Ask the Scholar

Document scope · 1 page
doc
Scholar
Ask about this object, its catalog metadata, its source description, or the page inventory. For page-specific OCR and visual context, open one of the page chats.

Scholar Source Context

Document identity
localId
34428479
label
[AARP (American Association of Retired Persons) Medicare Restructuring Forum Papers] [loose]
core
doc
dtoType
document
pageCount
1
Source metadata
Source extras
naId
34428479
levelOfDescription
fileUnit
otherTitles
42-t-7422560-20140536S-004-010-2016
recordType
description
ocrSource
nara-archive
Single page context
seq
1
pageIndex
0
type
document
mediaId
ecdcd156d9e43f10
ocrText
July 31, 1995 TO: Distribution FROM: Chris Jennings and Jen Klein Attached are the papers presented at AARP Medicare Restructuring Forum held last week. The first set attached is the document prepared by Henry Aaron and Bob Reischauer. We thought this information might be helpful in preparation for tomorrow's meeting with Henry Aaron, Bob Reischauer, and Uwe Reinhardt (3:30-5:30, OEOB Room 211). Suggested format includes: 1) Aaron and Reischauer short presentation outlining key points of their paper; 2) Reindhardt's response and his own insights; 3) Vladeck outlining the consistency/inconsistency of the three previous presentations; 4) Rasco and Tyson open discussion among the attendees. Please feel free to direct any questions or concerns about this memo to Chris Jennings as soon as possible. Summary Medicare: Where to from Here? by Henry J. Aaron and Robert D. Reischauer¹ Medicare is at the core of two current policy debates. One centers on the steps needed to balance the federal budget over the next seven to ten years. The other focusses on how current entitlement programs might best be modified to cope with the retirement of the baby boom generation which will begin at the end of the first decade of the 21st century. Cuts in Medicare undertaken to balance the budget will probably make no more than a modest contribution to meeting in the second and more serious challenge; realistically, measures designed to deal with the consequences of the baby boom generation's retirement will contribute little to the short-run deficit reduction effort. We describe a reform proposal designed to address the second of these issues, the long-run future of Medicare. Popularity and Success: A Brief History of Medicare The changes needed to prepare Medicare for the future are so fundamental that it is worth reflecting on how we got where we now are, on the original premises of the program and whether they should be reconsidered, on the underlying factors that are contributing to the program's financial stress, and the on program's popularity and accomplishments. Before the enactment of Medicare, few retired elderly or non-working disabled enjoyed the protection offered by health insurance. The private insurance available to these groups was expensive, often inadequate, and uncertain in the sense that renewal was not guaranteed. In one view, the government should provide assistance only to those who are too poor to buy the service. The principle behind this approach to assistance is that, with few exceptions, markets should allocate all goods and services. The other view holds that the best way, on balance, to provide some types of assistance is through a universal entitlement, aid available without regard to personal income or 1 Henry J. Aaron is the Director of the Economic Studies Program at the Brookings Institution. Robert D. Reischauer is a senior fellow in that program. The views expressed in this paper are those of the authors and should not be attributed to the Brookings Institution, its staff. trustees, officers or funders. The authors wish to thank Linda T. Bilheimer and Joshua Wiener for their helpful comments and absolve them of the responsibility for any errors remaining in the paper. --1-- means based on some more or less objective indicator of need. When Congress approved Medicare, it embraced the second philosophy by establishing a universal entitlement to stipulated benefits for everyone over age 64 who has worked for at least a modest period in a job subject to payroll taxation or is a spouse of an entitled worker. The great popularity of Medicare almost certainly guarantees that the reforms adopted by Congress will not compromise the fundamental entitlement nature of the program. According to public opinion polls, Medicare ranks next to social security as the federal program with the widest popular support. Nor is this support of recent vintage. For many years the public has declared its support for Medicare, its hostility to cutting benefits, and its willingness to support increased taxes, if necessary, to sustain benefits. Medicare's popularity is understandable. Together with supplemental policies, it has provided the elderly and disabled with heavily subsidized access, on a non discriminatory basis and with unrestricted choice of providers, to the same health services that workers and their families enjoy. The explosive growth of Medicare costs is neither new nor mysterious. While growth has been rapid, Medicare spending per-enrollee rose less rapidly than national or private health spending from 1984 to 1993. Some spending growth is attributable to an increases in the number of beneficiaries by about 2.5 percent a year since 1968. Legislation and administrative actions have also expanded the list of covered services. While benefits have increased somewhat over the program's thirty years, the Medicare benefit package remains comparatively parsimonious, a fact that should be kept in mind when considering changes to reduce program costs. Overall, the Medicare benefit package is less generous than about 85 percent of private health insurance plans. The program covers only about 45 percent of the total health bill of the elderly who, on average, spend considerably more out-of-pocket on health care than the non-elderly. Medicare benefit cuts would only increase these disparities. The major contributor to the explosive growth of both Medicare and other health costs has been the increased capabilities of medicine. The proliferation of medical technology has surpassed all imaginations and has produced new diagnostic tools, procedures, treatments, and drugs. Few advocate policies that might dampen this source of cost growth nor would it be reasonable to deny Medicare participants the fruits of medical advances available to those with employer-sponsored or private health insurance. --2-- Some Considerations Relevant for Reform Successful and sensible Medicare reform must build on an appreciation of the way the current program operates and of certain central characteristics, and from an understanding of the health care environment of which Medicare is a part. Reformers must realize the limited contribution Medicare now makes to covering the health expenditures of the aged and disabled, the program's bifurcated structure, the limited role managed care plays in Medicare, the potential under the existing payment structure for pernicious risk-selection competition, the uncertainty of cost projections, and the interaction between Medicare and Medicaid. Public Medicare and "Real Medicare" The federally financed Medicare benefit package is not the health insurance plan that most Medicare participants actually live with. Over 80 percent of elderly and disabled Medicare participants are covered by employer-sponsored retiree plans, private medigap plans, or both, or by Medicaid. Those covered one way or another by Medicaid, like those with medigap and supplemental retiree policies, face negligible deductibles and cost sharing and are covered for additional services not included in the Medicare benefit package. In thinking about reform of Medicare, most discussion focusses on the public program alone, not the hybrid system. But because the large majority of beneficiaries live with the coverage and incentives of the hybrid system, sensible policy change can only be formulated by considering how best to transform the entire system. In most private insurance plans, deductibles and coinsurance serve to discourage demand for low-benefit care. Deductibles also spare insurers the administrative cost and the insured the added premiums necessary to cover the cost of processing many small claims. Because of medigap and Medicaid, however, Medicare deductibles and coinsurance do not serve their intended purposes and increasing them would not have the intended effects. While doing little to reduce low-benefit services and nuisance claims, increases in Medicare coinsurance and deductibles would shift costs from the Medicare trust fund to Medicare eligibles (for those with medigap insurance), their former employers, or another account in the federal budget and to states. Reform of Medicare should not be confined to the public program, but should encompass rules affecting the sale and pricing of supplemental policies. --3-- The A and B of Medicare Reform Whatever rationale may once have existed for the distinction between Part A and Part B, medical technology, the development of new forms of service delivery, and new payment structures have rendered it obsolete. No good case can be made, in our judgment, for perpetuating the separate parts of Medicare. The distinct methods of financing Part A and Part B -- the former with a payroll tax and the latter through premiums and regular appropriations -- have created very different political and economic dynamics within the two parts. Any major reform of Medicare should unify parts A and B. The consolidation should not, however, be used as an opportunity to relax the limited fiscal constraints under which the program now operates and tap into general revenues to support Part A. Fee-for-service and risk-contracts. Medicare is rapidly becoming the last remnant of relatively unmanaged, fee-for-service care in the United States. Two percent of Medicare enrollees are served by a managed care plan reimbursed on the basis of reasonable costs. Only 7 percent are served by capitated managed care plans, although the number is increasing. Overall, about three- quarters of the Medicare population live where a managed care option is available. Nevertheless, there is no significant enrollment by risk contractors in 28 states. Reforms that propose to move Medicare beneficiaries into managed care plans will therefore take time. Institutional capabilities will have to be developed and participants will have to learn more about and become more comfortable with managed care delivery systems. The "Cherry Picking" Potential in Medicare. Insurance carriers or managed care plans can increase profits either by providing services more inexpensively or by insuring those with lower- than-average health costs. Medicare risk contractors are now required to accept any Medicare participant who wants to join their plan. Nevertheless, those who up for risk plans available have lower-than-average costs. As a result, risk contracting raises costs to the government, although it may be saving money for the nation. The extent to which vendors currently try to enroll only better-than-average risks is not known, but the profit-boosting potential of such efforts is large because health costs are highly skewed. If Medicare is to move from fee-for-service to capitated reimbursement, payments to plans will have to be calibrated to health risks better than they now are, and steps to prevent plans from engaging in subtle practices to ward off high risk participants will be vital. --4-- Cost Variations. The Medicare benefit package is nationally uniform, but its cost is not. In 1992 reimbursements averaged $4,194 per person served, but varied from a low of $2,942 in Nebraska to a high of $4,977 in Louisiana. Within states, the variations are even larger. Although variations based on differences in input prices can be rationalized, it is doubtful that a persuasive case could be made for per person reimbursements which are 69 percent higher in Louisiana than in Nebraska. Uncertainty about costs. Actuarial projections that Part A of Medicare will be insolvent in a few years add urgency to the call to "do something" to fix Medicare. The impossibility of balancing the budget in seven (or even ten) years without large reductions in Medicare raises this sense of urgency to almost frantic levels. Two aspects of the actuarial projections are relevant to the process of reform. The first is that the actuarial projections are extremely volatile, changing by many years the period until insolvency projected in one trustees' report from the projections of the preceding year or two (see table 1). The second important aspect of the projections is that the gap between current revenues and current benefits is 80 huge that even the large fluctuations in projected costs associated with the shifts shown in table 1 give no hope that the current system can be sustained. The 25-year actuarial projections indicate that Part A outlays would have to be cut 30 percent or revenues increased 43 percent to balance the program, and since any legislative changes would be small at first, the changes in the terminal years of the projection period would have to be considerably larger. The growth of spending under Part B is considerably higher than under part A, in large measure because many services that were once provided in hospitals can now be provided on an outpatient basis or in physicians' offices. As a share of gross domestic product, Part A is projected to increase from 1.62 percent in 1995 to 2.83 percent in 2020, a 75 percent increase. Part B is projected to increase from 0.93 percent of GDP to 3.18 percent over the same period, a 242 percent increase. Medicaid and Medicare. Without Medicaid most of the limited income of sick low-income Medicare participants would be absorbed by out-pocket health expenditures. But Medicaid is a heavy burden on states which, on average, pay 43 percent of the costs of Medicaid. As Medicare evolves or is changed, the logic of dividing the responsibility for health care for the low-income aged and disabled between these two programs will weaken. If poor Medicare enrollees require premium subsidies, equity suggests that they should be nationally uniform. The prospect that Medicaid --5-- payments per recipient may be capped (the President's proposal) or transformed into a block grant (as proposed in The Contract with America) reinforces the logic of having Medicare assume full responsibility for covering the acute health care costs of the low-income aged and disabled. Goals and Principles to Guide Medicare Reform We believe that the following four principles and goals should direct the changes that Medicare will require. I Reforms should deal not just with "public" Medicare but with the full package of benefits most Medicare beneficiaries use. 2 A reformed Madicare program should not result in the provision of health care to the olderly and disabled of a quality materially different from that available to the general population; nor should the delivery system for the elderly and disabled be segregated from that of the rest of the population (other than for definable services where medical reasons justify separate delivery, as with geriatric care). 3 Medicare should create incentives for beneficiaries to seek care from efficient plans and should encourage physicians and hospitals to provide care of given quality at lowest possible cost. 4 Medicare beneficiaries should have a degree of choice among health plans similar to that enjoyed by the rest of the population. A Program for Medicare Reform We propose converting Medicare from a "service reimbursement" system into a "premium support" system. Rather than paying for all services on a stipulated menu, Medicare would pay a defined sum toward the purchase of an insurance policy that provides a defined set of services. As with private insurance for the working population, plans could reimburse any provider the patient chooses on a fee-for-service basis (the current method Medicare uses for most beneficiaries), contract with a preferred provider organization, or operate through a health maintenance organization. Plans could manage care in any of the ways now in use or that might arise in the future. Ultimately, all Medicare beneficiaries would receive a predetermined, geographically variable amount that could be applied toward the purchase of a health plan providing defined services. Plan Outline. Health plans would be required to offer defined services. The marketing of insurance to Medicare beneficiaries would follow the principles developed for the operation of --6-- managed competition. The first step would be the definition of health market areas. These marketing areas could cover portions of more than one state. Next, entities interested in bearing the risk of providing health care for the Medicare population -- insurance companies, independent practice associations, health maintenance organizations, preferred provider organizations -- would be invited to submit bids on the price at which they would be willing to provide the defined benefit package for the "average" Medicare enrollee within a particular market area. The federal Medicare payment in each market area would be the same regardless of which plan the Medicare enrollee selected. Since some Medicare enrollees are frail or querulous and most may be confused by complex literature on health insurance, local marketing organizations would be established to handle the sale of insurance, to discourage insurers from using marketing to attract superior risks, and to hold down marketing costs. Enrollees would select an insurance plan for the coming year, but would be permitted to switch plans during an annual open enrollment period. Because some participants have expected health costs that are much higher or lower than average, plans would receive risk-adjusted payments from Medicare based on age, sex, disability status, and other health indicators. We advocate the approach described above not because we necessarily accept the claims of the more enthusiastic advocates of managed care and managed competition regarding potential savings, but because we think such a framework will encourage people to choose the plans whose style of care matches their preferences and because enrollees should bear the financial consequences of their choices. Furthermore, the framework described above lends itself to budget control in ways that the current Medicare system does not. Many questions would have to be answered in implementing such a\study, including the following. 