This Article Abstract Submit a response to this article Services E-mail this article to a friend Alert me to new issues of the journal

H E A L T H  T R A C K I N G : 
M A R K E T W A T C H
W E B  E X C L U S I V E   

02 April 2003  
  

Can Managed Care And Competition 
Control Medicare Costs?

It will take more than managed care and competitive options
to ensure that Medicare can meets its obligations in the future.


by Marsha Gold

ABSTRACT:

Medicare+Choice (M+C) was conceived to bring managed care and competitive forces to bear on Medicare. Ultimately, M+C could not thrive under the conditions of the marketplace and the Balanced Budget Act of 1997. Here I review what went wrong and the lessons from the experience, concluding that M+C is a tool, not a strategy. While managed care in a multiple-choice environment may have the potential to generate limited savings, promoting managed care and competition alone will not preempt the need for a debate on Medicare’s obligations and how to finance them.

Medicare’s considerable fiscal stress will only increase as baby boomers age, technology continues to evolve, and the pressure to deal with the program’s benefit limitations mounts.¹ In this paper I examine whether managed care and competition can help relieve this stress, although I recognize that the evidence for their potential to do so is weak at best.

Managed care and competition were major strategies employed by purchasers to control health care costs in the 1990s. Competition here means the offering of multiple plan choices that may differ in price to the enrollee. Even though the models used now both in the public and private sector depart substantially from pure price-competitive models, they provide important insights on real-world applications of competition.² Unfortunately, the early positive experience with these models has given way to doubts about their ability to control costs, particularly in the long run. The movement toward managed care in the private marketplace initially led to slower growth in premiums. However, a shift in the underwriting cycle, the backlash from consumers and providers in response to constraints imposed by managed care, and continued growth in medical technology and demand for care have emerged as key driving forces, and now premiums are rising much more rapidly.³

Experience under the Medicare risk-contracting program has been similar. In the mid- to late 1990s Medicare health maintenance organizations (HMOs) appeared to offer a competitive product with expanded benefits at low cost, attracting a substantial, if still limited, share of Medicare beneficiaries.⁴ But after the rules were changed and the Medicare risk program was folded into the Medicare+ Choice (M+C) program, the growth in enrollment slowed, halted, and ultimately reversed course, as the range and attractiveness of products diminished.⁵

I begin by reviewing the experience with M+C and the lessons learned. I then consider the likely cost savings for Medicare with managed care and competition under various conditions. I draw on empirical research and on experience with the policy and market environment in which managed care and competition are likely to operate in this country.

Analysis Of The Medicare+Choice Experience

Few would disagree with the basic facts about the experience under M+C. It is the interpretation of those facts that is likely to generate debate, particularly about how Medicare should be structured in the future.

The Balanced Budget Act (BBA) of 1997 established the M+C program, folding the Medicare risk-contracting program into beneficiaries’ broader set of choices.⁶ Prior to the BBA, the number of beneficiaries in Medicare managed care—basically, Medicare HMOs—was growing rapidly (Exhibit 1). More than five million beneficiaries were enrolled in 1997, although penetration was only 14 percent (slightly higher if enrollment in demonstrations and other related products is included). Availability of M+C plans reached a high of 74 percent in 1998, when well over a third had access to five or more plans.⁷ Enrollment growth was spurred by increasingly broad benefits, including prescription drug coverage, often offered for little or no additional premium (Exhibit 2). When M+C was adopted, the Congressional Budget Office (CBO) predicted that M+C penetration would be 34 percent by 2005.⁸

Reality was much the opposite. After M+C was implemented, enrollment continued to grow through 1999 but at a slower rate; by 2000 it began to decline (see Exhibit 1). There are now fewer beneficiaries in managed care than before the program began. In 1999 ninety-seven plans withdrew from the market or reduced their service area, affecting 407,000 enrollees in plans that left the market. In 2000 ninety-nine more plans dropped out, affecting 327,000 enrollees. In 2001 about 934,000 enrollees were affected, and 536,000 more were affected in 2002.⁹ In 2002 only 156 contractors remained in the program, down substantially from the 346 contractors in 1998.¹⁰

