|
Items in cart [0] |
Questions? Call 888-624-8373 |
| Copyright © 2009. National Academy of Sciences. All rights reserved. Terms of Use and Privacy Statement |
Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 173
FINANCING OF MEDICAL AND GRADUATE MEDICAL EDUCATION EDUCATION:
ISSUES IN PRIMARY CARE EDUCATION SUPPORT*
Ruth S. Hanft, Ph.D.
Research Professor and Consultant
George Washington University
Introduction
The financing of graduate medical education is complex and evolved
pragmatically with the historic development of teaching hospitals and patient care
financing. For almost all of this century, medical and graduate medical education
clinical instruction has been concentrated in the hospital setting. Graduate and
undergraduate medical education are intertwined because of the jointness of
clinical activities of faculty/residents and M.D. students.
Although the medical school, as an institution, is the focus of undergraduate
medical education, graduate medical education organizationally evolved as hospital,
program/specialty based education. It is only in recent years that the majority of
programs are affiliated with a medical school but control remains decentralized on
a departmental or program basis and the accreditation processes are separated for
undergraduate and graduate education.
While there are data on the sources of revenue that support medical schools
and their faculty, and sources of data on support of hospitals, there are no
comprehensive data on the funding streams for medical and graduate medical
education that: allow disaggregation by discipline/specialty; separate the funds
flow between undergraduate and graduate medical education; and separate
inpatient and ambulatory care financing. Many of the funding sources are
fungible and their specific use is departmental or hospital specific. There have
been special one-time studies, cited later, that provide some fragmentary data on
primary care education financing, particularly family medicine graduate programs
and ambulatory care training.
* I would like to thank Catherine White for her ongoing assistance; Lawrence
Clare for assistance on federal funding of primary care; and Jessica Townsend for
her helpful comments.
173
OCR for page 174
The discussion that follows provides:
o
o
o
o
o
A brief history of financing medical and graduate medical education
A discussion of the general federal support streams
Role of the states
Discussion of primary care and ambulatory care education financing
Potential sources of support for primary care and ambulatory care
education
History of Support
Until the development of a variety of federal programs that indirectly and
directly support medical education, didactic and clinical undergraduate medical
education was supported similarly to other higher education through state
appropriations for public institutions, tuition and endowments. Public and private
medical schools continue to have different mixes of sources of support. Graduate
medical education and clinical faculty were supported by hospitals and fees
_1 _ ~ 1~__ A_ __ 11 1~ · · ~ ~ · 1
generated oy faculty trom provision ot patient care services to private patients.
The majority of clinical education was conducted in hospitals with indigent
patients until the advent of Medicare and Medicaid and since that time large
programs continue to be concentrated in public hospitals and/or hospitals with
large indigent caseloads (Hanft, unpublished). As late as 1955, residents had
virtually no responsibility for the day-to-day management of private patients of
moderate means and even less contact with patients on the hospitals' "gold coast"
(Rabkin, 19601.
Prior to World War II, medical schools relied on a small full-time faculty,
mainly in the basic sciences, and a "volunteer" or geographic full-time clinical
faculty which received no compensation or modest stipends from the school and/or
hospital with the quid pro quo being the prestige of the affiliation with a medical
school and/or teaching hospital and admission privileges for private patients at a
teaching hospital. Residents and interns were provided with meals, housing and
sometimes small stipends.
Medicine started developing a scientific base about 1890 and medical schools
and some hospitals began to evolve into research institutions with the institutions
receiving funds from individuals, foundations and state appropriations (Dietrick
and Berson, 19531. The Ransdell Act of 1930, created the National Institutes of
Health (Mider, 19761. The National Cancer Institute was established in 1937
(IOM, 1976; Hanft, 19841. World War IT marked a turning point with substantial
federal investment in biomedical research funding. This funding became a major
174
OCR for page 175
source of indirect support for medical schools and their faculty with the clinical
research focused in the teaching hospital.
In 1945, at the end of World War II, President Truman explicated his views
of the deficiencies in health and medicine. Among the recommendations were
federal support of medical education and biomedical research. In 1947, the
President's Scientific Research Board recommended a substantial increase in
federal funds for medical research and medical education (Bardley and Harvey,
1976).
Though the AMA opposed federal support for education, financing of health
care services and expansion of social insurance to health care, it did not oppose
funding of biomedical research. Biomedical research funding became a major
source of revenue for the development of medical schools and the expansion of
facula with clinical faculty involved in graduate medical education as well as
undergraduate education. NTH grants awarded for the conduct of research and
research training included funds for salaries of faculty who conducted research
and also spent time in teaching. These funds enabled the schools to expand the
numbers of full-time faculty (Relman, 1984). By 1955, one third of medical school
revenues were derived from research grants and contracts (Stevens, 1971; JAMA,
1987).