1 What should be the benefit package in the new plan? The new plan should not use the current Medicare benefit package, but should combine the current Medicare package with some modest coverage of prescription drugs and catastrophic protection. A standard benefit package and standardized cost-sharing regimes are important, at least initially, because they will reduce risk segmentation among plans and help participants compare the cost and quality of the different plans. 2 How should the federal payment be set? We suggest that the initial federal payment be set at 95 percent of the cost of the current Medicare package in the market area, adjusted to remove indirect medical education, direct medical education, and disproportionate share payments. During a phase-in period of perhaps five to ten years, the federal payment should grow more slowly than --7-- projected baseline costs. In the long run, the federal Medicare payment should grow at the same rate as per capita spending on health care for the non-elderly. This formula is mechanical and may require periodic adjustment since the per capita cost of care depends on the average age of each population, the age-specific gradient in health care costs, and the age bias of new medical technology. Initially, payment levels would have to reflect historical spending patterns, but differences should be gradually narrowed until payments varied only for differences in wage rates and other input prices. If some plans submit bids below the federal payment, we suggest that during the initial few years it would probably be best to require that these dividends be devoted to supplemental services such as eyeglasses, routine dental care, or other services. Over time, as the competitive marketplace developed and participants became familiar with their options and the consequences of choosing one plan over another, differences between the federal payment and plan cost could be rebated to participants as non-taxable income or split between the government and the participants. 3 Will Medicare enrollees segregate themselves or will plan administrators market plans in ways that result in economic segregation among plans? One of the strengths of the current Medicare program is that all participants--rich and poor alike--come under the same basic plan and providers are relatively indifferent to the income of their Medicare patients. In a world of competing plans, economic segregation could occur in various ways. Because health status and income are positively correlated, such segregation will aggravate the risk adjustment problem. To have some choice of plans, low-income participants would require some sort of premium supplement to replace Medicaid assistance. 4 How should a new system be phased in? This question is the most important and most difficult. Before the plan described here can operate, it is necessary to create an environment of competition among health plans, to construct the apparatus necessary to regulate marketing of insurance and the risk-adjustment procedures for allocating payments based on enrollment, and to overcome the practical difficulties in transferring costs to many current Medicare enrollees. Even in the most prepared regions, it will take several years to establish the necessary institutional structure. The plan also hinges on the availability of a sufficient number of plans in each market area to create meaningful choice and competition. For this, and many other, reasons, Medicare savings can not come as fast as called for in the 1996 budget resolution if real Medicare benefits are to be sustained. --8-- The most difficult transitional issues involves how to phase Medicare enrollees into the new system. This problem would be complex, but quickly solvable, if enrollees could be enticed into the new system by payments of increased generosity; but the current budget situation forecloses that option. The simplest practical method would be to run the current Medicare system along side the new system. The new system would be mandatory for everyone who turns 65 and becomes eligible for Medicare after a certain date. It would be optional for everyone enrolled in Medicare before that date. 5 What should be done with DRGs and RBRVS? Most current Medicare payments are well below reimbursements from other payers. In addition, providers are not permitted to bill participants for charges beyond certain specified and limited amounts. A plan that offered the choice of providers currently available under Medicare without these price discounts would either have to charge much higher premiums or leave participants exposed to significant balance billing. For this reason, the various Medicare fee schedules should be maintained for a transitional period. 6 How much regulation will the new system require? What type of organization should do it? Initially, at least, the new system will require a good deal of regulation and much of it will have to emanate from Washington. The medical capabilities, quality of care, administrative competence, and financial soundness of plans will have to be certified. State agencies or specialized non-profit organizations could enforce these standards. State, local, or not-for-profit organizations will be responsible for organizing and assuring the orderly functioning of the health plan market in each area. Conclusion Budgetary and demographic developments mean that Medicare as currently structured, is unsustainable. The approach we have presented builds on the strengths of the current program while converting it into a system similar to the employer-sponsored insurance now available to most Americans. Our plan would strengthen participants' incentives to obtain services from cost effective delivery systems and providers' incentives to operate efficiently. It would enable Congress directly to control the per capita budget cost of Medicare. Nonetheless, we believe that after heroically trying to cut costs, Congress will recognize that the American people want to sustain Medicare benefits and that to do SO some increase in payroll or other earmarked taxes will have to be part of a plan to restore financial balance to Medicare. --9-- TABLE 1. Number of Years from Trustees' Projection until Insolvency Year of Trustees' Report Year of Insolvency 1970 2 1971 2 1972 4 1973 none indicated 1974 none indicated 1975 about 20* 1976 about 15* 1977 about 10* 1978 12 1979 13 1980 14 1981 10 1982 5 1983 7 1984 7 1985 13 1986 10 1986 amended 12 1987 15 1988 17 1989 none indicated 1990 13 1991 14 1992 10 1993 6 1994 7 1995 7 . Projections for 1975, 1976,, and 1977 put the dates of insolvency, respectively, in "the late 1990s," "the early 1990." and the late 1980." Source: Intermediate projections of various Hospital Insurance Trustees' Reports, 1966-95. --10-- THE HENRY KAISER THE KAISER-HARVARD PROGRAM ON FAMILY FOUNDATION THE PUBLIC AND HEALTH/SOCIAL POLICY HARVARD UNIVERSITY: ROBERT 1. BLENDON. ScD DIRECTOR PROFESSOR OF HEALTH POLICY AND POLITICAL ANALYSIS. SCHOOL OF PUBLIC HEALTH AND KENNEDY SCHOOL OF GOVERNMENT. EMBARGOED FOR RELEASE UNTIL: CONTACTS: Matt James HARVARD UNIVERSITY Thursday, June 29, 1995, 9:30 AM EST Tina Hoff JOHN BENSON DEPUTY DIRECTOR (415) 854-9400 KAISER FAMILY For NDATION: DREW E. ALTMAN. PHD PRESIDENT. KAISER FAMILY FOUNDATION NEW SURVEY FINDS MOST AMERICANS OPPOSE SLOWING MATT JAMES. VICE PRESIDENT OF THE GROWTH OF MEDICARE TO BALANCE THE BUDGET OR CUT TAXES, COMMUNICATIONS AND MEDIA PROGRAMS KAISER FAMILY FOUNDATION BUT WOULD SUPPORT CHANGES TO AVOID BANKRUPTCY -Public Favors Incremental Rather than Sweeping Reforms- -Significant Generational Differences on Medicare Reform-- Washington, D.C. -- A new survey has found that close to three out of four Americans (738) support reducing the rate of growth in Medicare spending if the goal of the reductions is to avoid the bankruptcy of the Medicare program. However, less than half the public supports major reductions in Medicare spending growth if the goal is to balance the Federal budget (448) or provide a tax cut (28%). The Kaiser/Harvard/Harris Survey on Medicare also found that most Americans (70%) know that the Medicare program is in danger of going bankrupt, and many (48%) express a high level of concern over that possibility. Nearly half (49%) say they are aware that Medicare "has been going bankrupt" for a long time. REDESIGNING MEDICARE The survey also found that while most Americans support Medicare changes to ensure the fiscal solvency of the program, most are leery of a major redesign of the program. A plurality of the public, 45%, support Congress making changes in the Medicare program as long as the changes still "preserve Medicare basically as it is." Two out of five adults (40%) support a complete redesign of the program, and a minority (14%) say Medicare should be left as it is. When given a choice between the current system, in which you get a Medicare policy directly from the government, and the option of receiving a voucher to purchase private insurance, the public supports the current system (65%) over a voucher (328). The public is evenly split (49% favoring, 48% against) on the idea of enrolling "most" Medicare beneficiaries in Medicare managed care. But, 72% favor government incentives to enroll in managed care, and 55% favor raising premiums for those who stay in fee-for-service. "Just as in health reform, the American people are leery of sweeping change, and generally opt for more modest incremental reforms. Medicare is no exception," said Drew E. Altman, president of the Kaiser Family Foundation. more A JOINT PROGRAM OF THE HENRY J. KAISER FAMILY FOUNDATION AND HARVARD UNIVERSITY KAISER FAMILY FOUNDATION: 2400 SAND HILL ROAD, MENLO PARK. CA 94025 415 854-9400 FAX 415 854-4800 HARVARD SCHOOL OF PUBLIC HEALTH: 677 HUNTINGTON AVENUE, BOSTON, MA 02115 617 432-4502 FAX 617 432-0092 but higher income retirees should pay more; but fewer than one in four (23%) would means test the program and require all those with higher incomes to buy their own health insurance. One third (34%) believe that if people have paid Medicare taxes, they are entitled to benefits when they retire no matter how well off they are. THE MEDICARE GENERATION GAP The survey found significant differences in how people of different generations view the Medicare program. For example, support for a complete redesign of the Medicare program is higher among those under 50 (44%) than among those over 65 (23%). Over one-third of those over 65 (35%) want Congress to leave Medicare alone, while few people under 50 (8%) hold this view. Those under 50 are almost twice as likely to support turning Medicare into a voucher system (40%) as are those over 65 (22%). When it comes to managed care, a majority of those under 50 (60%) favor encouraging the greater use of low-cost managed care plans to avoid bankruptcy and increased taxes, a position favored by less than half of those 65 and over (42%). In addition, while a majority of those under 50 (57%) favor having most Medicare beneficiaries enrolled in private managed care plans rather than traditional Medicare fee-for-service (41%), seniors by an overwhelming margin (70% to 23%) prefer having most beneficiaries remain in fee-for-service. Although most seniors (56%) recognize that the program is in danger of bankruptcy, many are cynical about the motives of Congress in the current Medicare debate. By a two-to-one margin, seniors think that members of Congress are trying to gain political advantage from the issue (63%), rather than genuinely trying to respond to a crisis in the program (31%). By contrast, a majority (52%) of adults under the age of 50 think members of Congress are genuinely trying to respond to a crisis. The survey also shows that, as a group, seniors are politically more active on Medicare and Social Security issues than other adults. MEDICARE KNOWLEDGE Most Americans blame rising Medicare costs on: -- excessive charges by doctors, hospitals, and other health providers (80% say this is a very important cause for rising costs); -- poor management by the government (70%), and fraud and abuse by doctors; -- hospitals (68%). Reasons often cited by Medicare experts for rising costs appear less important to the public, such as the increased number of retirees (50%) and new drugs, tests and treatments for the elderly (31%). Most Americans (77%) are aware that Medicare is primarily a Federal government, rather than a state or private program; that the current Medicare program pays for doctor bills for individuals age 65 and older (75%); and, that Medicare is a program that principally serves the elderly (74%). 3 American Association of Retired Persons (AARP) Conference on the Future of Medicare PUBLIC OPINION STRATEGIES Prepared by: Bill McInturff, Partner July 20, 1995 PARADIGM OF CHANGE The filters on which the American people need to be satisfied before undertaking significant policy change: 1. Which is scarier -- the status quo or major change? 2. Does it affect someone else, and not you? 3. Is it consistent with American values and traditions? 4. Is there a transition message? 5. Is it inherently incremental? 6. Can you prove that it has worked someplace else? Medicare IS Social Security. Not really, but fully 64% of respondents say that if a Member of Congress votes to cut Medicare, they have broken their word not to cut Social Security. If your Member of Congress ran for office this fall promising to not cut Social Security but then votes this year to cut Medicare spending, would you say that Member of Congress broken their campaign promise? HAS 64% DK / REF 12% HAS NOT 24% PUBLIC OPINION STRATEGIES -- Frederick/Schneiders National Survey, N = 800 Likely Voters January 24 - 26, 1995 Medicare is very well perceived. Eighty-one percent of seniors have a favorable rating of Medicare (49% very favorable) while 86% of seniors have a favorable rating of Social Security (54% very favorable). Favorability Ratings of Various Federal Programs Social Security 36% 97% Medicare 29% 85% Medicaid 17% 53% Very Favorable Mostly Favorable PUBLIC OPINION STRATEGIES -- Frederick/Schneiders National Survey, N = 800 Likely Voters January 24 - 26, 1995 To seniors, Medicare is not an entitlement. 67% see it as a "government program where people get the benefits their taxes have already paid for" while 74% of seniors say Social Security is a program they have already paid for. Are Social Security and Medicare entitlement programs or do people get benefits that their taxes have already paid for? % Responding "Paid For" by Age 72% 74% 74% 67% 67% 53% 49% 52% Total 18 - 34 years 35 - 64 years 65+ years Social Security Medicare PUBLIC OPINION STRATEGIES -- Frederick/Schneiders National Survey, N = 800 Likely Voters January 24 - 26, 1995 And, seniors do not want to see it changed. When asked to assess the Medicare program and its financial health, only 20% say it requires either "radical" or "significant change" compared to 44% responding "no change" or "only minor change." Because it's not broken. There is a limited perception of the current financial risk that the Medicare program is facing. A majority of seniors believe Medicare either faces no financial problems until baby-boomers retire or believe because it is paid for by a special trust fund it will not face financial problems at all. Medicare Bankrupt by 2002 % of those surveyed who believe this to be true 71% 66% 67% 59% 45% 18 - 29 years 30 - 44 years 45 - 54 years 55 - 64 years 65 + years American Viewpoint, Inc. National Survey, June 1995 However, the public debate is working. Americans have become increasingly familiar with reports of Medicare's financial status. Awareness of Medicare Trust Fund's financial status % seen/heard/read 34% 14% April June American Viewpoint, Inc. National Survey, June 1995 As coverage increases, more Americans are convinced of bankruptcy. Perceptions of the future of the Medicare program change as Americans become aware of threats to its solvency. As coverage of the Medicare Trustees Report increased, more people found it likely that Medicare will, in fact, be bankrupt in seven years. Medicare Bankrupt in 2002 (% of those surveyed who believe this to be true) 63% 43% 37% 32% April June Agree Disagree American Viewpoint, Inc. National Survey, June 1995 Awareness heightens with age. Americans aged 55 - 64 and 65 + are the most attuned to public reports of the financial status of the Medicare Trust Fund. Awareness of Medicare Trust Fund's financial status % seen/heard/read 57% 46% 42% 23% 19% 18 - 29 years 30 - 44 years 45 - 54 years 55 - 64 years 65 + years American Viewpoint, Inc. National Survey, June 1995 Belief in bankruptcy prompts support for reform. Among those who believe the report that Medicare will be bankrupt, there is strong support for dramatically changing the program. Among the 63% who find believable the statement "The Medicare program will be bankrupt in 7 years." 70% Agree that Republicans in Congress are trying to do what's best for all Americans because the Medicare fund is going broke and we have to act responsibly now to preserve the fund for future generations. 59% Agree with the Republican rhetoric of continuing to increase Medicare spending but at a slower rate, rather than the Democrat rhetoric of cutting Medicare spending. 58% Believe President Clinton and the Democrats know that Medicare is in deep trouble, but won't do anything about it because they want the Republicans to pay the political price for fixing the system. 