Copyright 1999 Federal Document Clearing House, Inc.
Federal Document Clearing House Congressional Testimony
May 12, 1999
SECTION: CAPITOL HILL HEARING TESTIMONY
LENGTH: 3729 words
HEADLINE:
TESTIMONY May 12, 1999 MURRAY ROSS EXECUTIVE DIRECTOR SENATE
FINANCE CHANGES TO THE MEDICARE SYSTEM
BODY:
May
12, 1999 Medicare's Special Payments and Patient Care Costs Murray N. Ross,
Ph.D. Executive Director Medicare Payment Advisory Commission before the
Committee on Finance U.S. Senate Chairman Roth, Senator Moynihan, members of the
Committee. I am Murray Ross, Executive Director of the Medicare Payment Advisory
Commission (MedPAC). I am pleased to participate in this hearing looking at
Medicare's special payments and patient care costs. My testimony today is
intended to provide you with background information about Medicare's policies
and not to support or oppose any particular policy option under consideration.
For this hearing, the Committee asked MedPAC to describe Medicare payments to
providers that are not directly linked to patient care services for
beneficiaries. Policymakers' interest in this topic stems from questions about
how provider activities supported by these special payments should be financed
if the Medicare program were put on a more market-based footing. Where
Medicare's special payments support activities that benefit society at large,
they raise program spending and beneficiaries' premiums above what they would
otherwise be. Beneficiaries might be unwilling to bear the costs of those
activities through the premiums they paid to private health plans in a
restructured program. Classifying special payments Several reasons make it
difficult to describe Medicare payments not specifically linked to patient care.
First, some payments that are commonly asserted to be for things other than
patient care may in fact cover patient care costs. The payments Medicare makes
to teaching hospitals for the direct costs of graduate medical
education may fit in this category. Second, some payments that look
like patient care--because they are made for specific units of service to
Medicare beneficiaries--may cover costs other than patient care. Some portion of
the indirect medical education adjustment that Medicare makes in setting
payments to teaching hospitals for inpatient hospital stays fits in this
category. Finally, in some situations Medicare may pay providers more than their
average cost of providing care to Medicare beneficiaries. Medicare's payments to
disproportionate share hospitals cover a portion of the cost of patient care for
people other than Medicare beneficiaries. Special provisions for rural providers
and payment floors in the Medicare+Choice program reflect the higher costs of
producing services at low volume and the difficulties of operating health plans
when both enrollment and provider supply are low. These policies are often seen
as a way of maintaining beneficiaries' access to those providers and fostering
the availability of choices. My testimony today discusses these examples in more
depth and points out the factors that policymakers need to consider as they
weigh alternatives for reshaping the Medicare program. Medicare's direct medical
education payments to teaching hospitals Medicare pays teaching hospitals an
amount, separate from payments under the inpatient prospective payment system
(PPS), for the direct costs of operating residency programs. These
payments--known as graduate medical education (GME) payments--
reflect salaries and benefits for residents and supervising physicians, office
costs, and other overhead. The Congressional Budget Office (CBO) estimates that
Medicare GME payments totaled $2.5 billion for fiscal year 1998, of which $2.2
billion was paid for residency training and $300 million was paid for nursing
and allied health training. When PPS was first enacted, Medicare paid its share
of hospitals' full GME costs. Since the late 1980s, however, payments have been
based on hospital-specific per-resident amounts, calculated using 1984 costs
updated for inflation and based on Medicare's share of inpatient days, not its
share of costs.(1) Several additional rules affect what is actually paid. First,
residents in their initial residency period--up to five years--are counted in
full toward payment, while those beyond the initial period are counted as half
time. Second, the Balanced Budget Act (BBA) of 1997 capped the number of
residents hospitals may include in their count at the 1996 level (although a
3-year rolling average of resident counts is now used to cushion the effect on
hospitals that reduce the size of their residency programs). Finally, the
per-resident amounts are set slightly higher for residents in primary care and
related specialties. Many observers view payments for the direct costs of
graduate medical education as a subsidy to teaching
hospitals--and ultimately residents--unrelated to the costs of care for Medicare
beneficiaries. But economic theory suggests why this may not be so. In
preparation for our forthcoming report on graduate medical
education, MedPAC's Commissioners have considered whether hospitals'
training costs are borne by residents in the form of lower salaries. If that is
the case, the direct costs actually represent the costs of patient care rather
than training costs. This conceptual approach, however, does not tell us whether
the current level and distribution of GME payments is appropriate. This idea
stems from an accepted proposition in economics that in competitive labor
