Medicare Physician Payments: Spending Targets Encourage Fiscal
Discipline, Modifications Could Stabilize Fees (14-FEB-02,
GAO-02-441T).
Congress implemented a physician fee schedule and a fee update
formula to moderate spending growth relative to specified
Medicare spending targets. These spending targets increase
annually to account for growth in the costs of providing
physician services, the growth in the overall economy, and
changes in the number of Medicare beneficiaries. Physician fees
are adjusted for changes in the costs of providing services and
on actual cumulative spending compares to the cumulative targets.
Physician fees are updated to reflect increased costs of
providing services with the updates adjusted up or down on actual
spending either falling below or exceeding the targets. In
November 2001, the Centers for Medicare and Medicaid announced
that updating Medicare's fees will decline 5.4 percent from what
was paid in 2001, despite an estimated 2.6 percent increase in
the cost of physician inputs. This reduction accounts for
historical cumulative spending that exceeded the target by $8.9
billion, or 13 percent of estimated 2002 spending. Several
factors contributed to the disparity between actual and targeted
spending, including the correction of substantial errors in past
spending estimates and the revision of targets for prior years.
The current update mechanism could be modified to moderate
fluctuations in physician fees and to ensure adequate payments,
while retaining the fiscal discipline created by having a
spending target. Such modifications would need to balance
concerns about preserving fiscal discipline on physician spending
with the need to maintain adequate payment rates to ensure that
beneficiaries have access to physician services. Because the
paramount consideration in setting payment rates is ensuring
appropriate beneficiary access to services, timely and detailed
data on Medicare beneficiary service use are essential to
achieving this balance.
-------------------------Indexing Terms-------------------------
REPORTNUM: GAO-02-441T
ACCNO: A02766
TITLE: Medicare Physician Payments: Spending Targets Encourage
Fiscal Discipline, Modifications Could Stabilize Fees
DATE: 02/14/2002
SUBJECT: Beneficiaries
Health care cost control
Medical fees
Medical services rates
Payments
Physicians
Health insurance
Health insurance cost control
Medicare Economic Index
Medicare Program
Medicaid Fee-For-Service Program
Gross Domestic Product
Sustainable Growth Rate System
******************************************************************
** This file contains an ASCII representation of the text of a **
** GAO Product. **
** **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced. Tables are included, but **
** may not resemble those in the printed version. **
** **
** Please see the PDF (Portable Document Format) file, when **
** available, for a complete electronic file of the printed **
** document's contents. **
** **
******************************************************************
GAO-02-441T
GAO United States General Accounting Office
a
Test i mony Before the Subcommittee on Health, Committee on Energy and
Commerce, House of Representatives
For Release on Delivery Expected at 8: 30 a. m. Thursday, February 14, 2002
MEDICARE PHYSICIAN
PAYMENTS Spending Targets Encourage Fiscal Discipline, Modifications Could
Stabilize Fees
Statement of William J. Scanlon Director, Health Care Issues
GAO- 02- 441T
Page 1 GAO- 02- 441T Mr. Chairman and Members of the Subcommittee: I am
pleased to be here today as you discuss modifications to Medicare?s method
for updating its payments to physicians, which in 2001 totaled
about $41 billion. 1 As you know, more than a decade ago to control rapid
increases in Medicare spending for physician services, the Congress
implemented a physician fee schedule and a fee update formula to moderate
spending growth relative to specified spending targets. These spending
targets increase annually to account for growth in the costs of providing
physician services, the growth in the overall economy, and
changes in the number of Medicare beneficiaries, while physician fees are
adjusted for changes in the costs of providing services and how actual
cumulative spending compares to the cumulative targets. In November
2001, however, the Centers for Medicare and Medicaid Services (CMS)
announced that updating Medicare?s fees for 2002 with this formula will
cause the fees to decline 5.4 percent from what was paid in 2001. 2 The
Congress has been concerned that fluctuations in physician payments and
payment reductions may over the long run jeopardize beneficiary access to
physician services. As a result, it asked the Medicare Payment Advisory
Commission (MedPAC), which advises the Congress on Medicare payment issues,
to study the possibility of eliminating spending targets and
modifying the method for updating physician fees. As you consider
refinements to Medicare?s method of updating physician payments, it is
important to remain mindful of the need to ensure Medicare?s sustainability
for future generations of beneficiaries. In view of the coming surge in the
Medicare- eligible population through the aging of the baby boom generation,
projected program spending threatens to absorb ever- increasing shares of
the nation?s budgetary and economic resources. Furthermore, the slowdown in
Medicare spending growth we saw in recent years appears to have ended. At
the same time, the fiscal discipline imposed on provider payments continues
to be challenged, and
interest in modernizing the Medicare benefit package to include 1 The $41
billion represents total Medicare payments to physicians, including
beneficiary cost sharing. This statement refers to both calendar and fiscal
years. We will use ?fiscal year? where appropriate; other references to
years, except where noted, are to calendar years.
