong to GME consortia, and another 12 will participate jointly with one or more other hospitals.
    Phase-out of GME Reimbursement. The reimbursement associated with cut residents will be phased out over the six years of the demonstration according to the declining percentages shown below.(see footnote 9) Reimbursement is completely eliminated after six years. Hospitals will use these declining levels of reimbursement, or transition amounts, to meet patient care needs in the absence of residents by hiring replacement personnel, such as full time physicians and other appropriate health professionals, and re-organizing the way that care is delivered to accommodate a smaller resident complement.

Table 5

    The phase-out of reimbursement accelerates as the number of cut residency positions grows. That is, a participant will absorb a greater percentage of the loss associated with a larger number of cut residents as the demonstration proceeds.
    Penalties for Missed Targets. HCFA will annually measure the performance of each participant to determine if it is in compliance with achievement of the final requirements and interim annual goals. If a participant deviates from the ''glide path'' that will bring it to the final reduction numbers, it forfeits all transition amounts for that year. If a hospital fails to maintain its reduced resident complement in the sixth year of the demonstration, it will forfeit transition amounts in that year and experience a retroactive halving of the amounts retained in the fifth year. And, if it fails to meet its reduction and specialty mix targets by the end of five years, it will be required to pay back all transition amounts retained since the inception of the demonstration.
 Page 132       PREV PAGE       TOP OF DOC
    Medicare Savings. Medicare will save about $300 million over the life of the demonstration and about $200 million per year after that. Specifically, the loss to hospitals from residency reductions is projected under current payment rules to be about $700 million over six years. As illustrated by the figure below, about 57% of that loss will be retained by the hospital as transition amounts during the six years, while the other 43%, or $300 million, is immediately taken as Medicare savings. After the sixth year, all savings will accrue to the Medicare program.
    Other Terms and Conditions. The demonstration imposes three significant caps on participants. First, the operating indirect medical education (IME) intern-and-resident-to-bed ratio (IRB) and capital IME intern-and-resident-to-average daily census ratio (IRADC) are capped at the 1996 level for all residents in training. In addition, the number of residents in programs sponsored by a participant and rotated to other teaching sites is capped at the current level, preventing participants from increasing rotations elsewhere as they reduce residents at their sites. And, hospitals in GME consortia that are not participating in the demonstration are nonetheless required to cap the number of residents trained at the current level.
"The Official Committee record contains additional material here."

STRIP OFFSET FOLIO 12 HERE

International Medical Graduates

    The demonstration design very appropriately does not target IMGs for reduction. However, a central evaluation question will likely focus on the practical effects of the project on these residents. Many anticipate that the overall reduction in available positions may well result in lowered proportions of IMGs because American medical school graduates will continue to be more competitive, but for a smaller number of available slots. As IMGs currently fill about half of the positions in New York, there will be ample room for American graduates.
 Page 133       PREV PAGE       TOP OF DOC
    However, the demonstration offers crucial flexibility to hospitals to continue to train IMGs based upon the quality of their candidacies and in underserved areas that continue to experience difficulty recruiting American medical school graduates. This flexibility is essential to ensure that health care services to needy populations are not compromised.
    As in the past, GNYHA is absolutely opposed to targeting IMGs for reduction or loss of Medicare support. Drawing a bright line distinction based upon medical school location has a superficial appeal as an easy way to cut GME programs. However, states that rely heavily upon IMGs, such as New York, New Jersey, Nevada, North Dakota, Illinois, Connecticut, Michigan, and others, would be hard hit by such an approach, which would strip their accredited training programs of the support given to identical programs elsewhere in the country. Such an approach would inappropriately deprive communities served by IMG residents of the services of fine and fully qualified American and foreign physicians.
    The concern that IMGs contribute to a physician oversupply results in large part from the pressures that managed care is placing on all health care providers. Whether a mild physician oversupply is harmful is the subject of different perspectives on which I offer no opinion. I will note that at Montefiore, we have witnessed a changing physician environment that has benefited the communities we serve. We have been aggressively pursuing the development of ambulatory capacity in both low income and middle class neighborhoods, and are grateful to find that physician recruitment has become easier as it appears that physicians are exploring a wider array of options than they had in the past. IMGs are not the sole cause of or solution to the problems that have been raised, and we urge that you not support proposals to withdraw Medicare's support for their training.

Process for Making Resident Reductions

 Page 134       PREV PAGE       TOP OF DOC
    We believe that the demonstration combines important elements of market forces with responsible government intervention. While the project imposes clear parameters for residency reduction and primary care/OBGYN training, teaching hospitals and their faculty will be free to make their own decisions about how best to meet these criteria and adjust their academic mission to a market being reshaped by managed care. If done properly, it should preserve programs that train physicians in specialties for which the long term need is high and downsize those for which it is not.
    Participants have established new internal processes to make these decisions, which will likely evaluate a combination of factors such as the quality of programs, job placement experience of recent graduates, service needs within the hospital, the availability and cost of strategies to replace or re-organize services to accommodate the loss of residents, and other variables. The crucial phase-out of GME reimbursement will allow participants to make these decisions as part of a transition to a different type of hospital system.
    The demonstration is not perfect in every respect. Some might find its requirements somewhat rigid and difficult to meet, and others might worry about meeting future patient care needs. However, from our perspective, the demonstration represents a real opportunity to explore ways to respond to national concerns about physician supply and achieve Medicare savings while we also adjust our own activities and structures to adapt to a changing health care system, maximize the quality of patient care and training, and ensure that the needs of our local communities continue to be met.

Critical Need for a National GME Trust Fund

    GNYHA strongly supports Senator Daniel Patrick Moynihan's ''Medical Education Trust Fund Act of 1997'' (S. 21) and urges Congress to include Senator Moynihan's bill in its FY 1998 budget proposals. Congresswoman Nita Lowey recently introduced the companion bill in the House (HR 881). And, of course, this Committee also included a trust fund provision in the Balanced Budget Act.
 Page 135       PREV PAGE       TOP OF DOC
    A Trust Fund is essential to ensure that the responsibility for paying for the training of our nation's future doctors is shared by all payers. The Trust Fund should be funded at levels adequate to ensure that we can maintain our preeminent training system, and all payers should pay their proportionate share. While the details of how the Trust Fund would be operated need to be worked out, and under the Moynihan/Lowey bill would wisely be determined by a new advisory commission, we believe funds should continue to flow directly to the institutions that incur the costs of training according to current Medicare payment rules. In addition, we would suggest that you consider fashioning payment rules that would avoid imposing harsh financial losses for reducing the numbers of residents trained. This could facilitate a process for downsizing and rebalancing GME programs where it makes sense.
    I would also like to emphasize the need for both a carve out of GME from rates paid to Medicare managed care plans, discussed below, and creation of a Trust Fund. The carve-out is needed to maintain Medicare's effort and the Trust Fund is needed to ensure that non-Medicare payers contribute. The proposals, therefore, are not the same and instead are distinct and necessary components of the support needed for GME. The Moynihan/Lowey bill, of course, contains both of these important proposals.
    You may be aware that through historic legislation passed last summer, New York State established a Statewide GME pool to which private payers contribute about half of the costs of GME attributable to the private pay population. The existence of this pool does not in any way diminish our enthusiasm for a national solution, however, because the mechanisms that must be relied upon at the State level to implement such pools are extremely complex. This is because States do not have the flexibility of the Federal government to develop simple funding mechanisms, and instead must negotiate a thicket of complicated Federal ERISA, provider tax, and State insurance law requirements in the design of such pools. We have nothing but admiration and praise for our Governor and State Legislature for meeting these challenges, but believe that a Federal solution would greatly simplify the structure and administration of such a GME fund.
 Page 136       PREV PAGE       TOP OF DOC

Medicare DSH and GME Payments

Congress Should Not Impose More Than A Fair Share of Reductions On Teaching and DSH Hospitals

    While we share the national concern over the solvency of the Medicare Part A Trust Fund, we urge that the Subcommittee not impose more than a fair share of any Medicare payment reductions on teaching hospitals, and especially on teaching hospitals that also receive DSH payments. This would be the result if three payment areas were targeted: the annual inflation adjustment, or update factor, to all Prospective Payment System (PPS) rates; the indirect medical education factor and direct medical education payment; and DSH payments. If Congress were to seek Medicare savings in these areas, it would impose a multiple blow to all teaching hospitals with even greater harm inflicted on teaching hospitals that also rely upon DSH payments.
    This is because the IME and DSH add-ons in particular are calculated upon the base of PPS rates. Reducing PPS payments through reductions in the update factor would ripple through to reduce projected IME and DSH payments. Separately reducing IME and DSH payments would impose real cuts, not reductions in the rate of growth, on a class of hospitals that continues to be least able to withstand them.
    Of course, through the GME Demonstration Project, New York's hospitals will produce significant permanent savings to the Medicare program through reductions in the GME reimbursement associated with cut residency positions. Despite these savings, demonstration participants are not in any way protected from further payment policy changes made for teaching hospitals everywhere. We urge you to limit further reductions as you continue your deliberations.

 Page 137       PREV PAGE       TOP OF DOC
Disproportionate Share Hospital Payments

    As noted earlier, virtually all of GNYHA's member hospitals are eligible for payments under the DSH program, and most of them, including Montefiore, qualify as ''high DSH'' hospitals. The program is an essential component of our ability to maintain the quality of care to Medicare beneficiaries and meet the needs of low income patients. Particularly given the increasing incidence of the loss of insurance and uncertainty regarding the impact of welfare reform, we urge Congress to support proposals that would maintain DSH at current levels and allow it to vary to accommodate changing needs.
    The DSH adjustment compensates hospitals for the limitations of the Medicare PPS case-mix adjustment with respect to recognition of the higher clinical and non-clinical costs associated with caring for poor Medicare patients. Because hospitals that care for a high proportion of low income patients tend to bear disproportionate financial burdens, the DSH adjustment is necessary to ensure their continued access to quality care. The current proxy that Medicare uses for measuring indigence is the sum of two ratios: the ratio of Medicare SSI to total Medicare inpatient days and the ratio of Medicaid to total inpatient days.
    Our principal concern regarding the current DSH proxy is the growing inability of hospitals to identify Medicare- and Medicaid-sponsored patients enrolled in managed care plans, which is caused because insurance identification cards denote the patient's health care plan, but not the sponsor. Thus, we strongly endorse a proposal by ProPAC to require a sponsorship code in each patient's insurance identification number. While states have the authority to mandate such a change, a Federal requirement is far preferable to ensure national compliance. In this regard, we note that the President's proposed budget would freeze each hospital's disproportionate patient percentage factor, which is used to calculate DSH payments, for two years while new approaches to identifying low income patient populations are identified, an approach we endorse.
 Page 138       PREV PAGE       TOP OF DOC
    In its March 1997 report, ProPAC also recommends modifying the DSH proxy by incorporating recipients of other government subsidies as well as the uninsured and underinsured. While this approach seems sound, we believe that any new measures should be thoroughly researched and the effects on providers evaluated before changes are enacted. There is sufficient need to measure services provided to the other populations to warrant permanent, additional data collection even if a change in the DSH proxy is not ultimately recommended.
    In addition, ProPAC recommends measuring indigence through inpatient and outpatient charges rather than through inpatient days. Since both are a proxy for cost, we would urge that new research center on cost rather than charges. Cost can be estimated through application of the ratio of cost to charges. This extra step is value added because there is considerable variability in the mark up of cost to charges around the country. We endorse the inclusion of outpatient costs in the research.
    Finally, given that hospitals rely greatly on the current DSH adjustment, we urge that any future changes in payment policy that reallocate funding be phased in appropriately to give providers sufficient time to adjust their programs.

Ensuring Continued Support for Teaching and Disproportionate Share Hospitals in Medicare Managed Care

Need For AAPCC Carve Out

    GNYHA believes that the benefits of managed care experienced by commercial populations should be increasingly available to Medicare beneficiaries, but urges that this be done in a way that preserves Medicare's commitment to high quality training and research and services to underserved populations.
 Page 139       PREV PAGE       TOP OF DOC
    As you know, health maintenance organizations (HMOs) receive monthly payments for each Medicare beneficiary enrolled in their plan; such payment is called the Adjusted Average Per Capita Cost (AAPCC). The AAPCC is calculated under a formula that includes Medicare payments for GME and DSH in the rates paid to Medicare contractors. However, HMOs are not required to pass these public benefits payments on to the hospitals that provide the services and incur the costs, or even to contract with them. Medicare managed care enrollment is increasing at a rapid pace in the New York City area, growing by almost 30% in the past year. It is essential that the transfer of GME and DSH payments to HMOs be stopped, not only to sustain the important activities supported by these special payments, but also to ensure that teaching hospitals, upon whom the Medicare population relies, are able to serve as an important component of the infrastructure to expand Medicare managed care.
    We therefore urge Congress to enact provisions to carve out 100% of GME and DSH payments from the AAPCC and pay them to hospitals that provide these services according to current payment rules. To the extent that further adjustments are made to the managed care rate paid to HMOs, we urge that regional variations in cost of living and similar factors be recognized. Not all hospitals are alike, and not all regions of the country are alike. If we are to preserve the best in our system, we need to recognize those differences in the way we pay for Medicare managed care.