Direct federal support for education did not emerge until the 1960s,
although throughout the fifties numerous individuals and commissions expressed
concern about the slow increase in the number of medical students. Federal direct
support for schools and undergraduate medical education began in 1963, in
anticipation of the increased utilization of services that would result from the
expansion of private insurance and the enactment of health insurance for the
elderly (Stevens, 1971; Hanft, 1982; Relman, 1984). This support was relatively
brief.
Patient care funding began to increase with the growth of private health
insurance in two ways. Hospitals incorporated education costs into their charges,
and for a growing number of middle class patients, insurance covered physicians'
fees. Teaching programs, however, continued to rely principally on indigent
patients until the advent of Medicare and Medicaid (Relman, 1984; Stevens, 19711.
A recent study shows a continued strong link between graduate medical education
and indigent care (Hanft unpublished).
Today support for graduate and undergraduate medical education comes
from a multiplicity of sources. Although data are available on medical school
support and some data are available on resident support, there are only very gross
estimates of hospital support, mainly Medicare estimates. There are no national
175
OCR for page 176
estimates on support for outpatient education. There is no way to separate
precisely medical school support by graduate and undergraduate training except
through special case studies.
Although sources of financing are similar for undergraduate and graduate
training, the proportion contributed by each revenue source differs for each type
of education. In the undergraduate years biomedical research, state appropriations
and practice plan revenues dominate the financing of medical schools. During
residency years patient care funds from both hospital and faculty practice plan
revenues dominate, paying an estimated 80 percent or more of the total costs of
graduate medical education, with the majority of funds from hospital revenues.
Table ~ shows trends in revenues; Table 2 shows sources of revenue for
public and private medical schools.
As Table ~ shows, the fastest growing source of revenues for medical
schools are medical practice plans. Much of this revenue is generated from the
joint teaching/patient care activities of graduate medical education. These
revenues have risen rapidly as a Pronortion of total medical school revenues.
~ , 1 1 ~ 1 ~
~ ederal revenues, other than research, have declined substantially and revenues
from research have also declined as a proportion of total revenues, although they
constitute the second largest source of support.
Table 2 shows the most recent published data by ownership of medical
schools. In reviewing the data, it should be noted that almost 60 percent of the
schools are public schools. Ownership is a key factor in the distribution of
revenues, with almost 30 percent of public school revenues derived from state and
local government. Private schools are more heavily dependent on professional fee
income. Payment from hospitals account for one-fifth of revenues and professional
fee income for another fifth or a combined total of 43 percent. In contrast, public
universities derive less than ~ percent of their revenue from hospital payments
and a slightly smaller amount than private schools from professional fees.
However, the difference in hospital payments may be an artifact of the way states
provide funds to their medical schools and teaching hospitals. For example, where
the state owns its own teaching hospital, the full faculty salary may be paid
directly to the medical school through the appropriation to the school rather than
passing the supervisory salary through the hospital books, or the hospital may pay
the department with the revenues reported as practice plan revenues.
176
OCR for page 177
Table 1
Trends in U.S. Medical School Revenues
Selected Years 1971-1987
Revenue Source
1970-1971 1975-1976 1986-1987
Percent Percent Percent
Federal research 25.6 24.3 19.9
Other federal 18.8 11.7 3.8
State and local government 18.9 23.8 18.5
Tuition and fees 3.7 4.6 5.3
Medical service 12.2 18.0 37.6
Other income 20.9 17.6 14.8
Total* 100.0 100.0 100.0
* Totals may not add due to rounding.
Source: "Medical Education in the United States, 1986-87" Journal of the
American Medical Association, August 26, 1988, Vol. 260, No. 8,
Table 4, p 1080
177
OCR for page 178
Table 2
Revenues of Public and Private Schools by Source of Funds. 1986-1987
Percent Distribution
Source of Funds Public Private
State and local government 29.7 2.2
Professional fee income 19.4 22.5
Recovery of indirect costs 5.0 S.3
Tuition and fees 3.2 7.6
Endowment 0.2 2.3
Gifts 0.2 1.2
Income from college services i.7 0.S
General university funds 2.4 1.0
Reimbursement from hospitals 7.S 21.2
Research and teaching training 1.5 1.1
Sponsored programs* 25.9 29.0
Miscellaneous 2.9 2.7
Total ** 100.0 100.0
* Mainly biomedical research.
** Totals may not add due to rounding.
Source: Same as above - Table 9 p. 1083.
178
OCR for page 179
It is important to note that these tables represent averages and there is
wide variation among medical schools as to source and mix of revenues. The
mixes of revenues depend not merely on public or private ownership, but factors
including: university ownership of the teaching hospital; the types of teaching
hospitals affiliated with the medical school; the patient payor base for generation
of fees; the research emphasis of the school.
Comparable aggregate data are not available for teaching hospitals,
undergraduate medical education separated from other missions or for graduate
medical education separated from other missions.
Federal Support*
Biomedical Research
Federal direct investment in support of medical education began in the
early sixties but indirect support from biomedical research was in place.
Biomedical research funding, predominantly from the National Institutes of
Health, began to grow rapidly after World War II, and provided a base of indirect
support for medical schools, particularly for undergraduate medical education.