2 Until June 2001, CMS was known as the Health Care Financing Administration
(HCFA). We will continue to refer to HCFA when referring to the
organizational structure and operations associated with that name.
Page 2 GAO- 02- 441T prescription drug coverage and catastrophic protection
has increased. Together, these developments will impede efforts to achieve
the fiscal restraint that the Comptroller General and others have warned is
essential
to the program?s sustainability. In the context of these broader interests,
I will discuss (1) Medicare?s use of spending targets as a means of
moderating the growth in physician service expenditures, (2) the factors
used in computing those targets that resulted in the reduced fees for 2002,
and (3) adjustments to determining and applying spending targets that could
moderate swings in physician fees, while ensuring payments are adequate to
maintain physicians? ability to provide high- quality care to Medicare
beneficiaries. My comments are
based on previous and ongoing work on Medicare spending trends and Medicare
payment methods, including the physician fee schedule. In brief, moderating
Medicare?s spending growth on physician services
while setting payment rates adequate to ensure beneficiary access to care is
not a straightforward matter. Medicare spending on physician services grew
rapidly through the 1980s, at an average annual rate of more than 12
percent, even though physician fees were subject to some limits. The
spending growth was driven by increases in the volume of services provided
to each beneficiary and by increases in the intensity of services provided.
3 Recognizing that expenditure growth of this magnitude was not sustainable,
the Congress attempted to impose fiscal discipline through a
physician fee schedule and a payment update mechanism that incorporates
spending targets. Physician fees are updated to reflect the increased costs
of providing services with the updates adjusted up or down depending on
whether actual spending has fallen below or exceeded the targets. The
targets themselves are adjusted annually to account for changes in the costs
of providing services, the number of Medicare beneficiaries, and the gross
domestic product (GDP). Since the introduction of this fee system in 1992,
annual increases in the volume and intensity of services provided per
beneficiary have moderated significantly. In 2002 the system resulted in
Medicare?s physician fees being reduced 5.4 percent below the fees paid in
2001, despite an estimated 2. 6 percent increase in the cost of physician
inputs. 4 This reduction is to account for historical cumulative spending
that exceeded the target by $8.9 billion, or 13 percent of estimated 2002
3 The intensity of services is the quantity and quality of resources used in
providing them. 4 Inputs for physicians? services are, for example, staff
salaries and overhead.
Page 3 GAO- 02- 441T spending. Several factors contributed to the disparity
between actual and targeted spending, including the correction of
substantial errors in past spending estimates and the revision of targets
for prior years. The current update mechanism could be modified to moderate
fluctuations in physician
fees and to ensure adequate payments, while retaining the fiscal discipline
created by having a spending target. Such modifications would need to
balance concerns about preserving fiscal discipline on physician spending
with the need to maintain adequate payment rates to ensure that
beneficiaries have access to physician services. Because the paramount
consideration in setting payment rates is ensuring appropriate beneficiary
access to services, timely and detailed data on Medicare beneficiary service
use are essential to achieving this balance. Background Total Medicare
spending for physician services depends on actual payment
rates, the volume of services provided, and the mix of those services.
Medicare spending goes up when the price paid to physicians for each service
increases, when the number of services provided rises, or when more
intensive, and therefore more expensive, services replace less intensive
ones.
Since 1992, Medicare has paid for physician services using a fee schedule.
The fee for each service is a dollar conversion factor, adjusted to reflect
the resources required for that service relative to the resources required
to provide all other physician services, and the differences in the costs of
providing services across geographic areas.