At the same time plans were departing the program, the M+C product itself began to seem less attractive.¹¹ Plans remaining in the program started to increase premiums (Exhibit 3). In 2002 the average plan premium rose to $38 per month, up from $11 in 1999. Among plans charging premiums (62 percent), the average was $61 in 2002. Plans also reduced benefits. Those offering any form of drug coverage dropped from 73 percent in 1999 to 66 percent in 2002. More importantly, by 2002 more than half of the plans covered generic drugs only. Among those also covering brand-name drugs, 29 percent provided annual coverage of $500 or less, and 73 percent covered $1,000 or less. Copayments for inpatient hospital care increased: Only 9 percent of plans had any such requirement in 1999, but 73 percent did in 2002. Despite the authority given by the BBA to include expanded managed care products, such as preferred provider organizations (PPOs) and provider-sponsored plans, they were seldom offered. Also, despite increased payments for plans in rural areas, more choices did not emerge there; indeed, there are now fewer choices per county than before, and the choice that does exist is concentrated in urban areas (Exhibit 4).

The notable decline in the M+C program is best viewed in historical context: During most of its existence, Medicare managed care has been relatively limited in enrollment and in benefits. Only in the mid- to late 1990s did it grow rapidly—a period coinciding with substantial growth in plans and comprehensiveness of benefits offered. M+C enrollment still remains higher than in 1996, although it is unclear how long it will stay that way, given current trends (see Exhibit 1).

Although M+C accounts for a small share of the overall Medicare market, it accounts for a much larger share of Medicare beneficiaries in some communities and market segments. In sixty-six major markets with M+C, penetration was 23 percent in 2002.¹² Estimates from the Medicare Current Beneficiary Survey (MCBS) show that 39 percent of beneficiaries who live in a county with an M+C choice and who do not get supplemental coverage through a private employer or Medicaid are enrolled in M+C; this compares with 38 percent who have Medigap and 24 percent who only have Medicare.¹³ Among those without supplemental coverage from employers or Medicaid in 1998, penetration was 71 percent in Southern California, 58 percent in Northern California, 56 percent in Philadelphia, 48 percent in New York City, and 43 percent in South Florida. About 40 percent of African Americans with no group or Medicaid coverage are in M+C plans, as are 52 percent of Hispanics and 40 percent of those with incomes between $10,000 and $20,000, regardless of race or ethnicity.

Although its coverage has diminished, Medicare managed care remains a good value compared with Medigap. Premiums remain at least comparable to early Medicare HMO rates, although there is much variability across plans and markets (Exhibit 3). Preventive benefits continue to be offered, but the value of drug coverage, especially for those with extensive needs, has shrunk because of exclusions (for example, brand-name or off-formulary drugs) and tighter annual limits. First-dollar coverage, especially for hospital care, is less available.

M+C now provides beneficiaries with much less protection against high out-of-pocket health care costs than it used to provide. Estimates by Mathematica Policy Research show that out-of-pocket costs for beneficiaries in M+C plans grew by a projected average of 83 percent overall from 1999 through 2002 (Exhibit 5). Particularly notable is that they rose 71 percent for those in good health but 102 percent and 116 percent, respectively, for those in fair or poor health. The M+C benefit package now provides less protection to beneficiaries with extensive health care needs because plans have been reluctant to raise premiums too high and therefore now expose enrollees to more cost sharing at the point of service. In 1999, for example, the Part B premium and M+C premium contributed 62 percent of estimated average enrollee spending. By 2002 this had dropped to 57 percent. While point-of-service cost sharing (for hospital and physician charges and for prescription drugs) accounted for 28 percent of estimated spending for the average enrollee in good health, it accounted for 79 percent for those in poor health in 2002.

What Went Wrong?

Without a doubt, some of M+C’s enrollment decline was the result of natural market evolution and shakeout after a period of rapid growth, as the industry learned how its products would compete in particular markets.¹⁴ However, a shakeout alone cannot explain the extent of decline in M+C. Other forces at work included Medicare policy changes, the backlash against managed care, and growth projections that were probably always unrealistic.

Slowdown in M+C payments. Historically, M+C payments have been linked to Medicare fee-for-service (FFS) spending. In the mid-1990s such spending was rising rapidly, leading to strong annual increases in payments for M+C (Exhibit 6). Medicare’s pricing system also made relatively crude adjustments for health status differentials between those enrolled in managed care and traditional Medicare. As a result, researchers estimated that favorable selection led the Medicare program to pay HMOs more than it would have spent for traditional Medicare enrollees.¹⁵ HMOs in turn used the excess payments, together with their savings from efficiency, to support an expanded benefit package that greatly reduced beneficiaries’ out-of-pocket costs. The availability of such coverage was a major inducement to join an HMO, but it meant that government did not get the 5 percent savings it sought.