While biomedical research funds are granted for the conduct of biomedical
research, these funds also support faculty salaries and graduate fellowships. This
support enabled medical schools to expand the size and expertise of their faculty
and led indirectly to the further development of technology, the demand for
specialist training and the consequent specialization of the delivery of medical care
in the United States. The combination of an increasing reliance on federal
support for research and the opposition by the American Medical Association to
direct federal involvement in the support of education, led to the domination of
biomedical research and research/specialty oriented faculty in medical education
over a twenty year period from the 1940s to the 1960s. The influence of the
domination of this support ran and runs counter to the public goals of increasing
the supply of primary care physicians relative to other specialties. State
appropriation support of public schools and more recently patient care funding
have tended to counterbalance the research domination, as did the direct federal
capitation payments (support per student) in the 1970s (Hanft, 1982).
*Drawn extensively from R. Hanft "Impact of Changes in Federal Policy on
Academic Health Centers Health Affairs Vol 1 No,. 4 Fall 1982:67-81.
179
OCR for page 180
Though biomedical research funds are still a major source of revenue for
medical schools, they have declined as a proportion of total revenues since the
1960s. However, the proportion of these revenues to total revenues varies
substantially by institution with some institutions receiving much higher than the
average revenues from this source and some schools deriving little revenue from
this source.
Education Support
In the late 1950s and early 1960s, increased availability of health insurance
stimulated public demand for health care services. In addition, there was a public
perception of a shortage of health professionals. Efforts to provide hospital
insurance for the elderly and increased federal involvement in the care of the
indigent elderly through the Kerr-Mills Act contributed to growing public and
congressional fears that the increased demand for services could not be met due to
a shortage of health professionals. Although some private and public commission
reports urged federal support for health professions education, organized medicine
stood firm in its opposition to such support until the early 1960s.
In 1963, at the onset of the "Great Society" programs, the Health Profession
Education Assistance Act was passed (P.~. 88-1291. It provided for matching
grants to assist in the construction of teaching facilities for schools of medicine,
dentistry, osteopathy, public health, optometry, pharmacy, podiatry and nursing
(Health Resources Administration, 1980). The Act was the start of a stream of
legislation which enlarged the federal direct commitment to health professional
education and culminated in the 1971 Comprehensive Health Manpower Training
Act (P.L. 92-157) and the Nurse Training Act (P.L. 92-168). These laws provided
a new type of support for health professions schools - "cavitation" - which was
based on the number of students enrolled in health professions programs.
Although the manpower legislation continued to require the expansion of
enrollment until 1980, during this later period the emphasis was on issues of
geographic and specialty distribution of physicians rather than on gross numbers
(Health Research Administration, 19801.
By 1974, disagreements over the need for capitation support developed.
The executive branch was concerned about a potential surplus of physicians, while
Congress was concerned about geographic shortages of personnel and specialty
distribution. In fiscal year 1980, the capitation support level for medicine was
reduced to less than one half the 1972 amount, and in fiscal year 1982 it was
completely eliminated. Special grants were included from the early 70s to spur
the development of family practice programs and beginning in the late 70s to
support primary care residencies in medicine and pediatrics. These special grants
180
OCR for page 181
continue at very modest levels and include support for family medicine, general
internal medicine, pediatrics and geriatrics. Grants for family medicine residency
programs began in 1972. These programs grew from 117 to 381 in allopathic
medicine. About 45 percent of these programs received federal support between
1972 and 1988 and in 1988 total federal grants were 20.3 million dollars (Health
Resources and Services Administration, 19881. Grants for general internal
medicine and general pediatrics began in 1977 and currently are funded at about
13 million dollars.
During the 1970s there were changes in support from the Veterans
Administration (VA) and the Department of Defense (DOD) including: Veterans
Administration support to increase enrollment and to improve the quality of
instruction in existing medical schools; the development of new 'Veterans
Administration" medical schools; and the creation of a federal medical school to
train physicians for the uniformed services. The VA began to develop affiliations
with medical schools in 1946 under legislation whose goal was the improvement of
the quality of care in VA hospitals and clinics. Medical school faculty and
residents have been provided salary support through these arrangements since
that time. The Veterans Administration Medical School Assistance and Training
Act of 1972 (P.L. 92-541) enabled the VA to assist in the establishment of new
state medical schools to be operated in conjunction with and located near VA
hospitals, and the expansion of existing schools. Five new schools were developed
and the first students were enrolled in 1977 and 1978 (Health Resources
Administration, 1980). Despite the direct educational support provided by the
federal government, biomedical research and patient care dollars continued to
provide higher proportions of financial support to institutions during this period.
Budget proposals from the Administration since 1981, have tried to
terminate all direct education support for health professions schools except the
support of faculty and residents provided through the VA deans' committee
arrangements and the funding of the Uniformed Services Medical School and
military residencies. About 12 percent of the residencies are supported through
these programs. The Congress however, has repeatedly rejected termination of
grant support for primary care and family medicine education.
