Assessing the New Federalism is a multi-year Urban Institute project designed to analyze the devolution of responsibility for social programs from the federal government to the states, focusing primarily on health care, income security, job training, and social services. Researchers monitor program changes and fiscal developments. In collaboration with Child Trends, Inc., the project studies changes in family well-being. The project aims to provide timely, nonpartisan information to inform public debate and to help state and local decisionmakers carry out their new responsibilities more effectively.
Key components of the project include a household survey, studies of policies in 13 states, and a database with information on all states and the District of Columbia, available at the Urban Institute's Web site. This paper is one in a series of occasional papers analyzing information from these and other sources.
The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors, and should not be attributed to The Urban Institute, its trustees, or its funders.
Notes: Those wishing to print this report may find it easier to use the PDF Version.
Contents
Introduction
The managed care revolution of the past decade can be characterized as the aggressive pursuit of better value by purchasers of health services. Reduced rates of premium growth for private purchasers, especially large employers, attest to the apparent success of self-interested, best-price buying. Most purchasers have used managed care organizations (MCOs) to carry this out and to promote a systematic suppression of the cost-shifting that has been rife in the health care arena. In fact, it appears likely that most savings achieved to date are the result of fierce price and payment negotiation, rather than significant changes in patterns of care delivery.
Concerted efforts have been under way for more than a decade, and in a few states much longer than that, to adopt and adapt features of commercial managed care for the Medicaid program. Justifications for pursuing managed care are variously attributed to access expansion, quality enhancement, or improved accountability. But it is difficult not to conclude that cost-savings (and by implication, increased value if quality does not decline) have also been a principal motivator for Medicaid agencies to undertake major managed care initiatives. The Medicaid program's tradition of being the most penurious of payers makes the embrace of managed care with its cost-saving potential a natural fit.
But there are many reasons why managed care in the Medicaid context could hold less promise and encounter greater challenges than observed in the private sector. Some of these reasons relate to demographic, social, and clinical differences between Medicaid beneficiaries and employed groups. There are also significant differences in benefit packages and the expected duration of enrollment between commercial and Medicaid-covered persons. Medicaid beneficiaries have less discretionary income, which may interfere with using cost participation to shape choice of plans and demand for services. The traditionally low provider payment levels that have resulted in the limited and skewed nature of provider participation may affect the ability of plans to develop and negotiate with provider networks. Moreover, the fact that the Medicaid program has been saddled with numerous responsibilities beyond just purchasing care for those it covers poses an even greater threat to enabling MCOs to use leverage to get better prices for services.
Medicaid agencies face significant constraints in promoting managed care and competition among managed care plans. In fact, much of the recent history of Medicaid managed care illustrates the delicate balance of the program trying to be an aggressive buyer on behalf of its beneficiaries, without colliding with many other roles that Medicaid agencies have assumed or been assigned. These roles include protecting the safety net for persons without health insurance coverage; funding many community-based human service providers that bill Medicaid directly for client care; financing many state-sponsored services for mentally ill and developmentally disabled persons; and substantially funding medical education in public academic health centers. In addition, Medicaid agencies in many states play an integral advocacy role in assuring access to care for vulnerable populations and in advancing policies to address within-state disparities in access to care. This list of responsibilities weighs heavily on agencies as they discover that Medicaid's version of managed care requires greater caution and more deliberate attention to secondary consequences than the private-sector version or, for that matter, the Medicare experience. Furthermore, Medicaid managed care is being carried out in a politically charged and highly visible public arena that magnifies the contentious nature of negotiations and transactions in managed care.
This paper examines, through the findings of case studies conducted in 13 states as part of the Urban Institute's Assessing the New Federalism (ANF) project, how states have approached a number of the key areas in which adaptation of commercial managed care to the Medicaid program is necessary. The states cover a broad spectrum ranging from the very largest (California, New York, Florida, and Texas) to mid-sized (Michigan, New Jersey, Massachusetts, Wisconsin, Minnesota) to a number of other states (Washington, Colorado, Mississippi, and Alabama). They constitute, by design, wide geographic and political variation and varied development of public health and human service capacity.