markets, rational employers will be unwilling to pay for the costs of general
training--training that makes workers more productive in all settings, not just
that of a particular employer. This result occurs because employers cannot
recoup the costs of such training through workers' higher future productivity;
if they tried to do so, workers would move to other employers where their
training was equally valuable. Workers who want general training must therefore
pay for it by accepting lower wages; they are willing to do so because acquiring
training allows them to earn higher wages in the long run. If this general
proposition holds in the context of teaching hospitals, then all of the direct
costs of graduate medical education can be attributed to
patient care. Although Medicare might appear to be paying for costs that are not
directly related to patient care--salaries for supervising faculty, overhead,
and the like--the payments it makes for the costs of residents' stipends are
lower by that same amount. In practice, the matter is considerably more complex,
and reality does not always conform to economic theory. But as a general
concept, this proposition implies that discussions about whether Medicare should
pay for direct GME should not center on the issue of whether the program is
subsidizing residents' educations. Rather, the focus should be on whether the
additional costs of care from having residents reflect a difference in product
for which society is willing to pay. (The next section discusses this point
further.) Medicare's indirect medical education payments to teaching hospitals
In addition to GME payments, Medicare adjusts teaching hospitals' operating
payments to reflect their higher costs per discharge that cannot be directly
attributed to teaching activities. These indirect medical education (IME)
payments totaled $4.1 billion in fiscal year 1998, according to CBO. The IME
payment amount depends on hospitals' teaching intensity, as measured by the
ratio of residents to beds. When PPS was enacted, the adjustment was set at 11.6
percent for each 10 percent increment of teaching intensity. This adjustment was
double the estimated relationship between residents per bed and Medicare
operating costs per discharge. Since then, the IME adjustment has been reduced
several times, most recently by the BBA. The BBA reduced the adjustment from 7.7
percent in 1997 to 7.0 percent in fiscal year 1998, 6.5 percent in 1999, 6.0
percent in 2000, and 5.5 percent in 2001 and later years.(2) (For comparison,
MedPAC's most recent estimate of the effect of a 10 percent rise in residents
per bed on costs per discharge is 4.1 percent.) The BBA also established a
separate IME payment to teaching hospitals that treat Medicare beneficiaries who
are enrolled in Medicare+Choice plans. That payment is being phased in over a 5-
year period beginning in 1998. Medicare's IME payments have been justified on
the grounds that they compensate teaching hospitals for several factors that
raise their costs but which cannot be separately identified: a more severe case
mix that is not reflected in Medicare's DRG payments, special capabilities, such
as the presence of trauma centers and burn units, unsponsored clinical research,
and higher quality of care related to teaching hospitals developing--or being
early adopters of--new diagnostic and therapeutic technologies. In reviewing
Medicare's payment policies, MedPAC believes that, other things being equal,
Medicare's payments should reflect the costs an efficient provider would incur
in providing patient care. By this standard, Medicare's IME payments clearly
reflect patient care costs to the extent they correspond to a more severe case
mix than is found in other hospitals. Where teaching hospitals' higher costs
reflect a different product, or when payments finance social missions other than
patient care, policymakers may ask whether those payments should be made by
Medicare or some other way. Medicare payment policies intended to maintain
access and foster choice A number of Medicare payment policies are intended to
maintain access to care for Medicare beneficiaries and to foster choices among
different providers and types of private health plans. These policies include
disproportionate share (DSH) payments made to hospitals that treat large numbers
of low-income patients, provisions for special payments to hospitals and other
providers in rural areas, and the floor payments established in the BBA for
Medicare+Choice plans. These policies may be justified in different ways. DSH
payments are intended to compensate hospitals that provide above-average amounts
of care to low-income patients. If Medicare and other payers' payment rates
covered only the costs of patient care for their own enrollees, hospitals would
not be able to make up for the uncompensated costs of care furnished to
low-income patients. Consequently, hospitals might seek to treat fewer
low-income and uninsured patients. Special payments to rural providers and the
floor payments to Medicare+Choice plans in some counties reflect a slightly
different rationale. Because rural providers and plans must generally operate on
a smaller scale, they cannot exploit economies of scale. Accordingly, their
average costs will be higher. If Medicare paid only the costs of an efficient
provider in average circumstances, its rates might not be sufficient for
low-volume providers to continue in operation or to induce health plans to