Patient Care Support
Patient care activities are an integral part of the educational process.
Prior to the 1960s, clinical instruction was supported primarily through small
stipends from the hospitals for support of residents and interns (Stevens, 1971;
Relman, 19841. Full time clinical faculty were supported by tuition and state and
county appropriations to state universities and their hospitals and salaries from
181
OCR for page 182
county hospitals. Volunteer and part-time faculty found their support through
private practice earnings. Most joint patient care/teaching activities centered on
indigent patients and were conducted in federal, counter, municipal, university, and
philanthropically supported hospitals (Rabkin, 1960~.
With the advent of Medicare and Medicaid and the rapid growth of private
insurance, new sources of funding became available for support of clinical training.
This support took two forms: additional salary support of residents and
supervisory teaching physicians in hospitals, and support for patient care services
to individuals newly covered by public or private insurance. These new sources of
revenue enabled teaching hospitals to expand their residency programs, increase
substantially the stipends paid to residents, pay faculty for supervision of
residents, and allowed these programs to keep pace with the expansion of
undergraduate medical school programs. In fact, the number of residency
positions have exceeded the number of medical school graduates for several
decades. In addition, these funds provided an additional stream of support for
schools and faculty, allowing for continuing rapid expansion of clinical faculty.
Teaching physicians who heretofore had provided individual patient care
service for indigent patients without reimbursement were now able to bill fees for
these services, for those covered by Medicare and Medicaid. Many medical schools
and their individual departments and divisions developed or expanded organized
"practice plans" for collection and disbursement of these fees. This source of
income has grown rapidly since the early 1970s (Hanft, 1982; JAMA, 1988~. The
amount of revenue generated from faculty practice plans varies widely among
institutions, depending on many factors, including the payment sources for
patients' care, the structure of the practice plans. The plans now account for
more than 19.4 percent of the gross revenue of public schools and almost 22.5
percent of private schools' revenue. Hospital payments in addition, amount to 7.8
and 21.2 percent respectively (see Table 21. It should be noted that not all
practice plan revenues flow through the medical schools and the amounts reported
in the literature are understated. These revenues are usually distributed by the
practice plan, after expenses are paid, to faculty in the form of salary supplements
and fringe benefits. Depending on the structure and charter of the plan, some of
these funds are also used by the department for recruitment, travel, seed money
for research and development, support of fellows and in a few instances, for
support of residents It should be noted that many plans provide a percentage of
gross or net revenues to the dean/vice president, which can be used for a variety
of purposes.
With the increased flow of third party payments, in the 1970s, issues
relating to both geographic location and types of specialty training began to arise,
as well as issues of the "fairness" of financing. Specifically, reimbursement from
~2
OCR for page 183
third party payers for inpatient services has financed a greater proportion of the
costs and charges than for outpatient services. Until recent changes designed to
reduce health insurance costs, private hospital insurance rarely required cost
sharing by the consumer. In contrast, reimbursement for outpatient services from
third parties is usually structured to include deductibles (payment by the patient
before the third party will pay) and coinsurance (a percentage of the bill paid for
by the patient) and does not cover preventive services. It is therefore easier to
support specialty training oriented toward inpatient specialties and inpatient care
than primely care specialties oriented toward outpatient training (IOM, 1976;
GMENAC, 1980).
Additionally, the usual and customary fees billed by physicians and paid by
public and private insurance programs tend to reward procedure oriented practice
at a higher rate than cognitive activities. Specialties such as surgery and
radiology are able to generate much larger revenues than pediatrics. A number of
commentators: the Institute of Medicine (IOM, 1976), the Health Care Financing
Administration's Office of Research, Demonstration and Statistics (HSAIO and
Stason, 1979) and the Graduate Medical Education National Advisory Committee
(GMENAC, 1980) and the Council on Graduate Medical Education (COGME,
1988) report that these sources of revenue have contributed greatly to the
emphasis on specialty and subspecialty training in medicine. The recent study for
the Physician Payment Assessment Commission, has recommended a new
structure for physician payments that would increase fees for primary care and
cognitive activities and reduce fees for surgery and other procedures (Hsaio, 1988).
The fee structures and the sites of care clearly influence the amount of practice
plan revenue different departments of a medical school can generate.
Financing of education through third party payments has different economic
burdens and benefits than financing through general revenues of government.
The benefits and burdens also fall in geographically uneven patterns since the
location of graduate medical education programs is not related to per capita
population in a state nor to the number of undergraduate medical student
positions within a state.
Private health insurance premiums are fixed dollar contributions based on
the health care experience of the particular group. Small employer groups, and
high risk industries pay higher premiums than large, young, white collar groups.
Employers can deduct the full cost of their contributions from their gross income
for tax purposes. Large industry, particularly unionized industry, tends to have
more comprehensive health insurance coverage than small or low wage industry.
The overall effect is regressive with smaller/low wage industry paying higher
prices for the same insurance and potentially a higher subsidy for graduate
medical education. This factor, however, is counterbalanced to some extent by
183
OCR for page 186
hospitals. In addition, some states provide specific funds for training in family
medicine and primary care residencies.