The BBA reversed these trends in three ways. First, it included numerous provisions to control FFS spending, leading to lower estimated aggregate costs. Because of the way payment rates were set, this translated into a slower growth rate for M+C payments. Second, the M+C provisions of the BBA modified the effect of payment reductions in the interest of geographic redistribution and cost savings. M+C plans were paid the greater of a minimum payment update of 2 percent, a national floor rate, or a phased-in blend of national and local rates, subject to budget-neutrality provisions and therefore only effective in 2000. Third, and subject to a phase-in, plan payments also were reduced (on average) as greater risk adjustment was introduced and graduate medical education (GME) payments were removed, the latter a policy with highly concentrated effects in certain counties.

The result was that the BBA’s effect on individual rates paid to health plans differed across plans and areas. While some received relatively high rates of increase (especially if they were in a floor counties), most plans received a 2 percent increase in M+C payments annually for most years after the BBA. This growth was more than many plans might otherwise have received if payment methods had not been modified. But it was much lower than it had been pre-BBA. It also was much lower than the industry had anticipated as it expanded its Medicare product line and, most importantly, much lower than plans’ rising costs. M+C plans also have problems planning because it is difficult to predict payment levels in advance—both because payments typically are set prior to congressional consideration of payment increases and because the annual update is adjusted not just by changes in anticipating future costs but also to correct for any differences between published and current estimates of growth of the prior year’s costs.

Shifting market conditions. As I and others have highlighted, the same rapid development of managed care in 1990s that spawned the M+C program also generated a backlash against managed care that made it harder for M+C and others to compete in the marketplace.¹⁶ Viewing the growth of managed care as inevitable, providers entered into many contracting arrangements and took on more risk than many could handle or absorb prudently.¹⁷ Providers’ displeasure was conveyed to their patients, already ill at ease from mandated changes in providers and from reports that some care was being denied.¹⁸ Policymakers and major purchasers who had vigorously endorsed managed care as an alternative to more formal means of cost control reacted to growing resistance from providers and the population at large by distancing themselves from the very models they had previously encouraged and endorsed.¹⁹

These forces encouraged providers to be more aggressive in negotiating contracts and terms, asking for rate hikes while the slowdown in payments to M+C plans made them less able to support such demands. As Joy Grossman and colleagues found, the drop in M+C payments occurred as overall health care costs were increasing and the private underwriting cycle was shifting in ways that accentuated the disparity between public and private payments to plans.²⁰ The backlash fueled what industry leaders viewed to be a “piling on” of new requirements for M+C, reinforcing plans’ growing wariness about the M+C program.²¹

Unrealistic initial projections. Projections of rapid M+C growth probably were unrealistic even if the market and payment rate growth had stayed the same. The segmentation of the M+C market, combined with many beneficiaries’ unwillingness to change from the traditional Medicare program, limited competition.

One-third of all Medicare beneficiaries have some form of retiree group coverage.²² Former employers typically subsidize the cost of supplemental coverage, providing less incentive for retirees to consider a change based on pricing. While employers offer managed care products, they might not necessarily offer an M+C product to every Medicare retiree; even if they do, incentives to select such a plan might be limited by the employer’s contribution strategy. If an M+C plan is not directly offered to the group, choosing it results in forfeiture of earned retirement benefits. While some employers encourage M+C enrollment to reduce their costs, many do not because such changes would violate labor-force contracts, require that separate benefit offerings and processes be established for active workers and retirees, generate instability in offerings, or raise other problems. Few data are available, but Jessica Mittler and I found that only 13 percent of those with retiree coverage were in a Medicare HMO in 2000—27 percent of M+C enrollment.²³ Other recent data suggest that group enrollment could be a smaller share of the total.²⁴

M+C enrollment thus is most relevant to Medicare beneficiaries in the individual market, where the main alternative to a Medicare HMO is a Medigap plan or “going bare” with only traditional Medicare. But even among this group, some will not live where an M+C plan is offered. Others may have access to subsidized forms of supplemental coverage—such as Medicaid or military health programs—which lower incentives to choose such a M+C plan. In addition, some beneficiaries are poorly positioned to make a managed care election, particularly if they are old, have cognitive impairments, or are unfamiliar with managed care choices.²⁵

