Medicare Reform: Modernization Requires Comprehensive Program
View (14-JUN-01, GAO-01-862T).
The Medicare Program faces many challenges. Clearly, the
overarching issue is how to ensure that Medicare remains
sustainable for future generations of beneficiaries. Meeting that
challenge will involve difficult decisions that will likely
affect beneficiaries, providers, and taxpayers. However, the
financing issue should not obscure other important Medicare
challenges. Medicare's current cost-sharing arrangements do not
encourage the efficient use of services without discouraging
necessary care. Moreover, the lack of catastrophic coverage can
leave some beneficiaries liable for substantial Medicare
expenses. Finally, some aspects of Medicare's program management
are inefficient and lag behind modern private sector practices.
Changes in Medicare's program management could improve both the
delivery of health care to the beneficiaries and the program's
ability to pay providers appropriately. Some view restructuring
of the relationship between parts A and B as an important element
of overall Medicare reform. Fundamentally, assessing the program
as a whole is an important first step in addressing Medicare's
challenges. Solutions to many of these challenges could be
crafted without restructuring. However, restructuring may provide
opportunities to implement desired reforms--with or without
unifying the Hospital Insurance and Supplemental Medical
Insurance trust funds--while undoubtedly raising issues that will
have to be considered.
-------------------------Indexing Terms-------------------------
REPORTNUM: GAO-01-862T
ACCNO: A01194
TITLE: Medicare Reform: Modernization Requires Comprehensive
Program View
DATE: 06/14/2001
SUBJECT: Health insurance cost control
Program management
Medicare Program
Medicare Hospital Insurance Program
Hospital Insurance Trust Fund
Medicare Supplemental Medical Insurance
Program
Medicare Choice Program
Supplemental Medical Insurance Trust
Fund
HCFA Common Working File
******************************************************************
** This file contains an ASCII representation of the text of a **
** GAO Testimony. **
** **
** 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-01-862T
Testimony Before the Subcommittee on Health, Committee on Energy and
Commerce, House of Representatives
United States General Accounting Office
GAO For Release on Delivery Expected at 10: 00 a. m. Thursday, June 14, 2001
MEDICARE REFORM
Modernization Requires Comprehensive Program View
Statement of William J. Scanlon Director, Health Care Issues
GAO- 01- 862T
Page 1 GAO- 01- 862T
Mr. Chairman and Members of the Subcommittee: I am pleased to be here today
as you continue to consider how the Medicare program might be modified to
better serve beneficiaries, providers, and taxpayers. Discussions about how
to reform and modernize Medicare have, in part, focused on whether the
structure that was adopted in 1965 remains optimal today. In that context,
questions have been raised about the desirability of maintaining Medicare?s
division into two distinct parts, part A for hospital and other
institutional care and part B for physician, outpatient, and other
noninstitutional services. This bifurcated structure is no longer common
among private insurance, as it was in the 1960s when insurers marketed
separate policies for different services.
Problems with financing, beneficiary cost- sharing, and program management
have been linked with the fragmented structure of the program. Yet merging
parts A and B may not be the only way to make progress in addressing these
problems. To assist the Subcommittee as it considers restructuring Medicare,
my remarks today focus on how reforms based on a more unified view of the
program might affect (1) program financing and assessment of the program?s
financial health, (2) cost sharing requirements, and (3) program management,
including administration and promotion of quality care. These observations
are based on previous and ongoing GAO work on Medicare and private sector
insurance, as well as other published research.
In summary, rethinking the relationship between parts A and B may encourage
use of a more comprehensive measure of Medicare?s financial health. The
commonly used measure, part A trust fund solvency, does not include the
growing share of program spending on part B services. While a more complete
picture of Medicare?s financial health can be obtained in a number of ways,
the desire for a better picture of the program?s financial prognosis is one
argument for a single trust fund. Establishing a single trust fund would
require agreement on how funds from payroll taxes, general revenues, and
beneficiary premiums would flow to the program. It would require consensus
on what measure would be used to track program finances and spur action to
increase revenue or curb spending when needed. It also would require
assessment of whether different beneficiary eligibility standards, similar
to those currently specified for parts A and B, would be maintained.
Rethinking the relationship between parts A and B also could facilitate
development of better cost- sharing requirements. The current cost- sharing
structure fails to promote prudent use of services and protect beneficiaries
Page 2 GAO- 01- 862T
from high out- of- pocket costs. These concerns could be addressed under the
current part A and B structure or a more unified structure. Unifying the
program completely would require some beneficiaries who now have other
coverage and are enrolled in only one part of the program to pay additional
premiums for coverage they already have. It also would increase costs to the
government for care that is now covered privately. Alternatively, partial
benefits could be extended to those who chose not to fully participate in a
unified program.