Experiences With New York State's New GME Pool

    I am aware that the Chief Executive Officer of a New York area HMO last month reported to this Subcommittee that it pays teaching hospitals rates that are 20% higher than those paid to other hospitals. He also charged that New York's hospitals are being paid twice for GME, once from our new Statewide private payer GME pool, and second, by trying to include GME in their negotiated HMO rates. Finally, he claimed that the effect of the State pool has been inflationary and that it will likely result in an increase in the premiums he charges to customers. Because these viewpoints have implications for your deliberations on the AAPCC carve-out, I would like to offer the following observations.
 Page 140       PREV PAGE       TOP OF DOC
    First, under our new State system, private payers are making a distinct contribution for GME to the new pool in addition to paying hospitals for direct patient care rendered to privately insured patients. When an HMO calculates that the combination of pool and patient care payments for teaching hospitals is higher than the patient care payments made to non-teaching hospitals, it simply demonstrates that the pool is having its intended effect. That is, the pool has somewhat leveled the playing field between teaching and non-teaching hospitals by allowing teaching hospitals to compete on the basis of patient care, not teaching, costs.
    With respect to patient care costs, many factors might well lead to higher payments at teaching hospitals. Teaching hospitals have higher cost structures for capital and high technology services. They provide very specialized, and sometimes unique, services such as multiple organ and bone marrow transplants or treatment in neonatal intensive care units. The severity of illness treated is naturally higher than that in community hospitals. It is not surprising that an HMO would pay more for such care.
    Second, the statement that teaching hospitals are being paid twice for GME in New York is patently untrue; the State's GME pool is only funded at half of teaching hospitals' GME costs. The State made it clear from the outset that hospitals are free to try and achieve full funding in their negotiations with payers. The degree to which hospitals have been successful is a different matter.
    Finally, HMO premiums may indeed rise in New York as a result of de-regulation, but not because there is a GME pool contribution requirement. Until January 1 of this year, HMOs in New York enjoyed an exclusive franchise on negotiating hospital rates, when all other payers had their rates set by the State. Under the rate-setting system that just expired, the State established fixed rates and caps on prices that could be charged to a host of payers, including private indemnity insurers, Medicaid and Blue Cross plans. These rates were calibrated to keep hospital margins low and to prevent cost shifting to the private sector.(see footnote 10)
 Page 141       PREV PAGE       TOP OF DOC

    HMOs, which were not significant payers at the time the rate-setting system was established in the early 1980s, were exempt from this rate-setting system and, consequently, were the only payers permitted to negotiate hospital rates. In the event that they lacked contracts with hospitals, they were required only to pay the State-set rates established for Blue Cross plans, which in turn were lower than the rates set for commercial indemnity plans. This allowed HMOs to underbid traditional insurers such as Blue Cross each and every time. The pressure to end the regulatory system came principally from regulated Blue Cross and commercial payers, who wanted to gain the same ability to negotiate. HMOs were the only payers to advocate the continuation of rate regulation.
    According to the KPMG Peat Marwick 1996 Health Benefits Survey, average HMO premiums in the New York City area were in fact low compared to neighboring regions, and especially compared to conventional insurance plan premiums. This seems to confirm the value to HMOs of their negotiating franchise. The 1996 average single premium for an HMO plan in the New York City area was $161.47, compared to $175.65 in New Jersey, a lower cost area across the river. In contrast, the conventional insurance product premium was higher in New York City, at $220.61, than it was in New Jersey, where it was $217.04.(see footnote 11) Thus, HMO premiums were actually lower in New York City, one of the most expensive areas of the country, than in New Jersey, and the disparity between the HMO and conventional insurance rate was much greater in New York.

    This suggests that New York HMOs have in fact enjoyed a competitive advantage over indemnity insurers and/or, as some have observed, that there may have been an artificial depression in HMO premiums for other competitive reasons. With de-regulation, conventional insurers are now on a level playing field with HMOs and some narrowing of the gap between their premiums would not be surprising.
 Page 142       PREV PAGE       TOP OF DOC
    Mr. Chairman and members of the Subcommittee, thank you very much for the opportunity to testify today. I would be pleased to answer any questions you may have.
    INSERT OFFSET FOLIO 13 HERE
    [The official Committee record contains additional material here.]

—————


    Mrs. JOHNSON. Thank you, Dr. Foreman.
    Mr. Starr.

STATEMENT OF GERALD STARR, CHIEF EXECUTIVE OFFICER, KERN MEDICAL CENTER, BAKERSFIELD, CALIFORNIA; ON BEHALF OF NATIONAL ASSOCIATION OF PUBLIC HOSPITALS & HEALTH SYSTEMS

    Mr. STARR. Thank you, Madam Chairman.
    I am Jerry Starr, chief executive officer of KMC, Kern Medical Center, which is a 270-bed hospital in Bakersfield, California. By volume, KMC is the largest provider of inpatient, outpatient, emergency, and high-risk obstetric services in Kern County. KMC is the community's primary safety net provider, providing $50 million annually in uncompensated care. With a longstanding tradition of excellence in medical education, KMC trains nearly 100 residents a year.
    I am testifying today on behalf of NAPH, the National Association of Public Hospital & Health Systems. I want to thank Mr. Thomas and the Subcommittee for inviting me to testify on a range of issues of concern to NAPH safety net providers.
    As the health care market undergoes revolutionary change in the delivery and payment for health services, two missions, training physicians and providing health care to low-income, uninsured individuals, will remain the responsibility of a few hundred hospitals that have shouldered this burden for decades. Eighty-five percent of NAPH members are teaching hospitals. Issues regarding graduate medical education are described more fully in my written testimony. Today, I will focus my remarks on the Medicare DSH Program.
 Page 143       PREV PAGE       TOP OF DOC
    NAPH member hospitals provide large volumes of care to low-income populations. Over 70 percent of their inpatient care is to Medicaid and self-pay patients, which means no pay in our hospitals. When Medicare patients are added, the proportion jumps to 90 percent. NAPH hospitals provide over 20 percent of the bad debt and charity care in the Nation. NPAH safety net hospitals rely on Medicaid and Medicare revenues to subsidize the huge amounts of uncompensated care they provide.
    Local government appropriations pay for 51 percent of uncompensated care, with Medicaid and Medicare DSH payments supporting the remainder.
    Medicare DSH has been and will continue to be an essential part of the funding that enables NAPH members to provide critical health services.
    The Medicare DSH Program is in need of updating. Since the DSH requirement was originally enacted, major changes in the way hospitals are being paid have put safety net hospitals at a disadvantage. Commercial payers no longer subsidize hospital costs of treating low-income patients or their costs of training physicians.
    New market realities are driving other providers to compete for Medicaid patients for the first time, provided they are low-cost, low-risk Medicaid patients. Competition will leave safety net hospitals with an increasing burden of uncompensated care.
    The current formula of calculating Medicare DSH payments is based on the hospitals' ''disproportionate patient percentage,'' a proxy for determining the amount of low-income care provided. The percentage is based on the proportion of care provided to Medicare SSI or low-income Medicare and Medicaid patients.
    There are a number of serious problems with this formula that warrant reexamination. The statutory low-income proxy does not include uncompensated care. This problem will be exacerbated as welfare reform legislation takes effect, especially in California and in States that choose to delink Medicaid and welfare eligibility.
 Page 144       PREV PAGE       TOP OF DOC
    Many hospitals are finding it difficult, if not impossible, to identify Medicaid patients in States implementing Medicaid managed care because the Medicare DSH formula relies on Medicaid utilization being unable to account for all Medicaid patients translates into reduced dollars.
    Safety net hospitals are finding it difficult to retain their share of less costly Medicaid populations; for example, healthy mothers and children. The burden of uncompensated care and care to high-risk, chronically ill populations is increasing. Their ability to cross-subsidize that care with lower risk Medicaid volume is diminishing.
    NAPH hospitals must reorient to provide more preventive and primary outpatient care and less episodic acute inpatient care. The DSH formula should include outpatient as well as inpatient services.
    The Prospective Payment Assessment Commission has proposed changing the low-income proxy to include all elements of low-income care, including uncompensated care. NAPH strongly supports this change to the low-income proxy and to targeting Medicare DSH funds on the highest volume providers of low-income care.
    ProPAC's analysis of the impact of the formula shows that Medicare DSH funds would be redistributed to major public teaching hospitals, other public teaching hospitals, and rural hospitals. NAPH believes this redistribution is appropriate. Limited Federal DSH funds need to address the problem that the health care market does not address, access to individuals with inadequate or no health insurance and very sick individuals who, because of their high cost, have limited access to care due to cherrypicking on the part of providers and plans.
    We support ProPAC's recommendation to include inpatient and outpatient services in the measure of low-income costs. We support their recommendation to pay DSH and GME payments directly to hospitals that incur those missions.
    Last, even though it is not in the jurisdiction of this Subcommittee, we ask that as you consider the budget, further Medicaid reductions are not part of the package. Growth in Medicaid costs have already been curbed in the last year. Providers are facing reductions in Medicaid revenues and increased costs due to welfare reform. The impact of further cuts or the imposition of per capital caps would be devastating.
 Page 145       PREV PAGE       TOP OF DOC
    We ask that the targeted groups of hospitals providing the highest volumes of low-income care are protected from cuts in the Medicaid DSH Program.
    Thank you for this opportunity to testify before you.
    [The prepared statement and attachments follow:]

Statement of Gerald Starr, CEO, Kern Medical Center, Bakersfield, California; on Behalf of National Association of Public Hospitals & Health Systems

    I am Jerry Starr, CEO of Kern Medical Center, which is a 270-bed hospital in Bakersfield, California. With over 14,800 discharges and 129,000 outpatient visits in 1996, by volume, KMC is the largest provider of inpatient, outpatient, and emergency services in Kern County. Over 90 percent of the services are provided to Medicare, Medicaid, and low income uninsured and underinsured patients. As primary safety net provider in the community, Kern provides $50 million annually in uncompensated care, which represents 29 percent of total charges. In addition, Kern operates the highest acuity Emergency Care Center in Kern County with 60,000 visits annually, a renovated Center for Women, Infants and Children with high risk obstetrics services, the County's highest level neonatal intensive care unit, and the County's only high security inpatient psychiatric unit. With a long-standing tradition of excellence in medical education, Kern trains nearly 100 residents a year in Emergency Medicine, Family Practice, Internal Medicine, OB/GYN, and Surgery.
    I am testifying today on behalf of the National Association of Public Hospitals & Health Systems (NAPH), which represents over 100 of America's metropolitan area safety net hospitals. Just like Kern Medical Center, these hospitals and systems are uniquely reliant on governmental sources of financing to support care to Medicare, Medicaid, and uninsured, low income patients. They also provide many preventive, primary and costly tertiary services to their entire communities, not just to the poor and elderly. These services include a wide variety of around-the-clock standby services such as trauma units, burn centers, neonatal intensive care, poison control, emergency psychiatric services, and crisis response units for both natural and man-made disasters.
 Page 146       PREV PAGE       TOP OF DOC
    I am pleased to have the opportunity to testify before the Subcommittee today on a range of issues of concern to safety net providers including Medicare disproportionate share hospital (DSH) payments and graduate medical education (GME) payments. As the health care market undergoes revolutionary change in how it delivers and pays for health care services, two missions—training physicians and providing health care to low income, uninsured individuals—will remain the responsibility of a few hundred hospitals and health systems that have shouldered this burden for decades. Governmental payers, like Medicare, have long recognized the importance of these missions and the need to support them with separate DSH and GME payments. As the health care market becomes more competitive and as payment rates increasingly reflect whatever plans or payers can negotiate with providers of care, the importance of explicit, adequately-financed funding streams for these missions will be essential to the survival of safety net and teaching hospitals and health systems.
    My comments today are summarized in four areas:
    First, I will describe for you the situation of safety net hospitals today, using data from NAPH's 1995 annual survey to highlight the important mission of these hospitals and how their ability to meet that mission is being impacted by changes in the health industry. Recent market trends have indicated increasing competition for Medicaid business (particularly low cost Medicaid business) while an ever-shrinking group of safety net providers shoulder most of the uncompensated care burden—a burden that is growing steadily.
    Second, the Medicare DSH formula needs to be changed to reflect new circumstances in the changing health care system. In particular, NAPH recommends that Congress adopt the approach suggested by the Prospective Payment Assessment Commission (ProPAC), which would include all elements of low income care costs, including uncompensated care, in the qualifying formula. Additionally, Medicare DSH payments must continue to be adequately financed for those providers which truly shoulder a disproportionate burden of low income care. Medicare DSH and GME payments must be carved out of the capitated amounts paid to risk contractors and made directly to hospitals that are fulfilling those missions.
 Page 147       PREV PAGE       TOP OF DOC
    Third, NAPH supports a ''shared responsibility'' approach to financing graduate medical education which treats GME as a public good. This approach would require contributions from all payers of health care, not just Medicare, and, thus, should distribute GME funding based on all patient care. In addition, any physician workforce reforms that Congress considers as part of proposals to restructure GME financing must be balanced with maintaining access to care in underserved communities. Proposals to limit financing for international medical graduates (IMGs) or specialists must recognize the unique role that residents supervised by attending physicians serve in safety net hospitals for communities with limited or no access to private practice physicians.
    Fourth, even though it is not in the jurisdiction of this Committee, we ask that as you consider the budget that Medicaid cuts not be part of the package. Medicaid program growth has slowed considerably in the last year. Due to the implementation of welfare reform legislation, safety net hospitals are faced with significant losses of Medicaid revenues as legal immigrants lose Medicaid and Medicare SSI eligibility. Further, states' delinking of the enrollment process for welfare and Medicaid is going to result in fewer healthy Medicaid recipients in the risk pool in states with Medicaid managed care. The impact of further cuts in the program or a shift to per capita caps would be devastating. At a minimum, we ask that a targeted group of hospitals treating the highest volumes of low income patients be protected from cuts in the Medicaid DSH program.

NAPH Members Provide Remarkable Levels of Both Inpatient And Outpatient Care

    Perhaps the most striking characteristic of safety net hospitals is the tremendous volume of both inpatient and outpatient services they provide. On the inpatient side, in NAPH's most recent member survey, 90 hospitals reported total staffed beds of almost 40,000 for an average of 442 per hospital, total admissions of 1.4 million and total inpatient days of 10.9 million. To place this volume of care in perspective, in comparison to the average hospital in the 100 largest cities in the U.S., the average NAPH member reported 30 percent more admissions, 9 percent more inpatient days, and an occupancy rate (75 percent) that was 11 percent higher.(see footnote 12)
 Page 148       PREV PAGE       TOP OF DOC

    Contrary to a sometimes-held misconception of safety net hospitals as primarily inpatient facilities, these institutions have always been the family doctor for large numbers of low income and uninsured patients, providing large amounts of primary and preventive care. In 1995, just 67 NAPH members provided an astounding total of 22 million outpatient visits, only 4 million of which were emergency room visits. Compared to the average hospital in the 100 largest cities, NAPH members provide a full 68 percent more outpatient visits.

Care To Low Income And Uninsured Patients Is On the Rise in Safety Net Institutions

    In addition to providing large volumes of care generally, safety net hospitals and health systems tend to provide a huge proportion of care to Medicaid, Medicare and uninsured patients in particular. Over 70 percent of inpatient care provided in NAPH member hospitals in 1995 was for Medicaid and so-called ''selfpay'' individuals. For safety net institutions, these patients are for the most part medically indigent individuals who cannot afford to pay for the services they receive.(see footnote 13) When Medicare patients are added, the proportion jumps to 90 percent. For outpatient and emergency care, the proportion of visits for Medicaid and selfpay individuals was 77 percent (Figure 1).