The paper begins with a brief review of the general approach to Medicaid managed care taken by states. Next it covers some of the areas in which private-sector managed care meets distinctive challenges in Medicaid, and it derives several related questions that can be examined from the unique perspective of the ANF state-specific case studies. Variation among states on these questions is described to illustrate how attentive states have been to them and, in some instances, how successful they seem to have been in finding solutions. Finally, the paper draws broad conclusions about further developments and implications of Medicaid managed care, based on the accumulated experience of the 13 states in their efforts to adopt and adapt managed care.
Medicaid Managed Care-Is It More Medicaid than Managed Care?
The dramatic growth of Medicaid managed care, particularly in the past five years, has been well documented.1,2 Only a little over half of the enrollment is in capitated managed care in conventional HMOs, illustrating that Medicaid continues to rely on multiple models to gain some of the benefits of managed care for both the program and its beneficiaries.3 There is considerable descriptive evidence of the administrative challenges, setbacks, controversies, and successes states have encountered as part of the rapid buildup.4 There is far less evidence on the impact of Medicaid managed care, and much of it is from early models undertaken well before these alternative delivery systems emerged as the dominant form of care, at least for the commercial population.5 Moreover, evidence of the impact of HMO enrollment on the cost-savings for Medicaid is particularly inconclusive.6 In most cases, presumed savings have not been assessed relative to the favorable selection into health plans that may occur in voluntary enrollment situations, the most common form of HMO experience in Medicaid.
It is perhaps a testament to a near-blind faith in managed care that it has grown so rapidly with little convincing evidence as to its ability to provide states with more value for their expenditures. While premiums of private purchasers have leveled off or even declined, savings in Medicaid are more difficult to confirm and may also be more elusive to realize. A number of states anticipating that substantial savings could be recycled for other purposes, such as eligibility expansion, have found that savings have not fully materialized.7 Low-base Medicaid rates suggest that either savings have to be limited or serious provider or plan participation problems will arise-an area in which Medicaid has had ample and painful experience in the past. Indeed, problems with the willingness of many health plans to accept rates or participate at expected levels have slowed the pace of implementation or planned expansion.8 Finally, some aspects of Medicaid, such as limited cost-sharing by beneficiaries, prevent HMOs from achieving savings by influencing use of services through financial incentives for members as they do with private membership.
Ironically, in recent years there seems to be more reticence around continued expansion of Medicaid managed care because it might be saving dollars that have been used to achieve other public purposes or that come out of rates that have been viewed as marginally adequate at best. The widespread concern that disproportionate share hospital (DSH) dollars should not be incorporated into capitation rates to MCOs because they might not reach providers of indigent care is a good example of this ambivalence. More generally, safety net providers have voiced their fears that they may not be incorporated in MCO networks, or that, once in the networks, their payments may be sharply reduced, compromising their ability to render uncompensated care.9
Interagency tensions are growing as state efforts to use federal matching funds to finance mental health or developmental disability programs are being jeopardized by managed care initiatives.10 At the local level, after more than a decade of "Medicaidization," local health and human service agencies (which have been counseled to respond to financial pressures by obtaining a Medicaid provider number and billing for covered services) face an uncertain future when Medicaid funds are channeled through private MCOs that might not credential these providers for their networks.11 For academic health centers, major recipients of both DSH dollars and medical education subsidies, the stakes for a Medicaid conversion to managed care are substantial, particularly given the limited experience many of these centers have had with commercial managed care.12
These challenges raise several questions about how state Medicaid programs can adopt managed care initiatives and adapt the experience from commercial managed care. It is reasonable to expect that, as major health service purchasers, Medicaid agencies will approach the task of converting from payers to aggressive, value-based purchasers by emulating many of the activities of private employers who have embraced capitated managed care. Medicaid agencies will have to make a number of judgments regarding the suitability of local health service market conditions to accept and respond to the demands of managed care. There will have to be adequate capacity in terms of MCOs and providers with the experience and sophistication to develop products that will attract membership and contribute to successful cost and care management. For public and private purchasers buying care in multiple and varied markets, purchasing strategies will need to be adapted across markets. In addition, a managed care strategy must also be linked to other benefits-related developments that a purchaser may have planned or be implementing to ensure that managed care initiatives are not in conflict with other program objectives. Thus, the first question to be examined from the experience of the thirteen states in the ANF project is as follows: To what extent have Medicaid agencies customized their development of managed care initiatives to local market conditions and consciously positioned their approaches in a broader context of relevant policy changes?