enroll beneficiaries in some areas. Disproportionate share payments The
disproportionate share (DSH) adjustment was implemented in 1986, the third year
after PPS began. An estimated $4.5 billion was spent on the DSH adjustment in
fiscal year 1998. The BBA reduced DSH funding by 5 percent, in single percentage
point increments implemented from 1998 through 2002. DSH payments are
distributed through a percentage add-on to Medicare's DRG payments for inpatient
hospital stays. The add-on hospitals receive is determined by a complex formula
and the share of their services provided to low-income patients. The low- income
share is the sum of two ratios--patient days for Medicaid recipients as a share
of total patient days and patients days for Medicare beneficiaries who are
eligible for Supplemental Security Income as a percentage of total Medicare
days. The adjustment was originally justified on the assumption that because
poor patients were more costly to treat, hospitals with substantial low-income
patient loads would have higher Medicare costs per case than would otherwise
similar institutions. That assumption has not borne out, however, and the DSH
adjustment has increasingly been viewed as serving the broader purpose of
protecting access to care for low-income Medicare and non- Medicare populations
by assisting the hospitals they use. In both its March 1998 and March 1999
Report to the Congress, MedPAC has relied on this premise in recommending
changes to the DSH adjustment. The Commission believes that DSH payments could
be made more equitable by using a better measure of care to the poor and by
using a distribution formula that more consistently links hospitals' DSH
payments to their low-income share. Under MedPAC's proposal, the low-income
share would be broadened to encompass all low-income groups by including
uncompensated care and measures of care covered by local indigent care programs.
The same distribution formula would be used for all hospitals, in contrast to
the current 10 formulas that provide a wide range of payments for hospitals
serving the same proportion of low-income patients. Special payments to rural
hospitals Several provisions of Medicare payment policy increase operating
payments for certain classes of rural hospitals above what they would otherwise
receive under the PPS. These classes include sole community hospitals, small
rural Medicare-dependent hospitals, reclassified hospitals, and rural referral
centers. Some rural hospitals may benefit from more than one of these
provisions. Sole community hospitals. Sole community hospitals are
geographically isolated providers representing the only readily available source
of inpatient care in an area. These hospitals are paid the highest of three
amounts: the PPS operating payments that would otherwise apply; a
hospital-specific amount per discharge based on their operating costs in 1982,
updated to the current year; or an amount per discharge based on their operating
costs in 1987, updated to the current year. About 700 facilities are designated
as sole community hospitals. Small rural Medicare-dependent hospitals. These are
rural hospitals with fewer than 100 beds and whose Medicare share of days or
discharges exceeds 60 percent for the cost reporting period that began during
fiscal year 1987. For discharges occurring in fiscal years 1998 through 2001,
these hospitals receive PPS operating payments plus 50 percent of the difference
between their updated hospital-specific base year amounts (1982 or 1987) and the
PPS rate. About 370 hospitals meet the qualifying criteria. Reclassified
hospitals. Hospitals that meet certain criteria may be reclassified by the
Medicare Geographic Classification Review Board to an area other than the one in
which they are physically located. In most cases, hospitals are reclassified
from a rural area to an urban area or from an other urban area to a large urban
area. Reclassification may affect either the standardized payment amount (the
basic payment rate under PPS) or the wage index (an adjustment made to the labor
component of the standardized amount to reflect local labor market conditions).
Even though the standardized payment amount does not vary between rural and
other urban areas, hospitals reclassified for this purpose may benefit by
qualifying for DSH payments (or for higher DSH payments) as urban hospitals.
Rural hospitals reclassified for the purpose of the wage index receive a higher
adjustment to the labor component of their standardized rate. In fiscal year
1998, 314 rural hospitals were reclassified for one or both of these reasons.
Rural referral centers (RRCs). Rural referral centers are rural hospitals that
meet criteria regarding the number of beds, annual discharge volume, case-mix
index, or proportion of care furnished to patients referred from outside their
local area. The standards RRCs must meet for geographic reclassification are
less stringent than for other hospitals, allowing many to qualify for a higher
wage index and for DSH payments as urban hospitals. Each of these provisions
raises PPS payment rates for RRCs relative to what they would otherwise receive.
Special payments to other rural providers In addition to special payments to
rural hospitals, Medicare payment policy includes provisions for special
payments to other providers, including rural health clinics and physicians
providing services in Health Professional Shortage Areas (HPSAs). Rural health
clinics. To promote access in rural areas with scarce medical services, P.L.