    As safety net providers, NAPH members have historically provided large amounts of uncompensated care in their communities and their share of the uncompensated care burden is steadily increasing. In 1995, 67 hospitals reported incurring $5.8 billion in uncompensated care (defined as bad debt and charity care) for an average of just over $86 million per hospital. For these institutions, bad debt and charity care charges represented a full 25 percent of total gross charges. According to data from AHA, all hospitals nationwide provided $28.1 billion in bad debt and charity care. While NAPH member hospitals represent less than two percent of hospitals, they provide over 20 percent of bad debt and charity care.
 Page 149       PREV PAGE       TOP OF DOC
    Moreover, in a trend with sobering implications for safety net institutions, uncompensated care is increasingly concentrated among an ever-shrinking number of providers. AHA data on public general hospitals in the 100 largest cities (a subset of total NAPH members) from 1980 and 1993 indicate that the category of selfpay (or no-pay) patients increased from 16.8 percent of gross charges to 22.2 percent, or an increase of over 30 percent. Among private general hospitals during the same period, the proportion of patients with no insurance decreased from 7.4 percent of gross charges to 5.5 percent, a 26 percent decrease. At the same time, private hospitals' share of Medicaid patients grew by 15 percent, reflecting increasing competition for less costly Medicaid patients, such as healthy pregnant women and children.(see footnote 14)

    Further, the number of uninsured Americans continues to rise. The passage of welfare reform legislation in the last Congress is the single most sweeping rollback in Medicaid coverage since the program's establishment. The bill eliminated Medicaid and Medicare SSI coverage for substantial numbers of legal immigrants, thereby not only significantly increasing the rolls of the uninsured, but placing a particular burden on safety net providers and the state and local governments that support them. Legal immigrants will continue to need medical care in times of sickness or accident, and will seek that care in safety net hospitals who treat all regardless of ability to pay. Many of these hospitals in high immigrant states, like California, will be overwhelmed by the burden of providing yet more uncompensated care.

Safety Net Providers Depend on Medicaid and Medicare to Finance Uncompensated Care

    Unlike most community hospitals that can tap commercial patient revenues to subsidize uncompensated care, urban safety net hospitals rely on Medicare and Medicaid revenues to subsidize the huge amounts of uncompensated care they provide. While Medicaid and Medicare combined represented 55 percent of the overall care provided by NAPH members in 1995, they accounted for 71 percent of net patient revenues (Figure 2).(see footnote 15)
 Page 150       PREV PAGE       TOP OF DOC

    Appropriations from local government and other revenues intended to cover indigent care costs amounted to 12.3 percent of total revenues. In effect, state and local subsidies cover just over half of the cost of uncompensated care provided at NAPH member hospitals. To make up the difference, these hospitals rely on Medicaid disproportionate share hospital (DSH) payments (40 percent) and Medicare DSH payments (9 percent) (Figure 3). While Medicare DSH payments may not appear significant by comparison, Medicare is a key payer in the fragile partnership of federal/state and local governments that currently finances uncompensated care, particularly in the face of serious proposals to cut Medicaid DSH and declining support by state and local governments (local subsidies have decreased 46 percent over the last eight years). In 1995, 53 NAPH hospitals received a total of $316 million in Medicare DSH payments, roughly 8 percent of the $3.8 billion DSH payments nationwide. Medicare DSH has been and will continue to be an essential piece of the patchwork funding that enables NAPH members to provide critical health services to the elderly, disabled and poor.

The Medicare Disproportionate Share Hospital Formula Should Be Changed to Reflect Uncompensated Care and Payments Must Be Targeted To High Volume Providers of Care to the Poor

    Medicare provides supplemental reimbursement to hospitals that treat a large portion of low income individuals through its disproportionate share hospital (DSH) adjustment. The original disproportionate share requirement was enacted in the early 1980s as part of Medicare's move away from cost-based hospital reimbursement to a prospective payment system to recognize higher than average costs incurred by some hospitals due to the disproportionately high number of low income patients they serve. Congress understood that those hospitals that treat large numbers of low income patients would be at a disadvantage under a system that is not cost-based. Similarly today, changes in the way hospitals are paid are putting safety net hospitals at a disadvantage. Private commercial payers are no longer willing to subsidize hospitals' costs of treating low income patients, or their costs of training physicians. In addition, competition is forcing providers to seek any patient with health insurance, including Medicaid—but only the low cost, low risk Medicaid patient. Thus, the market is leaving safety net hospitals with an increasing burden of uncompensated care, and the sickest of Medicaid and Medicare patients, as well as a diminishing base of revenues to subsidize those costs.
 Page 151       PREV PAGE       TOP OF DOC
    The current formula for calculating Medicare DSH payments is based on a hospital's ''disproportionate patient percentage,'' which is intended to be a proxy for determining the amount of low income care provided. The percentage is based on the proportion of care provided to Supplemental Security Income (SSI or low income Medicare) and Medicaid patients.

There are a number of serious problems with this formula that warrant reexamination.

    •  In relying on measures of SSI and Medicaid populations, the statutory low income proxy does not include the significant uncompensated patient care load that some hospitals are currently bearing.(see footnote 16) This problem will be exacerbated as the impact of welfare reform legislation begins to reduce Medicaid eligibility in states with high numbers of immigrants or in states that choose to de-link Medicaid and welfare eligibility.

    •  Many hospitals are finding it difficult, if not impossible, to identify Medicaid patients in states that have moved to implement Medicaid managed care—Medicaid patients show up with an insurance card from a managed care plan, which may not identify them to be Medicaid recipients. To the extent the Medicare DSH formula relies on Medicaid utilization, the inability to account for all Medicaid patients translates into reduced Medicare DSH dollars.
    •  Hospitals with significant uncompensated care burdens are finding it increasingly difficult to retain their share of the less costly Medicaid populations (for example, healthy mothers and children) as market competition intensifies. Their burden of uncompensated care and care to high risk chronically ill populations is increasing while their ability to cross-subsidize that care with lower risk Medicaid volume is diminishing.
 Page 152       PREV PAGE       TOP OF DOC
    •  The DSH formula needs to reflect the change in health care delivery from inpatient to outpatient services. As hospitals reorient to provide more preventive and primary outpatient care and less episodic, acute inpatient care, the DSH formula should include inpatient and outpatient services as part of its measure of low income costs.
    For all of these reasons, changing the Medicare DSH low income proxy is imperative to protecting access in hospitals that serve large numbers of low income patients.
    The Prospective Payment Assessment Commission (ProPAC) has evaluated this problem and has proposed changing the low income proxy to include all of the elements of low income care. Their proposed low income cost variable includes Medicare SSI patients, Medicaid patients, care to patients supported by local indigent care programs, and uncompensated care. NAPH strongly supports this approach to incorporating all of the components of low income care and to targeting Medicare DSH funds on the highest volume providers of low income care.
    ProPAC's analysis of the impact of changing the formula shows that Medicare DSH funds would be redistributed to major public teaching hospitals, other public teaching hospitals, and rural hospitals—essentially to hospitals that provide a lot of uncompensated care. NAPH believes that this redistribution is appropriate, because limited federal DSH funds need to address a problem that the health care market is not addressing—access for individuals with no or inadequate health insurance and individuals who are so sick that they have limited access to care due to ''cherry-picking'' on the part of providers and plans. We strongly support ProPAC's recommendation that payments be targeted to high volume providers of low income care and that progressivity in DSH payments be achieved by allocating payments based on hospitals' proportion of low income care, including uncompensated care.
    Lastly, we support ProPAC's recommendation to include outpatient as well as inpatient services in the measure of low income costs to be used in the Medicare DSH formula.

 Page 153       PREV PAGE       TOP OF DOC
Medicare Graduate Medical Education Payments Are a Critical Source of Financing Physician Training

    NAPH member hospitals play a significant role in training residents and health professionals. Over 85 percent of NAPH members are teaching hospitals, and they trained nearly 18 percent of all residents in 1994. In 1994, 62 NAPH hospitals trained 12,531 residents, or an average of 202.
    The Medicare program provides substantial funding for graduate medical education (GME), and is in fact the only payer to make explicit payments for GME (although some state Medicaid programs provide GME payments; this funding totals less than $1 billion annually.(see footnote 17)) As such, adequate Medicare GME payments are critical to support the training of our nation's physicians. The Medicare program designates two separate payments to cover GME costs—the Direct Graduate Medical Education (DGME) payment and the Indirect Medical Education (IME) payment.

    In 1995, 63 NAPH members received $158 million in DGME payments from Medicare and nearly double that or $261 million in IME payments. While these payments represent substantial support for training programs, the formula used to calculate these payments distributes the funds inequitably among teaching hospitals. Basing the formula on Medicare inpatient volume alone means that many major teaching hospitals with significant training commitments receive relatively little DGME payments due to their low Medicare volumes. For example, DGME payments as a percent of total net revenues for NAPH member hospitals average 2.2 percent, while for members of the Council of Teaching Hospitals (COTH), they are 5.2 percent. The difference in funding is attributable primarily to lower Medicare volumes at NAPH member hospitals—18 percent of gross revenues compared to 29 percent at COTH hospitals.

GME is a Public Good Which Should Be Financed By All Parts of the Health Care System
 Page 154       PREV PAGE       TOP OF DOC

    NAPH supports a ''shared responsibility'' approach to financing graduate medical education which treats GME as a public good. This approach would require contributions from all payers of health care, not just Medicare, and, thus, should distribute GME funding based on all patient care volume. Alternatively a trust fund approach could be financed with general revenue contributions or a broad-based tax.
    The level of financing for GME is critical. As other payers negotiate ever lower rates, teaching hospitals are losing their ability to cross-subsidize medical education costs. In addition, as more Medicare patients move to managed care, GME funds, which are currently based on the volume of Medicare fee-for-service patients, will diminish considerably. These trends threaten to undermine the viability of our nation's teaching hospitals and their ability to train physicians.

Medicare GME and DSH Funds Should Be Carved Out of the AAPCC and Made Directly to Hospitals

    The current methodology for distributing DGME, IME and DSH payments is seriously flawed in the Medicare managed care context. For Medicare patients enrolled in managed care, these supplemental payments are incorporated into the average adjusted per capita cost (AAPCC) which is the capitation payment made to managed care plans. The plans do not necessarily pass these payments along to the hospitals which incur the costs that justify the payments. In fact, some plans receive the payments and do not even contract with such hospitals. As Medicare increases the use of capitated risk contracting, the amount of DGME, IME and DSH funds that go to teaching hospitals will diminish considerably unless this payment policy is changed. In essence, payments intended to support the costs of teaching or low income care are being diverted from the hospitals that provide the care to managed care plans that are not fulfilling this mission. For this reason, the GME and DSH payments must be carved out of the AAPCC rate and made directly to the hospitals that incur those costs.
 Page 155       PREV PAGE       TOP OF DOC

Physician Workforce Reforms Must Be Balanced With Maintaining Access to Care in Underserved Communities

    Current Medicare GME financing policy does not address the size or type of physician workforce that hospitals are training. Yet often physician workforce reform proposals go hand in hand with discussions about reform of medical education. Many policymakers believe that the nation's medical schools and teaching hospitals have done a poor job of training the kind of physician workforce needed by our health care system—training too few primary care physicians, too many specialists, and perhaps too many physicians overall. Certainly the focus on specialty care, as well as on acute, episodic intervention, as opposed to primary and preventive care, has raised costs and compromised the health status of many communities. Additionally, the medical education system has trained too few minorities and too few doctors willing to practice in underserved communities. The shortage of physicians in these communities in particular requires that policymakers act cautiously in reforming physician workforce policy, because many of our nation's safety net providers rely on residents and supervising attending physicians to provide otherwise unavailable care in these areas. Further, NAPH member systems tend to train more minorities and more physicians who will most likely continue to serve those communities and/or who will have a unique understanding of the needs of vulnerable populations in their private practices
    This concern about workforce policy and access to care in safety net hospitals is particularly relevant to two common GME reform initiatives: reducing the number of international medical graduates (IMGs) and changing the mix of primary care and specialty training slots. Most of the growth in resident training positions over the last ten years has been the result of increases in slots for individuals who trained in medical schools outside the U.S., called IMGs. While reducing the number of or financing for IMG slots is often seen as the solution to a perceived over supply of physicians—and, in fact, was recommended two weeks ago by AAMC, the AMA, and other medical societies—in the case of safety net hospitals, such proposals could have dire consequences. These residents, supervised by attending physicians, provide a considerable amount of care to low income patients with otherwise limited access to services (Figure 4). Not all safety net hospitals or systems rely heavily on IMGs, however, the phenomenon is not localized in any particular state or region. Replacing these physicians is costly—NAPH members estimate that it would cost two to three times more to replace a resident with some combination of physicians and non-physician providers—and difficult, since replacements are not easily found.
 Page 156       PREV PAGE       TOP OF DOC
    The same care should be taken with respect to proposals to limit or reduce the number of specialty training slots or financing for such slots. While we support the expansion of primary care capacity in physician training, specialty residents and fellows in safety net hospitals may be the only physicians available to serve many low income, vulnerable communities—where no private physicians practice. NAPH urges that access to care be an important criterion in decisions about reducing or restructuring the physician workforce.
    INSERT OFFSET FOLIOS 14 TO 17 HERE
    [The official Committee record contains additional material here.]

—————


    Mrs. JOHNSON. Thank you, Mr. Starr.
    Mr. Jaffe.