Private companies that convert their health benefits programs to managed care models have considerable latitude in the planning, timing, and sequencing of development and implementation. Often conversion to managed care is undertaken because of perceived financial pressures due to rapidly rising health premiums. Firms typically exhibit deliberateness and caution to minimize employee disgruntlement, friction with unions, or perhaps adverse publicity and backlash from community providers. This caution seems to be grounded in recognition that change is sometimes difficult, especially when it relates to something as personal as health care. A measured approach also reflects a need to signal providers that change will be forthcoming and they will need to respond accordingly, such as developing MCO affiliations. Implementation must be done with deftness and foresight to deflect resistance and promote behavioral adjustment on the part of all affected parties, particularly when an initiative is new and threatening. Whether states approach these challenges differently from private purchasers is important to explore. Medicaid agencies have many diverse constituencies whose aims may be conflicting, they operate in a highly public and sometimes politicized environment, and they are frequently under tight schedules to implement initiatives to achieve savings. Do states have overall strategies to design and implement managed care initiatives, and how well are they able to phase in long-term strategies given the policy and political environments in which they operate?
The desire of purchasers to have more customized products and financing methods for their health benefits than were offered by traditional indemnity insurers has been a major driver of the managed care revolution. Recent evidence indicates that managed care companies continue to diversify their product portfolios to respond to buyers who may want a full spectrum of arrangements to offer employees. Product proliferation suggests that a broader choice set is available, allowing employees to select the models that will provide them with maximum value. Sophisticated self-insured purchasers may provide options that go beyond what capitated plans may offer and that, in some instances, are intended to protect the purchaser from experiencing the consequences of adverse selection in non-risk options. Compared to private purchasers, Medicaid agencies have an even more diverse clientele for whom they wish to do value purchasing, including persons with chronic disease and disability who have health and human service needs beyond those found in the typical managed care plan. Given the lack of cost participation by beneficiaries, the types of models offered and the bases for choosing plans will also be different from commercial experience. These contrasts pose the following question: How and why do states adopt the particular models of managed care they have chosen, and on what bases have they determined which populations to cover through various models?
Private purchasers commonly select MCOs using, among other criteria, so-called geo-access matching of network providers with employees' place of residence. In addition, large and influential purchasers may have the clout to induce or encourage MCOs to incorporate into their networks providers that are frequently used by the employee workforce or, conversely, to avoid deselection of such providers when networks are being downsized or reconfigured. MCOs may protest that these demands compromise their negotiating leverage with providers, but buyers that are big enough are likely to prevail. A similar experience is evident in the Medicaid program, as states normally require MCOs to incorporate traditional providers in networks to comply with access or cultural compatibility standards. Given the nature of private-sector managed care, the aims of most employers extend only to meeting the needs of active workers, retirees, and their dependents-not those of other consumers of services. But, as noted above, Medicaid agencies are perceived by many as protectors of safety net providers and advocates for access to care for indigent persons not eligible for Medicaid, substantially complicating pursuit of the lowest cost possible for beneficiaries. Consequently, plans selected by Medicaid agencies may face significant constraints on their ability to do best-priced buying as they act as agents for states that have multiple goals. From this discussion arises the following question: What kinds of accommodations are states making when selecting MCOs to avoid conflicts with the multiple roles that Medicaid agencies play?
Historically, private purchasers have paid prices for health care services well in excess of the costs to produce them, largely because of provider-based cost-shifting.13 It is hardly surprising then that aggressive negotiation has been the major instrument employed by MCOs to obtain better prices. Selective contracting, steering of volume, realignment of incentives, explicit service standards, and ambitious utilization management practices have been variously employed to extract cost and service concessions from providers who fear the consequences of not being included in networks. The success of MCOs in building membership volume can be translated to still greater leverage at the time of contract renegotiation.