Rethinking the relationship between parts A and B would not fundamentally
address challenges the Health Care Financing Administration (HCFA) faces in
efficiently managing the disparate services Medicare covers. HCFA?s outdated
information technology (IT) systems have hindered its ability to develop
data to improve payment methods and the quality of care beneficiaries
receive. Further, as a large public program, Medicare is limited in its
ability to incorporate innovations that private insurers have used to
influence care delivery. These include targeted beneficiary education,
preferred provider networks, and coordination of services. The National
Academy of Social Insurance (NASI) has reviewed these private sector
practices and concluded that they could potentially improve Medicare.
However, they would need to be tested to determine their impacts and
evaluated to ascertain how well they might be adapted to reflect the
uniqueness of Medicare as both a public program and the largest single
purchaser of health care. Full implementation of many of these innovations
would require statutory changes to the program.
At its inception, Medicare?s design mimicked the structure of existing
private insurance, which commonly included different policies for different
sets of services. It also was designed, like private insurance at the time,
as a passive bill payer that did not try to influence how care was
delivered. In fact, because of concerns about the potential influence of
such a large government program, the original Medicare statute requires that
Medicare not influence providers? practice of medicine and gives
beneficiaries access to all participating providers.
Medicare is administered by the HCFA, and pays for some $200 billion in
health care benefits each year for about 40 million elderly and disabled
Americans. Individuals who are eligible for Medicare automatically receive
Hospital Insurance (HI), known as part A, which covers inpatient hospital,
skilled nursing facilities (SNF), certain home health, and hospice care.
Beneficiaries generally pay no premium for this coverage, having Background
Page 3 GAO- 01- 862T
previously contributed payroll taxes from covered employment, but they are
liable for required deductibles, coinsurance, and copayment amounts. (See
tables 1 and 2.)
Medicare- eligible beneficiaries may elect to purchase Supplementary Medical
Insurance (SMI), known as part B, which covers physician, outpatient
hospital, laboratory, and other services. Beneficiaries must pay a premium
for part B coverage, currently $50 per month, and are also responsible for
part B deductibles, coinsurance, and copayments.
Most of Medicare?s 40 million beneficiaries are enrolled in both parts A and
B. However, approximately 2 million are enrolled only in part A. Another
400, 000 are enrolled only in part B. Those enrolled in only one part of the
program often have private insurance from an employer or other source to
make up the difference.
Approximately 14 percent of Medicare beneficiaries enroll in Medicare+
Choice plans. These plans include health maintenance organizations and other
private insurers who are paid a set amount each month to provide all
Medicare- covered services. Beneficiaries must be enrolled in both parts A
and B to join these plans, which typically offer lower cost- sharing
requirements and additional benefits compared to Medicare?s traditional fee-
for- service program, in exchange for a restricted choice of providers.
Page 4 GAO- 01- 862T
Table 1: Medicare Part A and Part B Coverage, Eligibility, and Funding Part
A Part B Coverage -Inpatient hospital.
-Skilled nursing facility (SNF). -Home health. a -Hospice.
-Physician services. -Laboratory services. -Outpatient hospital. -Home
health. a -Durable medical equipment
Eligibility -Individuals and their spouses over 65 who paid the Medicare
payroll tax for 10 years (40 quarters). -Individuals over 65 who paid the
Medicare payroll tax for 30 to 39 quarters and who pay a $165 monthly
premium. -Individuals over 65 who paid the payroll tax for less than 30
quarters and who pay a $300 monthly premium. -Individuals eligible for
Social Security disability benefits. -Individuals with end- stage renal
disease.
-Individuals over age 65, disabled, or with end- stage renal disease who pay
a monthly premium ($ 50 in 2001).
Funding Medicare payroll taxes. Premiums cover 25 percent and general tax
revenue covers 75 percent. a Part A covers up to 100 home health visits
following an inpatient hospital or SNF stay. Part B covers
other home health visits. Source: Medicare & You 2001, HCFA.