55% Have more trust in Congress to reform Medicare in a responsible way, compared to 30% who trust the President. American Viewpoint, Inc. National Survey, June 1995 THE BOTTOM LINE: If it "ain't broke, don't fix it" sums up the data and serves as a powerfully cautionary note as we approach the budget battle. Medicare works (at least for the people on it), is well received, and given its importance and people's sense of security it will be difficult to change. We need to recognize the message about the current financial crisis facing Medicare needs to be delivered by other parties as well to help be convincing. Republicans need to vocally focus on protecting, improving, and preserving Medicare not solely on balancing the budget. And we need to recognize the significant amount of time that will be required to raise these points with the American public (one to three years or more). We are literally at "ground zero" in terms of people having the information they need to be receptive to a message about the level of change needed in the program. THE FEHBP AS A MODEL FOR A NEW MEDICARE PROGRAM Stuart Butler and Robert Moffit The problems of the current Medicare system derive from its two central characteristics - it is a defined benefit program and for its organization and cost constraint it relies on central planning and elaborate price controls. This structure invites cost escalation and inefficiency because it lacks strong incentives for beneficiaries to seek value for money and because centralization slows the introduction of innovative management and delivery innovations to reduce costs while increasing efficiency. To control costs while improving value and choices for beneficiaries, Medicare should be converted into a program in which the government provides beneficiaries with 56080 a defined contribution which may be used to purchase a Medicare-approved health plan. Such a system would be, in effect, a modified version of the Federal Employee Health Benefits Program (FEHBP), which currently makes 400 competing private plans available to nine million active and retired federal employees and their family members. The FEHBP population is not an ideal insurance pool, and on the face of it the program should not be successful. Enrollees tend to be older than the general population and the proportion of retirees covered is 40 percent and growing. Enrollment is optional and eligibility requirements liberal. Beneficiaries can switch plans annually without regard to their health condition and without any waiting period or exclusions, and the 1 plans must community rate their premiums. Yet despite its seeming vulnerability to adverse selection. the FEHBP works remarkably well. As the Congressional Research Service in a comprehensive review of the program in 1989. "That the FEHBP has continued to 'work' over the years, despite major changes in the environment in which it has operated, reflects the soundness of its basic design." The FEHBP functions in ways that could be incorporated in a reformed Medicare system. Federal workers and retirees in any particular area typically have access to between one dozen and two dozen plans. Nobody has a choice of fewer than seven plans. These range from traditional fee-for-service plans to HMOs, PPOs, IPAs and various point-of-service plans. All the plans cover basic hospital and physician services but beneficiaries can pick plans with different additional benefits, such as dental, prescription - drug, or mental health benefits. Whatever the cost of their chosen plan, the government pays a fixed amount (with the proviso that it will pay no more than 75 percent of the premium of any plan). Retirees, like active employees, are given a great deal of assistance in choosing plans in the FEHBP. Local Members of Congress commonly sponsor "health fairs" to review plans in their districts. Checkbook's Guide to Health Insurance Plans for Federal Employees. published by a consumers organization. outlines plan features and costs, gives general advice on picking a plan, and provides "consumer satisfaction surveys." These surveys rate plans in such areas as the ease of getting appointments, access to specialists. 21 waiting times in doctors' offices and the quality of care. The National Association of Retired Federal Employees also provides information and recommends plans for retirees with particular ailments. The law governing the FEHBP is just 26 pages long. The program is run by OPM, which has authority to contract with plans for inclusion in the system. OPM's small staff "negotiates" prices and benefits - best characterized as "jawboning" - but does not impose price controls, fee schedules or benefit requirements. OPM also provides beneficiaries each year. before an "open season". with standardized information on costs, services etc. for each plan, and a form for indicating their choice. The authors envision a new Medicare system structured much like the successful 788000 FEHBP. but with modifications to refine the government contribution, to make the new program even less susceptible to destabilizing adverse selection than the FEHBP, and to provide beneficiaries with better information on which to make choices. This new program would have four core elements. 1) Entitlement to a defined contribution. Elderly and disabled Americans would have an entitlement not to a defined set of benefits, but to a voucher worth an amount based on a number of factors. The total expenditure on the voucher system would be limited to a program budget, with the voucher amount adjusted each year according to the budget. 3 2) Voucher amount. The base for the voucher would be budgeted Medicare expenditure (the combined net expenditures on part A and B) divided by the eligible population. This base would then be adjusted up or down according to three categories. The first would be primary risk factors, namely age, sex, reason for eligibility (age or disability), institutional status and ESRD status. The second would be an income adjustment applied to one-third of the voucher, to be the equivalent of means-testing today's Part B premium. The third would be a local market variance, to reflect the weighted average enrollee cost of a "basket" of typical plans. We envision the basket as comprising "typical" plans, such as the Medicare Standard Plan (see below), a catastrophic/MSA plan, a Blue-Cross standard plan, and a comprehensive HMO plan. This is a refinement of the "big six" formula used by OPM to set the government contribution to the FEHBP. 3) Standards of participation. To be permitted to sell to Medicare participants. plans would have to meet certain threshold requirements. Beyond these they could offer varieties of benefits and delivery systems. There would not be a restriction on the number of plans - what one might call an "any willing plan" arrangement. The requirements would include basic solvency standards and a service area acceptable to HHS. Plans would have to provide medically necessary acute care services (hospital and physician) and catastrophic coverage for included services. In addition, plans would have to provide HHS with standardized information on benefits, rates, and supply consumer 4 information as determined by a consumer advisory board. Plans would have to accept any Medicare-eligible person during the annual open season, and specify premiums using limited underwriting principles equivalent to the actuarial categories used to determine the voucher amount (that is, age, institutional status etc.). 4) The government's role. In the new system, HHS would no longer regulate the prices charged by providers, and instead would take on functions more like those carried out by OPM in the FEHBP system. The government would have three important roles: a) The government would establish a federal corporation, governed by an appointed board, to run a Medicare Standard Plan similar to the current Medicare program. The Standard Plan would be available in all markets and the board would set premium prices to meet long-term solvency requirements. Subject to congressional approval, the board could adjust benefits, out-of-pocket costs and payment levels in the Standard Plan. b) HCFA would calculate the voucher amount for each beneficiary, setting that amount after the plans had filed their price and benefit information for the following year. c) HHS would conduct a Medicare open season, much as OPM does for the FEHBP. Before open season. each Medicare beneficiary would receive an information kit from HHS, including the amount of their voucher and the standardized information on 5 prices, benefits and consumer satisfaction for Medicare-approved plans in their area. including the Standard Plan. Beneficiaries would also receive a selection form on which to indicate their choice. Once the selection had been made, HCFA would send the beneficiary's voucher to the chosen plan. The beneficiary would be responsible for any difference between the voucher and the premium costs, but could elect to have the government pay that difference and reduce the beneficiaries Social Security check (similar to the part B option today). If the voucher amount exceeded the plan premium, the difference would be deposited by HCFA into a Medical Savings Account of the beneficiary's choice. Disbursements from MSA accounts could be used only for medical expenditures eligible for the Schedule A tax deduction. Thus Medicare would operate much like the FEHBP serves retired federal workers and retirees. Medicare beneficiaries would be able to pick a private plan which included the services they wanted (beyond the core package), delivered in the way they wanted, and, if they wished, perhaps through an organization with which they were affiliated (as many FEHBP enrollees do). Or they could decide to put their voucher towards the premium of the Medicare Standard Plan. Because beneficiaries would receive a voucher of a specific amount (paid directly to the plan of their choice), they would have a strong economic incentive to pick the plan that best met their objectives of price, quality, and services. 6 The organization of services, the selection of benefits, and payments to providers would be in the hands of the plan managers competing for enrollees. Unlike the federal officials managing Medicare today, these managers would have the freedom and the financial incentive to experiment with new ways to deliver care at a competitive price. And the voucher approach, in contrast to the AAPCC payment system, would give all plans an incentive to strive for the best pricing, and not consider the voucher as a floor price. In stark contrast to today, HCFA would have no role in setting the provider reimbursement rates, deductibles, or cost-sharing levels of any private plan, nor any role in requiring benefits. However, while the authors remain skeptical of the price-maker theory of the FEHBP's success, HCFA could carry out for Medicare the "jawboning" role in premium setting and agreeing on service areas that OPM undertakes for the FEHBP. Since Medicare would become a defined contribution program, the government's share of Medicare spending could be held in check by a budget. But would this merely shift costs to beneficiaries because average premiums grew much faster than the voucher? There are good reasons to believe that the incentives in the new Medicare system would. in fact, force a moderation of health care costs for Medicare beneficiaries. Both the CRS and Lewin-ICF reported in their studies that FEHBP cost increases were lower during the 1980s than those of the private sector. More recently, FEHBP beneficiaries have seen a 7 drop in average premiums. The GAO notes also that programs with some similarities to the FEHBP. such as CalPERS, have succeeded in controlling premium increases. Even though adverse selection is not a severe problem in the FEHBP, the design of the proposed Medicare program would be even less susceptible to that problem. The main reason for this is that plans could vary their premiums according to age, institutional status and other major risk factors, which FEHBP plans cannot do. The ability to vary premiums in this way would also help protect the "residual" Medicare Standard Plan from adverse selection. And because private plans could vary benefits and would provide catastrophic coverage, it is by no means clear that sicker individuals would gravitate to the Standard Plan. dogor A refined FEHBP structure for Medicare would also provide beneficiaries with far more usable information for choosing plans than they are given to pick doctors or hospitals today. With the advice of a Consumer Board, HHS and private organizations could circulate before Open Season such standardized price and benefit information on available plans, benchmark treatment costs, and patient report cards. The proposed Medicare program would have implications for some existing health programs and plans. It would be wise to eliminate the overlap between Medicare and the FEHBP. to avoid the potential for adverse selection effects by enrollees (a feature of overlapping eligibility today). One way to do this would be to give federal retirees a 8 one-time choice between Medicare and the FEHBP when they become eligible for Medicare. Another option would be to end FEHBP eligibility for those eligible for Medicare. The authors would expect the Medigap market to shrink considerably, and perhaps disappear, under a reformed system based on our proposal. The main reason for this is that the primary function of Medigap - virtual catastrophic protection - would become a standard feature of all plans. Another attractive feature of Medigap coverage today is that it allows Medicare beneficiaries to purchase additional insurance for certain services, such as prescription drugs or preventive care, on top of their Medicare coverage. But such additional features typically would become optional features of Medicare- approved plans under the reform. Further, Medigap plans today must by law reimburse Medicare beneficiaries from certain routine copayments (and many categories of plan must also reimburse deductibles). This reduces cost consciousness. No such utilization- increasing requirements would accompany to plans offering additional services in the system we propose. The proposed Medicare reform likely would not have a significant impact on Medicaid, however. whether or not that program undergoes significant reform this year. Just as states may today "buy in" to Medicare, states could decide to supplement a Medicaid-eligible person's Medicare voucher to enable that person to afford coverage under a certain Medicare plan. 9 DRAFT July 7, 1995 PRESERVING AND STRENGTHENING MEDICARE Summary Marilyn Moon* and Karen Davis** * Senior Fellow, Urban Institute, 2100 M Street, N.W., Washington, D.C. 20037 ** President, The Commonwealth Fund, 1 East 75th, New York, N.Y. 10021 Paper prepared for the American Association of Retired Persons conference on The Future of Medicare, Washington, D.C., July 20-21, 1995. The views are those of the authors and not the institutions with which they are affiliated. PRESERVING AND STRENGTHENING MEDICARE Marilyn Moon and Karen Davis As Medicare turns thirty, it is an opportune time for a critical reexamination of the program. Medicare has opened the door to health care and greater economic security for the nation's elderly and disabled populations. It has improved access to health care and contributed to improved health and quality of life of millions of vulnerable Americans. Without Medicare, many of its chronically ill beneficiaries would quickly exhaust their financial resources. Changes in the program, however, are dictated by the projected insolvency of the Medicare Hospital Insurance Trust Fund in 2002, the rapid growth in Medicare outlays in an era of federal budget deficits, and the looming retirement of the baby boom generation. Medicare now faces twin problems -- brought on in part by its own success -- of continuing to guarantee access to care and financial security for its beneficiaries while stemming an unsustainable rate of cost growth. Grappling with the choices will be extraordinarily difficult. Yet, a reexamination affords an opportunity for creative restructuring of the program to meet the growing health and long- term care needs of an aging population, while being cognizant of competing demands for the nation's health care and budgetary resources. This paper argues that it is possible to build on the strengths and structure of the Medicare program, while undertaking long-term changes to assure its fiscal soundness and preserve its mission of affording health security for an aging population in the 21st century. We attempt to offer pragmatic suggestions on how to improve the program while recognizing the substantial fiscal responsibilities that will need to be faced. THE ROLE OF MEDICARE Any discussion of Medicare restructuring should begin with an understanding of why it came to exist, its role in assuring the well-being of America's oldest and most disabled citizens, and its strengths as a program. as well as areas which require strengthening. Medicare is the largest public health care program in the United States, providing the major source of health insurance for acute care for 37 million elderly and disabled beneficiaries. It came into being in 1965 because the nation's elderly lost their health insurance when they retired. Without health insurance, the nation's elderly and their families were at risk for financially ruinous health care expenses. Why is Medicare so costly? The answer lies not in inefficiency, or a failure to adopt modern techniques of cost-savings. Its administrative costs, at two percent of program outlays, are far lower than those of private insurance and managed care plans.¹ It is a leader in electronic submission of claims -- 90 percent for hospital services and 67 percent for physician services -- I Health Care Financing Administration, Medicare: A Profile (Washington, D.C.: HCFA, 1995). I far in excess of private health insurers.² It imposes price discounts on hospitals and physicians unmatched by even the best managed care plans. It offers beneficiaries a choice of enrollment in health maintenance organizations (HMOs) or coverage under fee-for-service with unlimited choice of physician; 70 percent of all HMOs offer Medicare risk contracts.