STATEMENT OF DAVID JAFFE, EXECUTIVE DIRECTOR AND CHIEF EXECUTIVE OFFICER, HARBORVIEW MEDICAL CENTER, SEATTLE, WASHINGTON; ON BEHALF OF AMERICAN HOSPITAL ASSOCIATION

    Mr. JAFFE. Thank you.
    Madam Chair, I am David Jaffe, executive director and chief executive officer of Harborview Medical Center in Seattle. I am pleased to present the views of AHA, the American Hospital Association, and the nearly 5,000 hospitals, health systems, networks, and other providers of care that the AHA represents.
    Harborview serves patients from Washington, Alaska, Montana, and Idaho. Our longstanding and unwavering mission is to care for all persons, regardless of their ability to pay, and as a key component of the University of Washington School of Medicine, to provide the finest in clinical education.
 Page 157       PREV PAGE       TOP OF DOC
    Since its inception, Medicare, through graduate medical education payments, has supported the additional cost that hospitals incur in conducting residency training programs, and through disproportionate share of hospital payments, the government targets funds to hospitals like mine that serve large numbers of low-income people, ensuring these people have access to at least basic care.
    GME and DSH are separate, but related, issues that serve two vital missions, teaching future physicians and providing a lifeline to health care for people with no other place to turn.
    The reason GME and DSH are so often lumped together in policy and budget discussions is simple. Because most teaching hospitals are located in urban areas, they provide care to greater numbers of low-income patients. Ninety-three percent of all DSH payments go to large hospitals in urban areas, and teaching hospitals receive about 65 percent of all DSH payments.
    It is critical that our Nation maintain its historic commitment to clinical education and caring for the poor.
    I would like to offer a few recommendations to help us maintain that commitment. Let me start with clarifying the way DSH payments are calculated.
    Current law states, simply, that Medicaid days should be used in the DSH calculation, along with the number of Medicare inpatient days.
    The Health Care Financing Administration has interpreted this to mean that the DSH payment should be based on the inpatient days Medicaid pays for. The problem is, as more Medicaid patients are covered by managed care, more days are provided than are actually paid for by Medicaid. The result is that many hospitals are receiving lower DSH payments than otherwise intended.
    The Department of Health and Human Services has already lost in four Federal Courts of Appeal on HCFA's interpretation. In each case, the court has sided with hospitals, determining that the Medicare DSH adjustment should be based on the number of days provided to Medicaid patients.
 Page 158       PREV PAGE       TOP OF DOC
    We are concerned the President's budget might attempt to lock us into 2 more years of the use of the days-paid interpretation rejected by the courts, rather than the days-eligible approach embraced by the courts.
    The AHA urges this Subcommittee to clarify the law so it requires the calculation of the Medicare DSH payment to be based on Medicaid-eligible days.
    On the GME issue, one item in the President's budget that we wholeheartedly agree with is the carve-out. Medicare payments to managed care plans are based on what is called the adjusted average per capita cost, or the AAPCC. The AAPCC includes what Medicare spends on GME and DSH payments, but very often, the plans being paid by Medicare do not pass GME and DSH payments on to the hospitals they have contracted with.
    The AHA strongly supports removing, or carving out, the GME and DSH amounts that are included in the AAPCC and making those payments directly to the hospitals that incur the costs of these programs.
    We also believe, Madam Chair, that support for graduate medical education must be shared through a Federal GME trust fund, a concept not included in the President's budget proposal.
    A Federal GME trust fund should be supported not by general revenues, but by public and private payers. Unless such a fund is established and adequately supported, teaching hospitals will have to choose between reducing educational responsibilities in order to be price-competitive or meeting those responsibilities and being priced out of the market.
    One final issue. The President's budget proposes considerable savings in Medicare spending, some of which target GME and DSH payments. Recent ProPAC hospital performance data may encourage the belief that these reductions can be achieved without inflicting pain. They cannot.
    Forty percent of the Nation's hospitals lose money when they treat Medicare patients, and 20 percent of the hospitals have negative total margins, meaning that overall, they lose money on all patients served.
 Page 159       PREV PAGE       TOP OF DOC
    For both inpatient and outpatient services, while Medicare pays less than the full-dollar value, Medicaid pays even less; but for about 1,000 hospitals, these two programs account for more than two-thirds of the revenues. Clearly, it is inaccurate to assume that all hospitals are faring well under Medicare, especially those carrying the burden of both teaching and providing uncompensated care.
    Again, Madam Chair, I appreciate the opportunity to be here and will be glad to answer any questions you may have.
    [The prepared statement follows:]

Statement of David Jaffe, Executive Director and CEO, Harborview Medical Center, Seattle, Washington; on Behalf of American Hospital Association

    Mr. Chairman, I am David Jaffe, Executive Director and CEO of Harborview Medical Center in Seattle, Washington. I am pleased to present the views of the American Hospital Association (AHA) and the nearly 5,000 hospitals, health systems, networks and other providers of care that the AHA represents.
    Harborview is owned by King County and managed by the University of Washington's School of Medicine under a contract that has existed since 1967. With a handful of exceptions, every person working at Harborview is a University of Washington employee. Harborview is the only Level 1 trauma center in the state, and, along with its regional burn center, serves patients from Washington, Wyoming, Alaska, Montana and Idaho.
    About 48 percent of Harborview's patients are reimbursed through Medicaid (one of the highest percentages in the country), 19 percent through Medicare, 25 percent through private insurance, and about 9 percent are uninsured. Harborview's longstanding and unwavering mission has been to care for all persons regardless of their ability to pay. It is important to note that Harborview receives no operating support from King County.
 Page 160       PREV PAGE       TOP OF DOC
    Harborview is central to the unique role of the University of Washington's School of Medicine. It is the only medical school in the region, and the salaries of full-time faculty are mostly supported through the medical services they provide at Harborview. The faculty is complemented by 123 residents in general and specialty fields.

GME and DSH

    GME (Graduate Medical Education) and DSH (Disproportionate Share Hospital) are separate but related issues that, during budget discussions, are usually stirred into the same ''alphabet soup'' as other budget line items. But GME and DSH are more than just acronyms in the federal budget. In America's communities, especially urban communities, these programs serve two very important missions: teaching our future physicians, and providing a lifeline to health care for people with no other place to turn.

Graduate Medical Education

    There is a full array of teaching hospitals in this nation: university-based academic medical centers, hospital-based independent academic medical centers, affiliated community teaching hospitals, VA medical centers, and military hospitals. These teaching hospitals incur significant direct and indirect costs in operating physician training programs. The direct costs include stipends and benefits for residents, salaries and benefits of supervising faculty, costs directly associated with providing the GME program, such as clerical personnel working exclusively in the GME administrative office, and allocated institutional overhead costs, such as maintenance, and depreciation.
    Indirect costs are those incurred in providing an appropriate environment for clinical education. They include the higher cost of patient care that is inherent in treating a much higher proportion of severely ill patients. Teaching hospitals also maintain a broader scope of specialized services and stand-by capacity, often on a round-the-clock basis. And, because teaching hospitals are usually in large urban areas, land, labor and operating costs are usually higher than elsewhere. Indirect costs also include the reduced productivity of the hospital staff, because they are helping educate residents, and the diagnostic tests and other services that residents may order during their learning experience.
 Page 161       PREV PAGE       TOP OF DOC
    Since its inception, Medicare has recognized and supported the additional costs these hospitals incur in sponsoring and conducting residency training programs for physicians. Medicare's support for GME can be broken down into two parts: direct graduate medical education (DGME) payments, which help cover the direct costs of providing clinical education and are based on the number of residents; and indirect medical education (IME) payments, which cover the extra patient care costs that teaching hospitals incur, and are based on the ratio of residents to hospital beds.

Disproportionate Share Hospitals

    Hospitals care for patients without regard for their ability to pay. The federal government, through Medicare and Medicaid, has explicitly supported such uncompensated care by targeting additional funds toward certain types of hospitals that serve a disproportionate share of the poor. A decade ago, Congress mandated an explicit payment adjustment in the Prospective Payment System for hospitals that serve large numbers of low-income patients. One rationale for this DSH adjustment was to compensate hospitals for the costs they incur in treating poor patients, who often are unable to get routine care or early intervention and, as a result, are sicker when they reach the hospital.
    Another rationale for the adjustment: Congress had become increasingly concerned that certain hospitals were at risk of closing as a result of treating large shares of low-income patients, and began to view the DSH payment as a way to mitigate that concern. The payments were seen as helping to maintain access to care for low-income Medicare beneficiaries and other patients.
    The reason that GME and DSH are so often lumped together in policy and budget discussions is simple: because most teaching hospitals are located in urban areas, they provide care to greater numbers of low-income patients. Ninety-three percent of all DSH payments go to large hospitals in urban areas. And teaching hospitals receive about 65 percent of all DSH payments.
 Page 162       PREV PAGE       TOP OF DOC

Threats to the Dual Mission

    The federal government is under pressure to reduce the rate of growth in health care spending, especially in light of the predicted bankruptcy of Medicare's Part A Hospital Trust Fund in 2001. But proposals to slow the growth in Medicare spending pose a direct threat to the funding of uncompensated care and clinical education.
    At the same time, the marketplace for medical care is changing: from competing on the basis of service to competing on price; from fee-for-service to capitation; and from inpatient care to ambulatory and community care. Each of these changes has significantly affected the current structures that support uncompensated care and clinical education. In addition to federal budget constraints, these marketplace trends will continue to undermine historical private sector support for the important community service role that GME and DSH hospitals play.

Maintaining the Funding Commitment

    If the Medicare program were to reduce its historic commitment to support hospital costs for physicians-in-training and uncompensated care, other payers might use the Medicare policy as justification for reducing or eliminating support for these costs. Therefore, it is critical that Medicare continue to provide a benchmark for support of the educational and uncompensated care costs of teaching hospitals.
    However, Congress and the Administration have often viewed Medicare DSH and GME payments as targets for program savings. This year is no different. In the Administration's current budget proposal, the President proposes to reduce spending for clinical education by capping the number of resident positions and by reducing IME payments to hospitals.
 Page 163       PREV PAGE       TOP OF DOC
    Unfortunately, recent ProPAC hospital financial performance data may encourage the belief that reductions in payments to hospitals can be achieved without inflicting pain. This is not true.
    Many hospitals are struggling financially—so reductions in Medicare payments to hospitals will hurt. First, it's important to note that ProPAC's findings apply solely to Medicare inpatient services. Second, at the same time ProPAC reported these Medicare PPS inpatient margins, it also estimated that approximately forty percent of the nation's hospitals lose money when they treat Medicare inpatients.
    More important, 20 percent of hospitals have negative total margins, meaning that, overall, they lose money on all patients served. Government payments cover less than the cost of providing care. In the aggregate (including both inpatient and outpatient services), Medicare pays only 97 cents on the dollar, according to ProPAC. Medicaid pays even less. For roughly 1,000 hospitals, representing one in five of the nation's hospitals, Medicare and Medicaid combined represent more than two-thirds of total revenue. Seventeen percent of these hospitals are sole community providers; another 16 percent are located in the urban core of metropolitan areas. Many are already in weakened financial positions, with roughly 10 percent of these hospitals experiencing bottom-line losses for three years in a row, considering all sources of revenue.
    Clearly, it's inaccurate to assume that all hospitals are faring well under the Medicare program. Any Medicare reductions will have an adverse impact on a significant number of hospitals, including teaching hospitals.

Changing the Structure of GME/DSH Support

    Also threatening clinical education and uncompensated care: the shift of Medicare beneficiaries from fee-for-service to managed care arrangements. The Congressional Budget Office projects that almost one-quarter of all Medicare beneficiaries will be enrolled in a managed care plan by the year 2002. As a result, a growing share of Medicare payments to hospitals will likely come through the managed care gateway, having an ever-increasing effect on teaching hospitals and other caregivers that will feel the pinch of reductions in Medicare managed care payments.
 Page 164       PREV PAGE       TOP OF DOC

The ''Carve-out''

    When Medicare beneficiaries join managed care plans, Medicare pays an up-front, monthly, per-person amount based on the adjusted average per capita cost (AAPCC). The AAPCC is a formula by which Medicare determines the average cost of providing care to beneficiaries in a particular area. The payment also includes what Medicare traditionally spends on DGME, IME and DSH payments.
    However, the rates that a plan negotiates with a hospital do not necessarily include these DGME, IME or DSH payments that the hospital would traditionally receive. In addition, the plan may direct patients away from the hospitals to a lower-cost site of care—because the plan receives the same AAPCC amount regardless of the provider with whom it contracts. In either case, there is no requirement that the health plan use the portion of the AAPCC that results from clinical education and uncompensated care payments to support these provider costs. As a result, the health plan often benefits financially if it can avoid using hospitals that support medical education.
    The hospital—which is directly incurring the costs of providing clinical education or uncompensated care—does not receive the funds that Medicare intends to help pay for those costs. That is why the AHA strongly supports removing the GME and DSH payment amounts included in the AAPCC and making those payments directly to the entities that incur the costs of graduate medical education programs. We applaud the administration for including such a carve-out in its budget proposal.

GME Trust Funds

 Page 165       PREV PAGE       TOP OF DOC
    The president's budget proposals do not, however, address the need to fundamentally change the way that medical education is financed in this country.
    Recognizing that payment systems are changing, Congress in 1995 proposed new trust funds for medical education, which would be supported by Medicare payments as well as general revenues. While this approach was an important step, the use of general revenues could result in funding being affected by federal budgetary pressures rather than the needs of patients and graduate medical education programs.
    The AHA believes a trust fund for graduate medical education is an appropriate vehicle for supporting a broader array of training sites that are better suited to contemporary needs of residency programs. Residency programs began in the inpatient units of teaching hospitals. Over the past two decades, an increasing amount of residency training has moved to ambulatory training sites, both hospital-based and elsewhere within a community. Medicare has recognized hospital-based training, both inpatient and outpatient. It has also recognized hospital-supported programs in non-hospital sites. Nevertheless, there is a need to expand support for residencies in ambulatory training sites, home and community service sites, and long-term care sites. In addition, the training provided under such a fund should be broadened to include not just physicians, but nurses and other health care practitioners.
    A federal trust fund for graduate medical education should be supported not by general revenues but by public and private payers. Unless such a fund is established and adequately supported, teaching hospitals will have to choose between being price-competitive by reducing their educational responsibilities, or retaining their responsibilities and being priced out of the market.

Resident allocation

 Page 166       PREV PAGE       TOP OF DOC
    In the past several years, a number of reports and recommendations have urged the federal government to limit the number of residents-in-training and to allocate positions by specialty to teaching hospitals. While this approach may have been consistent with the health planning legislation of the early 1970s, it is inappropriate to the market-driven changes occurring today. Hospitals are adapting to changes in their local communities. Any effort to superimpose a national resident allocation structure on local, market-based reforms will produce unintended and harmful results. Hospitals are already revising residency training programs to better match the needs of their markets. The AHA believes there is no need to enact a federal system that allocates residencies to local communities or teaching hospitals.