Page 5 GAO- 01- 862T
Table 2: Medicare Beneficiary Cost- Sharing for 2001 Part A services:
Copayments, coinsurance, and deductibles:
Inpatient hospital $792 deductible per admission a $198 copayment per day
for days 61 through 90 $396 copayment per day for days 91 through 150 b All
costs beyond 150 days Skilled nursing facility (SNF) No cost- sharing for
first 20 days
$99 per day copayment for days 21 through 100 All costs beyond 100 days Home
health No cost- sharing
20 percent coinsurance for durable medical equipment Hospice $5 copayment
for outpatient drugs
5 percent coinsurance for inpatient respite care
Part B services c : Physician and medical $100 deductible each year
20 percent coinsurance for most services 50 percent coinsurance for mental
health services Clinical laboratory No cost- sharing Home health No cost-
sharing
20 percent coinsurance for durable medical equipment Outpatient hospital
Coinsurance varies by service and may exceed 50
percent a No deductible is charged for second and subsequent hospital
admissions if they occur within 60 days
of the beneficiary?s most recent covered inpatient stay. b After the first
90 days of inpatient care, Medicare may help pay for an additional 60 days
of inpatient care (days 91 through 150). Each beneficiary is entitled to a
lifetime reserve of 60 days of inpatient coverage. Each reserve day may be
used only once in a beneficiary?s lifetime. c No cost- sharing is required
for certain preventive services- including specific screening tests for
colon, cervical, and prostate cancer and flu and pneumonia vaccines. Source:
Medicare & You 2001, HCFA.
Medicare pays for services out of two separate trust funds. Part A services
are paid for out of the HI Trust Fund. It is primarily financed through the
Medicare payroll tax that is exclusively dedicated to this trust fund. Part
B services are paid for out of the SMI Trust Fund. This trust fund is
financed in part through the part B premium, which is adjusted each year to
equal 25 percent of expected part B spending. The remaining 75 percent is
paid for out of general tax revenues.
Page 6 GAO- 01- 862T
Medicare?s two parts have distinct financing and participation arrangements.
Modifying these arrangements could promote the use of a more comprehensive
measure of Medicare?s financial health and help policymakers anticipate
future fiscal imbalances. In addition to selecting such a measure or
measures, Congress could also decide to establish thresholds that would
trigger corrective actions designed to rebalance Medicare revenues and
spending. Unification of the now separate HI and SMI trust funds would
require consideration of these issues, but even without such a merger,
comprehensive financial measures and associated triggers would be useful.
Unification would also require Congress to determine how the current mix of
payroll taxes, beneficiary premiums, and general revenues might be modified
to fund the program, as well as whether beneficiaries would be obligated to
participate in the full program or could obtain coverage for subsets of
services.
In the past, Medicare?s financial status has been generally gauged by the
projected solvency of the HI trust fund. Looked at this way- and based on
the latest annual report from the Medicare Trustees- Medicare is viewed as
solvent through 2029. Solvency is a popular measure, in part because the
consequences of insolvency are clear. If there is no money in the HI trust
fund, the government cannot pay hospitals or other providers of part A
services. Thus, the threat of insolvency can be a powerful driver for
action. In 1997, the Medicare Trustees estimated that the HI trust fund
would become insolvent in 2001. The HI trust fund had not been so close to a
crisis since 1972. Following the Trustees? 1997 report, Congress enacted the
Balanced Budget Act of 1997, which contained substantial payment and other
reforms designed to slow Medicare?s cost growth. These reforms, coupled with
a strong economy, helped to increase the life expectancy of the HI trust
fund.
However, HI trust fund solvency is an incomplete measure of Medicare?s
fiscal health. It does not reflect the cost of the part B component of
Medicare, which covers outpatient services and is financed through general
revenues and beneficiary premiums. Part B accounts for more than 40 percent
of current Medicare spending and is expected to account for a growing share
of future total program dollars. The concept of solvency does not apply to
the trust fund for part B, SMI, because increases in expenditures are
automatically matched with increases in general revenues and beneficiary
premiums.
In addition, HI trust fund solvency does not mean that Medicare?s part A
component is financially healthy. Although the trust fund is expected to
Restructuring Raises
Financing and Beneficiary Participation Issues
Focus on HI Trust Fund Provides Misleading View of Medicare?s Financial
Health
Page 7 GAO- 01- 862T
remain solvent until 2029, HI outlays are projected to exceed HI revenues
beginning in 2016. As the baby boom generation retires and the
Medicareeligible population swells, the imbalance between outlays and
revenues will increase dramatically. Thus, in 15 years the HI trust fund
will begin to experience a growing annual cash deficit. At that point, the
HI program must redeem Treasury securities acquired during years of cash
surplus. The government will then need to increase taxes, increase borrowing
(or retire less debt), impose spending cuts, or implement some combination
of these actions.
When part A expenditures outstrip payroll tax revenues, it may be tempting
to reallocate some expenditures from part A to part B. This would extend the
solvency of the HI trust fund, but would do little to improve Medicare?s
overall financial health. For example, BBA reallocated a portion of home
health spending from part A to part B. Although that action- phased in over
time- reduces HI expenditures and extends that trust fund?s solvency, it
also increases SMI expenditures. Consequently, the home health reallocation
increases the proportion of Medicare funded by general revenues and
beneficiary premiums.