This more lives-more leverage notion would also seem to apply to public purchasers, except that Medicaid, in particular, has already been engaged in these practices as a fee-for-service purchaser. In some respects, Medicaid could be facetiously referred to as the "mother of all preferred provider organizations (PPOs)," dealing with a limited subset of providers who have agreed to accept deeply discounted fees and to subject themselves to utilization management to minimize inappropriate use. Thus, lowering rates and shrinking networks to expand leverage are far more difficult to engineer when Medicaid has already taken these steps. Medicaid agency leverage is further weakened by the conditions they impose on plans concerning whom to include in their networks, as noted previously. Looming over this entire relationship is the political and practical uncertainty of doing business in the public domain with state governments. Such concerns point out the need to address the following question: In what ways are states approaching the rate-setting and contracting processes, given their tradition of historically low payment rates in Medicaid and the skepticism of MCOs and providers about the adequacy of payment and the reli-
ability of public purchasers as business partners?
Private purchasers of health benefits for employees and their dependents display a mix of paternalism and passivity, especially when they can offer an array of options that honor choice and preference for the sponsored individuals. Employers do this against a backdrop of both public regulatory and private accreditation mechanisms to afford a measure of comfort with what and from whom they are buying. Beyond this, employers typically invest in brokers and consultants to facilitate plan and product selection. Relatively few of them invest substantial resources in performance-assessment efforts, presuming that employees will literally vote with their feet if plan performance disappoints. Medicaid approaches may differ substantially on this dimension for several reasons. Compared to privately insured persons, beneficiaries may be far more vulnerable, their needs more diverse, and their experience with and capacity for exercising choice more limited. They may also lack resources to go "out of plan" if dissatisfied with access or care. Thus, oversight of plan performance for Medicaid beneficiaries takes on added importance. However, resources needed to mount oversight and monitoring initiatives may be unavailable or severely restricted, particularly during a period of downsizing in public sector employment. These concerns underscore the importance of considering what states are doing to gear up for these distinctive responsibilities: How are states performing and adapting to their expanded responsibilities of MCO oversight and consumer protection brought on by the rapid expansion of Medicaid managed care?
These six questions provide a template for exploring several themes and variation on these themes across the
states being examined as part of the ANF program.
Data Source: Assessing the New Federalism
On-site case studies were conducted in the aforementioned 13 states as part of a much larger project of the Urban Institute to examine the issues and implications of devolution and the evolving relationship between state and federal health policy.14 Medicaid managed care was one of several areas in which data were collected through interviews with multiple informed parties within and outside of state government. While all states had some experience with managed care, the variability was substantial.
The site visits were conducted between late 1996 and mid-1997. The interviews examined the evolution of Medicaid managed care in each state, including the goals of managed care initiatives, the selection and design of models, and the application of these models to various eligible populations. Specific issues included choice of providers and plans as well as contracting terms and rate development in states with capitated models. Efforts to develop oversight and program monitoring were also reviewed. In addition, specific attention was focused on how managed care initiatives take into account the needs of safety net providers. The following discussion draws on the summaries of these interviews.
State Variation in Adapting and Adopting Managed Care
The 13 states reflect enormous variation in their experience with and use of Medicaid managed care. Mississippi, Texas, and Alabama have limited or narrow experience with pilot programs of relatively recent duration, while Wisconsin, Minnesota, and Michigan have more than a decade of large-scale experience. New York, California, Massachusetts, and Florida have had a wide range of long-standing activity in Medicaid managed care, but now have under way or are on the verge of starting new and far more ambitious initiatives with statewide enrollment. It is this spectrum of experience that permits the six questions developed earlier to be examined.
To what extent have medicaid agencies customized their development of managed care initiatives to local market conditions and consciously positioned their approaches in a broader context of relevant policy changes?
Market Readiness
Although many of the 13 states can cite long-standing experience with Medicaid managed care-back to 1967 (New York) or 1972 (Michigan and California)-most had little experience with programs of substantial scale until the late 1980s. Accelerating enrollment is an even more recent development. In many instances, the early pilot projects were small, and were attempts at symbolic HMO participation with little interest or intent to expand, largely because managed care was so undeveloped, even for private purchasers. Medicaid managed care has consistently lagged behind the commercial market, in part because of the difficulties Medicaid has had in generating provider participation, and in part because of the lack of interest among commercial MCOs because of traditionally low rates.