³ In the 1980s Medicare expenditures per beneficiary grew more slowly than private health insurance per enrollee. After adjusting for the packages of health services covered by both private insurance and Medicare, Medicare has done better than private insurance between 1990 and 1992 and nearly as well in 1993. In short, Medicare's costs have increased for the same reasons that health care costs generally have risen: health and quality of life enhancing technological change, excess capacity in the health care system, overspecialization, and an open- ended payment system that encourages doing more. STRENGTHS OF THE MEDICARE SYSTEM Medicare has many strengths that deserve to be retained, including some of the basic principles on which the program was founded. Universal Social Insurance Medicare provides universal health insurance coverage to nearly all people age 65 and over and those who are permanently and totally disabled for two years or more. At 98 percent of all the elderly, for example, this represents a percentage far in excess of the rate of coverage found for any other group in the population. By offering a uniform benefit package and ready access to most health care providers in the U.S., Medicare has achieved its promise of offering mainstream medical care even for the sickest and lowest income populations. Medicare's Risk Pools One of Medicare's strengths is its sharing of risks across a large group of the population. With 37 million beneficiaries, Medicare is the largest risk group for health insurance in the U.S. Effectively, costs for chronically ill 85 year-olds are averaged in with healthy 70 year-olds, making insurance less expensive than if these 85 year-olds were seeking insurance in the private market from a company that covered a much smaller number of persons. In addition, once a commitment has been made to treat all these beneficiaries alike, the costs associated with underwriting and differentiating risks can be foregone. Medicare's Discounts While Medicare has been criticized for not promoting aggressively enough managed care ²Health Care Financing Administration, Medicare: A Profile (Washington, D.C.: HCFA, 1995). ³Health Care Financing Administration, Medicare: A Profile (Washington, D.C.: HCFA, 1995). "Marilyn Moon and Stephen Zuckerman, "Are Private Insurers Really Controlling Spending Better than Medicare?" Discussion paper, Urban Institute, 1995. 2 alternatives for its beneficiaries, the program is itself similar to a preferred provider managed care plan. With the recent reforms in provider payment, Medicare sets prospective prices for hospitals and physicians at a substantial "discount" to usual charges. Medicare's physician payment fees, for example, average 68 percent of fees paid under private health insurance plans. 5 All providers who are willing to participate at these rates are permitted to enroll. Even when compared with formal preferred provider organizations (PPOs)--which is the area of managed care now growing fastest in the U.S.--Medicare's payment levels remain substantially lower. Similarly, hospital payments are below costs while private insurers continue to pay a premium for care. Medicare has already achieved the discounts that are just now helping to hold down premiums for private insurers. BUILDING ON MEDICARE'S STRENGTHS At present, too little attention is being focused on how to improve the functioning of the basic Medicare program, rather than departing radically from its basic structure. The goal should be preserving genuine choice for all Medicare beneficiaries to be cared for by physicians or a health system of their choice while guaranteeing quality care at a reasonable cost to beneficiaries and to taxpayers. Fee-for-service care has the disadvantage of creating incentives for too much care at too high cost; capitated managed care has the disadvantage of creating incentives for too little care at substandard quality. Providing a genuine informed choice for beneficiaries of both options may counter the harmful consequences of either extreme. Major issues include: 1) how to improve the fee-for-service option within Medicare; 2) how to expand Medicare managed care choices while assuring quality standards; 3) how to minimize the difficulties posed by risk selection; and 4) what financial contribution Medicare beneficiaries and taxpayers can reasonably be expected to make. Improving Medicare's Fee for Service Option Medicare's greatest strength as a social insurance program is the efficiency of its fee for service option. It has low administrative costs and low provider payment rates. It could appropriately be characterized as a PPO that takes any willing provider that meets quality standards. And it does so without sacrificing beneficiary choice of physician. It is not surprising that Medicare beneficiaries rate it highly, and that it has broad popular support. Where the fee for service option falls short is the adequacy of its benefit package, and the difficulty of coordinating benefits with Medicaid and private supplemental coverage. This creates complexity and confusion among beneficiaries and providers. Any comprehensive examination of options for Medicare restructuring should include: ⁵Physician Payment Review Commission, Annual Report to Congress (Washington, D.C.: PPRC, 1995). ⁶Stephen Zuckerman and Diana Verrilli, "The Medicare Relative Value Scale and Private Payers: The Potential Impact on Physician Payments." Prepared under Health Care Financing Administration Contract No. 500-92-0024, D.O. #4, 1995. 3 Merger of Medicare Part A and Part B to simplify financing and administration of the program. Creation of two standardized Medicare benefit packages: the current benefit package with a unified deductible and ceiling on out-of-pocket costs and a comprehensive benefit package that covers current services with little or no cost-sharing and prescription drugs.⁷ Beneficiaries preferring the comprehensive benefit package would pay an additional premium to cover the differential cost.⁸ Federalization of Medicaid supplemental acute care coverage for Medicare beneficiaries; Medicare cost-sharing and premiums for poor and near-poor Medicare beneficiaries would be financed from federal general revenues and administered jointly with Medicare.⁹ Improvements in Medicare's fee-for-service environment also requires attention to further cost containment activities. This part of the program needs to improve its payment and oversight activities to remain a viable option. In some cases, Medicare could adopt innovations from the private sector to improve its operations. Thus, changes in Medicare could include: Expansion of prospective payment methods to all Medicare benefits, including consideration of expenditure targets that link future prices to performance in controlling the outlays Use of sophisticated computer techniques for profiling use of services, and thus identifying outliers and possible abuse Establishing and applying strict principles for appropriateness of care High cost case management. Expanding Medicare's Managed Care Options Medicare currently permits federally qualified HMOs and state-accredited competitive health plans to enter into Medicare risk contracts. Plans may make a profit on their Medicare beneficiaries, but only to the extent to which they make a profit on non-Medicare enrollment. Any additional savings must be returned to beneficiaries in the form of improved benefits, ⁷A voluntary long term care insurance supplement financed by an income related premium might be another option that could be considered as well. For further description, see Karen Davis and Diane Rowland, Medicare Policy: New Directions for Health and Long-Term Care (Baltimore, Md.: The Johns Hopkins University Press, 1986). ⁸While such an option has considerable appeal to allow beneficiaries who wish to do so to opt for just one simple public program, there is the danger that sicker beneficiaries will opt for comprehensive benefits while healthier beneficiaries select basic coverage. If so, premiums will need to be risk adjusted--encountering the same methodological problems that managed care options create. See Marilyn Moon, "Adding a Long Term Option to Medicare," in A Call for Action, Supplement to the Final Report, The Pepper Commission. U.S.G.P.O., September 1990. °For more details see Diane Rowland and Barbara Lyons, Medicare's Poor (Baltimore, Md.: Commonwealth Fund Commission on Elderly People Living Alone, 1987). 4 reduced cost-sharing, or reduced premiums. Medicare could take several steps to expand its managed care options: Including PPOs and POS plans that meet quality and fiscal soundness standards and provide either the basic or comprehensive benefit package Informing Medicare beneficiaries of all managed care plan options available in their geographic area, providing beneficiaries with information on which to make an informed choice, and having beneficiaries enroll annually in either Medicare's fee for service option or a managed care option Beneficiaries could receive all or a portion of the savings of managed care plans with premiums below the Medicare fee for service option or pay all or a portion of the incremental cost of managed care plans with premiums above the Medicare fee for service option. Minimizing Risk Selection The primary problem with expanding Medicare managed care options is that it creates incentives to sort out healthier risks into managed care plans while poorer risks elect the fee for service option. If beneficiaries bear all of the financial cost of higher premiums, sicker, chronically ill patients may bear additional financial burdens not because they enroll in a less efficient plan but because the kind of care they need is only available in the fee for service option. The community rated character of Medicare could be destroyed, with each subgroup bearing its own risk. Given the extreme variability in health outlays among Medicare beneficiaries, there is great leeway for plans to select relatively healthier beneficiaries for whom capitated rates exceed true costs. If managed care plans succeed in attracting and retaining relatively healthier Medicare beneficiaries which they have very strong incentives to do, Medicare will be overpaying for those under managed care, and yet paying the full cost of the sickest Medicare beneficiaries who are unattractive to managed care plans. Medicare HMOs currently have the option of switching to a fee-for-service method of payment from a capitated risk contract if they 1: 7 experience adverse selection and would receive higher payment under Medicare's cost reimbursement rules. Monthly disenrollment by Medicare beneficiaries also means that managed care plans can encourage sicker patients to leave the plan and be cared for on a fee-for- service basis. In the case of network-model HMOs the same physician might even continue to care for the patient when he or she disenrolls. This is such an overwhelming downside that it warrants a gradual approach to managed care, gaining experience with mechanisms used by plans to avoid bad risks. The following approaches may help to minimize adverse risk selection: Insurance market reforms such as open enrollment, same premium for all enrollees (or with age-gender adjustment only), no underwriting or exclusion of poor risks Regulation of marketing, enrollment, and disenrollment practices with Medicare managing the enrollment process and providing standardized information on plan choices Quality standards that assure that chronically or seriously ill patients get appropriate 5 specialty care Requiring all managed care plans to enter into risk contracts. The most fundamental way of eliminating risk selection is setting capitation rates in a way that reflects the risk of enrolled populations. This is easily said, but we do not do this well now and there is, as yet, no consensus on how to solve this problem. The current method of paying HMOs for Medicare patients is seriously flawed. Its primary weakness is that it does not adequately adjust for differences in the health status of beneficiaries. Unfortunately, a good method of setting capitation rates to adjust for differences in beneficiary health status seems years away. Financing Options These improvements in Medicare's fee for service and managed care options may not generate substantial savings over the long term. They represent program improvements, and should be considered on their merits quite apart from their savings potential. Yet, their extensive uncertainties make them an unreliable foundation on which to assure the future fiscal soundness of the Medicare program. There are two basic alternatives for financing Medicare benefit payments: direct beneficiary contributions and taxes. Ultimately, this is a public policy choice that the nation must make. There is no magic right combination of beneficiary financial responsibility and governmental financial responsibility. If beneficiaries are asked to pay more, some options are more equitable than others. Consider cost sharing changes. While cost sharing makes theoretical sense as a means for controlling use of services, in practice it often results in lower use of services only for low income families 10 In the case of Medicare beneficiaries, higher cost sharing will likely have only an indirect effect on most people since they have additional supplemental insurance that would insulate them for the direct impact. In addition, Medicare coinsurance and deductibles represent a complicated collection of mismatched requirements, some of which could be increased and some of which should be decreased. For example, the Part A hospital deductible ($716 in 1995) is very high by any standards while the Part B deductible ($100) is quite low. Hospital and skilled nursing coinsurance are also unreasonably high. It makes sense to rearrange this cost sharing along more rational lines, with perhaps a small net increase. 11 Premiums for Medicare beneficiaries could be increased, and this is a reasonable way to require higher contributions if special efforts are made to protect low income beneficiaries. In particular, the Qualified Medicare Beneficiary (QMB) and the Specified Low Income ¹⁰Newhouse, Joseph. Some Interim Results from a Controlled Trial of Cost Sharing in Health Insurance. The RAND Corporation, Santa Monica, 1982. ¹Marilyn Moon, Medicare Now and in the Future (Washington, D.C.: Urban Institute Press, 1993). 6 Beneficiary (SLIMB) programs should be moved into Medicare. 12 This would likely help with low participation rates in this program. In addition, the eligibility cutoffs could be raised if premiums go up to ensure that moderate income families would not be hit too hard. An income related portion of a premium increase also could be considered¹, but there are practical implementation problems that may render this a less desirable initial step. Implicitly, Part A already has an income-related premium via the taxation of Social Security benefits, a portion of which is dedicated to the Part A trust fund. This element could be retained, not eliminated as some have suggested.¹⁴ It seems likely, however, that the fiscal soundness of Medicare can only be guaranteed as the baby boom population reaches retirement if the tax base of the program is improved. To be dynamically sound, payroll tax revenues will need to be supplemented with other sources of revenues that grow with the aging of the population, such as premiums paid by beneficiaries. The regressive shift in financing by a greater reliance on premium financing could be offset by subsidies for poor and near poor beneficiaries, or through income-related premiums. With merger of Part A and Part B, a single trust fund could be established to receive payroll tax. premiums, and general revenue contributions. One of the most fundamental long-term issues that must be considered is the extent to which costs of retired persons are borne by current workers. By the year 2030 under current projections, there will be two covered workers for every Social Security beneficiary. 15 The cost of Social Security and Medicare per worker could be staggering. The age of Social Security eligibility is gradually being increased to age 67; Medicare could do the same. The difficulty, however, is that those who are involuntarily retired because of limited job opportunities or health reasons can take Social Security at age 62, but at reduced actuarial rates. Medicare currently has no such option, and many early retirees become uninsured and are at great risk. 16 If Medicare retirement age were to be increased, consideration would need to be given to permitting early "They are currently part of the Medicaid program and likely to be at considerable risk if Medicaid becomes a block grant or even if states are given more discretion concerning coverage and eligibility. See Marilyn Moon, Medicare Now and in the Future (Washington, D.C.: Urban Institute Press, 1993). "Karen Davis and Diane Rowland, Medicare Policy: New Directions for Health and Long-Term Care (Baltimore, Md.: The Johns Hopkins University Press, 1986). "Marilyn Moon, "Taxation of Social Security Benefits," Hearing before the U.S. House of Representatives Committee on Ways and Means, January 19, 1995. is 1995 Annual Report of The Board of Trustees of the Federal Old Age and Survivors Insurance and Disability Insurance Trust Funds, April 3, 1995. ¹⁶Karen Davis, "Uninsured Older Adults: The Need for a Medicare Buy-in Option," testimony before the U.S. House of Representatives, Committee on Ways and Means, Subcommittee on Health, June 12, 1990; and Marilyn Moon. "Expanding Medicare Coverage to the Near Elderly," testimony before the U.S. House of Representatives Committee on Ways and Means, Subcommittee on Health, March 19, 1991. 