Clearly, it is total program spending- both part A and part B- which
determines whether Medicare is sustainable over the long haul. Whether the
program remains in its current configuration, or the relationship between
parts A and B are restructured, a more comprehensive measure of Medicare?s
financial health could help Congress anticipate future fiscal imbalances. A
variety of such measures exist now. For example, the Medicare Trustees
report total Medicare spending as a share of gross domestic product (GDP).
This measure clearly shows that total Medicare expenditures will likely
consume an increasingly larger share of the national economy. Currently,
combined HI and SMI expenditures account for 2.3 percent of GDP. This
percentage is expected to rise to 4.5 percent in 2030 and 8.5 percent in
2075. Another comprehensive indicator measures Medicare spending relative to
the entire federal budget. We estimate that Medicare?s share of the federal
budget will increase from 10 percent in 2000 to over 23 percent in 2030 if
the program?s spending growth continues unchecked. 1
1 Medicare: Higher Expected Spending and Call for New Benefit Underscore
Need for Meaningful Reform (GAO- 01- 539T, Mar. 22, 2001). Comprehensive
Measures
Could Better Indicate Program Sustainability
Page 8 GAO- 01- 862T
The adoption of new financial health indicators for Medicare would be one
step; the next would be to decide what should trigger congressional action.
Congress could agree that it would take action to rebalance Medicare
spending and revenues whenever a comprehensive measure reached a
predetermined level. Possible actions could include increasing general
revenue contributions, payroll taxes, or beneficiary premiums; reducing
benefits; cutting provider payments; or introducing efficiencies to moderate
spending. The 1999 Breaux- Frist Medicare reform proposal provides one
example of a potential trigger. Under that proposal, the two trust funds
would be unified and congressional action would be required in any year when
general revenue contributions exceeded 40 percent of total Medicare
expenditures.
The need for measures of program sustainability and thresholds that would
trigger congressional action would be most acute if the trust funds are
unified. Such a reconfiguration could remove the powerful signal of the HI
trust fund insolvency and reduce the apparent urgency of corrective actions.
If the trust funds remain separate, comprehensive measures of Medicare?s
financial health and associated triggers could avoid the shortcomings that
arise from a focus on the HI trust fund?s solvency.
Improved measures of Medicare sustainability and agreed- upon thresholds
will not, however, alter the difficult decisions facing this and future
Congresses. A growing Medicare population and advances in expensive medical
technology will increase future demands for health care spending.
Policymakers will need to find ways either to control Medicare?s spending
growth or obtain additional revenues to pay for it. Any solution to address
the financial imbalance will affect beneficiaries, taxpayers, providers, or
some combination of the three groups. Better measures of Medicare?s
financial health may help identify the need for action, but will not lessen
the difficulty of implementing a solution.
Creating a unified trust fund for Medicare parts A and B would raise several
new issues Congress would need to address. One is program financing-
Congress would have to specify Medicare?s revenue sources and the share that
each source would contribute. Under the current arrangement, revenues come
from the Medicare payroll tax, general revenues, and beneficiary premiums.
Broadly speaking, the amount financed from each revenue source depends upon
the amount spent on Medicare services and the classification of services
into parts A and B. The payroll tax supports part A services. The amount of
general revenues Fiscal Measures Could
Trigger Congressional Action
Unification of Trust Funds Raises Questions About Financing, Premiums, and
Participation
Page 9 GAO- 01- 862T
devoted to Medicare is set equal to 75 percent of part B expenditures.
Beneficiary premiums are collected to pay for the remaining 25 percent of
part B spending. If the trust funds were unified, Congress would have to
specify the funding mechanism. It could, for example, determine the share
that general tax revenues, payroll tax revenues, and beneficiary premiums
would each contribute to total Medicare spending. Alternatively, it could
adopt an allocation formula similar to the present one by designating some
services to be supported by the payroll tax and others to be supported by
general revenues and beneficiary premiums.
Beneficiary participation issues would also arise under a restructured
program with a unified trust fund. Currently, about 2 million individuals (5
percent of beneficiaries) are eligible for Medicare part B but do not
participate in the voluntary program. A smaller number of individuals do not
qualify for coverage under part A, although provisions allow certain
individuals to buy into the program by paying a monthly premium. Under a
restructured program, Congress would need to determine beneficiary
participation and premium options. For example, should participation in the
full program and payment of any associated premium be mandatory? If full
participation is mandated, program costs could increase and some
beneficiaries would receive Medicare coverage for services covered by
existing private policies. If full participation is voluntary, what coverage
should be provided to those individuals who choose less than full
participation? Would individuals who had made payroll tax contributions but
decline to pay the premium not receive coverage? Or would reduced benefits-
for example, coverage only for current part A services- be available for
such individuals?