The reduction of federal funds to state governments with the end of general
revenue sharing and reductions in funding for many social programs has led states
to scrutinize increasingly, medical education and teaching hospital support.
Competing priorities for state revenues, combined with the separation of Medicare
and Medicaid reimbursement in 1982, are stimulating reevaluation of state
support. State Medicaid programs no longer follow the Medicare reimbursement
principles and many do not recognize indirect education costs. In addition,
Medicaid reimbursement can be at rates below the costs of teaching hospitals, and
Medicaid providers can be limited by competitive bidding (Schramm, 19831. State
appropriations decreased from 23.S to 18.5 percent of medical school revenues in
1986-87.
The Lewin Wisconsin study surveyed a number of states regarding state
support for graduate medical education. For example, California in 1985-86,
provided 40 percent of the support of resident stipends and fringe benefits at five
University teaching hospitals. They provided operating subsidies to these
hospitals. In addition, there is a special grant program for support of family
medicine residencies. The state has required cuts in residencies.
Illinois supports the residency positions in its state University hospital but
these funds are not earmarked. There is also support for residencies in primary
care specialties. Indiana University hospitals receive no operating subsidies or
payment for resident stipends. The state does subsidize residency programs in
community hospitals. In addition, there is a special grant program for family
Practice residencies. Virmnia provides a subsi~v to its University hospitals for
~ ~ , ~
· ~ · . ~ . ~ ·~ ~ · · · ~ · ~- · ~ ~ ~ ~
indigent care and supports tamlly medicine residencies ~Lewln 1YbU).
The states varsr widely in how they support graduate medical education,
their teaching hospitals and the degree of control they maintain over the number
of residency positions in their own hospitals with some states, notably New York,
attempting to control the total number of residencies.
State university hospitals are a major training base for residents, providing
approximately 15 percent of all of the graduate medical education positions.
~6
OCR for page 187
Primary Care Residencies and Ambulator Care Training
Historically, support of graduate medical education has come from hospital
financing. While prior to the Flexner report preceptorships with individual
physicians were a common method of training, the rise of the hospital and the
concentration of technology in the hospital led to the focus of undergraduate
clinical and graduate medical education in hospital settings (Rosenberg, 1987).
Traditionally, hospitals incorporated education costs in their cost base and these
costs were recognized during the implementation of Medicare. Blue Cross cost
based plans also recognized these costs and hospitals incorporated these costs in
charges to charge payers.
The evolution of the family medicine residency and the development of the
community based medical schools in the 1960s and 1970s stimulated the initial
interest in a change in focus of residencies from the traditional large teaching
hospital to community based hospitals and ambulatory care settings. The advent
of competition in the late 1970s and prospective payment, stimulated changes the
nature of the hospital and the delivery of health care services and has led to
increased pressures to expand education sites to ambulatory care settings. While
most pronounced in the primary care specialties, ambulatory care training is an
increasing need in specialties like ophthalmology, radiology, and general surgery.
The financing of graduate medical education however, has not changed
accordingly, except for the recent Medicare change which recognizes the direct cost
the hospitals pays when the resident is in an outpatient setting, including
outpatient settings outside the hospital if the hospital is willing to support these
costs.
There are no national data on financing of graduate medical education in
ambulatory care settings. There are data on federal primary care grants and data
can be aggregated from the states where there are appropriations for primary care
residencies. Family medicine residences and faculty were funded by 32 states in
1985. In dollars in 1982, the total state funds were 54 million (Ricketts, DeFreize
and Wilson 1986). There are fragmentary studies primarily in family medicine on
sources of financing but not in total dollars.
Family medicine residencies are structured differently than other
residencies. In general, the first year of education is based in a hospital with the
financing from the hospital. In subsequent years the education takes place in an
ambulatory care group setting with support from grants and fees for service
generated by faculty and residents. Residents who are licensed can have their
services billed for in these settings, although not in the hospital setting. On
187
OCR for page 188
average, about thirty percent of the revenues come from fees from patient care
(Ciriacy, 19851. Perkoff cites estimates of a maximum of one-third of the expenses
of primary care education could be generated from patient fees (Perkoff, 19861.
Schroeder has noted that shifts in the site of training from the hospital to
the ambulatory care setting are not followed by the revenues "Thus the medical
schoolts designated sources for clinical teaching (tax supported state money for
public) and some private schools ... tend to remain at the teaching hospital even
while some of the teaching function shifts out of it..." (Schroeder, 1988 p. S121.
There are several generic problems in financing prlmaly care reslaencles
outside of the hospital setting. These problems may be of lesser magnitude in
support of general surgery or other procedural specialties where patient charges
tend to be substantially higher for services. The problems are summarized as
follows:
. _
In the hospital setting the resident and supervisory
physician are paid salaries from hospital revenues with
education costs separately recognized by Medicare and
Medicaid and historically included in hospital charges.
a personal and identifiable service is prodded by the
teaching physician, a fee can be charged to the patient
or insurer. Residents may not bill fees.