Along with the fee schedule, the Congress enacted a system of spending
targets designed to control growth in total spending for physicians?
services. The Sustainable Growth Rate (SGR) system was created in the
Balanced Budget Act of 1997 and revised in the Medicare, Medicaid, and SCHIP
Balanced Budget Refinement Act of 1999 (BBRA). 5 It replaced the first
system of spending targets, implemented in 1992, known as the Volume
Performance Standard. The SGR system sets spending targets for physician
services and adjusts payment rates to bring spending in line with those
targets. The SGR target for total spending is based on spending in an
initial, or base, year and the estimated growth in real per capita GDP each
5 Pub. L. No. 105- 33 sect. 4503, 11 Stat. 251, 433 (to be codified at 42 U. S.
C. sect. 1395w- 4( f)). Pub. L. No. 106- 113, Appendix F, sect. 211, 113 Stat. 1501,
1501A- 345 (to be codified at 42 U. S. C. 1395W4)).
Page 4 GAO- 02- 441T year and three other factors that affect overall
spending on physician services- the changes in the cost of inputs used to
produce physicians? services (as measured by the Medicare Economic Index
(MEI)), the
number of Medicare beneficiaries in the traditional fee- for- service
program, and expenditures that result from changes in laws or regulations.
The spending target for physician payments is applied by incorporating it
into the adjustment to the conversion factor that determines the payment
amount per service. The conversion factor is determined annually by
adjusting the previous year?s conversion factor by the change in the MEI, to
account for the cost of inputs for physician services, and adjusting this
product on the basis of the relationship between the cumulative SGR target
and Medicare physician spending. The conversion factor update is greater
than the MEI when physician spending has been below the targets and is less
than the MEI when physician spending has been higher. Spending Targets
Established to Moderate Rapid Rise in Outlays for Physician Services
In response to escalating Medicare expenditures, the Congress made major
changes in Medicare payment policies, beginning first by enacting the
hospital inpatient prospective payment system, which was implemented in
1983, and then the Medicare physician fee schedule, implemented in 1992.
When enacting the fee schedule, the Congress recognized that setting fees
alone would not sufficiently restrain physician spending growth. Despite
some constraints on physician fees since the 1970s, spending on physician
services had grown dramatically in the 1980s as a result of increases in the
volume and intensity of services provided. The Congress, therefore, provided
that annual physician fee increases would depend upon whether total Medicare
physician spending exceeded or fell short of cumulative
spending targets. Since the implementation of the fee schedule and spending
targets, the rise in Medicare spending for physician services has slowed
significantly, reflecting lower growth in the volume and intensity of these
services.
Spending on Physician Services Grew Rapidly Before 1992
Before the physician fee schedule was implemented, Medicare payments for
physicians? services were largely based on historical charges. Although
during the 1970s the Congress introduced some controls on annual payment
rate increases, Medicare spending for physician services continued to rise.
This was also true in the 1980s- between 1980 and 1990, for example,
Medicare spending per beneficiary for physician services
Page 5 GAO- 02- 441T grew at an average annual rate of more than 12 percent,
tripling from $278 to $890 (see fig. 1). Figure 1: Medicare Spending for
Physicians? Services, per Beneficiary, 1975- 1990
Note: Amounts represent Medicare spending, net of beneficiary cost sharing,
for the year ending June 30. Sources: Board of Trustees, Federal
Supplementary Medical Insurance Trust Fund, 1998 Annual
Report of the Board of Trustees of the Federal Supplementary Medical
Insurance Trust Fund (Washington, D. C.: Apr. 28, 1998), pp. 51- 2; and
HCFA, A Profile of Medicare: Chartbook 1998 (Washington, D. C.: 1998), p.
64.
Much of the spending growth resulted from increases in the volume of
services provided to each beneficiary and the substitution of more intensive
and expensive services for less intensive and expensive ones. The Physician
Payment Review Commission, which was charged with advising the Congress on
Medicare physician payment issues, observed,
?[ b] y the late 1980s. . . volume and intensity growth had become the
primary cause of higher program spending. In fact, from 1986 until 1992,
while physician payment rates grew by less than 2 percent annually, the
volume and intensity of services rose by almost 8 percent per year.? 6 6
Physician Payment Review Commission, 1995 Annual Report to Congress
(Washington, D. C.: Physician Payment Review Commission, 1995).