Rethinking the relationship between parts A and B could facilitate
rationalization of cost- sharing requirements and help make Medicare more
like private sector and Medicare+ Choice plans. Medicare?s benefit design
has changed little since its inception 35 years ago, and in many ways has
not kept pace with changing health care needs and private sector insurance
practices. Medicare?s current cost- sharing requirements in particular are
not well structured to promote prudent use of discretionary services. At the
same time, they can create financial barriers to care and Beneficiary
CostSharing
Could Be Improved
Page 10 GAO- 01- 862T
leave beneficiaries with extensive health care needs liable for high out-
ofpocket costs. 2
Health insurers today commonly design cost- sharing requirements- in the
form of deductibles, coinsurance, and copayments- to ensure that
beneficiaries are aware there is a cost associated with the provision of
services and to encourage them to use services prudently. Ideally,
costsharing should encourage beneficiaries to evaluate the need for
discretionary care but not discourage necessary care. Optimally, costsharing
would generally require coinsurance or copayments for services that may be
discretionary and could potentially be overused, and would also aim to steer
patients to lower cost or better treatment options. Care must be taken,
however, to avoid setting cost- sharing amounts so high as to create
financial barriers to necessary care.
The benefit packages of most Medicare+ Choice plans illustrate costsharing
arrangements that have been designed to reinforce cost containment and
treatment goals. Most Medicare+ Choice plans charge a small copayment for
physician visits ($ 10 or less) and emergency room services (less than $50).
Relatively few Medicare+ Choice plans charge copayments for hospital
admissions. Plans that offer prescription drug benefits typically design
cost- sharing provisions that encourage beneficiaries to use cheaper generic
drugs or brand name drugs for which the plan has negotiated a discount.
Medicare fee- for- service cost- sharing rules diverge from these common
insurance industry practices in important ways. For example, as indicated in
table 2, Medicare imposes a relatively high deductible of $792 for hospital
admissions, which are rarely optional. In contrast, Medicare requires no
cost- sharing for home health care services, even though historically high
utilization growth and wide geographic disparities in the use of such
services have raised concerns about the potentially discretionary nature of
some services. 3 Medicare also has not increased the part B deductible since
1991. For the last 10 years the deductible has
2 Medicare: Cost Sharing Policies Problematic for Beneficiaries and Program
(GAO- 01- 713T, May 9, 2001). 3 See Medicare Home Health Care: Prospective
Payment System Will Need Refinement as Data Become Available (GAO/ HEHS- 00-
9, Apr. 7, 2000). Cost- Sharing Requirements
Are Not Well Structured
Page 11 GAO- 01- 862T
remained constant at $100 and has thus steadily decreased as a proportion of
beneficiaries? real income.
Also unlike most employer- sponsored health plans for active workers,
Medicare does not limit beneficiaries? cost- sharing liability.
Employersponsored plans typically limit maximum annual out- of- pocket costs
for covered services to less than $2,000 per year for single coverage. 4 In
Medicare, however, current estimates suggest that the combination of cost-
sharing requirements on covered services and the cost of services not
covered by Medicare leaves beneficiaries liable for about 45 percent of
their health care costs. The average beneficiary is estimated to have
incurred about $3,100 in out- of- pocket expenses for health care in 2000-
an amount equal to about 22 percent of the average beneficiary?s income. 5
Some beneficiaries face much greater financial burdens. For example,
lowincome single women over age 85 in poor health and not covered by
Medicaid are estimated to have spent more than half (about 52 percent) of
their incomes on health care services. 6
The average beneficiary who obtained services had a total liability for
Medicare- covered services of $1,451, consisting of $925 in Medicare
copayments and deductibles in addition to the $526 in annual part B premiums
in 1997, the most recent year for which data are available on the
distribution of these costs. The burden of Medicare cost- sharing can,
again, be much higher for beneficiaries with extensive health care needs. In
1997 slightly more than 3.4 million beneficiaries (11.4 percent of
beneficiaries who obtained services) were liable for more than $2,000.
Approximately 750,000 of these beneficiaries (2.5 percent) were liable for
more than $5,000, and about 173,000 beneficiaries (0. 6 percent) were liable
for more than $10, 000.
Different approaches could be taken to address concerns about current cost-
sharing requirements. Cost- sharing for less discretionary services could be
reduced or eliminated. Catastrophic protection could be added to
4 The Kaiser Family Foundation and Health Research and Educational Trust,
Employer Health Benefits: 2000 Annual Survey. 5 Stephanie Maxwell, Marilyn
Moon, and Mesha Segal, Growth in Medicare and Out- Of- Pocket Spending:
Impact on Vulnerable Beneficiaries (Urban Institute, Dec. 2000). 6 Maxwell,
Moon, and Segal. Beneficiary Liability Is
Unlimited Options for Addressing Cost- Sharing Concerns
Page 12 GAO- 01- 862T
the benefits package. In addition, the part B deductible could be raised, or
the part A and B deductibles could be combined.