7 retirees to purchase Medicare on a subsidized basis. Expansion of job opportunities for older people, or opportunities to earn Medicare premium credits through voluntary service to their communities should be considered. CONCLUSION What should be preserved is the essential role that Medicare plays in guaranteeing access to health care services and protecting from the financial hardship that inadequate insurance can generate for our nation's most vulnerable elderly and disabled people. No American should become destitute because of uncovered medical bills nor be denied access to essential health care services. Medicare is a model of success. It should not be hastily jettisoned in an ill-conceived and short-sighted effort to obtain federal budgetary savings. Instead a full array of options needs to be carefully analyzed, critiqued, and debated. We are pleased to be a part of this conference on the Future of Medicare that helps set this public debate in motion. - 8 Bill Frist U.S. Senator (R-TN) ill Frist is the first practicing physician elected to the U.S. B Senate since 1928. Elected in 1994, Senator Frist serves on the Senate Labor and Human Resources Committee; the Banking, Housing and Urban Affairs Committee; the Budget Committee; and the Small Business Committee. A nationally acclaimed heart surgeon, Senator Frist started the Vanderbilt Transplant Center at the Vanderbilt University Medical Center in 1986. He served as director of the Heart and Heart-Lung Transplantation Program prior to his legislative campaign. At Vanderbilt, Senator Frist pioneered innovative diagnostic techniques and performed the first pediatric heart transplant and first lung transplant in Tennessee. He is the author of more than 100 articles and abstracts on medical research and policy. His book, Transplant, published in 1989, examines the social and ethical issues of transplantation. Senator Frist also served on the faculty at Vanderbilt University Medical Center for eight years. In 1992, he organized a statewide grassroots campaign to return the organ donation card to the Tennessee driver's license, and received the Distinguished Service Award from the Tennessee Medical Association for his efforts. Also in 1992, he was asked to serve as chairman of the Governor's Medicaid Task Force. Jim McDermott U.S. Representative (D-7th WA) im McDermott has served in the U.S. House of Representatives J since 1988. Rep. McDermott is especially interested in health care issues. In Congress, he is active in health care reform issues, and he was appointed to the Ways and Means Committee's Health Subcommittee in 1992. He also founded and chairs the Congres- sional Task Force on International HIV/AIDS and introduced the AIDS Housing Opportunities Act, a new program enacted into law in 1990 authorizing $156 million in FY 92 for special housing assistance for people with AIDS. Rep. McDermott, the co-author of single payer health care legislation, is leading the fight in the House of Represen- tatives for guaranteed comprehensive health care coverage for all Americans. Before being elected to Congress, Rep. McDermott, a physician, served as a Foreign Service medical officer based in Zaire, providing psychiatric services to Foreign Service, AID, and Peace Corps personnel in sub-Sahara Africa. Prior to going to Africa, he spent 15 years as a legislator, first as a state representative from the 43rd legislative district in Washington, and then as a state Senator. While in the state legislature, Rep. McDermott developed the Washington Basic Health Plan, the first state program in the country to provide low-cost health insurance to the unemployed and working poor. COMMENTS ON BUTLER/MOFFIT PAPER "THE FEHBP AS A MODEL FOR A NEW MEDICARE PROGRAM" Prepared by: Peter D. Fox When Medicare was enacted in 1965, its structure reflected the prevailing Blue Cross and Blue Shield plans in effect at that time. For example, the typical plan covered only 60 or 90 days of inpatient care. Enrollees faced at most a fixed deductible, and the plan paid hospitals directly based on their actual costs. Private sector practices, both purchaser and provider, have changed dramatically since then. Benefit packages are structured differently than they were 30 years ago. Medical technology has advanced considerably, vastly increasing the dilemmas surrounding payment of high cost, marginally effective technologies. And perhaps most importantly, managed care -- hardly known in 1965 -- has become the norm in employer-based coverage. Increasingly, the employed population does not have access to indemnity coverage. Medicare, however, has changed little. To be sure, coverage has been extended to a limited number of disabled; some benefits such as hospice have been added; and new fee-for-service based payment methodologies, such as for hospitals and physicians, have 2 been adopted. These, however, do not represent fundamental changes. Medicare has contributed much to the financial and medical security of 38 million Americans, a trite thought but one that can be easily forgotten. Beyond that, it has been successful in some respects but not others. In my opinion, these successes and failures are intrinsic in the current fee-for-service based program. On the positive side, the cost of administration is low. Medicare operates with administrative costs of roughly 2 percent. In contrast, the annual reports of publicly traded managed care companies commonly proclaim retention rates of more than 20 percent. To be sure, the HMOs bring value in terms of how they organize the delivery system -- but at a cost. Where Medicare has largely failed has been in controlling the delivery system. In short, the Medicare program is able to address price -- albeit awkwardly at times -- but not utilization of services, and the government knows little about the quality of care that is delivered. While Medicare could rely more on managed care under its fee-for-service program, I believe, in common with the authors of the paper, that a federal agency is institutionally incapable of entering into many of the types of arrangements that private sector plans have adopted. The government has difficulty recruiting large 3 numbers of technically qualified staff, its decision processes are overly slow, and society imposes due process requirements on government agencies that far exceed those under which private businesses operate. Also, government officials are often constrained in making judgments and look bad when policies are not administered uniformly. Examples of private health plan activities that government would have difficulty emulating include how physicians are selected and deselected, the nature of the information providers receive on their practice patterns and how to change them, and the highly imaginative and varied reimbursement structures that exist. Butler and Moffitt have proposed a major restructuring of Medicare entailing, first, moving from a defined benefit to a defined contribution approach and, second, emulating the Federal Employees Health Benefits Program (FEHBP). I regard these two issues as separable. The major advantage of a defined contribution program is that it would limit the financial exposure of the federal government. However, it would transfer that exposure to the beneficiary. The authors have confidence that adopting the FEHBP model will generate efficiencies that will allow existing benefit levels to be sustained, albeit in a more managed and restricted environment. They may be correct, but I view their approach as speculative. Moving to define contributions is simply 4 not necessary to bring about the changes in the relationships between the Medicare program and the delivery system that they seek, including enhancing competition and consumer choice. Their more interesting proposal is the movement to the FEHBP model, fundamentally a voucher approach that would allow a wide variety of health care plans to participate, competing against each other for enrollment and charging premiums in excess of the government contribution. I believe that the approach has merit, but only if some issues are attended to. I would like to offer five points. The first question is whether significant biased selection can be avoided if broad latitude of plan choice is allowed, ranging from highly limited coverage to very comprehensive. The factors that they suggest to correct for biased selection will not compensate for the phenomenon of individuals in a given demographic grouping electing coverage that is limited to catastrophic expenses if they are well, electing drug coverage if they have a chronic condition, and electing dental coverage if they have bad teeth. Certainly, the current FEHBP has not solved the adverse selection problem, particularly with regard to the choice between high and low options, although arguably it has not tried very hard. 5 Second, their model requires informed consumer choice. They, appropriately, call for extensive processes for collecting and disseminating information. However, I believe that evidence from the existing Medigap and HMO markets, and from the FEHBP program itself, suggest that considerable consumer confusion will occur if benefits are not standardized. Third, they have not addressed the issues of transitioning from the current system. Does one simply announce that at some not- too-distant date all beneficiaries will be required to select among plans, which are presumed to exist and be capable of absorbing the new (and, I might add confused and scared) enrollees? Furthermore, the impact of having to switch plans and physicians or face significant financial penalties holds far more trauma for the typical 80 year old than the 40 year old. It is nice to have a vision of paradise. But, is there a way to get from here to there? Fourth, to a significant degree the authors have assumed away realities related to the functioning of government. With far larger stakes than at present, issues such as the drawing of geographic boundaries for purposes of setting contribution rates and the choice of measures to address biased selection have the potential for becoming highly politicized. For example, moving from counties as a basis for determining contribution levels to market areas would, on 6 average, increase payments on behalf of suburban residents at the expense of the inner-city. This is hardly a noncontroversial shift. Also, the authors anticipate that their proposal will restrain costs as a result of the government bargaining premium levels with individual plans; they have more faith than I that the government is capable of bargaining with the myriad of plans that are expected to be offered in a manner that is informed, effective, and devoid of politics. They propose to address this matter, in part, by having the Medicare program administered by a new federal corporation. However, one must ask what this accomplishes besides shifting the locus of bureaucracy. Finally, the authors appear to have an insufficient understanding of the existing Medicare HMO risk contracting program, and this misunderstanding colors their thinking. I suspect that current HMO risk contractors would be relieved to hear that they do not compete on price or benefits, as the paper suggests. However, such is simply not the case. The Medicare program in effect provides a voucher based on estimated fee-for-service expenditures that results in intense competition among plans. Second, although still small, the Medicare risk program has been growing rapidly, with the number of participating plans increasing 40 percent this past year, and the number of enrollees 30 percent. Arguably, the rate setting methodology is flawed because of its reliance on fee-for-service 7 expenditures, which vary widely and unaccountably geographically. Also, many, including myself, view current law as overly rigid in some of its requirements on HMOs such as not allowing them to offer PPO-type arrangements. Another change I would recommend is to remove the requirement in current law that at least 50 percent of enrollees in HMOs with Medicare risk contracts be other than Medicare or Medicaid. This change would foster competition and increase the plan choices available by allowing, in particular, integrated delivery systems to serve Medicare beneficiaries without having, also, to market to employers, most of whom are not interested in contracting with new plans. In conclusion, I believe that the Medicare system needs to change and share many of the objectives of the authors of the paper. However, there are ways in which the existing structure can be built upon to enhance competition without suddenly requiring that some 38 million aged and disabled Americans make choices that they fear and ill-understand. THE HENRY ]. KAISER NEWS RELEASE FAMILY FOUNDATION EMBARGOED FOR RELEASE UNTIL: CONTACTS: Wednesday, July 19, 1995, 9:30 am EST Kaiser Family Foundation Matt James (415)854-9400 Chris Ferris (202)347-5270 The Urban Institute Marilyn Moon (202)857-8691 Stephen Zuckerman (202)857-8679 NEW STUDY FINDS SIMILAR PER CAPITA GROWTH RATES IN MEDICARE AND PRIVATE INSURANCE WASHINGTON, D.C. A new study conducted by the Urban Institute and supported by the Kaiser Family Foundation reports that per capita spending for services provided by both Medicare and private insurers grew at about the same rate from 1984 to 1993. As policy makers struggle to reduce Medicare spending by $270 billion over the next seven years--the figure in the budget plan approved by the Congress last month--some believe that Medicare should follow the lead of private insurers because the private sector is viewed as more successful in controlling costs. This study examines Medicare's performance and sheds new light on how Medicare's growth rate compares to that of the private sector. Designed to compare growth across a consistent set of services, the study reports that in 1993, per capita expenditures for Medicare grew by 7.4 percent, relative to 7.1 percent for private health plans. When comparing Medicare and private insurance over time, Medicare shows a slower growth rate than the private sector in eight of the ten years ending in 1993. The authors note that preliminary data indicate that the private sector outperformed Medicare in 1993 and 1994, but are unsure whether this constitutes a trend. "History shows that those who search for a magic bullet to control health care spending are almost always disappointed," said Drew E. Altman, president of the Kaiser Family Foundation. "The private sector is a good source of ideas for Medicare reform, but this study suggests caution in comparing the growth rates of Medicare and the private sector." The authors of the Urban Institute study, Marilyn Moon and Stephen Zuckerman, warn that measuring aggregate costs is not the most accurate way to compare Medicare and the private sector. This comparison, the authors explain, accounts neither for changes in the number of people insured from one year to the next nor for critical differences in services covered under Medicare and private health plans. Moon and Zuckerman calculate the cost per enrollee for both Medicare and private insurance to adjust for the 2 percent rise in Medicare beneficiaries in recent years. They also limit their analysis to categories of health care services covered by both. The authors calculate the following differences in --more-- 2400 SAND HILL ROAD MENLO PARK, CALIFORNIA 94025 415 854-9400 FAX 415 854-4800 WASHINGTON OFFICE: 1450 G STREET NW, SUITE 250 WASHINGTON, DC 20005 202 347-5270 FAX 202 347-5274 spending growth for 1993, the most recent year for which the government's national health expenditure data are available: Using aggregate spending, the growth in Medicare expenditures in 1993 was 11.6%, compared to 6.6% for private plans. Comparing per capita spending, which accounts for changes in the number of people covered by health insurance, the differences narrow-to 9.5% for Medicare and 7.1% for private plans in 1993. Limiting the comparison to similar health care services (by eliminating home health and skilled nursing facility services under Medicare and outpatient prescription drugs under private plans) shrinks the difference even further--with Medicare and private insurance growing at 7.4% and 7.1%, respectively, in 1993. Some analysts suggest that home health and skilled nursing facility services should not be excluded from the analysis because they are integrally linked to the growth in Medicare spending and have served as substitutes for inpatient hospital care. According to the study, even when these services are included, the per capita difference between Medicare and private plans was only about 2.5 percent in 1993. However, Moon and Zuckerman note that it is appropriate to exclude these services because they are not typically used by younger people insured by private plans. Moreover, they explain that home health and skilled nursing facility benefits have been used in recent years to provide long term care-type benefits, rather than as substitutes for inpatient hospital care. The analysis was based upon data from the National Health Expenditure (NHE) accounts collected by the Health Care Financing Administration, which draw on actual public spending and a variety of surveys of the private sector's health care providers. The Kaiser Family Foundation, based in Menlo Park, California, is a non-profit, independent national health care philanthropy and is not associated with Kaiser Permanente or Kaiser Industries. The Foundation's work is focused on four main areas: health policy, reproductive health, HIV, and health and development in South Africa. A copy of this report, "Are Private Insurers Really Controlling Spending Better than Medicare?" is being sent to you via regular mail. To receive a copy next day, members of the media may call the Kaiser Family Foundation's publication request line at 1-800- 656-4533 (order #1072). To receive a copy by messenger in the Washington, D.C., area call the Foundation's Washington office at (202)347-5270. Additional copies of the report are available by regular mail by calling 1-800-656-4533. THE HENRY J. KAISER FAMILY FOUNDATION ARE PRIVATE INSURERS REALLY CONTROLLING SPENDING BETTER THAN MEDICARE? Prepared by Marilyn Moon and Stephen Zuckerman of the Urban Institute Prepared for The Henry J. Kaiser Family Foundation July 1995 This research was supported by a grant from the Henry J. Kaiser Family Foundation in Menlo Park, California. The