In the outpatient setting not linked to a hospital (for
Medicare) and for outpatient settings in terms of other
insurers, the resident's salary and a supervisory salary
for the faculty must be generated from fees to the
patient/third party or from grants from government
and/or philanthropy. In the primary care specialties, the
fee levels as noted extensively in the literature, are
substantially lower than for procedure oriented
specialties. While there are two sources of patient care
support for hospital based or hospital outpatient linked
training there is only one in the non hospital ambulatory
care setting. Payments for physicians services as
distinguished from payments for hospital services,
historically did not incorporate education costs since
education was almost exclusively hospital based in
allopathic medicine.
-
The development of faculty practice plans has been on a
departmental/specialty basis similar to the organization
OCR for page 189
of residencies with the procedural specialties able to
generate substantially higher revenues than primary care
specialties, because of the Medicare and private
insurance charge structure. The revenues of these plan
flow to the department with some small percentage
flowing to the institution.
. . ~. . . .
Conceptually, all education,
both undergraduate and graduate medical education
should be an institutional responsibility. The
organization of medical schools on a departmental basis
and graduate medical education on a specialty/program
basis combined with the departmental flow of hospital
and practice plan revenues leave the medical school
institution with a paucity of flexible funds. Institutions
that do not receive public appropriations, or where the
appropriation is in the form of line items, unless the
institutional percentage of practice plan revenue is
. While
substantial, have little ability to cross subsidize
cross subsidies are endemic among the missions of a
medical school, they do not operate on an institution
Awe basis in the medical schools for graduate medical
education programs. High earning departments and
specialties retain the majority of their practice earnings
for departmental and even division rather than
institution wide goals.
Graduate medical education programs for more than fifty years were
focused in the hospital setting. With the development of third parer payments,
these payment systems incorporated the costs of education within the hospital at
first; imnlicitv With the enactment of Medicare graduate medical education. costs
~ O ~ ~
~ · · . ~ · ~ ~' ~ ~ . AT · ~ ~ · ~ ~ · · I · 11
were explicitly recognized as allowable costs. hospital relmDursement initially
under Medicare was cost reimbursement. Physicians' services however, were and
are, reimbursed on a charge basis, based on usual and customary costs within a
prevailing, with no mechanism and little need until the 1970s to recognize
education functions.
Supporting education through charge based reimbursement to physicians is
feasible and has been done by Mayo and in part by some primary care programs,
notably family practice residencies. There are however, both institutional and
structural constraints that limit the viability of relying on charge payments to
fully support ambulatory based programs summarized as follows:
- The charge and reimbursement structure that rewards procedural
activities at much higher rates than cognitive services.
~9
OCR for page 190
-
-
Potential reduced productinty in ambulato~ care education and
higher costs in contrast to hospital based education (see LJave paper).
The high indigent and Medicaid caseloads that reduce revenue
potential from fee billings where public clinics , community health
centers, and/or outpatient departments of public hospitals are the
sites of education.
The departmental structure of practice plans and the consequent lack
of flexible funds at the medical school level that could allow it to
cross subsidize across revenue generating and nonrevenue generating
graduate medical education programs.
Greater burden on patients since outpatient services generally have
higher proportinate deductibles and coinsurance than inpatient
services.
Currently Medicare permits the hospital to count residents serving in
outpatient settings and allows salary and fringes for them. This assumes
however, that the hospital will be willing to pay for services provided outside of
the institution. While helpful to the hospital in supporting patient services in
outpatient settings in its own facilities, the liklihood of this source of support
being available to freestanding ambulatory care settings is limited.
The Importance of Patient Payor Mix
As A Source of Financing of Primary Care Education
The scattered studies of the financing of primary care residencies,
particularly family medicine residencies, show the importance of clinic fees in
supporting these programs. The discussion above has pointed out that while
education costs have been incorporated into hospital payments, either implicily
through the charge structure or explicitly through cost pass throughs and indirect
adjustments, there is no similar support mechanism for ambulatory care training.
Even when directly funded through grants or the hospitals in their own clinics,
physician payment/fees for service are an important source of support. The
patient payor mix becomes a critical element in the ability to finance the
ambulatory care training outside of the traditional hospital setting.
,, ~
The lack of an education component in the fee structure is compounded by
other factors, based on the type of payor for the patient care. Where community
health centers serve as the ambulatory care site, they are heavily dependent on
Medicaid payments and federal grants under Section 330 of the public health
190
OCR for page 191
service act. These grants do not provide funding for residents. Furthermore,
except in states like New York where Medicaid payments to these types of
facilities are relatively generous, physician fee payments under Medicaid are
substantially lower than under Medicare or private insurance and in some states
are actually below average practice costs.
In highly competitive medical markets, HMOs must keep their costs low
and premiums competitive. Unless there is another source of support for
residents and faculty, the graduate medical education programs would need to
either substitute for current costs or generate sufficient income through enhanced
productivity to cover the graduate medical education costs in an HMO (see the
Lave paper).