$135.45 $278.30
$548.36 $889.57
0 100
200 300
400 500
600 700
800 900
1,000 1975 1980 1985 1990 Medicare spending in dollars
Page 6 GAO- 02- 441T The Congressional Budget Office in 1986 stated that ?[
b] oth the price and the volume of services must be controlled to constrain
costs . . ..? 7 Spending targets were needed to limit growth in volume and
intensity of physician services. In 1989 testimony, Health and Human
Services Secretary Louis W. Sullivan said ?Medicare physician spending has
increased at compound annual rates of 16 percent over the past 10 years. And
in spite of our best efforts to control volume and reign in expenditures,
Medicare physician spending is currently out of control. . . An expenditure
target. . . sets an acceptable level of growth in the volume and intensity
of physician services.? 8 Spending Targets Created
Incentives to Moderate Growth in Volume and Intensity of Services The
Congress introduced spending targets for physician services in conjunction
with the physician fee schedule in 1992 to help constrain the
rise in Medicare spending for physician services. The targets incorporated
limited growth in the volume and intensity of services and were revised each
year based on estimates of changes in the number of Medicare beneficiaries
and physician input prices. If actual spending exceeded the targeted
amounts, future payment rates would be reduced, relative to what they would
have been if actual spending had equaled the targets, to offset
the excess spending. If actual spending fell short of the targets, future
payment rates would be increased.
Since 1992, the growth in the volume and intensity of physicians? services
per Medicare beneficiary has moderated (see fig. 2). Between 1992 and 2000,
the average annual increase in Medicare spending due to changes in volume
and intensity of services per beneficiary was about 2 percent. In contrast,
between 1985 and 1992, immediately before the introduction of spending
targets, volume and intensity of services per beneficiary
increased at an average annual rate of about 9 percent. 7 Congressional
Budget Office, Physician Reimbursement Under Medicare: Options for Change
(Washington, D. C.: Apr. 1986). 8 Testimony before the Subcommittee on
Medicare and Long- term Care, Committee on Finance, U. S. Senate, 101st
Congress, 1st Session (June 16, 1989).
Page 7 GAO- 02- 441T Figure 2: Changes in Volume and Intensity of Medicare
Physician Services, per Beneficiary, 1975- 2000
Notes: Data are for beneficiaries in the traditional fee- for- service
program only. From 1975 through 1995, volume and intensity of services
changes are based on Medicare outlays for all physician services. From 1996
through 2000, volume and intensity of services changes are based on Medicare
outlays for physician services covered by the fee schedule.
Sources: 1998 Annual Report of the Board of Trustees of the Federal
Supplementary Medical Insurance Trust Fund, pp. 51- 2; A Profile of
Medicare: Chartbook 1998, p. 64; and 2001 Annual Report of the Board of
Trustees of the Federal Supplementary Medical Insurance Trust Fund,
http:// www. hcfa. gov/ pubforms/ tr/ smi2001/ tabiig2.htm accessed Feb. 9,
2002.
Several Factors Associated With 2002 Fee Reductions
The application of the SGR system in 2002 resulted in a 5.4 percent
reduction in physician payment rates, despite an estimated 2.6 percent
increase in the costs of inputs used to provide physician services. The
reduction occurred because estimated cumulative physician services spending
since 1996 exceeded the target for cumulative spending by
approximately $8. 9 billion, or 13 percent of projected 2002 spending. In
part, the payment update reflects adjustments made to the spending targets
for previous years for revisions in GDP estimates and for more accurate
actual spending statistics. Correcting these errors in previous years?
targets and spending totals to reflect more recent data resulted in larger
physician payment increases in those years than if accurate data had been
used, and they contributed to the size of the reduction in payments in 2002.