Calculating DSH payments

    The DSH payment that Medicare pays hospitals for treating large numbers of low-income patients is calculated using what is called the DSH Index—a formula that takes into account a hospital's Medicare inpatient days, and the number of Medicaid inpatient days for which Medicaid has paid a hospital, as proxies for the number of low-income people served by a hospital.
    The AHA believes that the Medicare DSH payment should represent the cost of caring for low-income people, and of maintaining access to health care for these people. In other words, the DSH payment should be based on the number of patient days furnished to Medicaid recipients—known as ''Medicaid-eligible'' days—and not on the number of Medicaid inpatient days that Medicaid actually pays a hospital—known as ''Medicaid-paid'' days, which is the current basis for DSH payment today. The number of days a hospital furnishes to Medicaid patients—which, especially as managed care grows, is often greater than the number of days Medicaid actually pays for—is the appropriate measure.
 Page 167       PREV PAGE       TOP OF DOC
    Current law states simply that Medicaid ''days'' should be used in the DSH calculation. The Health Care Financing Administration (HCFA) has interpreted this to mean that the Medicare DSH payment should be based on the inpatient days Medicaid pays for—which is defined differently in each state. However, the Department of Health and Human Services has lost in four federal courts of appeal—the Ninth, the Eighth, the Sixth, and the Fourth—on HCFA's interpretation. In each case the court has sided with hospitals, determining that the Medicare DSH adjustment should be based on the number of days provided to Medicaid patients. The Ninth Circuit, which includes my own state of Washington, was particularly pointed, saying that ''Patients meeting the statutory requirements for Medicaid do not cease to be low-income patients on days that the state does not pay Medicaid inpatient hospital benefits. Thus it is illogical to conclude that Congress intended that only Medicaid-paid days serve as proxy for low-income patient days.''
    The president's budget, unfortunately, seeks to keep in place for two years the current method of determining DSH payments. This, in effect, locks in for two more years the use of ''days paid'' rather than ''days eligible.'' The AHA urges this subcommittee to reject the administration's proposal and clarify the law so that it requires the calculation of the Medicare DSH payment to be based on Medicaid-eligible days.
    Another point on the DSH calculation: In the current environment of managed care, keeping track of hospital inpatient Medicaid days has become more difficult—resulting in often understated Medicaid burdens for hospitals. Under Medicaid managed care many hospitals do not know whether the patient seeking care is a Medicaid recipient. Also, state-based Medicaid waiver programs have changed certain eligibility rules, bringing new population groups into the program. And discussions about restructuring Medicaid adds uncertainty to a hospital's ability to track Medicaid inpatient days when calculating Medicare DSH.
    Therefore, we believe that keeping track of which patients are covered by Medicaid should be the responsibility of the managed care plan that has contracted with Medicaid to provide care for those patients. Because the plan receives payment directly from Medicaid for Medicaid-eligible patients, and then in turn pays the hospital for that recipient's care, it seems logical that the plan would be more easily able to track those Medicaid-eligible patients than the hospital.
 Page 168       PREV PAGE       TOP OF DOC
    Policy changes in Medicaid and welfare, along with more health care being provided on an outpatient basis, raise another question: Should the DSH adjustment itself be reviewed? A re-examination of the mechanics of the adjustment may be useful in order to ensure that the adjustment continues to meet its mission in light of the many dramatic changes health care is going through.

Conclusion

    Hospitals and health systems are directly affected by the dramatic changes in health care. In 1985, there were 5,732 community hospitals in the United States. By 1995, that number had fallen by almost 10 percent to 5,194 facilities. Over the same time period, trends in the use of hospital services have changed, primarily in a shift from inpatient care to outpatient services: annual admissions to community hospitals declined from 33.4 million to 30.9 million; the average length of a hospital stay fell from 7.1 days to 6.5 days; the number of outpatient visits almost doubled from 219 million to 414 million.
    Because demand for care will continue to increase as the number of beneficiaries grows and the average age of beneficiaries rises, further rationalization of our health care system must be undertaken carefully. Certainly, we cannot preserve Medicare by weakening the hospitals and health systems that serve Medicare beneficiaries, that teach our future physicians, and that take care of our nation's poor and uninsured.
    If the federal government turns away from its traditional role of supporting clinical education and uncompensated care, private payers may also try to avoid those costs. No one opposes hospitals providing these services, but few are willing to accept the responsibility of helping pay for them.
    As a result, federal initiatives to control health care costs, balance the budget, and extend the life of the Part A trust fund place government payments for uncompensated care and graduate medical education at risk. They also erode the likelihood of critical support coming from private payers. We can't let that happen.
 Page 169       PREV PAGE       TOP OF DOC
    These two missions are helping provide care to millions of Americans. It is important that the current budget discussion keep in mind our national commitment for hospitals like mine to care for people with no place else to turn, and to include improving the structure of the GME/DSH payment system so that we can maintain that national commitment for generations to come.

—————


    Mrs. JOHNSON. Thank you very much, all the members of the panel, and let me ask you this, Mr. D'Eramo. There are six associations supporting your statement, and we are interested in whether or not you all support reducing the number of international medical graduates, which would mean that as we reduce the number of residency positions, we would give preference to American citizens.
    Carrying your testimony a step further, do you support reducing the number of international medical graduates serving in residencies?
    Mr. STARK. Would the gentlelady yield?
    Mrs. JOHNSON. Yes.
    Mr. STARK. Just for a clarification of that point, we have been informed in previous hearings that many of the international medical graduates are, in fact, U.S. citizens.
    Mrs. JOHNSON. Right, and I am asking a double-edged question. I can separate them. They are two different questions.
    Do you support reducing the number of international medical graduates? But the follow-on question is clearly going to be, and the policy issue is going to be for us, as we downsize the number of residencies, and this is true for us, Dr. Foreman, in the New York demonstration project, Are you going to give preference to American citizens who have graduated from foreign residency programs?
 Page 170       PREV PAGE       TOP OF DOC
    Dr. D'Eramo.
    Mr. D'ERAMO. It is the official position of the AAMC that a method of approaching the budgetary driven need to reduce the expenditure associated with residency education, Madam Chairwoman, would be useful to reexamine the rationale from a policy point of view of providing financing for medical education for foreign graduates.
    Now, there are about 24,000 foreign medical graduates on an annual basis in terms of slots. And 4,200 of those fall into the category of being American citizens. So it is the smaller portion. That is a discernment we have not gotten into in depth, but we are not talking about not providing the opportunity for medical education for foreign medical graduates. Rather, what we are talking about is looking at the policy decision of should we continue to provide total financing for them.
    The other question that arises is the distribution question. There are areas of the country, apparently, where there is an undersupply of physicians, and what about the prospects of foreign medical graduates filling those positions? I think that is another area that has to be studied, but currently, the AAMC's position is that they suggest this is an avenue we can go down.
    Mrs. JOHNSON. Dr. Foreman.
    Dr. FOREMAN. The Greater New York Hospital Association has categorically been opposed to selective and targeted reductions of international medical graduates for some time, but the practical effect of reducing residency opportunities by reducing the number of slots is to give American residents an opportunity to compete. Both sets of residents compete for the remaining slots, but there is very little doubt in our mind that American residents will compete more successfully and that the resultant decline in residents will be much more heavily felt among the international medical graduates than the American medical graduates.
    In fact, I would be surprised if any American medical graduates lost an opportunity to train in New York as a result of downsizing these training programs. That is not true, however, of international medical graduates.
 Page 171       PREV PAGE       TOP OF DOC
    Mrs. JOHNSON. Have you found any interest from the governments that provide the majority of the international medical graduates in negotiating with hospitals for slots that they would subsidize instead of the American taxpayers subsidizing them? That would not be disallowed by any proposal that we have been talking about, and it would enhance the ability of those governments to require well-trained graduates to return to their home country and share their expertise with their own citizens, as was the original tension of training in American institutions.
    I was just wondering, as the conversation about all of this matter has gone forward, whether any of you have seen any interest amongst foreign institutions in providing home country subsidies for foreign residents in our institutions.
    Dr. FOREMAN. That, thus far, has not presented itself as an opportunity, but I must say, Madam Chair, in our experience, international medical graduates come to train in this country for their own benefit, rather than on behalf of their country. That is to say, the overwhelming majority of international medical graduates come here to develop their own careers, some returning to their home countries and some remaining in the United States, as a matter of fact, a considerable number remaining in the United States, as you well know.
    We certainly would be open to discussing this with another nation, but the idea being that it is not our intention to retain additional residents in the system. They would have to return to the country of origin, it seems to me, to fulfill the general objectives of this experiment.
    Mrs. JOHNSON. Dr. Foreman, in terms of the New York experiment, isn't it true that the subsidy does decline? It just doesn't decline. It, in a sense, lags by 1 year, a decline in the number of residents.
    Dr. FOREMAN. That is exactly correct.
    Mrs. JOHNSON. Because I would remind the Members that that is exactly the model we used in eliminating farm subsidies last year. We eliminated agricultural subsidies. It is a very dramatic policy change, and we lagged the subsidy cut by 1 year in order to provide capital for farmers to get in another business. In a sense, that is one of the things I actually like about your proposal because it should provide you with a capital to train, to hire physician assistants or other kinds of folks that could help you with the workload, but be affordable when you get to the end of the project.
 Page 172       PREV PAGE       TOP OF DOC
    Dr. FOREMAN. I am delighted to hear you say that. That is precisely the intent.
    Some of the initial responses in the media suggested that this was somehow a bribe to New York hospitals to get them to do what they should have done anyway.
    We simply have no way of meeting our uncompensated care responsibilities without some incremental financing during this transition, and that is precisely why the lag was built in.
    Mrs. JOHNSON. How are you going to manage the issue of which residencies get eliminated?
    Dr. FOREMAN. Well, there are three different models. Individual institutions volunteer to participate. Pairs of institutions volunteer to participate, and consortia volunteer to participate.
    In the case of the individual institutions, they must meet the targets they have agreed to, generally speaking, a 25-percent reduction in the number of residents and maintaining or increasing the ratio of primary care physicians to specialists.
    In order to encourage consortia, the experiment permits two things. One, within a consortium, reductions in the aggregate need to meet the targets, but they do not need to meet the targets in every individual institution. So that, five hospitals together, collectively, would have to reduce the number of residents by, in the case of consortia, 20 percent, but that any individual in that consortium could lower their number of residents by more or less than 20 percent, and in joint applications, it works somewhat like that.
    Mrs. JOHNSON. But it would free the consortium to have this money to help with some of the transitional costs submerging and increases in efficiency as well as changing the structure of the personnel to cover the needs for care.
    Dr. FOREMAN. Precisely.
 Page 173       PREV PAGE       TOP OF DOC
    Mrs. JOHNSON. Congressman Stark.
    Mr. STARK. Thank you, Madam Chair.
    Just some comments. I am inclined to like the New York demonstration. I think we should legislate and make it applicable to all States that choose to participate, and I think that would have been a more proper way to proceed.
    I do notice that Episcopal and North Shore, they get $165,000 and $148,000 a year per resident, did not participate in the demonstration, and I can see where they are getting that much, they wouldn't want to.
    In the February issue of Modern Health Care, a spokeswoman from the North Shore health system is quoted as saying, ''We don't think the demo is intelligently applicable, and we will make some reductions, but they will be done at our pace, where they will not impact the quality of care or impact the 148,000 bucks per resident that they are getting.'' I rather suspect their pace will be somewhat akin to that of a medical snail, but with some, perhaps, sticks in addition to the carrots; Dr. Foreman, New York, as usual, has put a creative twist on pork, and I would like California, Dr. Hopper, to get right in there with him and make sure we get our share.
    I do have a conundrum here for California, but I have a hunch, it applies to all of the gentlemen at the table. I am going to give you the bad one, Dr. Hopper, first.
    The Women's Medical Group of Tarzana publishes a memo, and it includes a preprinted form, both in English and in Spanish, telling pregnant women who are not yet eligible for, I am going to say, MediCal, which is Medicaid for those of you who are unaware, that they should have an ultrasound exam by the 18th week of pregnancy, and they could get it at the women's medical group by paying cash, which obviously, if they are going to apply for MediCal, they don't have, or they could go to USCA's Olive Branch. They don't say where presumably. They will get the exam as uncompensated care.
 Page 174       PREV PAGE       TOP OF DOC
    Now, my guess is the marginal cost of an ultrasound is 5 minutes of the same doctor that has got the woman on the examining table anyway, while the machine is sitting at that doctor's elbow, and I have been in an examining room a dozen times when these ultrasounds have been performed. The same OB performs it, and if he isn't performing or she isn't performing, the machine is sitting idle. So it is the cost of the electricity and a few minutes of the doctor's time, but they can't collect later from MediCal if they do it then.
    Now, it seems to me this is dumping on your hospitals by a bunch of guys who are going to college, a very generous—now generous MediCal payment for treating this pregnant woman. So I say shame on them. That sounds to me like the cost of your hospital, Dr. Hopper, admitting this patient, getting him qualified for MediCal, which they will probably do, and not collecting either, that we are shuffling an uncompensated—probably at no net saving to the system.
    So I say that shouldn't happen with the private or profit medical centers dumping. But not to be outdone, the good guys, presumably, at UC-Irvine, in California Medicine, in their December-January issue, you get a Dr. Philip DiSaia, and I want both of these, if I may, Madam Chair, in the record.
    Dr. DiSaia sends a memo to his medical group in our Cal State teaching facilities saying, ''As a tertiary referral medical center and as the only center of its kind in Orange County, these attending physicians are sought out by many patients with a variety of conditions. However, if we are to remain financially viable, we can no longer tolerate patients with complex, expensive-to-treat conditions being encouraged to transfer to our group.'' He doesn't say it, but he means without insurance or a way to pay for it.
    In fact, it is incumbent on every physician member to refrain from encouraging transfer to our group or changes in insurance status from indemnity to HMO, for example, that place us at risk for serious negative financial consequences.
 Page 175       PREV PAGE       TOP OF DOC
    Now, that is very good business advice, but it is very lousy social policy, Dr. Hopper. I hope that Dr. DiSaia's memo doesn't reflect the UC system's philosophy any more than these folks at the Tarzana women's medical group should reflect the position of well-meaning physicians around the country. But what I am saying is, Here we have in the same place, people dumping these poor people. I would bet you when we are all done, nobody is saving a whole hell of a lot of money.
    So I just raise this issue to say that, as we think about how we are going to compensate people, there has got to be a certain amount of cooperation here among the for-profit or the fee-for-service folk—on the one hand, they come and tell us they provide all this charity care, while somebody on their staff is sending a memo to the doctor saying, Don't let him in. You can't have it both ways.
    Now, the only other question I have, to your knowledge, Dr. Hopper, are any of the medical schools in the university system negotiated with Columbia Hospital Corp. at this point? That rumor abounds. Do you know of any?
    UC-Irvine is the one that is suggested here.
    Dr. HOPPER. Mr. Chairman, or Congressman Stark, let me respond to your first question which is the relatively easier one.
    Mr. STARK. No. Let's try the last one first because I am out of time. I really want to know about Columbia.
    Dr. HOPPER. I think it is fair to say, and I mention in my testimony that we are actively having to look at a number of mergers and affiliations with other health care organizations.
    In San Francisco, we were fortunate that Stanford——
    Mr. STARK. Stanford Cal.
    Dr. HOPPER. Yes, which has the same background, the same modus operandi that we do, and I believe that is pretty straightforward.
 Page 176       PREV PAGE       TOP OF DOC
    The Columbia, as you well know, has purchased a number of university hospitals around the country. I think that it is fair to say that at this point in time, at both Irvine and until recently, San Diego, we were actively discussing, having discussions with a number of potential partners, including Columbia.
    Mr. STARK. We got rid of them at Sharp, right? That is dead.
    Dr. HOPPER. Well, as you may know, Mr. Stark, the State intervened in the Columbia Sharp situation, in such a way, as I understand it, that Sharp is off the table.
    I believe it is also fair to say that the sentiment of the community in San Diego preferred our looking more carefully at the Scripps of possibility, but we are at the behest of the regions looking actively for partners with our Irvine situation, and that leads me into a response to your first observation, and that is that I am not familiar with the specific comments by Dr. DiSaia, but I would point out that Irvine has the largest proportion of indigent patients of any major teaching hospital in the United States, up until very recently, that is. We have lost large amounts of money, and I think we have been dumped on by lots and lots of people, and I think that Dr. DiSaia is cautioned, but it may not have been counted in language that you and I would prefer.
    I think that caution basically said that if we continue to take unlimited numbers of compensated patients from the entire community, we ultimately are going to go broke. I don't know of any other way of putting it.
    Mr. STARK. I understand what you are saying. So, at Irvine, you are negotiating with Columbia is what you said.
    Dr. HOPPER. We are having conversations with Columbia and at least one other major for-profit organization at this point in time. The expectation, of course, would be that whatever arrangement we would go into, if we went into an agreement, it would be one that would protect the mission of the medical school and also protect the major part of our mission of caring for the uncompensated and undercompensated in that community.
 Page 177       PREV PAGE       TOP OF DOC
    I was personally surprised——
    Mr. STARK. It could be the subject of another conversation, Dr. Hopper, but the chances of that happening with Columbia involved in it, that is a matter to be reserved for some other time.
    Dr. HOPPER. I was surprised to find out, Congressman Stark, that one of the entities that we have had conversations with provide larger amounts of undercompensated care to citizens of Orange County than do the nonprofit hospitals at this point in time.
    Mr. STARK. Thank you.
    [The information follows:]
    INSERT OFFSET FOLIOS 18 TO 26 HERE
    [The official Committee record contains additional material here.]