contents of this paper are solely the responsibility of the authors and do not neccessarily represent the views of the Urban Institute or the Henry J. Kaiser Family Foundation. The authors are grateful for the helpful suggestions of John Holahan and Len Nichols, and for the assistance of Crystal Kuntz. The Henry J. Kaiser Family Foundation is an independent national health care philantrophy and not associated with Kaiser Permanente or Kaiser Industries. THE HENRY J. KAISER FAMILY FOUNDATION ARE PRIVATE INSURERS REALLY CONTROLLING SPENDING BETTER THAN MEDICARE? Prepared by Marilyn Moon and Stephen Zuckerman of the Urban Institute Prepared for The Henry J. Kaiser Family Foundation July 1995 This research was supported by a grant from the Henry J. Kaiser Family Foundation in Menlo Park, California. The contents of this paper are solely the responsibility of the authors and do not neccessarily represent the views of the Urban Institute or the Henry J. Kaiser Family Foundation. The authors are grateful for the helpful suggestions of John Holahan and Len Nichols, and for the assistance of Crystal Kuntz. The Henry J. Kaiser Family Foundation is an independent national health care philantrophy and not associated with Kaiser Permanente or Kaiser Industries. Executive Summary Reducing the rate of growth of the Medicare spending is an essential policy goal both for putting the program on firmer financial footing and for achieving the budget savings outlined in the recently passed budget resolution of the Congress. The targets established in that resolution-of $270 billion in savings over the next seven years--will require major changes in the Medicare program. In searching for ways to achieve savings, some have suggested adopting principles developed in the private insurance market. Indeed, many Americans seem to take for granted claims that Medicare spending is out of control and do not question comparisons suggesting large differentials between Medicare and private insurance spending growth. However, these comparisons prove difficult to interpret because the private insurance estimates use data from surveys of employer health plan costs that are often affected by shifts in enrollment across types of plans as well as changes in costs sharing, service coverage and utilization review that are hard to quantify. Instead, we use the National Health Expenditure (NHE) accounts data which capture spending by category of service for both Medicare and private insurance and these data allow us to make consistent adjustments. It is important to calculate growth per capita and to limit the data to categories where both Medicare and private insurance provide coverage of services. While these are simple adjustments, they have a dramatic impact on the comparisons shown in the accompanying chart. Viewed over the years 1990 to 1993 (the last year when the full NHE data set is available), Medicare spending grows substantially slower than spending from private insurance until 1992, then private insurance gains a 0.3% advantage. Looking down the chart, it is easy to see how much the numbers can differ depending upon how growth is measured. On a per capita basis, Medicare still fares well as compared to private insurance even when all categories of spending are included, but in the last two years, its growth is higher than the private sector. This differential largely occurs because home health services and skilled nursing services (both of which are less important to those covered by private insurance) have been growing very rapidly under Medicare. Comparison of Growth Rates Per Capita Growth in Services Covered by Both Medicare and Private Insurance 13.0% 12.0% 11.0% 10.0% Private 9.0% 8.0% 7.0% Medicare 6.0% 5.0% 1990 1991 1992 1993 Per Capita Growth in Total Expenditures 13,0% 12.0% 11.0% Medicare 10.0% 9.0% 8.0% Private 7.0% 6.0% 5.0% 1990 1991 1992 1993 Aggregate Growth in Total Spending 13.0% 12.0% Medicare 11.0% 10.0% 9.0% 8.0% Private 7.0% 6.0% 5.0% 1990 1991 1992 1993 The bottom graph shows aggregated growth rates--the least desirable way to compare Medicare and private insurance, although it is often used (Freudenheim 1995). Since 1990, the numbers of persons covered by private insurance has been dropping, while the number of Medicare beneficiaries has grown by about 2% per year on average. In that case, much of the positive differential between private insurance and Medicare aggregate spending growth is driven by how many people receive coverage, certainly not a factor that should be included in any measure of how well the private sector is controlling spending nor one that is often acknowledged. But even with the consistent data methods we use in making comparisons, several other factors need to be kept in mind in thinking about how much Medicare can benefit from adopting the techniques of the private insurance industry. For example, one way in which private insurers are slowing growth rates is to pay less for services than in the past. But since the level they start from is much higher than the levels that Medicare already pays, this source of savings would largely be unavailable to Medicare. For example, government studies show that in 1993 hospitals receive payments equal to 89 percent of their costs of treating Medicare patients as compared to 129 percent of their costs from private payers. Medicare was well ahead of private insurers in realizing it could benefit from low levels of hospital and physician payment. Finally, this study uses data on actual experience and thus does not capture information for 1994. Preliminary numbers for 1994 growth rates do indicate a stronger showing for private insurance relative to Medicare. But we do not yet know whether such results can be sustained over time since some of the savings being achieved in the private sector represent one time gains as employees shift from expensive to less expensive plans. While Medicare could also achieve such savings, they are likely to be short term in nature and, thus, would not necessarily result in the low rates of growth needed year after year to achieve the federal budget targets set for Medicare. ii INTRODUCTION Part of the current debate over the federal budget focuses on the rapid growth in Medicare spending. Further, Medicare's growth is being contrasted unfavorably with some estimates of growth in private health insurance spending.¹ The argument becomes that "the Medicare program could be rescued if only the Government would adopt some of the cost controls that employers have imposed on their workers under the banner of "managed care''' (Freudenheim, 1995). Before concluding that enormous savings are readily available by simply moving beneficiaries into private plans, however, it is important to take a closer look at the numbers people cite and what they mean. How fast is Medicare growing? Why is Medicare growing so rapidly? Is the rate of growth so different between Medicare and the private sector? Many Americans seem to take for granted claims that Medicare is out of control and do not question comparisons suggesting large differentials between Medicare and private insurance. But careful answers require consistent data between the public and private sectors, a reasonable time horizon for meaningful comparisons, and careful discussion of what the various numbers mean. At worst, the comparisons may reflect an "apples vs. oranges" problem in which comparisons are made on noncomparable data. But even when total comparability is not possible due to data limitations, it is still useful to look closely at Medicare and the private sector for some lessons for the future. 'See, for example, Hage and Black, 1995; and The Heritage Foundation, 1995. i:\bp\mmoon\growth\ 7/17/95 4:37pm This paper begins with a discussion of some of the most commonly cited numbers on the growth in Medicare and private insurance; finding a broad array of differences that make comparisons potentially misleading. We then turn to a more consistent set of comparisons based on the National Health Expenditure data that allow us to examine changes over time in a number of subcategories of health care services. We conclude that growth rates between private insurance and Medicare have actually been quite similar--a very different finding than casual comparisons often suggest. Moreover, growth rate comparisons need to be viewed cautiously given the different payment levels for services under Medicare as compared to private insurance. Sharp declines in the private sector may reflect discounting off of a very generous level, for example. Finally, we conclude with a look at how changes in the mix of persons covered by Medicare and private insurance might affect rates of growth over time. COMMON PRIVATE SECTOR COMPARISONS When people try to assess the size of Medicare spending growth relative to the private sector, they are typically drawn to the results from two national employer-based surveys, one sponsored by Foster Higgins (1995) and the other by KPMG (1994). Although there are differences in the sampling frames for these two surveys, both are trying to estimate the change in the total cost of the employer-sponsored health insurance package, including both the employers' and employees' contributions. Foster Higgins estimates a 1.1% reduction in health benefits costs per employee between 1993 and 1994, while KPMG estimates a 4.8% increase. The KPMG number is probably more useful because it represents the average i:\bp\mmoon\growth\ 7/17/95 4:37pm 2 change in costs for the same plans at the same employers. Thus, it is essentially a genuine year to year "apples to apples" comparison. The Foster Higgins estimate, on the other hand, is hard to interpret because of the way it treats retiree health costs.² In addition, the 1.1% reduction is greatly influenced by shifts in enrollment among types of health plans. In fact, Foster Higgins explicitly cautions that a large part of this "favorable experience in 1994 is a one-time savings due to moving employees from a higher cost plan to a lower cost plan." What this means is that the underlying growth rate in health plan costs may have changed imperceptibly, but that the downward shift in the cost of an average plan is producing a short-term adjustment rather than a long term trend. A simple example based on premiums and the distribution of covered individuals across plans similar to that in the Foster Higgins data highlights this point. Suppose that there are two types of health plans in 1993--indemnity and managed care --and that covered individuals are evenly split among these plans. In addition, assume the indemnity plan has a premium of $4000 and the managed care plan a premium of $3500 in 1993. If the premium for the indemnity plan increases by 10 percent in 1994, the managed care plan premium increases by 5 percent, and the insured remain evenly split, then the average plan costs would change from $3750 to $4038, an increase of 7.7 percent. However, if the insured shift among plans in the direction of the lower cost managed care plan so that in 1994, say, 35 percent are The Foster Higgins data shows that fewer firms provided retiree benefits in 1994 than in 1993. This would tend to bias the change in health plan costs per active employee (the widely-cited 1.1 percent reduction downward by reducing aggregate health plan costs via the retiree portion without necessarily lowering the number of active employees covered. i:\hp\mmoon\growth\ 7/17/95 4:37pm 3 in managed care and 65 percent in indemnity, the average plan costs increase to only $3929, or by 4.8 percent. Thus, as individuals shift to lower cost plans, overall health care cost may grow at rates below those reflected in the experience of any single health plan. Moreover, when the shift toward lower cost plans stabilizes, the annual change in health care costs will likely increase. Can either of the Foster Higgins or KPMG estimates of private spending growth be used as a basis for comparison to Medicare? For several reasons, these estimates of private premium growth per employee may not be comparable to Medicare program spending growth. Since these surveys measure private spending from the perspective of a health plan's costs, their results will be affected by changes in health plan characteristics. Changes in deductibles, copayments, service coverage, and utilization review will all affect these estimates of spending growth. Changes as a result of some of these plan characteristics might be viewed as a "success" in holding down health care spending (e.g., utilization review), but others make claims about slowing spending difficult to interpret. For example, if increasing deductibles and copayments lower health premiums over time by shifting costs to patients, we should not conclude that the underlying growth in total health care spending has necessarily slowed. Unfortunately, it is difficult to determine just how much private sector cost sharing has changed. If health plans were still generally traditional indemnity insurance, it would be easier to correlate increasing deductibles and copayments with lower premium growth. However, in a world of rapidly changing types of plans, new cost sharing structures are being developed and the implications of these changes are more difficult to assess, particularly on a i:\hp\mmoon\growth\ 7/17/95 4:37pm 4 year-to-year basis. For example, KPMG data suggest that deductibles have been increasing in Preferred Provider Organization (PPO) and Point of Service (POS) plans--both for in-plan and out-of-plan users. While this implies that PPO and POS premium growth is lower than it would have been with constant deductibles, it is impossible to determine the actual average cost sharing within PPO and POS plans since that depends on the extent of in-plan or out-of- plan use and such information is not reported in the current surveys. Medicare, on the other hand, has maintained a fairly stable schedule of deductibles and copayments, particularly under Part B where most of the cost sharing occurs.³ The Part B deductible, originally $50, is still only $100. The copayment for Part B services has always been 20 percent. On the hospital side, cost sharing, particularly for the deductible has risen steadily over time, but it remains a small share of total cost sharing. Overall, the share of acute care spending covered by Medicare has remained relatively constant since the 1970s (Moon, 1993).4 In addition, although the core set of services covered (e.g., physician and hospital care) by private plans might appear reasonably stable over time, plans may add or subtract benefits such as dental care, vision services and prescription drugs annually. Mental health and substance abuse benefits represent another area where employers have been establishing ³Part B of Medicare covers physician and other ambulatory services, while Part A covers hospital, skilled nursing and home health care. 4The presence of private supplemental coverage and the Qualified Medicare Beneficiary program means that many of the elderly have nearly first dollar coverage for acute care services. This likely affects the level of Medicare spending, but not necessarily the rate of growth over time since the share of the elderly with such protection has remained relatively constant. i:\hp\mmoon\growth\ 7/17/95 4:37pm 5 special limits (Foster Higgins, 1995). Thus, if services paid through private health plans have been reduced, insurance is effectively a different product than it was several years ago and growth rate figures may thus be misleading.⁵ In contrast, we know that Medicare's coverage has changed little, with the exception of home health and skilled nursing services where regulatory control and other limitations have alternatively eased and tightened at various points in time. An area where changes in the private sector could legitimately be used to tout success is control over the use of services. Private plans differ markedly from Medicare in their adoption of utilization review, including pre-admission certification for non-emergency care and case management. The goal of utilization review is to reduce the volume of unnecessary services and hence to make the delivery of care more efficient. HMOs and other new forms of managed care have moved aggressively into these areas of control. Although Medicare has some review activities that occur through Professional Review Organizations (PROs), it does not have the prospective review employed in the private sector. Although KPMG data show that the share of plans with utilization review has been fairly stable over the last few years, there is little hard evidence to indicate how the criteria upon which this review is based may have changed. If these criteria are becoming stricter, they could explain some of the slowing of the growth in spending in both the Foster Higgins and KPMG surveys. Although changes in cost sharing, service coverage, and utilization review can all lead to a short-term slowdown in health care spending growth, they do not alter the determinants ⁵Such changes are more likely to show up as long term trends, however, and hence may also not be very important for annual growth rates. i:\hp\mmoon\growth\ 7/17/95 4:37pm 6 of the longer-run trends. Long-run trends in health care spending can only be lowered by slowing the rate at which new technologies are developed, adopted, and used or by continued reductions in the overall volume of service use. This would imply both an evolution of a new set of standards of care and lower expectations about the ability of the system to address health care needs. TRENDS WITH NATIONAL HEALTH EXPENDITURE DATA To effectively compare Medicare and private insurance, it is important to use a consistent data base. For this, we use the National Health Expenditure (NHE) accounts. Produced each year by the Health Care Financing Administration, these data draw on actual public spending and a variety of surveys of the private sector's health care providers to give detailed numbers of spending by type of service and by payer (Levit et al., 1994). While the data on payments by Medicare and private insurance plans in the NHE accounts do come from different sources, there is an attempt to produce numbers that conceptually track the same components of spending for payers. These are the best available data for this purpose. Further, since the number of Medicare beneficiaries is growing more rapidly than the number of persons covered by private insurance, a first step in this analysis is to focus on per capita numbers. This is essential because the number of enrollees in Medicare has been rising steadily while the absolute number of enrollees in private insurance plans has been falling since 1990. Without such an adjustment, aggregate Medicare spending would grow ⁶Since 1982 Medicare's beneficiary growth has averaged about 1.9 percent per year (Committee on Ways and Means, 1994). i:\bp\mmoon\growth\ 7/17/95 4:37pm 7 more rapidly than aggregate private insurance spending even if both payers experienced the same growth in costs per enrollee. If we begin with the standard overall personal health expenditure category, per capita growth rates in Medicare and private insurance spending for the most recent year, 1993, were 9.5% and 7.1% respectively-important differences, but not on the order of two-to-one (or more) as has sometimes been suggested.