The natural inertia of current enrollment also works against massive change. Our research shows that most beneficiaries say that they never thought seriously about the choice between joining a Medicare HMO or getting supplemental coverage (44 percent), or did so only when they first were eligible for Medicare (14 percent).²⁶ Only 15 percent reported in early 2001 that they thought seriously about their options in the 2000 open enrollment period. With more than four-fifths of beneficiaries still in the traditional Medicare program, the low level of interest works against the movement toward managed care and competition, particularly when no change is required and enrollment is voluntary.

Lessons From The Current M+C Program Experience

The M+C experience highlights the potential tensions between markets and the goals policymakers have in using those markets. There are at least four lessons from which to draw.

Legislating products or markets. Legislative authority alone cannot successfully develop a specific product or market if underlying market conditions are unfavorable. Through the BBA, Congress wanted to expand the choices available to Medicare beneficiaries and increase enrollment. But the industry apparently did not see—at least not under conditions of the BBA and the ensuing regulations—a market for many of these new products, and hence they did not develop them. Efforts by government to generate savings led to the minimum 2 percent annual increase for plans becoming instead a maximum increase, leading to the erosion in the existing managed care market. The M+C experience provides an important cautionary tale about the ability to mount competitive strategies in a politically charged environment where care for vulnerable beneficiaries is at stake, administrative protections are sought, and Congress specifies payment methods and limits the discretion of the Centers for Medicare and Medicaid Services (CMS) to negotiate.

Geographical diversity. Managed care and competition are not equally applicable across communities and market segments. For the most part, managed care and multiple-choice offerings are tools that work better in urban areas, where there is more competition among providers, than in rural areas, where competition is usually minimal or nonexistent.²⁷ Even in urban areas, the ability of managed care to thrive varies with local market conditions, including previous managed care history, the structure of provider networks, beneficiary demographics, and other factors.²⁸

Current incentives to switch plans. The current Medicare structure limits beneficiaries’ incentives to actively consider their health plan choices, an important condition of a functional market. The Medicare program provides a protected alternative with a standardized set of benefits available to all beneficiaries in the nation at the same price. To get beneficiaries to switch, plans need to develop products that improve upon Medicare benefits, particularly its exclusion of outpatient pharmacy benefits and the relatively high levels of cost sharing for some services. Unless government pays much more to M+C than to the traditional Medicare program, plans will find it difficult to offer attractive enough products to induce beneficiaries to switch. Many have no financial incentive to switch plans because they already have subsidized supplemental benefits through an employer or Medicaid. Others believe that it is simply not worth the trouble to switch or to educate themselves about available choices.

Market instability. There are inherent instabilities associated with market-based strategies because exit, along with entry, is a fundamental principle of well-functioning markets. For Medicare and managed care, when plans drop their program participation, political challenges result. Some beneficiaries are better off, then worse off, as managed care provides and then takes away a valued product. Because Medicare beneficiaries use more health care than the population at large, they are particularly vulnerable financially, and their continuity of health care is more likely to be jeopardized by a change in plan, benefits, or provider network.

However, instability on the scale experienced in M+C is not necessarily inevitable in markets. Instability, for example, is much more constrained in major employer programs such as the Federal Employees Health Benefits Program (FEHBP), which use negotiation or defined contributions to set payments, than in M+C, which relies on administered pricing in a legislative context. Legislators seeking to promote managed care and multiple-choice offerings from the private sector would do well to recognize the pitfalls of instability and to make program requirements and payments more predictable over time.

Is The Past Prologue? Developing Realistic Future Policy

Unable to bring costs down to match constraints on their revenue, M+C plans became more reluctant to participate in Medicare and cut back benefits. I consider here how this experience likely translates to the future.

Modest savings expectations. Recent experience and existing research both suggest that managed care in Medicare can generate at best modest savings through efficiency, even in its most developed HMO form. In their series of meta-analyses of the research on HMOs, the most tightly organized form of managed care, Robert Miller and Hal Luft show some potential for savings through the use of managed care, with less use of hospital and other expensive services, greater efforts at prevention, and some communitywide spillover effect.²⁹ Managed care scored lower on patient-reported access and satisfaction, but there were no consistent patterns of difference in quality, with substantial variability across plans and delivery sites. Changes in data, medical management, and clinical reengineering, which might encourage better results, appeared to be lacking in most cases.