—————


    Mr. MCCRERY [presiding]. I thank the gentleman from California for thoroughly discussing matters in California, and I thank Dr. Hopper——
    Mr. STARK. Well, you have got a lot of witnesses here from California.
    Mr. MCCRERY. I thank Dr. Hopper for being so forthright in his response.
    Mr. Jaffe spoke about a historic commitment to funding medical education. I am just curious. Would any of you like to tell us how medical education was funded prior to 1965 when Medicare came into existence? Does anybody remember?
    Dr. Foreman.
 Page 178       PREV PAGE       TOP OF DOC
    Dr. FOREMAN. I regret to say that I am old enough to have been in training before 1965.
    Hospitals were paid for prior to Medicare by a system that could only be called jury-rigged. Some people paid fees. Those people who were covered by insurance were covered largely by indemnity insurance, more often than not under Blue Cross or related sponsorship.
    Public moneys were made available to certain kinds of hospitals, and there were charitable contributions which were a substantial portion of the financial underpinning.
    From this mix of funding, hospitals conducted their business. Residents worked for little or no salaries. Sometimes they got laundry, room and board, and $50 a month. The faculty was largely unpaid voluntary physicians who gave whatever time they felt that they could.
    Graduate medical education was a catch-as-catch-can thing in many places, and the whole medical specialization explosion, which we are now living in, was really only beginning. The system was transformed primarily by the Medicare and Medicaid Programs, which suddenly put the financing, not merely of graduate medical education, but of the entire health care system on a much firmer footing, and particularly took care of large hospitals which had previously been teaching, and from the first day recognized in its cost-based reimbursement formula the cost of graduate medical education.
    In the prior panel, Mr. Stark and others spent a lot of time talking about the wide disparity in the payment, the direct medical education payments that run across the system.
    I think it is important for Congress to understand that it is not the cost of training physicians that varies so widely. What varies so widely is the proportion that part A pays of it. That is to say, when Medicare and Medicaid were plugged in, in 1965, the educational payments were cost based, and the costs on the day of the big bang reflected that portion of the cost the hospital paid. That did not reflect the portion that was paid for by State subsidy or by philanthropy or by private practice funds or whatever, and then, everything that has happened since 1965 is an outgrowth of that original cost-finding process.
 Page 179       PREV PAGE       TOP OF DOC
    One way or the other, even in 1984, when the direct medical education payments became prospective, the prospective payment was developed on an institution-specific basis by simply taking the existing cost, dividing through by the number of residents then on the payroll, generating an average cost, and then turning it forward every year since 1984.
    So we are, in fact, looking at the anomalies that were baked in early on as a result of various players paying their fair shares, but that is not to say all institutions, one way or another, are not financing graduate medical education, which is now much more expensive than it ever was because it is so much more complex and intense.
    Mr. MCCRERY. Well, at least one resident thinks we ought to go back to pre-1965. He wrote in the New York Times, so you should be interested, but he is from California, so you should be interested, Dr. Hopper.
    He writes recently that he believes teaching hospitals could provide better care if the Federal Government stayed out of the funding of graduate medical education. The doctor writes, ''Many hospitals receive more money from the program than they actually need to cover medical residency training. So residents are the cheapest possible hospital labor and are asked to perform many nonmedical tasks that ordinarily would be done by other personnel that the hospital must pay out of its own pocket.''
    As a result, residents ''spend less of their time treating patients and learning skills from experienced doctors.'' Now, this is a resident in the University of California system writing in the New York Times in the op ed page.
    Dr. Hopper, do you agree with this doctor's assessment of how residents are used and that they are the cheapest possible labor and that perhaps they would be better off without the subsidy?
    Dr. HOPPER. I certainly am hoping that he is 1 out of 9,000 in terms of those views.
 Page 180       PREV PAGE       TOP OF DOC
    Clearly, as Dr. Foreman has pointed out, residents provide a critical part of the work force in these hospitals and particularly in hospitals that treat large numbers of underreimbursed patients.
    I think the balance between the resident as a provider of care and as a student is one that we consistently have to deal with.
    I have no way of knowing this resident's particular experience that led him to the conclusion that things are wasteful, that Medicare money is being wasted, but I would hope from the broader perspective that we look at in terms of our overall product that Congress is persuaded that graduate medical education continues to be a public good, and frankly, we believe as efficient at this point.
    Mr. MCCRERY. A quick comment, Dr. Foreman.
    Dr. FOREMAN. A residency is an opportunity to learn a great many things, but not much about the economics of health care. I read that piece when it was published, and unfortunately, the writer really has a very primitive understanding of what the real costs of graduate medical education are and how they should be paid for.
    There is no doubt, however, that he understands how hard he works, because he does, and that a substantial portion of his energy and output does, in fact, convert into valuable and useful services on behalf, largely, of the poor.
    So I do not—I don't take issue with his view of the value of his services, but I would suggest to him and to others that if you did an analysis of the real costs of educating that doctor and how they were paid for, he is not paying for his own education.
    Mr. MCCRERY. Let me try to clarify what he is saying, lest we get lost. He is saying, basically, that since he is being subsidized by the Federal Government, he is valuable to the institution not only because of the subsidy, but because you can let other people go that aren't subsidized and have him perform their tasks. You don't do that?
 Page 181       PREV PAGE       TOP OF DOC
    Dr. FOREMAN. Mr. McCrery, I have the privilege of doing a number of things, and one of the things I do is to serve as a Commissioner on ProPAC. So I can cite the material that John Newhouse raised at the last hearing.
    If you look at Medicare margins for teaching hospitals, they do very well. If you look at the overall margins, they do very poorly.
    Now, in the end, all money is fungible, as you yourself pointed out. In the end, we take as many dollars in as we can and we pay out the dollars that we have to, and we get by, but very closely. That is, the teaching hospitals in this country have a 3-percent or less margin, and in New York, it is closer to 1 percent.
    So it is true that there are—if one looks at just one piece of the funding stream, some pieces are a little larger and some pieces are a little smaller, but when you add them all up, teaching is not a profitable business.
    Mr. MCCRERY. Dr. Hopper, in your testimony you discuss the agreement you reached with the State of California regarding reducing the number of nonprimary care residency physicians in your system. Did you get any Federal money for agreeing to that?
    Dr. HOPPER. No, sir. This was in response to, again, I think, first of all, a State initiative. I think, in all candor, Assemblyman Eisenberg pointed out something that we had been aware of for quite some time, that we needed to expand the number of primary care physicians. But we responded with a plan that, again, over a period of 8 years, would result in a total shift in that proportion.
    By the year 2000, 20 percent of our residency positions will be in family medicine, for example. We have not gotten additional dollars for doing this, and we are looking at, at this point in time, a subsidy in dollars, where the dollars that had gone into nonprimary care, into specialty care training, will be going into primary care.
    Mr. MCCRERY. So you don't plan to reduce the overall number of residencies?
 Page 182       PREV PAGE       TOP OF DOC
    Dr. HOPPER. As I pointed out, the president of the university commissioned a study on medical education for the university. That commission will be reporting back to him in June, and it is my anticipation they will recommend a rather substantial reduction in the overall number of residents for the University of California.
    Mr. MCCRERY. And without any Federal incentive?
    Dr. HOPPER. Without any Federal incentive, but I do have some concerns in that if there is not a policy mechanism in place in the State that approaches the downsizing across the board and not just the University of California, then, quite frankly, you could end up with other non-university-affiliated residency, just picking up the additional number.
    Earlier on, Dr. Wilensky talked about equity issues with respect to my colleague from New York, and if we end up having to downsize substantially, then, clearly, there will need to be some financial considerations for doing that.
    Mr. MCCRERY. Thank you.
    Dr. Foreman, one final question to you. You were asked by Mrs. Johnson about international graduates, and you stated at your conclusion that American graduates would fare much better in the competition for the reduced number of slots.
    In light of the fact that in New York you have such a high percentage of foreign graduates now in the system, how do you reach that conclusion?
    Dr. FOREMAN. What I meant to suggest is that right now, the system partitions itself into several sets. There are academic medical centers whose programs are filled largely with American medical graduates, and there are municipal and community-based teaching hospitals that vary considerably in the number of American graduates that they attract, but a very substantial number of institutions in New York attract no American graduates.
    As a matter of fact, about one-half of the teaching hospitals have nearly all or all of their residents as graduates of international medical schools.
 Page 183       PREV PAGE       TOP OF DOC
    We have a broad base of participation in the demonstration. The 42 hospitals and medical centers that are participating cut across all sorts of hospitals, including 11 public general hospitals run by the Health and Hospitals Corp., which have a very high proportion of international medical graduates.
    All of those hospitals are committed to reducing by 20 to 25 percent. Now, in a hospital in which all the slots are filled by international medical graduates, reducing by 25 percent leaves the residual of all international medical graduates, but 25 percent fewer.
    In institutions like academic medical centers, it seems very clear to those of us who run them that American graduates across the board will be given preference for the slots, as they are now, and that the 25-percent reduction that we institute at our place will remove very large numbers of residency slots, most of which, I think, are now filled by international medical graduates, but to the extent that we would displace an opportunity for an American medical graduate, there would be an abundant opportunity for that American medical graduate to essentially move to another institution and bump, as they say in union terms, bump the slots down the line until the last fellow out is almost certainly likely to be an international medical graduate.
    Mr. MCCRERY. Do you think that is a desirable result?
    Dr. FOREMAN. Well, it is not a policy result. We do not——
    Mr. MCCRERY. I understand that. Do you think it is a desirable result?
    Dr. FOREMAN. I think it is desirable. This is not an association position. It is my own. I think we should give preference to our own graduates, yes.
    Mr. MCCRERY. Mr. Starr.
    Mr. STARR. Thank you, sir.
 Page 184       PREV PAGE       TOP OF DOC
    I feel I ought to comment as we talk about the New York demonstration project. What we may be trading off here in some locations for an excess of physicians is a shortage of midlevel practitioners and a maldistribution of midlevel practitioners.
    I don't know in my particular situation who would backfill those positions very easily. I am not sure. Has anybody looked at what the supply is, the physician assistants, the nurse practitioners, the folks who we are going to have to put in place to replace these physicians that give the service?
    Mr. MCCRERY. I am assuming the institutions themselves will look at that. I think we have to assume the institution is not going to choose to deliver less health care. They are just going to do more with fewer people, and they will have to make those accommodations.
    Mr. STARR. Yes, sir, I am sure, but in those associations and professional groups, is the supply out there to accommodate this change, and how quickly can it do that?
    Mr. MCCRERY. Perhaps those in charge of making the accommodations can tell us.
    Mr. D'ERAMO. Well, there are two things that are lacking. One is the supply, and the other is the funding to increase that supply. Both are needed in order to supplant, in some cases, in many cases, the medical residency reduction, where service is still a component of what they do.
    I would like to hark back to an earlier point, though, and suggest to you that there is much greater rigor that is being imposed on academic teaching institutions relative to what residents can and should do vis-a-vis the educational experience they are supposed to be going through, and we have, over time, had to accommodate that. I think that is one of the things that overshadows what this one resident might have said about working.
 Page 185       PREV PAGE       TOP OF DOC
    As you look backward on history, I think the proportion of time of their having to work versus having pure opportunities for observation and education has really switched.
    Mr. MCCRERY. Dr. Foreman, did you have a comment on the availability?
    Dr. FOREMAN. We think the critical factor in sustaining the system as we downsize the residency training programs is money, not manpower. We believe that certainly in the New York area, we have or can generate sufficient manpower to fill the open slots.
    Mr. MCCRERY. Time will tell.
    Mr. Stark, since it is just the two of us, feel free.
    Mr. STARK. Thank you.
    I just wanted, Dr. Hopper, to have you talk with Dr. D'Eramo. You have a law in Connecticut, I believe, that does not allow for-profit hospitals. Is that correct?
    Dr. D'ERAMO. It is not a law, per se. The regulatory authority has not as yet seen fit to approve a for-profit institution.
    Mr. STARK. Are you comfortable with that?
    Dr. D'ERAMO. I am not sure it is going to hold up over time.
    Mr. STARK. In listening to Dr. Foreman, near as I can remember how hospitals seemed to operate back in the sixties, fifties, forties, thirties, it is different in the sense of teaching facilities now, but I do wonder, and I would make the same comment that we made to the previous panel in different ways, that I don't think we have heard today any recommended possibility in changing the way we educate physicians, but we have changed in the recent past, like 20 years ago, and now we are into almost a quantum change in the way medical care is being delivered with all the variety of so-called managed care plans.
    Now, to put the fear of God in Dr. Hopper, it isn't so much that we let Columbia Hospital into California, but it has been suggested that some managed care plans may, indeed, decide to train their own physicians. That ought to scare the hell out of you. They will train them to withhold care. That would be a great education, but I am saying that there is a ferment in the way you all are operating as medical care deliverers, and perhaps we may have to rethink, not because you are not doing a good job, but just maybe because you aren't going to get the patients from the managed care plans or you are not going to get the support you once had for financing the uncompensated care that you must provide.
 Page 186       PREV PAGE       TOP OF DOC
    Somebody—Dr. Wilensky, I think, earlier suggested that we aren't providing much training in the outpatient mode. It is an inpatient type of world, and that a lot of hospitals are shoving more and more procedures or medical provider plans out of the hospitals, steering them away from hospitals, and I just leave you, I guess, with the idea that maybe we are trying to shove a limited number of dollars into a program that won't be relevant to whatever new way medical care delivery is evolving, in spite of what we do here. I am just suggesting that all of these different plans, whether they are pushed by bargain plans with big employers, or whether they are pushed down by a change in the way we have classified groups of citizens as being eligible for any Federal assistance, the number of uninsured will grow. Where they will get treatment, I am not sure.
    In California, our community hospitals or our public hospitals are really going to be jammed if the Governor continues to have his way.
    I would hope that, perhaps, you all might come back to us and say maybe there is a different way to do this that accommodates the changes that are coming.
    I don't pretend to know what they are because we tend to operate much the way we have around here. We are still looking for proxies. We don't want to identify the hospital specifically, which I have always wanted to be the way. We just want a little simple formula. We want to leave here after we voted on the formula being fairly sure that all the deserving hospitals will get a little help, and those that aren't quite so deserving won't get so much.
    We are never sure because everybody complains. Once we are done, the formula doesn't make many people very happy, but I am a little nervous about having the for-profit managed care chains beginning to train physicians to serve in their particular brand of medical care delivery. That gets a little scary to me.
    So I would encourage you all, whom I trust more——
    Dr. HOPPER. Let me comment, Congressman Stark, on a couple of pieces of this. First of all, as you know, we have in California the oldest and perhaps the most prominent managed care organization at Kaiser.
 Page 187       PREV PAGE       TOP OF DOC
    Mr. STARK. Indeed, we do. Kaiser.
    Dr. HOPPER. Over the years and particularly in recent years, we have been getting the feedback from Kaiser, and frankly from others as well, that says they are having to retrain our physicians once they come into the organization.
    Mr. STARK. Let me take Kaiser out. Let me take Kaiser and the Marshfield Clinic and Johns Hopkins and a few of these other institutions that run managed care plans out of that. I am addressing myself to Prudential who is still under indictment for a variety of things and Columbia and——
    Mr. MCCRERY. Let's not talk about the good managed care plans.
    Mr. STARK. No, I am just suggesting if you have a 50-year-old plan, in my county, one-half of the people in the county belong to it, and those plans are an institution in some communities, as the charitable hospitals are in Louisiana. These are not the new entries into the Wall Street mix who may have an incentive for different reasons.
    Mr. MCCRERY. We are going to conclude so our witnesses can go back to their respective coasts, but I think the gentleman brings up an excellent point, not perhaps in the same way I would have, but it is an excellent point and one that we need to give some thought to, as the health care marketplace changes.
    Thank you all very much for your testimony and for your patience.
    [Whereupon, at 3:25 p.m., the hearing was adjourned.]
    [Submissions for the record follow:]