⁷ Preliminary projections by the Health Care Financing Administration indicate a wider difference for 1994 of 9.1% and 4.6%.⁸ These savings in the private sector relative to Medicare for the past two years may prove to be an important turning point. But we do not yet know whether they can be sustained over time, since two years represents a very short period for tracking health care spending. In terms of trends, Medicare fares very well over the last decade as compared to the private sector in which it bested the private sector in seven of the last ten years (See Chart 1).⁹ But it is somewhat misleading to look only at overall personal health expenditures since Medicare and private insurance often cover very different services. For example, home ⁷For example: "while private health insurance is not rising on average this year because of the competition, expenditures for Medicare is (sic) rising 10.5 percent" (Burrelle's Information Service, 1995). ⁸Arguably, there might be some interest in also deflating Medicare and private insurance spending trends to net out differences in price growth. However, there is legitimate disagreement regarding the appropriate price deflators to use with these data (see Levit, et. al, 1994 and Huskamp and Newhouse, 1994). Therefore, we choose to report all trends in nominal terms. ⁹Moreover, 1989 should be treated as an anomalous year since the short-lived catastrophic program led to modest increases in hospital spending, but a dramatic 261% growth in skilled nursing spending. It is also interesting to compare Chart 1 with Appendix Chart A which shows the same growth rates on an aggregate rather than per capita basis and illustrates how important just this one adjustment can be. i:\bp\mmoon\growth\ 7/17/95 4:37pm 8 Chart 1 Per Capita Growth Rates of Total Personal Health Expenditures 1976-1993 20.0% 18.0% 16.0% 14.0% Private 12.0% Growth Rate 10.0% 8.0% Medicare 6.0% 4.0% 2.0% 0.0% 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 Year health care and skilled nursing services--which are two of the fastest growing parts of Medicare--are much less important for younger families covered by private insurance. If these two benefits were truly post-acute care services and hence served as substitutes for hospital care, it might still be important to include them with other acute care coverage. But a post-acute care response to hospital changes should have occurred between 1985 and 1988 (when hospital growth was lowest). But both SNF and home health services showed little or even negative per capita growth during that period (See Appendix Table C). Instead, much of the growth of these benefits has occurred since 1989 and reflects an expansion of Medicare into long term care types of services.¹⁰ And, on the other side, private insurance often includes some prescription drug coverage, and sometimes dental insurance, while Medicare does not. Consequently, to offer a more accurate picture of differences in spending growth between these two payers, we focus on those services where both Medicare and private insurance play a substantial role. Our efforts are limited by the fact that the National Health Expenditure (NHE) numbers combine a number of categories that we would like to be able to disaggregate.¹¹ For example, vision ¹⁰In the case of skilled nursing facility services (SNF) a court case in 1988 and then the Medicare Catastrophic Care Act in 1989 loosened substantially the restrictions on receipt of these benefits (for example, by eliminating the three day prior hospital stay requirement). Skilled nursing benefits expanded substantially, in part substituting for Medicaid long term care (Liu et al., 1995). A similar legal challenge in 1989 for home health services seems to have opened the door for some beneficiaries to receive large numbers of visits--particularly for nonskilled home health aide services (Kenney and Moon, 1995; Bishop and Skwara, 1993). "In addition, 1994 numbers are only available for total spending, so the disaggregations we show below cannot yet be updated for 1994. When we are able to do so, the numbers will likely show a more favorable tilt to the private sector. Finally, it is not possible to pull out insurance that supplements Medicare ("medigap"), but if it were that would result in an i:\bp\moon\growth\ 7/17/95 4:37pm 9 care and durable medical equipment (DME) are combined in the NHE data. The first of these is less well covered by Medicare while DME represents an area of considerable growth for Medicare in recent years. Nonetheless, we concentrate much of our attention on four of the NHE categories: hospital services, physician services, other professional services, and vision/DME services. Altogether these services accounted for 90.6% of what Medicare covered in 1993 and 84.7% of what private insurance covered in that year. Home health and skilled nursing care made up the rest of Medicare services, while they accounted for only 1.7% of private insurance coverage. Drugs and dental services account for the rest of private insurance coverage not reported here and were negligible for Medicare. Table 1 indicates the importance of each of these categories to Medicare and private insurance. Two years of data are presented in Table 1 to also illustrate how these shares have shifted since 1975-another factor which is important to understanding growth in spending as well. Over time, for example, hospital services have become much less important to Medicare while all other categories of covered services have increased. The same hospital trend holds for private insurance, although hospital services have never been as important as under Medicare. The share of private insurance spending for drugs and dental services has more than doubled, two areas not covered by Medicare. By concentrating on the first four categories of Table 1, we are able to examine spending growth across the two sectors for a more consistent set of covered services. Medicare's spending growth in 1993 is much closer to that for private insurance among this improved estimate between Medicare and private insurance for younger families. i:\hp\mooon\growth\ 7/17/95 4:37pm 10 Table 1 Share of Total Personal Health Expenditures Medicare Private Insurance Health Expenditure Categories 1975 1993 1975 1993 Consistently Covered Services Hospital Services 73.3% 61.3% 62.0% 45.6% Physician Services 21.5 23.0 29.1 32.6 Other Professional Services 1.3 3.7 1.3 6.1 Vision and DME 0.8 2.5 0.6 0.3 Other Nursing Homes 1.9% 4.1% 0.3% 0.7% Home Health 1.2 5.3 0.2 1.0 Drugs and other Nondurables 0.0 0.0 3.4 7.1 Dental Services 0.0 0.0 3.2 6.5 Source: National Health Expenditures, Health Care Financing Administration. g:\moonm\6548Wotexp.11 July 5. 1995 set of services. Per capita numbers for the four combined NHE categories yield average rates of growth of 7.4% for Medicare versus 7.1% for private insurance. And, when comparing the patterns of these two sectors over time, Medicare does better than the private sector in eight of the last ten years (Chart 2). In some of those years, particularly during the late 1980s, Medicare's growth rates are substantially below those for private insurance. Even more interesting patterns emerge if we look at each of the four service groups separately. For both hospital and physician services, the effects of changes in Medicare's payment policies are quite clear. Rates of change in spending for hospital services (Chart 3) were comparable for Medicare and private payers prior to 1985. The last two years of this period cover the start-up period for Medicare's Prospective Payment System (PPS) during which Medicare rates were set quite generously. Between 1985 and 1988, Medicare corrected for these initially high payment rates by establishing very low update factors and kept hospital spending growth well below that of private payers.¹² This caused hospitals' PPS margins to fall from over 14 percent in 1985 to 1.4 percent in 1989. After 1988, Medicare continued with growth rates that were closer, but still below, private payers in all years except 1992. In contrast to the PPS, Medicare approached changes in physician payment policies in a more piecemeal fashion during the 1980s. It was not until 1992 that comprehensive 2 physician payment reform was implemented. Nevertheless, the effects of the 1980s policies can be seen in Chart 4. Until 1983 Medicare and private payer rates of change in physician ¹²The increase in the growth in privately-insured hospital spending may be due to cost- shifting, a revenue-enhancing strategy through which hospitals, as a whole, offset losses on some patients by earning more on others. However, there is wide variation across hospitals in the ability to undertake cost shifting (ProPAC, 1992). i:\hp\mmoon\growth\ 7/17/95 4:37pm 11 Chart 2 Per Capita Growth Rates of Services Covered by Both Medicare and Private Insurance 1976-1993 20.0% 18.0% 16.0% 14.0% Private 12.0% Growth Rate 10.0% 8.0% 6.0% Medicare 4.0% 2.0% 0.0% 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 Year Source: National Health Expenditure Data, Chart 3 Per Capita Growth Rate of Hospital Services 1976-1993 20.0% 18.0% 16.0% 14.0% 12.0% Growth Rate 10.0% 8.0% Private 6.0% 4.0% Medicare 2.0% 0.0% 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 Year Chart 4 Per Capita Growth Rate of Physician Services 1976-1993 20.0% 18.0% Private 16.0% 14.0% 12.0% Growth Rate 10.0% 8.0% Medicare 6.0% 4.0% 2.0% 0.0% 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 Year spending were bouncing around; in some years Medicare was higher and vice-versa. However, in 1984 Medicare froze physician fees for almost two years and achieved a dramatic slowdown in spending growth relative to the private sector. Once the freeze was lifted in 1986, Medicare spending growth exceeded that of private payers for two years. By 1988, there was general agreement about the likely shape of the reforms that would ultimately take place in 1992. 13 As such, Medicare gradually began to move fees for specific services toward the reform target, while keeping overall fee growth to moderate levels. These concerted efforts allowed Medicare physician spending to grow at lower rates than private payers in every year starting in 1988. Other professional services constitute another category where both Medicare and private insurance provide substantial coverage (Chart 5). These include spending for services provided by independently practicing licensed health care practitioners other than physicians and dentists. Such professionals include private duty nurses, psychologists, and podiatrists. They also cover services in freestanding outpatient clinics such as mental health and rehabilitation centers. After substantial declines earlier in the rates of growth in these services, they again picked up in the early 1990s under Medicare. The patterns are quite different for private insurance. Rates of growth were much higher there than under Medicare for most of the period before 1990. Since then, there has been a dramatic decline in private sector growth. Could this reflect some of the tightening on services such as mental health by 13 The basic notion was that fees for procedures and diagnostic testing were "too high" relative to fees for evaluation and management services and that this could be corrected if a payment system were established under which relative fees were based on relative resource costs. This objective was achieved through adaptation of the Resource-based Relative Value Scale developed at Harvard University (Hsiao, et. al, 1979). i:\hp\mmoon\growth\ 7/17/95 4:37pm 12 thanks Per Capita Growth Rate of Other Professional Services 1976-1993 45.0% 40.0% 35.0% Private 30.0% 25.0% Growth Rate 20.0% 15.0% 10.0% Medicare 5.0% 0.0% 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 Year National Health Expenditure Data private plans? If costs of care are shifted to consumers, there should be a concurrent increase in out-of-pocket spending in this category in the 1990s. Rates of growth in out-of-pocket spending for the services have not, however, shown a consistent trend that would substantiate a shifting of burdens onto consumers.¹⁴ It may be, however, that restrictions have occurred in this area without shifting costs onto patients. Vision services and durable medical equipment constitute the last area of comparison between private plans and Medicare. In looking at Chart 6, it appears there are few consistent patterns. Perhaps most impressive is the steady decline in the rate of growth of spending by private insurance in this area since 1987. Again, an examination of out-of- pocket spending growth does not indicate any clear evidence of shifting of burdens onto individual consumers in this category. The excluded categories of spending are also interesting (as shown in Appendix Table C). Part of the reason that Medicare looks so much better than the private sector in recent years within our framework is the exclusion of SNF and home health services. These two areas have grown very rapidly under Medicare, particularly since 1989, largely as a result of a relaxation in regulatory oversight.¹ But our choice of categories for inclusion in Chart 2 does not work exclusively in the favor of Medicare. The excluded categories have grown ¹⁴It is important, however, to view this with caution since we are not able to link the rise in out of pocket spending to those with private insurance. Rather our out-of-pocket numbers refer to all Americans. More detailed databases would be needed to determine whether the "success" of private plans in this spending category is merely a shifting of burdens onto consumers. 15 And, 1989 was an anomalous year for skilled nursing care under Medicare. Benefits were expanded and cost sharing changed for that year only as a result of the Catastrophic legislation. i:\bp\mmoon\growth\ 7/17/95 4:37pm 13 Chart 6 Per Capita Growth Rate of Vision and DME 1976-1993 35.0% 30.0% Medicare 25.0% 20.0% Growth Rate 15.0% 10.0% 5.0% Private 0.0% 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 -5.0% Year faster than the included ones for the private sector as well. In particular, private insurance spending on drugs and other nondurables have grown at a per capita rate greater than overall private insurance spending in every year since 1981. Dental services also grew rapidly in the early years, but their growth rates have slowed substantially since then. PAYMENT LEVELS AND GROWTH RATES Despite the low rates of growth private insurers have achieved in recent years, it is still the case that, on average, they pay more than Medicare for most services in most areas of the country. Given the current direction of the policy debate, an obvious question that would be useful to answer is "what would it cost to insure Medicare beneficiaries through private health plans?" While this is a complex question to answer precisely, based on what is well known about what private payers pay for hospital and physician services, Medicare payments would rise substantially if the program paid for these services at average private rates. The Prospective Payment Assessment Commission reports that revenues from private payers were 29 percent above hospital costs in 1993, down from 31 percent above costs in 1992. Although there may be reasonable expectation of a continued decline, it would take many years for these rates to converge to Medicare payment levels, which are, on average, 11 percent below the costs of treating its beneficiaries (ProPAC, 1995). A similar pattern exists in the physician services market. Based on 1993 data, the Physician Payment Review Commission estimates that Medicare fees were, on average, 38 i:\bp\mmoon\growth\ 7/17/95 4:37pm 14 percent below those of private payers (PPRC, 1995). 16 Although the precise size of this differential may be somewhat sensitive to the methodology used in the calculation (e.g., Miller, Zuckerman, and Gates, 1993), all studies suggest that average Medicare physician fees are below those of private insurers. Obviously, many Preferred Provider Organizations are entering into contracts with physicians that contain fees well below those that existed under payment systems based on, say, "usual, customary and reasonable" charge screens. A recent Urban Institute study shows that PPO discounts can be substantial--averaging 24 percent for one large national insurer in 1993. (Zuckerman and Verrilli, 1995). However, even with these discounts, Medicare's average physician fees were still about 30 percent below those paid by this insurer's PPO. The role these payment differences play in understanding the National Health Expenditure analysis presented above is important. The differentials imply that some part of the success that private insurers have achieved in controlling spending in the last two years is the result of negotiating discounts from historically-high provider payment rates. Given the levels that these rates were starting at and the excess supply of provider capacity in many areas, it is not surprising that large discounts can be obtained, enabling large reductions in spending growth. 