A major evaluation of the Medicare HMO program in the early 1990s found similar results.³⁰ Medicare HMOs used fewer resources than traditional Medicare FFS used to deliver care of comparable quality. When differences in health status measures, attitudes toward health care, and demographic variables were controlled for, HMO enrollees used fewer hospital days and received less intense care while being more likely to make some use of care; effects were greatest for the seriously ill. The overall impact of this was about 10.5 percent less spending by plans for Medicare-covered services compared with Medicare FFS. These estimates do not include administrative costs or the reduction in government savings due to favorable selection that led plans to be overpaid for their case-mix.

Although historical research suggests that savings from managed care have been modest, the issue is what to expect in the future. Some would argue, for example, that current performance is driven by incentives that encourage delivery of individual services, with little reward for better quality or delivery. But shifting incentives to encourage quality provides no guarantee of additional savings because poor quality can result from underuse or misuse of services, not just overuse.³¹ Some believe that greater savings, especially for the elderly and disabled, can come from reconfiguring systems to truly encourage better care management, not just cost management.³² But such reconfiguration is extremely challenging and time-consuming. Support for true change may be lacking among providers, who need to make it happen, and purchasers and policymakers, who need to allow the time and stability for it to occur.

More likely, the amount of savings expected from managed care is even more constrained today than it has been in the past. First, studies of documented savings are for the most part limited to HMOs, the managed care segment growing the most slowly.³³ Second, managed care already may have knocked out some of the easier savings (“low-hanging fruit”). Cost savings may be contingent on eliminating care known to be marginally effective and still expensive. Efforts to eliminate such care tend to be characterized as “medical rationing” and are strongly resisted. Third, costs reflect utilization and price. A major constraint on potential savings from Medicare managed care is the competition it faces from traditional Medicare in establishing payment rates for individual providers, because of Medicare’s market dominance and ability to set administered prices (within political limits).

Current versus future policy. Current M+C experience can help in judging the likely savings from managed care and multiple-choice offerings under two types of changes, one that greatly expands benefits under the traditional Medicare program, and one that introduces much more price competition within Medicare.

Expanded benefits through traditional Medicare. One proposed alternative for Medicare is to greatly expand benefits, including offering drug coverage as an integral part of the program and potentially adding a limit on total out-of-pocket spending. Such expansion could occur independently or as part of a restructuring of Medicare Part A and Part B benefits. The traditional Medicare program still would serve as the default option available on the same terms and conditions as today. A key issue will be whether expansion greatly increases the actuarial value of the Medicare benefit and how much new money is infused.

Assuming a substantial expansion in benefits, the demand for managed care and competitive products should drop, because their attraction has been based on their ability to offer more than traditional Medicare at a competitive rate. To the extent that benefit expansion makes Medicare more comprehensive, the price-benefit gain likely would not be sufficient to get a large number of beneficiaries to agree voluntarily to restrict their choice of providers. We know, for example, that voluntary enrollment in Medicaid HMOs was limited because the traditional Medicaid program offered comprehensive benefits with very limited cost sharing.³⁴

However, budgetary pressures are likely to limit the extent of benefit expansion, leaving beneficiaries still liable for sizable out-of-pocket payments. Under these conditions, Medicare managed care could remain attractive to beneficiaries. A Medicare HMO product with a modest premium and an integrated series of benefits that extend beyond those of Medicare could be especially attractive to beneficiaries who are more risk-adverse financially or less able to absorb the risk of fluctuating needs and health care spending. Under this scenario, one might expect, as now, modest savings as a result of the growth in managed care, although most of it would accrue to the beneficiaries in added benefits, not to managed care plans. Plans also would gain the advantage of not being expected to provide more benefits (because of Medicare’s limits) than they are capable of providing for the price.

Restructured Medicare program along a defined-contribution, competitive model. This scenario assumes that Congress restructures the Medicare program more in line with proposed competitive models. The traditional Medicare program still would be offered, but its premium would no longer be protected from competition. Medicare’s contributions to each plan would be standardized, using either competitive-bidding principles or some external benchmark. Benefits would be expanded, but the infusion of new funds would be limited, particularly over time. Funding would be based on principles more consistent with defined-contribution than with defined-benefit models, although some protections are assumed to be in place to assure some “minimum” benefit or actuarial equivalence.