Statement of American College of Preventive Medicine

Introduction

 Page 188       PREV PAGE       TOP OF DOC
    As Congress turns its attention to the physician workforce needs of the nation as it relates to reform of the Graduate Medical Education system, the American College of Preventive Medicine (ACPM) urges Congress to consider the unique aspects of physician training in preventive medicine and public health as it studies and recommends a system of financing graduate medical education that will meet the needs of the 21st century. ACPM is the national medical association of physicians whose primary interest and expertise are in disease prevention and health promotion. Specialists in preventive medicine are uniquely trained in both clinical medicine and public health. They have the skills needed to understand and reduce the risks of disease, disability and death in individuals and in populations groups. Physicians trained in preventive medicine work in public health and community agencies, in primary care settings, in workplaces and in academia. The College membership constitutes a major national resource of expertise in disease prevention and health promotion, areas vital to protecting and improving the nation's health.
    ACPM strongly supports payment for graduate medical education through a new accountable system separate and distinct from reimbursement for patient care. Such a system should provide for training in preventive and population-based medicine in the variety of settings where such training takes place and should be financed equitably by all health care payers, both public and private. Particular attention must be paid to supporting the training of physicians in disease prevention and health promotion, through residency training in the specialty of preventive medicine.

Do We Need More Physicians Trained in Preventive Medicine?

    Every major study of national health workforce needs has concluded that there is a shortage of physicians trained in preventive medicine. The October 1992 report of the Council on Graduate Medical Education (COGME) reported continued shortages in the field of preventive medicine and recommended increasing the percentage of physicians trained and certified in preventive medicine as a national goal. COGME's June 1995 report recommended upweighting Medicare graduate medical education payments for preventive medicine residents.
 Page 189       PREV PAGE       TOP OF DOC
    In addition to the need for more physicians trained in the specialty of preventive medicine, there is a need for more training in prevention in all the other medical specialties, especially in primary care. In many cases, those specialties are initiating efforts to strengthen prevention education, particularly in relation to individual patient care. Specialists in preventive medicine, who have skills in population-based prevention as well as individual preventive interventions, can assist the other specialties in the further development of education in prevention for residents and medical students alike.
    The importance of placing greater emphasis on prevention of disease and disability in our nation's health system is universally acknowledged and the need for more physicians prepared to provide leadership in prevention has been consistently documented. The Third Report of the Pew Health Professions Commission in particular recommends that all health professions schools include the social and population-based health sciences, which are the underpinnings of preventive medicine, in an evidence-based approach to clinical work, and that the traditional public health disciplines and clinical medicine be brought together to address the needs of integrated systems of care. These objectives are identical to those of preventive medicine, which long has worked to bridge the gap between the individual and population health perspectives. Preventive medicine residency programs, medical school departments of preventive and community medicine, and schools of public health have a long history of experience in teaching and research in these disciplines and much to contribute to educating a physician workforce equipped to practice medicine well in managed care environments. However, our current system of financing physician training severely impedes our ability to produce physicians skilled in prevention and the population-based health sciences.

How Can We Train More Physicians in Preventive Medicine?

 Page 190       PREV PAGE       TOP OF DOC
    The current system of financing for graduate medical education is inconsistent with training needs in preventive medicine. Of particular importance is the fact that Medicare reimbursement for graduate medical education costs is available only to those few programs where residents can be counted toward a hospital's training costs. Most training takes place in non-hospital based settings. Therefore, most preventive medicine residency programs do not receive funds from Medicare or other traditional sources of graduate medical education financing.
    Many preventive medicine residency programs report that they turn away qualified applicants because they lack the funds for residents' stipends. The capacity of existing programs to train physicians is only about two-thirds funded. It is, therefore, essential that reforms to graduate medical education financing take into account the distinctive characteristics of training and practice in preventive medicine.

What Is Preventive Medicine?

    Prevention, in its broadest sense, is practiced by all physicians and other health professionals who help their patients stay healthy. Preventive medicine, however, is also a distinct medical specialty, one of 25 recognized by the American Board of Medical Specialties.
    The specialty of preventive medicine is based on our knowledge that promoting health and preventing disease require work with both individuals and communities. It is the only medical specialty whose objective is to equip physicians to care for both individuals and populations. The distinctive aspects of preventive medicine include knowledge and competence in:
    •  biostatistics
    •  epidemiology
 Page 191       PREV PAGE       TOP OF DOC
    •  environmental and occupational health
    •  planning, administration, and evaluation of health services
    •  the social and behavioral aspects of health and disease
    •  the practice of prevention in clinical medicine
    The American Board of Preventive Medicine grants certificates to physicians who have successfully completed three years of supervised training, one year of work experience, and a written examination in any one of three areas: general preventive medicine/public health; occupational medicine; or aerospace medicine. Specialists in general preventive medicine/public health focus their skills on population groups, such as the residents of a particular community or state, or the patient population of a managed care plan, health center or hospital. Occupational physicians focus on health and safety in the workplace, while the community associated with aviation, including passengers, is the domain of the aerospace physician. There currently are about 4200 physicians board-certified in preventive medicine.
    Preventive medicine specialists work in a wide variety of settings, including public health and community agencies, organized health plans and health maintenance organizations, outpatient and primary care settings, and academia. These physicians usually engage in multiple activities, including disease surveillance, planning, administration and evaluation of disease prevention and health promotion programs, quality management and outcomes measurement, research, teaching, and direct patient care.

How Are Physicians Trained in Preventive Medicine?

    There are 87 preventive medicine residency programs accredited by the Accreditation Council for Graduate Medical Education. These include 43 in general preventive medicine/public health, 41 in occupational medicine, and three in aerospace medicine. The first of the three years of training is in clinical medicine. Most preventive medicine residency programs do not offer this first year, so residents complete it in training programs in other specialties, usually a primary care specialty. The second and third years consist of academic training in the fundamental disciplines of preventive medicine and supervised practicum experiences. Rotations take place in a wide variety of public and private settings, including outpatient facilities, health departments, managed care organizations, and worksites. About 500 residents currently are engaged in some phase of training in an accredited preventive medicine residency.
 Page 192       PREV PAGE       TOP OF DOC
    Physicians in specialties other than preventive medicine also acquire skills in prevention and the population-based health sciences by studying in academic programs in schools of public health or medical schools. There are 27 accredited schools of public health and 14 accredited graduate programs in preventive medicine/public health located in medical schools.
    Accredited preventive medicine residencies are located in a variety of institutions. About half of the programs in general preventive medicine/public health and occupational medicine are located in medical schools. The department conducting the training may be called a department of preventive medicine, or family and community medicine, or occupational and environmental medicine. About 20% of residencies are located in schools of public health. The remainder are in health departments, military or other federal facilities, or hospitals. The three aerospace residencies are run by or closely connected with military facilities.
    It is important to note that all but a handful of accredited residencies are not in teaching hospitals. Residents learn through working under supervision in community and outpatient settings, where prevention is most effective. Therefore, traditional sources of financing for graduate medical education through reimbursement for hospital inpatient services are not routinely available to preventive medicine residencies.

Recommendations

    We support strongly two key recommendations of both the Pew Commission and the Institute of Medicine. First, payment for graduate medical education should be severed from payment for patient care. In preventive medicine and public health, the ''patient'' is often the community or an enrolled population. The interventions that preventive medicine and public health physicians conduct measurably improve the health of groups of people, often by preventing disease, yet the Medicare GME financing system has never enabled payment for training in population-oriented practice. In order to account for the growing diversity of training sites and training activities, it is essential that funding for residency training follow the resident, not remain as a component of reimbursement for inpatient care.
 Page 193       PREV PAGE       TOP OF DOC
    Second, all payers, not simply the federal government, should contribute explicitly to funding of graduate medical education. This is important both as a matter of equity and as a means of establishing accountability for the numbers, types, and quality of physicians that are trained.

Conclusion

    The paroxysms of change that are sweeping our health care system require new breeds of health professionals trained to manage not just the care of individual patients, but also the health status of entire populations. Physicians trained in preventive medicine and public health are leaders in such efforts, and the programs that train them are committed to working to meet the needs of a rapidly evolving marketplace. These programs can thrive if they receive equal access to funds for the costs of training, from which they have historically been almost completely cut off. An investment of a tiny percentage of the six billion dollars that Medicare alone spends to subsidize inpatient graduate medical education will enable preventive medicine and public health to help meet the challenges of balancing cost, quality and access in new health systems.

—————


Statement of Integrated Healthcare Association, Pleasanton, California

    The rapid growth in managed care and the pressure on the Medicare system have combined to create an environment in which a serious re-structuring of both the focus and financing of graduate medical education is timely and necessary. The Integrated Healthcare Association (IHA), a statewide California policy group concerned with managed care and integrated health care issues (membership roster attached), believes that the following principles should guide the development of new public policies:
 Page 194       PREV PAGE       TOP OF DOC
    •  Graduate medical education (GME) should be refocused in order to reflect current health care practice, including both what is taught and where people train.
    •  Funding for GME should be separated from funding for patient care.
    •  Public funding should provide strong economic incentives for the development of an appropriate health professions work force
    •  Medical schools and academic medical centers should seek input from managed care organizations to assure that the right kinds of health professionals are trained with the right skills; and managed care organizations, for their part, should develop specific suggestions for the kinds of core competencies that are needed
    •  Reliable data is urgently needed on the true costs of GME, the appropriate overall level of public funding to assure an adequate supply of health professionals, and the impact of re-structuring the financing mechanisms.