Reducing or eliminating cost- sharing for less discretionary services, such
as inpatient hospital care, could be done within the current program
structure. Congress has already taken similar action by reducing and
eliminating cost- sharing requirements for various cancer screening tests
and vaccinations in order to ensure that affordability is not a barrier to
these important services.
Adding catastrophic protection by capping how much beneficiaries are
required to pay out- of- pocket also could be done under current program
structure. There would need to be agreement on how to allocate between parts
A and B the added cost to the program and recognition of the time and
resources needed to incorporate such a change into HCFA?s information
systems.
Raising the part B deductible or creating a combined deductible for part A
and part B services has been suggested to offset some of the additional cost
of providing catastrophic protection. It would also offset some of the real-
dollar decline in the part B deductible, which has not been adjusted for
inflation or raised in any way since 1990. These changes could be done under
current program structure as well, again with recognition of the time and
resources needed to incorporate the change into HCFA?s information systems.
Most beneficiaries who incurred cost- sharing would likely meet a combined
deductible through their use of what are now part B services. If the
combined deductible is set higher than the current part B deductible,
providing protection for low- income beneficiaries so that costs do not
become a barrier to needed services or an undue burden would be an important
consideration.
Combining the deductible or providing catastrophic protection would again
raise the issue of whether to maintain individuals? ability to participate
independently in A or B or to require full participation by all
beneficiaries in the entire program. Requiring full participation for
beneficiaries who now participate in only one part of the program could
result in additional costs for beneficiaries who have alternative coverage
as well as additional program costs. It also raises the issue of the
entitlement for persons who have paid the required payroll tax, but choose
not to pay the premium.
Partial benefits could be extended to those who do not fully participate in
the program. Alternatively, some of the effects of mandatory participation
Page 13 GAO- 01- 862T
could be muted by phasing in a unified program so that new beneficiaries
would participate in the full program while those who now participate in
only part of the program could continue to do so.
As noted earlier, the original Medicare statute reflected 1960s private
health insurance practices that often included separate policies for
different services as well as a passive bill paying approach. In contrast to
Medicare, which has not changed much since its inception, private insurance
has evolved over the last 40 years and now offers comprehensive policies and
employs management techniques designed to improve the quality and efficiency
of services purchased. Private insurers are able to undertake these efforts
because many have detailed data on service use across enrollees and
providers, as well as wide latitude in how they run their businesses.
Regardless of whether the relationship between parts A and B is
restructured, HCFA faces challenges in seeking to more efficiently manage
Medicare services due to its outdated and inadequate IT systems, statutory
constraints, and the fundamental need for public accountability that
accompanies a large public program. These limitations have hampered the
agency?s ability to administer the program and incorporate new innovations.
Private insurers have taken steps to influence utilization and patterns of
service delivery through efforts such as beneficiary education, preferred
provider networks, and coordination of services. NASI has reviewed many of
these private sector activities and concluded that they could have potential
value for Medicare. However, they would need to be tested to determine their
effects as well as how they might be adapted to reflect the uniqueness of
Medicare as both a public program and the largest single purchaser of health
care. In addition, HCFA would likely need new statutory authority to broadly
implement many of these innovations.
To effectively oversee claims administration and assess the effects of
innovative policies that private sector insurers have adopted, HCFA needs
timely and comprehensive information on services and payments in the
aggregate and for individual beneficiaries. HCFA lacks that capacity today,
not because it has separate contractors for parts A and B, but because of
deficiencies in its information systems. Some of the agency?s vital
information systems are decades old, with some operating software rarely
used today by any entity other than HCFA, and lack the capacity and
flexibility that newer technology can offer. Consequently, HCFA has had
difficulty assembling timely and comprehensive information about provider
billing patterns and beneficiary service use. Challenges for
Management and Promoting Care Quality Remain Regardless of Restructuring
Effective Program Management Depends on Comprehensive and Timely Information
Page 14 GAO- 01- 862T
Currently, data from parts A and B do flow to some common points- both
during claims processing and after. During claims processing, both part A
and part B claims are checked through a prepayment validation and
authorization system operated by HCFA- the Common Working File (CWF). Claims
approved for payment are ultimately complied in the National Claims History
(NCH) file, which can be analyzed to look at broader payment trends within
the program. The problem is that this compilation of information occurs long
after services have been delivered and claims paid.