Under this scenario, beneficiaries would be forced, at least in theory, to choose. The efficiencies generated by managed care, even if modest, probably would support products that were more attractive to beneficiaries because the premiums would be more competitive. (This assumes appropriate risk adjustment so that prices truly reflect efficiencies.) Thus, more beneficiaries may be in managed care plans, although there is also is some evidence that retirees’ willingness to switch plans is lower than that of active workers.³⁵

Under a competitive model, segmentation of the market by income, race, and ethnicity likely would grow. Current experience shows that moderate-income people, including minorities, would be particularly interested in these plans.³⁶ Higher-income people are more likely to have access to subsidized group coverage and less incentive to consider price. Segmentation will be larger if subsidies to reduce cost sharing based on income are limited. Disparities by location also could grow because there would be many more attractive alternative plans available in highly urbanized areas than in other areas.

Under a competitive model, competitors are needed, so that plan participation would become even more critical than it is today under the voluntary M+C program. M+C experience suggests strongly that plan participation would be a business decision, and the government would generate competition only if the terms meet market demand. The current M+C experience means that plans distrust the government as a business partner (because of M+C experience) and that providers distrust managed care (because of the backlash). Government policy will have to reverse these attitudes, and that is likely to cost money. M+C experience reinforces the importance of predictable and stable payments to minimize instability for beneficiaries. Extensive savings from efficiency should not be assumed because arguably neither the providers nor the beneficiaries would support it and plans would view reduced payment as making the Medicare market less attractive, particularly given their experience with M+C.

In sum, managed care and competition under this scenario promises to potentially increase the enrollment in Medicare managed care, but there is little reason to expect that the forces that limit savings from managed care today will be any weaker under a competitive system. That being the case, most savings for the Medicare program that result from a competitive model are likely to come less from the efficiency of managed care or competition among plans than from the cost shifting of moving from a defined-benefit to a defined-contribution model. The cost of such savings is likely to be absorbed by Medicare beneficiaries in the form of reduced benefits or higher supplemental premium costs. The extent of these effects obviously will vary according to how tightly Medicare controls its contributions to plans.

Conclusions: The Bottom Line

Current evidence suggests that neither managed care nor competition are likely on their own to generate sufficient savings through efficiency to address the fiscal concerns currently facing Medicare or the health system more generally, a point others have made before.³⁷ Any savings will probably not offset the growing pressures on Medicare to expand benefits and address the demands created by the growth of technology and the baby-boom generation’s coming retirement. While managed care has demonstrated some potential to be more efficient, the level of savings is limited by the difficulty of changing providers’ practice patterns and reconfiguring care delivery. Political opposition to explicitly limiting access to available care and technologies also is strong. There may be many reasons to pursue the development of more organized care systems, but realistically, large and rapid cost savings to underwrite desired policy change is not one of them. The cost-savings potential of managed care is limited, not so much by managed care’s technology or competence of the industry (although shortcomings exist) as by our societal expectations and a range of stakeholder interests that lessen support for change needed to hold down costs. Whether these expectations and interests can be modified—and what it would take to do so—is worthy of discussion. Historical experience is not at all encouraging.

This leaves policymakers with the broader question of what social obligation Medicare bears for aged and disabled beneficiaries. If the obligation is for a relatively comprehensive Medicare benefit package that provides access to high-quality care and protects those it covers from financial ruin, much additional funding is required, whether or not managed care is used and competitive forces are in play. Conversely, if one believes that responsible public policy necessitates limiting the general financial commitment to Medicare so that funds can be used for other important purposes, there are savings to be had (at least against what spending might otherwise be). The cost of such savings, however, would be borne by Medicare beneficiaries, particularly those with lower incomes and less access to other subsidized sources of supplemental coverage. Policymakers will disagree on the equity of these disparities. Unfortunately, managed care and competition will not prevent the need for a debate about Medicare’s obligations and how to fund them.