17 However, the spending growth slowdown that would be observed as a 2 16 The PPRC analysis of private fees uses a weighted average of fees from both indemnity and PPO payers, suggesting that it is not a worst case scenario for Medicare fee generosity. 17 However, our analysis of NHE data through 1993 does not suggest that private insurers have slowed spending growth below Medicare for comparable sets of services. Further declines may be noted in the future. i:\bp\mmoon\growth\ 7/17/95 4:37pm 15 result of this realignment of prices would represent only a transition from a high-level of spending to a moderate-level of spending, with little impact on underlying long-term trends. CASE MIX DIFFERENCES AND SPENDING GROWTH Not only are the populations served by Medicare and by private insurance quite different, but they may be changing over time in ways that affect rates of growth. For example, for the last few years, fewer people are being covered under private insurance each year; if those losing coverage have health problems and thus cannot get insurance, then their exclusion would move growth rates of private insurance spending downward. On the other hand, if those who drop coverage are younger workers who feel that coverage is not worth the premiums they would have to pay, this would shift growth rates upward. Unfortunately, data on the health status of persons with private insurance coverage is almost nonexistent so we are only left to speculate about the extent to which the reported growth rates are reflecting differences in the cost of insuring a changing mix of patients. Medicare data, on the other hand, do allow us to look at what impact a changing composition over time of at least age and a few other factors have on the costs of insurance. As part of the analysis for this project, we examined whether the aging of the Medicare population contributes substantially to Medicare's growth rates. A positive finding in this regard would suggest that Medicare spending growth should be higher than that for the rest of the population even after controlling for all other factors. We found that the aging of the population covered by Medicare does add to the rate of growth of spending since the i:\bp\mmoon\growth\ 7/17/95 4:37pm 16 proportion of oldest and sickest beneficiaries is rising, but by less than might be expected.¹⁸ Specifically, we compared Medicare spending in 1977 and in 1992 by detailed age groupings for those aged 65 and above and for the disabled as a group. We then considered whether overall spending would have been lower in 1992 if we allowed average spending for each age group to change but held the share of the population in each group constant at the 1977 level.¹⁹ This essentially allows us to consider what spending would look like if the age distribution of the population (and the share of disabled versus elderly) did not change over time. Under this exercise, average per capita spending for all Medicare beneficiaries would be lower in 1992 than the actual number, indicating that the aging of the population does slightly bias spending upward. The 1992 actual per capita amount was $3391, while our age- controlled simulation yielded a per capita average of $3324. Translating this into growth rates implies that spending each year is about 0.2 percent higher as a result of the changing demographics within the elderly population. CONCLUSION While we find little evidence to support the claims that the private sector is doing dramatically better than Medicare, this should not be interpreted as a claim that nothing can 1⁸Actually, this finding is consistent with other analysis suggesting that attributing the high costs of medicare to the very old or those at the end of life usually overstates that impact (Lubitz and Riley, 1993). ¹⁹We also conducted this exercise using 1992 age distributions as the control factor and the results are essentially the same. i:\hp\moon\growth\ 7/17/95 4:37pm 17 or should be done to try to slow the rate of spending in the Medicare program. Indeed, high growth rates in this program create problems "for federal government financing and for out-of- pocket burdens on older and disabled Americans who pay a share of these costs. Serious efforts will need to be made in the future to slow these growth rates. Medicare could and should do better; indeed even within its current structure a number of efforts could slow growth--particularly in the areas of home health, skilled nursing facility care and outpatient hospital services, for example. But such efforts will require us to face up to tough choices if health care spending is to be controlled. Unfortunately, some of the debate on slowing Medicare's growth has suggested that by simply adopting principles developed in the private insurance market, Medicare's problems can easily be resolved. Discrepancies in growth in spending between Medicare and private insurance are used to support such claims. Our findings indicate that growth rates in Medicare and private insurance are quite similar when carefully measured. Historically, Medicare stacks up very well with the private sector. And, even if private insurance does well in the next few years, no one knows very much about the sustainability of these low growth rates over time. In fact, it should not be surprising for Medicare and private insurance per capita rates to turn out to be very similar, since all health care spending shares technological change and improvement as a common major determinant of growth (Newhouse 1993). Reining in use of services will constitute a major challenge for both private insurance and Medicare in the future. If the private insurance market, through the expansion of managed care, is truly successful in restraining growth, Medicare may be able to benefit from adopting some or all i:\hp\mmoon\growth\ 7/17/95 4:37pm 18 of the techniques. However, the emerging forms of managed care (e.g. POS and PPO plans) have little experience in covering elderly and disabled populations. And as yet, techniques for determining the appropriate payments to make to plans that cover beneficiaries are not well developed. Careful consideration will be needed to determine how these plans control expenditures, how this affects patient outcomes, and whether these methods should be adopted by Medicare. For example, if managed care reduces spending by eliminating not only unnecessary services, but some necessary ones as well, it may be difficult for a public program to adopt such stringent controls (Newhouse, et al., 1982). As yet, however, there is little evidence to reassure us that such success has yet been established or will be painless in its implementation. i:\hp\mmoon\growth\ 7/17/95 4:37pm 19 BIBLIOGRAPHY Bishop, Christine and Kathleen Carley Skwara. 1993. "Recent Growth of Medicare Home Health." Health Affairs. 10 (Fall): 95-110. Burrelle's Information Services. 1995. Transcript from "Meet the Press" with Senator Sam Nunn and Senator Phil Gramm. January 29: 17. Foster Higgins. 1995. National Survey of Employer-Sponsored Health Plans/1994. New York: Author. Freudenheim, Milt. 1995. "Medicare, Jot This Down." New York Times, May 31: D1, D4. Gutterman, Stuart. 1995. "Prospective Payment, Medicare Spending, and Hospital Costs: The Impact of Medicare's PPS." Presentation at Health Care Policy and Regulation Workshop, Rutgers University. Hage, David and Robert F. Black. 1995. "New Surgery for Health Care." U.S. News & World Report, February 27: 68-9. KPMG Peat Marwick. 1993. Health Benefits in 1993. Newark: Author. KPMG Peat Marwick. 1994. Health Benefits in 1994. Newark: Author. Hsiao, William C. and William B. Stason. 1979. "Toward Developing a Relative Value Scale for Medical and Surgical Services." Health Care Financing Review (Fall): 23-38. Huskamp, Haiden A. And Joseph P. Newhouse. 1994. "Is Health Spending Slowing Down?" Health Affairs 13 (Winter): 32-38. Kenney, Genevieve and Marilyn Moon. 1995. "Medicare Subacute Care Services and Enrollee Characteristics." Prepared under Health Care Financing Administration Contract No. 500-89-0064. Levit, Katharine R., Cathy A. Cowan, Helen C. Lazenby, Patricia A. McDonnell, Arthur L. Sensenig, Jean M. Stiller, and Darleen K. Won. 1994. "National Health Spending Trends, 1960-1993." Health Affairs 13 (Winter): 14-31. Liu, Korbin, Genevieve Kenney, Douglas Wissoker, and Jill Marsteller. 1995. Final Report: The Effects of the Medicare Catastrophic Coverage Act and Administrative Changes on Medicare SNF Participation and Utilization: 1987-1991. Urban Institute Discussion Paper, June. Lubitz, James D. And Gerald F. Riley. 1993. "Trends in Medicare Payments in the Last Year of Life." The New England Journal of Medicine. 328: 1092-1096. Miller, Mark, Stephen Zuckerman, and Michael Gates. 1993. "How Do Medicare Physician Fees Compare with Private Payers?" Health Care Financing Review. 14 (Spring): 25-39. Moon, Marilyn. 1993. Medicare Now and in the Future. Washington, D.C.: The Urban Institute Press. Newhouse, Joseph P. 1993. "An Iconoclastic View of Health Cost Containment." Health Affairs 12 (Supp): 152-171. Newhouse, Joseph P. et al. 1982. Some Interim Results From a Controlled Trial of Cost Sharing in Health Insurance. Santa Monica, CA: The Rand Corporation. Prospective Payment Assessment Commission. 1992. Medicare and the American Health Care System: Report to the Congress. Washington, D.C.: Author, June. Prospective Payment Assessment Commission. 1995. Medicare and the American Health Care System: Report to the Congress. Washington, D.C.: Author, June. Physician Payment Review Commission. 1995. Annual Report to Congress. Washington, D.C.: Author. The Heritage Foundation. 1995. Rolling Back Government: A Budget Plan to Rebuild America. Scott A. Hodge, ed. Washington, D.C.: Author. U.S. Congress. House Committee on Ways and Means. 1994. 1994 Green Book: Background Material and Data on Programs within the Jurisdiction of the Committee on Ways and Means. Washington, D.C.: U.S. Government Printing Office, Zuckerman, Stephen and Diana Verrilli. 1995. "The Medicare Relative Value Scale and Private Payers: The Potential Impact on Physician Payments." Prepared under Health Care Financing Administration Contract No. 500-92-0024, D.O. #4. APPENDIX Most of the analysis in this paper uses data from the National Health Expenditure accounts (Levit et al. 1994). These data are available on disk and allowed us to examine eight categories of spending (as shown in Tables B and C) by type of payer. These data are gathered in different ways, but there is an attempt to assure that these figures conceptually track the same components of spending and that they are national in scope. As a result these are the most consistent data available. Most of our analysis concentrated on Medicare and private insurance. The private insurance data include all types of private insurance: employer-based and privately purchased plans for younger families and medigap policies for the elderly and disabled. Our preference would have been to take medigap out of these numbers, but that was not possible. Growth rates for aggregate data were calculated by dividing nominal dollars in year one by year two. That growth rate is shown as year two's growth rate. For purposes of this analysis we used data back to 1975. By that time, the disabled population had been assimilated into Medicare and it was during that period that interest in containing costs began. Per capita dollar spending values were created using numbers supplied to us by the Office of the Actuary of the Health Care Financing Administration (HCFA). The private insurance numbers were adjusted to create an unduplicated count of the number of persons with some type of private health insurance. After calculating per capita spending numbers, we then estimate annual growth rates for each of the various spending categories. We used two criteria for determining which categories to include in our comparative measure: the figure had to be positive for both Medicare and private insurance (which ruled out drugs and dental services as shown in Table 1), and the categories needed to primarily capture acute care benefits. This second criterion is somewhat more controversial since it omits nursing home and home health services, both of which are growing rapidly under Medicare at present. Moreover, in theory, these services might have grown because of the rapid changes in inpatient hospital lengths of stay after the introduction of the hospital prospective system under Medicare in 1984. But between 1985 and 1989, home health spending under Medicare declined in per capita terms and skilled nursing facility services grew quite slowly(see Appendix Table C). The rates of growth for both picked up at the end of the 1980s and continued into the 1990s. From other analysis in this area, we find that much of the recent growth in these programs is in a shift toward long term care, particularly in the case of home health services (Kenney and Moon 1995). Much of the growth in home health services, for example, is for home health aides rather than for skilled services. Our analysis of whether the aging of the Medicare population over time contributed to its rate of growth used HCFA data on per capita spending over four different years and used two different index calculations. Detailed results are not presented here since we found that the increasing share of Medicare beneficiaries over the age of 85 and under the age of 65 (the disabled) did not add substantially to growth over time. This is consistent with findings on other studies about expenditures at the end of life (Lubitz and Riley 1993). Appendix Chart A Aggregate Growth in Medicare and Private Insurance Spending 1976-1993 25.0% 20.0% 15.0% Growth Rate Private 10.0% Medicare 5.0% 0.0% 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 Year Appendix Table A Comparison of Growth Rates for Alternative Measures of Health Care Spending Aggregate Per Capita Per Capita Subtotal Total Personal Total Personal of Consistently Health Expenditures Health Expenditures Covered Services Dates Medicare Private Medicare Private Medicare Private 1976 20.5% 17.1% 17.2% 15.8% 17.0% 14.4% 1977 17.7% 17.0% 14.1% 12.3% 14.2% 11.9% 1978 15.9% 14.2% 13.0% 14.5% 13.1% 13.1% 1979 16.5% 18.2% 13.6% 17.0% 13.6% 16.5% 1980 21.3% 19.1% 18.7% 17.9% 18.9% 17.3% 1981 19.7% 17.8% 17.7% 16.4% 17.6% 15.7% 1982 17.5% 15.5% 15.5% 14.1% 15.3% 13.1% 1983 14.0% 10.7% 12.1% 10.1% 12.0% 9.4% 1984 11.3% 10.5% 9.4% 10.0% 9.3% 8.9% 1985 8.3% 12.3% 6.2% 11.9% 6,4% 11.0% 1986 6.8% 10.0% 4.5% 8.1% 4.8% 7.6% 1987 7.2% 12.9% 5.2% 11.7% 5.4% 11.1% 1988 8.3% 12.1% 6.3% 10.9% 6.0% 11.4% 1989 14.9% 11.9% 12.9% 10.7% 9.9% 10.7% 1990 9.5% 10.6% 7.5% 9.7% 8.1% 9.1% 1991 9.9% 8.3% 7.7% 9.2% 6.7% 9.0% 1992 12.4% 8.2% 10.2% 8.6% 8.1% 8.4% 1993 11.6% 6.6% 9.4% 7.2% 7.4% 7.1% Appendix Table B Per Capita Growth Rates for Medicare and Private Insurance for Consistently Covered Services Hospital Physician Other Professional Vision and Subtotal of Consistently Total Personal Services Services Services DME Covered Services Health Expenditures Dates Medicare Private Medicare Private Medicare Private Medicare Private Medicare Private Medicare Private 1976 17.7% 14.9% 13.6% 13.0% 32.2% 27.6% 18.6% 4.8% 17.0% 14.4% 17.2% 15.8% 1977 13.1% 11.7% 17.1% 11.6% 26,2% 25.0% 25.4% 13.9% 14.2% 11.9% 14.1% 12.3% 1978 12.3% 14.6% 14.4% 9.2% 22.2% 22.7% 30.1% 20.4% 13.1% 13.1% 13.0% 14.5% 1979 13.2% 16.0% 14.7% 16.8% 11.3% 23.0% 21.2% 27.0% 13.6% 16.5% 13.6% 17,0% 1980 18.8% 17.1% 19.1% 17.4% 12.7% 16.4% 29.5% 23.6% 18.9% 17.3% 18.7% 17.9% 1981 17.8% 14.0% 16.4% 18.1% 13.4% 41.8% 31.0% 6.4% 17.6% 15.7% 17.7% 16.4% 1982 14.4% 13.5% 17.2% 11.6% 24.3% 21.5% 26.3% 10.0% 15.3% 13.1% 15.5% 14.1% 1983 10.2% 7.1% 16.0% 13.9% 20.9% 16.8% 28.8% 5.3% 12.0% 9.4% 12.1% 10.1% 1984 9.1% 4.3% 9.4% 15.9% 13.9% 36.1% 14.1% 14.8% 9.3% 8.9% 9.4% 10.0% 1985 5.0% 5.2% 9.7% 19.3% 13.4% 33.8% 11.9% 14.3% 6.4% 11.0% 6.2% 11.9% 1986 1.8% 4.9% 11.9% 10.7% 17.9% 20.4% 11.6% -3.2% 4,8% 7.6% 4.5% 8.1% 1987 2.1% 9.6% 13.2% 12.1% 12.4% 22.5% 12.4% 10.6% 5.4% 11.1% 5.2% 11.7% 1988 4.5% 9.1% 9.3% 12.1% 8.1% 32.0% 9.6% 6.9% 6.0% 11.4% 6.3% 10.9% 1989 9.5% 9.6% 10.2% 11.0% 18.4% 19.4% 10.7% 4.6% 9.9% 10.7% 12.9% 10.7% 1990 7.2% 7.6% 8.0% 10.8% 25.6% 12.1% 15.8% 0.2% 8.1% 9.1% 7.5% 9.7% 1991 7.1% 8.3% 2.7% 8.9% 23.0% 15.8% 20.5% -0.9% 6.7% 9.0% 7.7% 9.2% 1992 10.2% 6.5% 1.1% 10.3% 20.9% 13.0% 10.4% -1.0% 8.1% 8.4% 10.2% 8.6% 1993 8.0% 8.5% 5.5% 6.2% 10.0% 2.7% 8.2% -2.4% 7.4% 7.1% 9.4% 7.2% Source: National Health Expenditure Data, Appendix Table C Per Capita Growth Rates for Medicare and Private Insurance for Other Spending Categories Nursing Home Drugs and other Dental Subtotal of Other Homes Health Nondurables Services Spending Categories Dates Medicare Private Medicare Private Medicare Private Medicare Private Medicare Private 1976 14.2% 17.6% 40.8% 70.8% 0.0% 13.1% 0.0% 56.5% 24.8% 34.5% 1977 -1.0% 15.6% 25.2% 38.9% 0.0% 6.6% 0.0% 23.2% 10.7% 16.7% 1978 -1.7% 31.4% 20.2% 56.7% 0.0% 25.1% 0.0% 31.6% 9.4% 30.0% ... 1979 2.7% 26.7% 21.0% 21.4% 0.0% 22.2% 0.0% 21.4% 12.9% 21.8% 1980 7.9% 22.8% 16.2% 20.5% 0.0% 24.4% 0.0% 22.9% 12.8% 23.3% 1981 8.1% 32.7% 27.8% 13.0% 0.0% 16.1% 0.0% 25.4% 20.2% 21.8% 1982 6.8% 37.0% 26.4% 22.9% 0.0% 28.2% 0.0% 17.1% 19.6% 21.8% 1983 4.3% 29.7% 20.7% 19.6% 0.0% 17.1% 0.0% 14.1% 15.6% 16.0% 1984 5.7% 25.3% 15.7% 18.1% 0.0% 20.7% 0.0% 15.3% 12.9% 17.8% 1985 1.4% 25.8% -0.6% 16.8% 0.0% 24.3% 0.0% 13.7% 0.0% 18.4% 1986 -2.4% 21.8% -6.5% 23.3% 0.0% 11.2% 0.0% 9.5% -5.4% 11.3% 1987 6.0% 20.2% -5.8% 13.8% 0.0% 13.4% 0.0% 16.6% -2.6% 15.5% 1988 47.6% 9.2% 6.3% 6.9% 0.0% 9.3% 0.0% 7.1% 18.6% 8.1% 1989 261.2% 3.2% 22.4% 30.8% 0.0% 16.4% 0.0% 6.2% 111.1% 10.9% 1990 -31.3% 16.0% 49.5% 29.7% 0.0% 21.0% 0.0% 5.0% -1.8% 13.3% 1991 16.1% 5.2% 38.5% 5.5% 0.0% 17.7% 0.0% 4.3% 28.5% 10.3% 1992 55.2% 2.7% 36.3% 25.1% 0.0% 8.7% 0.0% 9.4% 43.9% 9.5% 1993 32.7% -5.4% 33.9% 18.7% 0.0% 8.0% 0.0% 6.4% 33.4% 7.3%