As noted in the subsequent discussion of strategy, many Medicaid agencies have had to consciously encourage managed care development in order to have credible plans with which to contract. The current experience of Texas, where commercial managed care has only recently surged, illustrates the extent to which Medicaid depends on private buyers to create the market conditions to attract plans. In Alabama and Mississippi, which have the least-developed managed care markets among the 13 states, limited provider capacity and strong political sentiments appear to be significant obstacles to capitated managed care in general, and for Medicaid beneficiaries in particular. States with very high HMO penetration, such as California, Minnesota, and Massachusetts, have progressed further in Medicaid managed care, but they have not relied exclusively on capitated models, nor have they expanded beyond major urban locations.
As multi-site employers pursuing managed care across disparate markets have discovered, it is difficult for states to achieve program and coverage uniformity because of differences in market capacity and maturity. In particular, Medicaid agencies have had to wrestle with urban and rural variability. HMO enrollment remains a largely metropolitan phenomenon, as only in these markets can substantial health plan capacity be found. Much of Medicaid capitated managed care has stopped at the boundaries of urban areas. Urban parts of Florida, Massachusetts, Wisconsin, Minnesota, and Michigan have had long-term and large-scale experience with capitated models. Other areas in these states, however, often have no managed care, or perhaps only primary care case management (PCCM) programs. For example, Milwaukee and Minneapolis have had mandatory HMO enrollment for Aid to Families with Dependent Children (AFDC) beneficiaries for more than a dozen years, but state Medicaid agencies have only recently begun to expand this model to the remainder of their respective states.
Evolving Policy Context
The complexities and demands of large-scale implementation infrastructure are such that even states with considerable Medicaid managed care history struggle with leveraging this experience effectively. Most of the early programs were voluntary, meaning they did not entail autoassignment of nonselectors or exclusion of plans or providers. Nor did they use sophisticated rate setting or competitive bidding, which probably worked to the advantage of plans that could market themselves to produce favorable selection in voluntary programs. In recent years, states have learned a good deal about regulating market practices, managing enrollment (often through brokers), contract and rate development, and bidding and procurement processes. This means that states that are later developers of managed care initiatives have been able to accelerate the design and implementation phases by drawing on the experiences of others. For example, it is particularly notable how many states assert their aim to adopt a competitive bidding strategy similar to that in Arizona, once their own markets and systems mature sufficiently to support this. Likewise, many states now want to convert most or all of their Medicaid managed care to prepaid models as managed care markets have evolved and matured.
Many Medicaid managed care initiatives have been encased in broad state reform strategies. Washington, Minnesota, Colorado, Florida, and New Jersey intend to use savings from managed care to expand access for persons without Medicaid coverage. This kind of commitment sets high expectations for Medicaid managed care programs and creates a natural momentum to keep the expansion on track-often by gaining buy-in through the promise of expanded coverage from providers and advocates who might otherwise resist these developments. Another dimension of the evolving policy context is that a number of states (New York, New Jersey, California, and Colorado) have recently updated and expanded their MCO regulatory efforts to facilitate market maturation and consumer protection for commercial and Medicaid beneficiaries. Thus, to some extent, Medicaid can ride the crest of the managed care wave rather than attempt to be a market leader. A final contextual issue is the movement toward privatization of public human services. In many respects, expanding HMO enrollment in Medicaid has the effect of privatizing a public program one beneficiary at a time-a development highly compatible with broader contemporary trends.
Do states have overall strategies to design and implement managed care initiatives, and how well are they able to phase in long-term strategies given the policy and political environments in which they operate?