This paper was commissioned by the Council on Health Care Economics and Policy for the Ninth Princeton Conference: Can Health Care Spending Be Contained? Support for this paper was provided by the council and through funding from the Robert Wood Johnson Foundation to Mathematica Policy Research (MPR) for Monitoring Medicare+Choice and Its Early Effects on Medicare Beneficiaries. Several colleagues provided valuable feedback on earlier drafts, including Randall Brown, Deborah Chollet, Robert Hurley, Tim Lake, Glen Mays, and Bob Miller. Felita Buckner provided secretarial support. Lily Chin edited the manuscript. Two anonymous reviewers at Health Affairs provided advice that helped strengthen the paper. David Colby of the Robert Wood Johnson Foundation oversees the Medicare+Choice project for MPR.

NOTES

1. V.R. Fuchs, “Health Care for the Elderly: How Much and Who Will Pay for It?” Health Affairs (Jan/Feb 1999): 11–21.
2. Among the departures are the use of contribution strategies that cushion incentives to select based on price and, in Medicare, administered pricing policies. While some view limitations of current practice as a major constraint in lessons to be learned from current experience, I disagree because current experience provides a good indication of what may be feasible and how alternative policies may perform under real-world conditions.
3. Mercer/Foster Higgins, National Survey of Employer-Sponsored Health Plans, 2001: Report on Survey Findings and Tables on Survey Responses (New York: William M. Mercer, 2002); and J. Gabel et al., “Job-Based Health Insurance in 2001: Inflation Hits Double Digits, Managed Care Retreats,” Health Affairs (May/June 2001): 180–186.
4. R.S. Brown et al., “Do Health Maintenance Organizations Work for Medicare?” Health Care Financing Review 15, no. 1 (1993): 7–23; and C. Zaraboza, C. Taylor, and J. Hicks, “Medicare Managed Care: Numbers and Trends,” Health Care Financing Review 17, no. 3 (1996): 253–261.
5. M. Gold, “Medicare+Choice: An Interim Report Card,” Health Affairs (July/Aug 2001): 120–138.
6. S. Christensen, “Medicare+Choice Provisions in the Balanced Budget Act of 1997,” Health Affairs (July/Aug 1998): 225–231.
7. T. Kornfield and M. Gold, “Is There More or Less Choice?” Fast Facts no. 1 (Washington: Mathematica Policy Research Inc., 1999).
8. Christensen, “Medicare+Choice Provisions.”
9. M. Gold and J. McCoy, “Choice Continues to Erode in 2002,” Fast Facts no. 7 (Washington: MPR, 2002).
10. Kornfield and Gold, “Is There More or Less Choice?”
11. L. Achman and M. Gold, Trends in Medicare+Choice Benefits and Premiums, 1999–2002 (New York: Commonwealth Fund, 2002).
12. M. Gold and J. McCoy, “Medicare+Choice Withdrawals: Experience in Major Metropolitan Areas,” Operational Insights no. 8 (Washington: MPR, 2002). See also A. Cook, T. Lake, and B. Schmitz, Early Experience under Medicare+Choice: Final Summary Report (Washington: MPR, 2002).
13. K.E. Thorpe and A. Atherly, “Medicare+Choice: Current Role and Near-Term Prospects,” 17 July 2002,
www.healthaffairs.org/WebExclusives/Thorpe_Web_Excl_071702.htm (6 March 2003); and K.E. Thorpe, A. Atherly, and K. Howell, “Medicare+Choice: Who Enrolls?” (Washington: Blue Cross and Blue Shield Association, April 2002).
14. L. Achman and M. Gold, Medicare+Choice 1999– 2001: An Analysis of Managed Care Plan Withdrawals and Trends in Benefits and Premiums (New York: Commonwealth Fund, February 2002); and U.S. General Accounting Office, Medicare Managed Care Plans: Many Factors Contribute to Recent Withdrawals; Plan Interest Continues,Pub. no. HEHS-99-91 (Washington: GAO, April 1999).
15. Brown et al., “Do Health Maintenance Organizations Work for Medicare?”
16. M. Gold, “ISO Quick Fix, Free Lunch, and Share of Pie,” Journal of Health Politics, Policy and Law 24, no. 5 (1999): 973–983; and other papers in the journal’s special issue on the managed care backlash, edited by Mark Peterson.