The SGR system sets spending targets for physician services and adjusts
payment rates to bring spending in line with those targets. Conceptually, if
spending equals the targeted amount, physician payment rates are updated
to keep pace with the percentage change in input prices as measured by the
4.5 8.3
3.7 7.6
9.7 3.9
6.5 9.0 9.4
0.2 0.8
4.1 3.6
2.8 2.5 3.0
-0.2 0.6
-1 0
1 2
3 4
5 6
7 8
9 10
1975 1980 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997
1998 1999 2000 Percentage change
Page 8 GAO- 02- 441T MEI. If spending exceeds the target, the change in
payment rates is smaller than the change in input prices. If spending falls
short of the target,
payment rates are allowed to grow faster than the rise in input prices. By
adjusting payment rates when prior- year spending has been too high, the SGR
system moderates the growth in Medicare outlays for physician
services. The SGR adjustments to the input price update are determined by
how much the cumulative physician spending since 1996 differs from the
cumulative spending target since then. Spending and targets must both be
estimated from information available each November when payment rates are
set for the following year. Previously, those estimates were then used in
subsequent years. Based on requirements in the BBRA, however, HCFA
implemented a process for revising the most recent two years of spending and
target estimates. Because the annual targets are determined by changes in
four factors- the number of fee- for- service beneficiaries, real per capita
GDP, input costs, and the effect of changes in laws or
regulations- a revision to any of those factors, or to estimates of prior
spending, can change the spending estimate. The SGR adjustments to the input
price update can then take effect because of growth in the volume or
intensity of services delivered, resulting in spending deviating from
targets, or because of revised estimates for prior years? targets and
spending.
In setting payment rates for 2002, CMS updated its estimates of past
spending and recalculated past targets. The net effect of both revisions
indicated that Medicare outlays were an estimated $8.9 billion too high. The
SGR is designed to recover this excess amount by lowering payment rates in
2002 and future years. CMS? original estimates of spending since
1998 were too low, in part because the agency had not included all
appropriate claims in making these estimates. The original spending targets
for 2000 and 2001 were too high, largely because the Bureau of Economic
Analysis in the Department of Commerce revised its GDP and GDP growth
estimates for those years. To some extent, the reduction in payment rates
this year corrects for inaccuracies in previous estimates that produced
physician fees that were too high in 2000 and 2001. In both years, payment
rates increased by more than the change in input prices because the
information available at the time those rates were established suggested
that physician spending had
been held below the targets. In 2000, payment rates increased 5.4 percent,
while input costs increased 2.4 percent; in 2001, payment rates increased
4.5 percent, while input costs increased by 2.1 percent. The reason that
Page 9 GAO- 02- 441T 2002 payment rates fall 5.4 percent while input prices
increase by 2.6 percent is that the revised estimates revealed that spending
exceeded the targets in previous years. The reduction would have been almost
4
percentage points greater, but the SGR system limits how much fees can be
adjusted for the differences between actual and target spending. Avoiding
Large Payment Swings in System Based on Spending Targets
Several measures could moderate the fluctuations in physician payment rates.
Although these modifications to the SGR could mitigate the possible impact
of rate instability or reductions in beneficiary access to needed services,
doing so could also lessen the ability of spending targets to encourage
fiscal discipline. Available data indicate that access is adequate, but more
timely and detailed information is critical to promptly recognize potential
deteriorations in access.
Moderating Fluctuations in Physician Payment Rates The SGR system is
designed to limit the fluctuations in payment rates, but
its design could be modified to achieve greater rate stability. The BBRA
specified that adjustment to realign spending with the targets cannot cause
payments to fall by more than 7 percentage points below, or increase by more
than 3 percentage points above, the percentage change in input prices. In
addition, spending deviations from past targets are not corrected in a
single year but are spread over several years. Greater rate stability could
be achieved by narrowing the range over which rates could change
from one year to the next. Similarly, the corrections for spending
deviations could be spread over longer periods of time.
Modifying how spending targets are set could also reduce year- to- year
fluctuations in rates. Currently, the changes in GDP for a single year are
used to establish the spending target. The difficulty in accurately
estimating GDP has contributed to the problem of fluctuations in the target.
In addition, linking annual changes in the targets to annual changes in GDP
ties the target to the business cycle. GDP growth rates are higher during
periods of prosperity and lower during downturns- a commonly
used definition of a recession is a decline in real GDP for two successive
quarters. But health care needs of Medicare beneficiaries are not cyclical.
Neither significantly lowering spending targets during a downturn nor
unduly increasing them in a period of prosperity is appropriate. Linking the
determination of spending targets to average levels of GDP over several
years would help eliminate much of the cyclical variation.