These system limitations are unfortunate because changes in Medicare payment
policy for one type of service can have reverberations in other areas. To
understand these effects requires analysis across a range of services
beneficiaries may be receiving. A clear example of this occurred after the
implementation of a prospective payment system (PPS) for hospitals, which
pays hospitals fixed, predetermined amounts for each hospital stay that vary
according to patients? diagnoses. Prior to this innovation, hospitals were
paid on the basis of their costs, with little incentive to limit patient
stays or provide care efficiently. Paying a fixed amount for an episode of
hospital care creates incentives for hospitals to reduce lengths of stay and
to shift services that had been provided in the hospital to other settings.
Understanding these modifications in care delivery led to payment changes to
prevent Medicare from paying twice for the same service. More recent payment
changes for home health and SNF services, and the soon to be implemented PPS
for inpatient rehabilitation services, will likely cause similar kinds of
care shifts. It is essential that HCFA has the ability to monitor changes in
care delivery in a timely and objective manner to determine how these
payment policies may need to be adjusted in the future.
Recent experience has also demonstrated HCFA?s difficulties in developing
information to measure the effects of changing Medicare policies on
beneficiaries and providers in a comprehensive and timely manner. The
Balanced Budget Act of 1997 (BBA) payment reforms represented bold steps to
control Medicare spending by changing the financial incentives for
delivering care efficiently. Reforms affected hospitals, home health
agencies, SNFs, and providers of other services. Affected providers
presented anecdotal evidence asserting that the BBA?s payment reforms caused
them financial difficulties and would impair beneficiary access, urging
Congress to undo some of the act?s provisions. HCFA analysts were ill-
equipped to assess the validity of these charges because the necessary
program data were not readily available.
Page 15 GAO- 01- 862T
Better and more timely information is a prerequisite to more effective
program management. It is essential to the development and refinement of
payment methods for different service providers. It can also help
policymakers understand the desirable and undesirable consequences of
changes on beneficiaries, providers, and the trust funds. Generating these
data is not dependent on unifying part A and part B, but rather on merging
part A and part B data in a modern information system capable of supporting
timely, pertinent analyses.
An expert panel convened by NASI has suggested that Medicare may benefit
from moving away from its passive bill paying approach by adopting some
private insurers? practices designed to improve the quality and efficiency
of care. 7 The panel focused on provider and beneficiary education,
preferred provider networks, and coordination of services as potential
improvements in Medicare. Educating beneficiaries or providers could improve
the use of important preventive and other services currently being underused
and minimize questionable use of services. Developing a system of preferred
providers selected on the basis of quality as well as cost could improve
care and help achieve savings. More actively coordinating care across
provider settings for beneficiaries with chronic diseases like diabetes or
who have recently experienced heart attacks might also help improve quality
and efficiency. HCFA has begun to implement some innovations and experiment
with others. Broadly implementing the experimental innovations that prove
successful may require new statutory authority. Other private sector
innovations, however, may be difficult to incorporate, given Medicare?s size
and the need for transparency in a public program.
HCFA has been able to implement broad- based education efforts but has been
stymied in implementing approaches targeted to individual beneficiaries most
likely to need the help. For example, it has an extensive effort underway to
encourage colon cancer screening that includes dissemination of more than
23,000 innovative posters. The posters include tear- off sheets that
beneficiaries can hand to physicians to facilitate discussions that
otherwise might be avoided because of the unfamiliar words, sensitive
issues, and unpleasant options that can be involved.
7 From a Generation Behind to a Generation Ahead: Transforming Traditional
Medicare, Final Report of the Study Panel on Fee- for- Service Medicare,
National Academy of Social Insurance, Washington, D. C.: January 1998.
Quality Promotion Efforts
Could Reap Benefits but Face Many Obstacles
Page 16 GAO- 01- 862T
HCFA is also involved in a multifaceted effort to increase flu vaccinations
and mammography use among beneficiaries. However, HCFA may be less able to
undertake more targeted education efforts that some private insurers are
using, such as sending out reminders to identified enrollees about the need
to obtain a certain service. Because of Medicare?s size and status as a
federal program, beneficiaries and others might have concerns about HCFA
using personal medical information from claims data to target educational
efforts. Providers might also object to a government insurance program
advocating certain medical services for their patients.
HCFA is providing more information to physicians about service use and
typical practice patterns in an effort to educate them about how their
practice patterns compare to the norm. For example, the Medicare peer review
organizations encourage those who have unusual practice patterns to
reconsider their service provision. However, private insurers can go one
step further and terminate providers who continue to have aberrant practice
patterns. HCFA?s ability to terminate providers is much more limited because
of statutory requirements intended to protect beneficiaries? choice of
providers.