An ambulatory care setting in a well insured fee for service community on
the other hand, is in a far more favorable financial position. Yet even in these
instances ambulatory care training in primary care is more difficult to support
than surgical or subspecialty care because of the fee structure. Payment to cover
the education costs, assuming that graduate medical education is additive to the
cost of practice, would entail transfer of some costs directly to the patient because
of deductibles and coinsurance applied to most outpatient services.
Robert Walkington's paper, published in this volume, provides a number of
examples of the reliance on different types of program support, including
demonstration projects, AHEC funding, state and federal grants.
Who Should Support GME
The question of who benefits and who should pay for graduate medical
education has been a subject of debate for many years. The resident's role
increases from apprentice to quasi/student provider to independent provider during
the course of the residency. Despite the finding of the National Labor Relations
Board that house staff should be considered students, the fact remains that they
provide a substantial amount of patient care (IOM 1974, 1976; Arthur Young
1986). While the academic medical community has long sought to place graduate
medical education within the purview of the medical schools and universities, and
therefore might be expected to seek or provide funding for this period of
education, the fact remains that resident clinical training has several beneficiaries.
Analyses of the activities of house staff uniformly show that a large part of
resident/training is spent providing care to patients and frequently without direct
supervision (IOM 1974, 1976). The 1976 IOM study found that house staff spent
an average of 61 percent of their time providing patient care with or without
191
OCR for page 192
supervision. An additional 15 percent was spent in patient care activities that
included undergraduate medical students and other health professions students
(19761.
Is the resident a student or an employee? If a student, shouts he/she pay
tuition like other graduate students or can one argue that the salary earned is
lower than the income that could be generated in practice at the same stage of
the career, so in effect the student is paying an implicit tuition in the form of
foregone earnings (Yoder, 19801. No matter how the resident is classified, there is
no doubt that the resident provides a significant amount of patient care services
that might otherwise be prodded by physicians who render fees for these services,
or more likely services to patients who are uninsured.
From the studies that have been conducted, patients are major beneficiaries
of house officer activities. Furthermore, a large proportion of training takes place
in public and large teaching hospitals and their outpatient departments that
provide care for the indigent. Residents have traditionally and continue to
provide indigent care. In some states, the family practice residency clinics are
major providers of care to Medicaid and uninsured patients.
Medical schools also benefit from the teaching activities of house staff. The
studies show a significant teaching contribution by the residents to the education
of medical students. Yet no payment is made by the schools to the teaching sites
for these activities.
Faculty also benefit significantly. The resident provides an "extra" hand
and coverage for the teaching physicians, as well as contributing substantially to
the services that the teaching physician bills for. These fees flow to the medical
school department and faculty in addition to the hospital payment for supervising
faculty.
Hospital benefits accrue from the availability of round-the-clock physician
staffing by residents at lower cost than if provided by community physicians.
Teaching programs are also regarded as a qualitative asset for hospitals.
The resident also benefits, receiving the advanced training that provides the
skills to practice a high earning profession.
There are multiple beneficiaries. However, current practice places the
majority of financing burden on patient care funds. Patient care financing under
certain programs, notably insurance premiums and Medicare, is regressive.
Opponents of using patient care funds to finance graduate medical education
object on several grounds: That this constitutes a "sick tax" on teaching patients
192
OCR for page 193
because the inclusion of the costs of GME tends to increase costs to patients; that
because the general public contributes to medical care financing through insurance
premiums and social security taxes, payment for graduate medical education
through reimbursement may result in inequitable income transfers from people
with low incomes to physicians with high income expectations; because some
specialties, e.g. surgery, are better able to generate patient care income, there is
an incentive to support and expand programs in those specialties, thus affecting
the ultimate specialty distribution of physicians (Fruen and Korper, 1981;
Commonwealth, 1985~. In addition, the distribution of residencies unrelated to
either future location of physicians or per capita population, places
disproportionate financing burdens on certain states (Hanft, unpublished).
In fact, a case could be made for multiple sources of financing from the
multiple beneficiaries if there could be quantification of the benefits that accrue to
each. These sources of financing could include:
Tuition paid by the resident for education.
Salary support from health professions schools to account for
teaching activities of the resident.
Salaly support from the hospital for standby/coverage of services.
Fees or salary support from patients/third parties for the provision of
services.
Salary support or fee sharing from faculty.
Moving to increased fee support, to support increased primary care training
in the outpatient setting raises the following problems, and probably is not
feasible unless the financing of practice plans on a departmental basis is changed
to institution wide plans. The barriers to increased fee based support in summary
include:
Increased regressing of the financing.
Continued incentives to support residencies in the high fee/high
earning specialties and inability of the primary care specialties to
generate sufficient funds to cover education costs.
The concentration of GME in settings with a large number of
indigent, both Medicaid and uncompensated care, patients.