Page 10 GAO- 02- 441T Any changes to the SGR must balance the desire for
greater rate stability with the need for fiscal discipline. Spending targets
create a feedback mechanism between physicians? behavior and payment rate
increases.
However, spending targets do not create direct incentives for any individual
physician, since it is the collective behavior of all physicians that
determines the payment adjustments that result from missing the spending
targets. The primary value of spending targets in instilling fiscal
discipline
is the signal they send that affordability of the program is an important
concern in establishing Medicare policies. Limiting the size of the payment
adjustment for missed spending targets or to corrections in prior years?
targets, and lengthening the time over which those adjustments are
incorporated, could partially mute the signal targets send and erode some of
the fiscal discipline they encourage. On the other hand, excessive rate
fluctuations can be difficult for providers and may ultimately affect
beneficiaries? access to physician services if rate fluctuations cause too
many providers to decline to participate in the program. Monitoring
Beneficiary Access to Physicians
Ensuring that the use of spending targets does not compromise appropriate
access to services is a key concern. Spending targets that are updated
principally by the growth in GDP and other factors may not reflect fully
changes in medical care and the markets for these services. It is therefore
important to monitor service use to assess whether appropriate access for
beneficiaries is secured, especially if fees are reduced. Such monitoring
needs to involve recent experience so that if spending targets need
adjustments, those adjustments are done promptly to ameliorate any problems.
Information on physicians? willingness to see Medicare patients is dated but
overall does not indicate access problems. Data from the 1990s show that
virtually all physicians treated Medicare beneficiaries or if accepting new
patients, accepted those covered by Medicare. According to data from
the American Medical Association (AMA), 96.2 percent of all nonfederal
physicians (excluding residents and pediatricians who do not normally serve
Medicare patients) treated Medicare beneficiaries in 1996, an
increase from the 94. 2 percent AMA reported in 1994. A 1999 survey
sponsored by MedPAC found that 93 percent of physicians who were accepting
any new patients were accepting new patients covered by
Medicare. Payment rate decisions should not be made in a data vacuum. As
health needs change, technology improves, or health care markets evolve,
Page 11 GAO- 02- 441T spending targets and resulting payment rates may need
to be adjusted periodically, not by a formula designed for annual updates,
but by specifying a new base year target calibrated to ensure appropriate
access.
Payment rates that are too low can impair beneficiary access to physician
services, while payment rates that are too high add unnecessary financial
burdens to Medicare. Informed decisions about appropriate payment rates
and rate changes cannot be made unless policymakers have detailed and recent
data on beneficiaries? access to needed services.
Concluding Observations The SGR mechanism uses information about physician
spending in relation
to cost increases, changes in the number of beneficiaries, and growth in the
overall economy to impose fiscal discipline on Medicare outlays for these
services. This mechanism provides a signal when spending threatens to grow
out of control and in that sense is analogous to what the Comptroller
General has called for in testimony on several occasions with regard to the
entire Medicare program. The demographic changes facing the nation require
policymakers to look ahead and assess both current and future Medicare
spending in relation to the entire federal budget and the
economy- first to understand the urgent need for fiscal discipline, then to
make choices to ensure the sustainability and affordability of the program.
A mechanism like the SGR provides a benchmark for assessing the trend in
physician spending and can prompt actions to bring that spending in line
with overall program goals. In assessing the options for updating physician
payments, the program?s prospects for long- term sustainability should be
paramount. Meeting that challenge will involve difficult decisions that will
likely affect beneficiaries, providers, and taxpayers.
This concludes my prepared statement. I would be happy to answer any
questions that you or Members of the Subcommittee may have.
GAO Contacts and Staff Acknowledgments
For more information regarding this testimony, please contact me at (202)
512- 7114 or Laura A. Dummit at (202) 512- 7119. James Cosgrove, Kathryn
Linehan, Lynn Nonnemaker, and Hannah Fein also made key contributions to
this statement.
(290172)
United States General Accounting Office Washington, D. C. 20548- 0001
Official Business Penalty for Private Use $300
Address Correction Requested Presorted Standard
Postage & Fees Paid GAO Permit No. GI00
*** End of document. ***