HCFA?s ability to encourage use of preferred providers is also limited. The
Medicare statute generally allows any qualified provider to participate in
the program. HCFA has experimented with bundling payments for certain
expensive procedures performed by designated providers. For example, it
tested the impact of making single ?global? payments to hospitals for all
services- both hospital and physician- related to coronary artery bypass
graft surgery. The hospitals chosen for the experiment were among those with
the best outcomes for these surgeries. 8 The experiment cut program costs by
10 percent for the 10, 000 coronary artery bypass surgeries performed, and
saved money for beneficiaries through reduced part B coinsurance payments.
More important, compared to a group of beneficiaries not receiving this
bundled care, beneficiaries who were treated in one of the selected
hospitals had lower mortality rates, were more satisfied with the quality of
the nursing care, and appreciated the simplicity of a single coinsurance
amount. HCFA has begun a similar experiment at selected acute care
hospitals, which involves bundling payments for hospital, physician, and
other health care professionals?
8 A number of studies prior to this experiment have found that hospitals
with the greatest volume of these types of surgeries generally had better
outcomes, in regard to mortality and complications.
Page 17 GAO- 01- 862T
services provided during a beneficiary?s hospital stay for selected
cardiovascular and orthopedic procedures.
However, more wide scale Medicare implementation of such hospital and
physician partnership arrangements may be difficult. Providers have raised
concerns about a government program designating some providers as delivering
higher quality care than others. In addition, bundling services for
hospitals and doctors added administrative burdens to the hospitals and took
control of payments away from doctors. In the end, it is not the separation
of parts A and B that would impede efforts to promote such preferred
provider arrangements. Rather, it may be more deep- seated concerns about
government promotion of certain providers at the expense of others that
serve as a barrier to this and other types of preferred provider
arrangements.
HCFA has also been conducting demonstrations to test how to better
coordinate care for certain patients since the 1980s. In addition, BBA 9
mandates that HCFA find budget neutral ways to test methods of coordinating
a range of services for chronically ill beneficiaries in at least nine urban
and rural sites. The law authorizes the Secretary of Health and Human
Services to incorporate any components proven to be costeffective into
Medicare through regulations and to expand the number of demonstration
sites.
While there is increasing interest in efforts to coordinate care, it is not
clear that they are always cost- effective. Some experience in both the
private and public sectors suggests that such efforts can improve quality
and achieve savings. For example, the Group Health Cooperative of Puget
Sound and PacifiCare teamed with a senior citizens center to offer
supervised health promotion and chronic illness self- management
interventions to chronically ill seniors. The intervention included meetings
with geriatric nurse practitioners to develop individually tailored health
promotion plans, medication reviews, classes, support groups, and volunteer
mentors. Preliminary findings suggested that the case- managed group had
fewer health problems and lower costs compared to a group that did not
receive the services. However, other experiments, including those conducted
by HCFA, have failed to demonstrate either quality improvements or cost
savings. Furthermore, there would need to be statutory changes to implement
different coordination approaches in
9 Section 4016.
Page 18 GAO- 01- 862T
Medicare if they involved coverage of new services, such as care
coordinators, or involved control over the use of particular services or
providers.
The Medicare program faces many challenges. Clearly, the overarching issue
is how to ensure that Medicare remains sustainable for future generations of
beneficiaries. Meeting that challenge will involve difficult decisions that
will likely affect beneficiaries, providers, and taxpayers. However, the
financing issue should not obscure other important Medicare challenges.
Medicare?s current cost- sharing arrangements are not well designed to
encourage the efficient use of services without discouraging necessary care.
Moreover, the lack of catastrophic coverage can leave some beneficiaries
liable for substantial Medicare expenses. Finally, some aspects of
Medicare?s program management are inefficient and lag behind modern private
sector practices. Changes in Medicare?s program management could improve
both the delivery of health care to beneficiaries and the program?s ability
to pay providers appropriately.
Some view restructuring of the relationship between parts A and B as an
important element of overall Medicare reform. Fundamentally, assessing the
program as a whole is an important first step in addressing Medicare?s
challenges. Solutions to many of these challenges could be crafted without
restructuring. However, restructuring may provide opportunities to implement
desired reforms- with or without unifying the HI and SMI trust funds- while
undoubtedly raising issues that will have to be considered.
Mr. Chairman, this concludes my statement. I would be happy to answer any
questions that you or members of the Subcommittee may have.
For more information regarding this testimony, please contact me or Laura
Dummit at (202) 512- 7114. Sheila Avruch, James Cosgrove, and Paul Cotton
also made key contributions to this statement.
(290088) Concluding
Observations GAO Contacts and Staff Acknowledgments
*** End of document. ***