95-210, passed in 1977, authorized Medicare and Medicaid reimbursement to
nonphysician practitioners providing primary-care services in rural health
clinics. The clinics can be independent, or they can be part of a larger
facility, such as a hospital. Medicare payments are based on an all-inclusive
rate for covered services provided during each visit. These rates are based on
costs up to prospectively set limits. Small rural hospitals with fewer than 50
beds are exempt from these limits. According to a recent report from the General
Accounting Office (GAO), the number of rural health clinics has grown by 30
percent per year since 1989. There were 3,000 clinics in 1996. Physicians in
Health Professional Shortage Areas. A HPSA is an area designated by the
Secretary of Health and Human Services as having a shortage of primary-care
providers. The Omnibus Budget Reconciliation Act of 1989 authorized a 10 percent
bonus payment for services provided in HPSAs and reimbursed under Medicare's
physician fee schedule. According to the GAO, about 46 percent of the $106
million in bonus payments made in 1996 were for services provided in rural
areas. Floor payments for Medicare+Choice plans Until 1997, Medicare paid
private health plans in any county 95 percent of the average per capita cost of
care for fee-for- service beneficiaries in that county, adjusted for the
demographic characteristics of Medicare beneficiaries in that county. The BBA
broke the direct link between fee-for-service spending and payments to private
health plans. Now, payments are the highest of a floor beneath which payments
cannot fall, a 2 percent increase above the prior year's rate, or a blend of
local and national payment rates (but only if a so-called budget neutrality
condition is met). In establishing payment floors, the BBA effectively raised
monthly capitation rates in many counties above local fee-for- service costs of
patient care. The objective of these provisions was to encourage private health
plans to participate in areas (particularly rural areas) where they had not
previously done so. In 2000, 944 counties--about one-third of the total--will
have monthly capitation rates at the floor. Medicare's special payments and
market-based reform How might the activities supported by Medicare's special
payments for medical education, disproportionate share hospitals, rural
providers, and health plans in floor counties fare in an environment that relied
more heavily on market forces? A definitive answer cannot be provided for each
case, but analysis suggests that if the Congress is interested in continuing
support for the these activities, it may need to find new mechanisms for doing
so. In regard to graduate medical education and Medicare's
special payments to teaching hospitals, the answer hinges on the extent to which
beneficiaries observe and value the difference in the services these hospitals
provide. Just as consumers are willing to pay higher prices for goods and
services they perceive to be superior--from automobiles to college
educations--we can reasonably suppose that some Medicare beneficiaries would
choose plans that contracted with teaching hospitals. We observe this today
among the nonaged population and among Medicare+Choice enrollees whose health
plans contract with teaching hospitals. Whether beneficiaries' premiums would
provide the same level of support currently provided through Medicare cannot be
known. However, to the extent that part of Medicare's payments support social
missions beyond patient care, one would expect a decline. With respect to
Medicare's payments to disproportionate share hospitals, it is likely that
support would decline under a market- oriented program. In the past, hospitals
were able to offset at least some of the costs of uncompensated care by charging
more to insured patients. They have been less able to do so as the health care
market has grown increasingly competitive, and private payers have resisted
paying costs for people other than their own enrollees. Making Medicare more
competitive would reinforce this trend. While the likely direction of the impact
is clear, its magnitude is not. Health care markets are complex, and the ability
of providers to pass on the costs of uncompensated care to payers varies from
market to market. The impact that moving to a more market-oriented program might
have on support for providers in rural areas and health plans in floor counties
is less clear and would depend in large measure on what the new program looked
like. Discussions of market-oriented reform often assume that beneficiaries
living in high-cost areas would receive a larger contribution toward their
premium to reflect those costs. On the one hand, policymakers could provide
greater support for beneficiaries living in areas where low volumes meant high
average costs as they might do for beneficiaries living in areas where costs
were high for other reasons (such as high labor costs). On the other hand,
policymakers could choose not to recognize higher costs attributable to low
volumes. In that case, market forces would encourage the expansion of geographic
service areas if beneficiaries chose to incur greater travel costs in exchange
for lower premiums that reflected greater volumes handled by providers. 1.
Medicare's share has been calculated as the fraction of total inpatient days
accounted for by Medicare fee-for-service beneficiaries. Beginning in 1998, a
percentage of inpatient days accounted for by Medicare+Choice enrollees has been
included in the calculation. That percentage will increase gradually until all
days are taken into account in 2002. 2. In 1999, operating payments to a
teaching hospital with a resident-to-bed ratio of 0.6 (typical of an academic
medical center) are increased by about 33 percent. Payments to a teaching
hospital with a resident-to-bed ratio of .083 (typical of teaching hospitals
other than academic medical centers) are increased by about 5 percent.
LOAD-DATE: May 13, 1999 