17. T. Lake, M. Gold, and R. Hurley, “HMO Provider Networks in Medicare+Choice: Comparing Medicare and Commercial Lines of Business,” Managed Care Quarterly 9, no. 4 (2001): 15–22; and T. Lake et al., Health Plans’ Selection and Payment of Providers, 1999, Report no. 20 (Washington: Medicare Payment Advisory Commission, 2000).
18. M. Brody, L.A. Brady, and D. Altman, “Media Coverage of Managed Care: Is There a Bias?” Health Affairs (Jan/Feb 1998): 9–25.
19. Gold, “ISO Quick Fix, Free Lunch, and Share of Pie.”
20. J. Grossman, B.C. Strunk, and R.E. Hurley, “Reversal of Fortune: Medicare+Choice Collides with Market Forces,” Issue Brief no. 52 (Washington: Center for Studying Health System Change, 2002).
21. B. Fried and J. Ziegler, “The Medicare+Choice Program: Is It Code Blue?” (Washington: Shaw Pittman for the Health Insurance Association of America, 2000).
22. See M.A. Laschober et al., “Trends in Medicare Supplemental Insurance and Prescription Drug Coverage, 1996–1999,” 27 February 2002,
www.healthaffairs.org/WebExclusives/Laschober_Web_Excl_022702.htm (6 March 2002); and M. Gold and J. Mittler, “The Structure of Supplemental Insurance for Medicare Beneficiaries,” Operational Insights no. 3 (Washington: MPR, 2002).
23. Gold and Mittler, “The Structure of Supplemental Insurance.”
24. The CMS has recently started to survey M+C organizations on their group enrollment. Incomplete data from reporting contracts show that 18.4 percent of enrollees were in groups in 2002. G.R. Mileman et al., “Medicare+Choice Individual and Group Enrollment: 2001 and 2002,” Health Care Financing Review 24, no. 1 (2002): 145– 153.
25. M. Gold et al., Medicare Beneficiaries and Health Plan Choice, 2000 (Washington: MPR, 2001).
26. Ibid.
27. Medicare Payment Advisory Commission, “Bringing Medicare+Choice to Rural America,” in Report to the Congress: Medicare in Rural America (Washington: MedPAC, 2001), 113–124.
28. RS Brown and M.R. Gold, “What Drives Medicare Managed Care’s Growth?” Health Affairs (Nov/Dec 1999): 140–149; and Grossman et al., “Reversal of Fortune.”
29. R.H. Miller and H.S. Luft, “HMO Plan Performance Update: An Analysis of the Literature,” Health Affairs (July/Aug 2002): 63–86; R.H. Miller and H.S. Luft, “Does Managed Care Lead to Better or Worse Quality of Care?” Health Affairs (Sep/Oct 1997): 7–25; and R.H. Miller and H.S. Luft, “Managed Care Plan Performance since 1980: A Literature Analysis,” Journal of the American Medical Association 271, no. 19 (1994): 1512–1519.
30. Brown et al., “Do Health Maintenance Organizations Work for Medicare?”
31. Institute of Medicine, Crossing the Quality Chasm: A New Health Care System for the Twenty-first Century (Washington: National Academies Press, 2001).
32. See, for example, N.L. Whitelaw and G.L. Warden, “Reexamining the Delivery System as Part of Medicare Reform,” Health Affairs (Jan/Feb 1999): 132–143.
33. Mercer/Foster Higgins, National Survey of Employer-Sponsored Health Plans, 2001; and Gabel et al., “Job-Based Health Insurance in 2001.”
34. R. Hurley, D. Freund, and J. Paul, Managed Care in Medicaid: Lessons for Policy and Program Design (Ann Arbor, Mich.: Health Administration Press, 1993).
35. T.C. Buchmueller, “The Health Plan Choices of Retirees under Managed Competition,” Health Services Research 25, no. 5 (Part I) (2000): 949–976.
36. Brown et al., “Do Health Maintenance Organizations Work for Medicare?”; Thorpe et al., “Medicare+Choice: Who Enrolls?”; and Gold and Mittler, “The Structure of Supplemental Insurance.”
37. K.E. Thorpe and A. Atherly, “Reforming Medicare: Impacts on Federal Spending and Choice of Health Plans,” 10 October 2001,
www.healthaffairs.org/WebExclusives/Thorpe_Web_Excl_101001.htm (6 March 2002).

Marsha Gold is a senior fellow at Mathematica Policy Research in Washington, D.C.

©2003 Project HOPE–The People-to-People Health Foundation, Inc.