Large amounts of dollars, not including faculty fees, estimated at 8-10
billion dollars (4 billion of Medicare funds) flow for the support of graduate
medical education. The issue is not the amount of revenues but the distribution
of the revenues in support of graduate medical education priorities.
193
OCR for page 194
REFERENCES
Bardley, J. and Harvey, A.M. (1976) Two Centuries of American Medicine 1776-
1976 W.B. Saunders Company, Philadelphia 413-419.
Commonwealth Fund (19881. Report of the Task Force on Academic Health
Centers, "The Contribution of Academic Medical Centers to Clinical
Research" (unpublished).
Commonwealth Fund (1985) Prescription for Change, Report of the Task Force on
Academic Health Centers, The Commonwealth Fund, New York.
Council on Graduate Medical Education, (COGME 19871. Discussion Paper on
Financing of Graduate Medical Education Washington, DC.
Deitrick, d.E. and Berson, R.G. (1953), Medical Schools in the United States at
Mid CenturY, McGraw Hill, New York, Chapters 2 and 3.
Fruen, M. and Korper, S. (19811. Issues in Graduate Medical Education Funding
Journal of Health Politics. Policy and Law (1) 87-97.
GMENAC (1980) Report of the Graduate Medical Education National Advisory
Committee to the Secretary, Department of Health & Human Services,
Washingrton, DC, DHHS Publication No. (HRA) 81-651.
Hanft, R.S. (1982) The Impact of Changes in Federal Policy on Academic Health
Centers. Health Affairs 1~41.
Hanft, R.S.~1989~. An Analysis of Factors that Influence the Size of Graduate
Medical Education Programs: Implications for Financing (dissertation in
progress).
Health Resources Administration (HRA), U.S. Department of Health and Human
Services 1980. Chronology of Health Professions Legislation 1956-1979.
Government Printing Office, Washington, DC.
Health Resources and Services Administration (HRSA) U.S. Department of Health
and Human Services (19881. Overview of Program Activity for Graduate
Training in Family Medicine. Section 786(a) Public Health Service Act. An
Annual Report of Grant Program Activity.
194
Fiscal Year 1988, Rockville, MD.
OCR for page 195
Page 2
Hsaio, W. Braun, P.;
An Overview".
22353.
Dunn, D.; Becker, E. "Resources Based on Relative Values,
Journal of the American Medical Association 260~16) 2347
Institute of Medicine (19741. Costs of Education in the Health Professions Parts
~ and Il. National Academy of Sciences, Washington, DC.
Institute of Medicine (1976~. PaYment of Teaching Physician Medicare Medicaid
Reimbursement Policies, National Academy of Sciences, Washington, DC.
Journal of the American Medical Association (JAMA) 1988.
Issue. American Medical Association, Chicato 260~81.
Lave, J. (1984~. Hospital Reimbursement Under Medicare.
Fund Quarterly 62~2) 251-268.
L`ewin and Associates, Birnbaum I, Hanft, R. et al (1986~.
Education in Wisconsin. Report to the Division of Health, State of
Wisconsin.
Annual Education
Milbank Memorial
Graduate Medical
Lundy, J. (1984). "Payment for Medical Education Under the Medicare Program",
prepared for The Hearings on the Subcommittee on H~lt.h ~nr1 th"
Environment of the U.S. Senate Committee on Finance.
Mider, C.. (1976~. "The Federal Impact on Biomedical Research". Advances in
. . , . . ~ _
American Medicine: Assays at the Bicentennial, Josiah Macy Foundation,
New York.
Perkoff, G.T. (1986~. Teaching Clinical Medicine in the Ambulatory Setting, New
England Journal of Medicine 314~) 27-31.
Rabkin, M.T. (1960~. The Teaching Hospital and Medical Education: One Room
Schoolhouse, Multiversity Dinosaurs? Journal of Medical Education 60 92-
97.
Relman, A.S. (1984~. Who Will Pay for Medical Education in our Teaching
Hospitals, Science 226, 20-30.
195
OCR for page 196
Page 3
Ricketts, T.C.; De Friese, G.H. and Wilson, G. (1986~. Trends in the Growth of
Family Practice Residency Training, Health Affairs 5~4) 84-96.
Rosenberg, C.E. (19871. The Care of Strangers, Basic Books Inc., New York,
Chapters 7 and S.
Schramm, C.~. (19831. Teaching Hospitals and the Future Role of State
Government, New England Journal of Medicine 308~1) 41-45.
Schroeder, S. (19881. Expanding the Site of Clinical Education: Moving Beyond
the Hospital Walls, Journal of General Internal Medicine 3(S) S5-14.
Stevens, R. (19711. American Medicine and the Public Interest, University Press,
New Haven and London, Chapter 17.
Yoder, S. (19801. "Financing Graduate Medical Education" in Graduate Medical
Education Present and Prospective A Call to Action, Josiah Macy
Foundation, Inc., New York.
Young, A. and Company (1986~. Study of the Financing of Graduate Medical
Education, Washington, DC.
196
Representative terms from entire chapter:
graduate medical