Background

    Market and policy developments are testing the financial stability of teaching hospitals and academic medical centers (AMCs) that offer medical education. Decision makers, both public and private, are seeking new models for providing the revenue necessary to produce the health care work force of the 21st century, and serious questions are being raised about whether current education systems are producing the right mix and skills among health professionals of all kinds.
    IHA's view of this issue is driven by its core belief in promoting the evolution of integrated health care supported by financial mechanisms that align the incentives of purchasers, payors, and providers. From an analytic standpoint, IHA is also trying to look beyond the immediate ''funding crisis'' to identify opportunities for significant, systemic changes. What parts of the graduate medical education function represent a public ''good'' that requires public financing, and what parts should be appropriately left to private market forces? How could we effectively separate the financing of clinical functions from educational functions at teaching institutions? How do we reflect the need to shift at least part of the educational process out of the institutional setting and into the community? How do we structure incentives so as to produce the right kinds of physicians and mid-level practitioners for a managed care environment? And who should appropriately bear the cost of their education and training? These difficult questions must be addressed now, and ideally in a process that includes all stakeholders.
 Page 195       PREV PAGE       TOP OF DOC
    It should be noted that there are legitimate and serious issues around the future of basic and applied research, and about the appropriate balance of public and private sector support for that function in the future. At this point, however, IHA is concentrating on the clinical and educational functions surrounding graduate medical education.

Major Issues

    The problem of developing long-term solutions for funding undergraduate and graduate medical education is complicated by multiple factors. In a companion background paper, IHA has noted the following key factors affecting new policy development:
    •  Increased public scrutiny and reduced public confidence in these institutions' ability to achieve their missions and improve the public's health.
    •  The price-conscious and competitive health care environment is eroding traditional sources of support for the tripartite mission of academic medical centers.
    •  Those institutions in the business of educating the health care workforce currently incur higher costs in providing patient care, but the payor community is not convinced that these higher costs are adequately justified.
    •  Work force size and training are not meeting industry demands.
    •  Current funding policies have contributed to the oversupply and maldistribution of physicians.
    •  Strategic partnering by medical centers with for-profit and non-profit partners may significantly impact the current practice of cross-subsidization of the academic enterprise by the clinical enterprise.
    •  Federal support of graduate medical education is dominated by Medicare but includes significant other funding sources as well.
 Page 196       PREV PAGE       TOP OF DOC

Suggested Principles for Policy Development

    To date, policy development has been hampered by lack of hard data on the true costs of medical education and by the narrow focus of policy development, which has tended to look at individual components of a highly complex set of financing mechanisms. It is time both to simplify and reform this system. IHA suggests the following principles for new policy development:
    1) Graduate medical education (GME) should be refocused in order to be reflect current health care practice, including both what is taught and where people train. Managed care is rapidly becoming the dominant mode of delivering health care. Yet the educational systems are not producing health care professionals with the right skills and in the right balance as rapidly as is needed. Changes in the mix of generalist vs. specialist physicians and increases in the number of mid-level practitioners (physician assistants, nurse practitioners, etc.) will take time to effect, but the pace at which this change occurs must be accelerated. Curricular changes are also needed to ensure that the health care professionals of the future understand and are equipped to practice in managed care settings. The locus of graduate training must continue to shift from institutional to outpatient and medical group settings, and public financing should encourage and reflect this shift.
    2) Funding mechanisms should be structured so as to provide incentives for appropriate health professions work force development. Investments of public funds should produce appropriate returns to public policy objectives. If more generalists are needed, or more mid-level practitioners, then funding mechanisms should provide clear incentives for the education and training of such individuals, and with sufficient strength to change behavior.
    3) Public policy should separate payments for graduate medical education from payments for patient care. Part of the dilemma in restructuring the way in which graduate medical education is financed is that we have not aligned funding with discrete public purposes and have created cross-subsidies and hidden policy impacts. In the absence of a clear understanding of the true costs of medical education in academic medical centers, we created a surrogate funding mechanism through the indirect medical education (IME) adjustment to Medicare DRG payments. Proposals to restrain the growth in the use of foreign medical graduates are hampered by the fact that these individuals have become a surrogate for direct funding of urban safety net programs. We can only properly align funding and policy outcomes if patient care and medical education payments are separated.
 Page 197       PREV PAGE       TOP OF DOC
    IHA supports two significant changes: (1) Uncoupling the medical payments from the financing system for Medicare managed care by taking all GME costs out of the calculations used to compute the AAPCC for Medicare risk contracting; and (2) placing these and perhaps other GME funding programs into a new GME trust fund that is available to support residency programs more relevant to managed care, including non-institutional as well as institutional settings..
    4) Medical schools and academic medical centers should create mechanisms through which they can obtain the input of the health plan and provider community in designing both the right mix of health professionals to be trained and the curriculum and training sites to be utilized. However graduate medical education is financed, it is critical that this function be enriched by the experiences of the marketplace to ensure that it is responsive to the needs of those who are arranging and delivering care. By the same token, it is incumbent on the health plan and provider community to help identify the core competencies needed for effective practice in managed care systems. Only through a partnership between academic medicine and the managed care community can we create the right balance and skills in the health professional work force of the future.
    5) Efforts must continue to be made to develop appropriate, reliable data on true GME costs and on the impacts of alternative funding scenarios to guide policy development. There is little reliable data on the actual costs of the graduate medical education function, or the net impact of the movement to more Medicare risk contracting. It is vital that such data be developed, especially if the financing of patient care and education are to be separated.
    The major system changes that will be required to restructure both the focus and financing of graduate medical education affect a large number of influential constituencies. It is critical that new policy be developed in an open, inclusive process with all stakeholders represented so that real reform emerges and not legislative gridlock. IHA itself is attempting to model this kind of process within California, both by including a broad range of stakeholder interests and by trying to focus on the function of graduate medical education, not the economic interest of any one constituency. Some national equivalent of this mechanism should be formed to help guide national policy development.
 Page 198       PREV PAGE       TOP OF DOC
    The marketplace is a powerful driver of change, and there is no question that managed care is creating rapid change in the design of both health insurance products and health care delivery systems. Absent major change in public policy regarding graduate medical education, managed care dynamics will also force changes in the focus and financing of this function. But this change process will take time and will include significant, disruptive side effects that could be avoided if the entire function of graduate medical education were re-based in terms of public policy.
    Graduate medical education has many of the classic characteristics of a ''public good'' in its relevance to public health and its lack of a perfect natural market for its research and educational dimensions. On the other hand, the public financing of GME has evolved into a tangled web of cross-subsidies and mis-directed incentives that has contributed to the current ''disconnect'' of academic medicine from the marketplace. IHA believes that there is a legitimate public interest in the education and training of a qualified and balanced work force of physician and mid-level professionals. By separating clinical from educational financing, public policy can inject appropriate market forces into the clinical dimension of academic medicine while allow funding for education to be reinvested in ways that will better meet the needs of today's and tomorrow's health care markets.

—————


Integrated Healthcare Association 1997 Board of Directors

CHAIR: Shelley A. Horwitz, Administrator, Bay Valley Medical Group, Inc.
PAST CHAIR: John M. Cronin, President/CEO, Priority Health Services
CHAIR ELECT: Arthur M. Southam, President/CEO, Health Net
 Page 199       PREV PAGE       TOP OF DOC
SECRETARY: Richard Slavin, M.D., Executive Vice President, Camino Medical Group
TREASURER: William H. Gurtner, Vice President, Clinical Services Dev., Office of the President, University of California

Member Organizations:

ADVENTIST HEALTH Donald R. Ammon, Executive Vice President
ALAMEDA ALLIANCE FOR HEALTH David J. Kears, Chief Executive Officer
BAY VALLEY MEDICAL GROUP, INC. Shelley A. Horwitz, Administrator
BLUE CROSS OF CALIFORNIA Max Brown, Senior Vice President, Specialty Business
BLUE SHIELD OF CALIFORNIA Bruce G. Bodaken, President/Chief Operating Officer
CALIFORNIA PUBLIC EMPLOYEES' RETIREMENT SYSTEM Margaret T. Stanley, Associate Executive Officer, Health Benefits Services
CAMINO MEDICAL GROUP Richard Slavin, M.D., Executive Vice President
CATHOLIC HEALTHCARE WEST J. Michael Gallagher, Senior Vice Pres., Public Policy Catholic Healthcare West & St. Joseph Health System
CONSUMERFIRST Clark E. Kerr, President/Chief Executive Officer
FOUNDATION HEALTH, A California Health Plan Gary S. Velasquez, President/Chief Operating Officer
GOULD/SUTTER MEDICAL FOUNDATION Charles V. Wirth, Chief Executive Officer
HEALTH NET Arthur M. Southam, M.D., President & CEO
HILL PHYSICIANS MEDICAL GROUP Steve McDermott, Executive Director
JOHN MUIR/MT. DIABLO HEALTH SYSTEM J. Kendall Anderson, President/CEO
KAISER FOUNDATION HOSPITALS & HEALTH PLAN Bernard J. Tyson, Senior Vice President/Market Leader for East Bay
LOS ANGELES COUNTY HEALTH SERVICES DEPARTMENT Mark Finucane, Director
 Page 200       PREV PAGE       TOP OF DOC
MEDICAL CENTER AT THE UNIVERSITY OF CALIFORNIA SAN FRANCISCO William B. Kerr, Director
MEDPARTNERS, INC. Mark L. Wagar, Chief Operating Officer
PACIFICARE OF CALIFORNIA Rick Badger, Vice President & General Manager
PRIORITY HEALTH SERVICES John M. Cronin, President & Chief Executive Officer
ST. JOSEPH'S REGIONAL HEALTH SYSTEM Edward G. Schroeder, President and Chief Executive Officer
SCRIPPSHEALTH Bud Beck, M.D., Executive Vice President, Physician Services Division
SUTTER/CHS Charles H. Mason, Jr., Consultant to Sutter/CHS
UNITED HEALTHCARE Henry Loubet, Senior Vice President, CEO, Western Region
UNIVERSITY OF CALIFORNIA William H. Gurtner, Vice President, Clinical Services Development, Office of the President
UNIVERSITY OF CALIFORNIA, BERKELEY James C. Robinson, Ph.D., School of Public Health
VETERANS HEALTH ADMINISTRATION U.S. DEPARTMENT OF VETERANS AFFAIRS Marjorie S. Wolf, Director, Sierra Pacific Network
WOODLAND HEALTHCARE H. Bard Coats, M.D., Medical Director
EX OFFICIO William K. Piché, President, Management Alliance
STAFF Beau Carter, Executive Director











(Footnote 1 return)
See Appendix A for a detailed discussion of the roles and characteristics of teaching hospitals.

(Footnote 2 return)
IMGs are U.S. citizens and persons of foreign citizenship who attend foreign medical schools. Medicare does not currently discriminate in its support of residency programs on the basis of whether the resident is an IMG or a graduate of a U.S. medical college so long as the IMG resident is in a qualified residency program and has passed Parts I and II of the National Board of Medical Examiners test. This test is equally applicable to both American and foreign IMGs.

(Footnote 3 return)
The DSH adjustment was developed as a component of the Medicare Prospective Payment System (PPS) because rural and urban hospitals that treat large numbers of low income patients have higher costs than otherwise comparable institutions due to the provision of extra services such as translation and social work, increased security, and other factors such as increased severity of illness. A hospital's eligibility for DSH payment is calculated according to the proportion of inpatient days it provides to Medicaid and low income Medicare beneficiaries. This calculation produces a disproportionate patient percentage for each hospital; hospitals with a percentage of 15% or greater qualify for DSH payments. Hospitals with percentages of 20.2% or higher are considered ''high DSH'' hospitals.

(Footnote 4 return)
Medicare and the American Health Care System: Report to the Congress, June 1996, Prospective Payment Assessment Commission, page 63.

(Footnote 5 return)
U.S. Department of Commerce, Bureau of the Census; NYS Department of Social Services; 1995 Current Population Survey.

(Footnote 6 return)
The AMA data may differ from other sources due to their restriction to residents in accredited programs and different data collection techniques.

(Footnote 7 return)
The NRMP is a national computerized placement process intended to maximize medical student placement in programs of choice. Under the NRMP, all teaching program sponsors, as well as graduating medical students, agree to abide by the rules of the program and to submit rank-ordered lists of preferred programs in the case of students, and preferred students in the case of programs. On so-called ''Match Day'' (March 19 in 1997), the NRMP matches these lists to place students in residency programs around the country.

(Footnote 8 return)
Primary care is defined as under current Medicare law for direct medical education payment purposes, i.e., family medicine, general internal medicine, general pediatrics, general osteopathic medicine, geriatric medicine, and preventive medicine.

(Footnote 9 return)
The demonstration design includes a second phase to incorporate hospitals able to commence resident reductions in the 1998–1999 program year. Participants that begin in 1998 would be required to make the residency reductions in five years and reimbursement would be phased out over five years under a slightly different schedule. The demonstration thus would end at the same time for both phases.

(Footnote 10 return)
This strategy met its goals and has kept New York's rate of cost shifting to private payers the second to lowest in the country. According to ProPAC, the U.S. average private payment-to-cost ratio in 1994 was 124%, meaning that hospitals were able to charge private payers 24% more than cost to make up for shortfalls in Medicare and Medicaid reimbursement. The private payment-to-cost ratio in New York State was 108%. Rhode Island was the only state whose hospitals charged private payers less in relation to their costs.

(Footnote 11 return)
Characteristics of Employer Sponsored Health Plans, Health Benefits in 1996 (KPMG Peat Marwick LLP 1996).

(Footnote 12 return)
The 100 largest cities are ranked according to population and defined as central cities, not MSAs. The analysis was conducted using data from the 1994 AHA Annual Survey.

(Footnote 13 return)
Self-pay patients typically include both uninsured patients who can afford to pay some or all of their hospital bills out-of-pocket, as well as uninsured patients who can or do not. So, for NAPH members, ''selfpay'' patients are the equivalent of ''no-pay'' patients.

(Footnote 14 return)
National Public Health and Hospital Institute. Urban Social Health: A Chart Book Profiling the Nation's 100 Largest Cities. Washington, DC:1995.

(Footnote 15 return)
Since much of the overall care provided by NAPH members (as measured by their ''gross revenues'') is to uninsured patients who cannot afford to pay for their care, the actual revenues received (as measured by their ''net revenues'') that is represented by Medicaid and Medicare revenues is higher than the proportion of care provided to these patient populations.

(Footnote 16 return)
Uncompensated care is accommodated in the formula only indirectly, because payments are made to hospitals with at least 100 beds that receive at least 30 percent of their net revenues from state or local government payments for indigent care.

(Footnote 17 return)
Diane N. Plumb and T. Henderson, ''Medicaid Financing of Graduate Medical Education: A Survey of the States,'' GWU Intergovernmental Health Policy Project study for the Association of American Medical Colleges, October 1995.