The case studies also provide an opportunity to reflect on the extent to which states have crafted coherent and deliberate long-term approaches to moving large numbers of beneficiaries into managed care arrangements. Unlike their private-sector counterparts, Medicaid agency executives operate in a much more politically volatile climate and face greater constraints on their administrative discretion. On one hand, they are particularly susceptible to fiscal pressures and tight schedules that are budget driven and may be viewed with disfavor by providers, plans, and beneficiaries. On the other hand, there is ample evidence from many states that clear, well-reasoned planning can build credibility, solidify support, and signal the need to make transitions and change behaviors. Florida, New Jersey, Washington, and Texas had overt strategies to entice commercial and other MCOs to participate by offering relatively attractive rates with modest demands to build contractor capacity as a precondition to program expansion. Michigan, Massachusetts, and Colorado have nurtured low-intensity models like primary care case management as a means to promote provider readiness for subsequent conversion to capitated arrangements or, in the case of Texas and Mississippi, to deflect organized resistance to capitated models.
Using explicit timelines for implementation brings these strategies to life and puts traditional Medicaid providers on notice that they have to prepare for coming changes. Particularly in the case of safety net providers who, because of their payer mix, have not yet encountered extensive managed care contracting or conditions, unequivocal signals can provoke strategic responses on their part, as in Florida and Wisconsin. Responses include prepaid plan formation, consolidation with similarly situated providers, and internal clinical and administrative adjustments to enhance attractiveness to potential contracting plans. Phasing strategies may also be necessary because of problems with MCO network capacity. This is the case in New York City, where serious concerns exist about whether the participating plans have developed large enough primary care networks to serve the large number of enrollees that mandatory enrollment will bring.
Conversion of programs from voluntary to mandatory status has proven to be especially challenging. California and New York illustrate how simply making careful default assignments for large numbers of beneficiaries is a major undertaking. A few states have phased in mandatory HMO enrollment directly from small voluntary enrollment programs, as seen in some regions of California. More commonly, states make the transition through mandatory PCCM models, as in Washington, Colorado, and Michigan, as providers and beneficiaries become more experienced in and accepting of using managed care. New Jersey, Wisconsin, and Minnesota have committed to mandatory HMO enrollment but have done so in specific markets according to defined timetables. Decisions about excluding services and populations, as discussed in the next section, may be critical to devising plans that are realistic and can succeed by pursuing a course of least resistance. Nearly all of the states have taken the path of least resistance relative to rural areas where, at most, PCCM programs have been or are being implemented (Florida, Colorado, Massachusetts).
Notwithstanding the variety in approaches, a strategic plan appears to contribute to successful implementation. Explicit plans and timetables provide affected parties-purchasers, providers, plans, and beneficiaries-with important information as they deliberate, adjust, and adapt to forthcoming changes. Conversely, vacillation in development and implementation may foster skepticism, which in term may encourage resistance to change. Or it may discourage involvement of those anticipating change and new opportunities that change may bring. Significant delays in program implementation can literally drive potential contractors out of a market-a concern openly voiced in Colorado and Texas.
How and why do states adopt the particular models of managed care they have chosen, and on what bases have they determined which populations to cover through various models?
The Transition toward Full-Risk Models
As in the case of private-sector managed care, it is evident that one model cannot fit all populations or market situations in the Medicaid context. While some models, like full-risk HMO contracting, are similar across public and private purchasers, Medicaid has devised other distinctive managed care models, such as the PCCM arrangement with or, more typically, without any financial risk-sharing. Developed initially in Michigan and Utah more than 15 years ago, the PCCM model has proved to be a remarkably robust, albeit limited, model that has gained widespread use throughout the country. It is the main managed care manifestation in a few states, including Alabama and Mississippi, and is used in tandem with other models in many more states, including at least 9 of the 13 ANF states. At a minimum this model has provided beneficiaries with a contractually obligated guarantee of access to a medical home 24 hours a day, 7 days a week-not a trivial contribution. And there is credible evidence to suggest the PCCM-based model has produced modest net savings while paying fee-for-service along with a case management fee.15
Massachusetts and Florida have been strongly wedded to the PCCM model as a competitive alternative to HMOs for Medicaid enrollees and to benchmark HMO performance. Michigan, Washington, and New York (particularly upstate) have relied on partially capitated PCCM programs as transitional strategies until markets have matured to the point where full-capitation models become available and attractive to providers. Even more commonly, PCCM has held strong appeal in states such as Alabama and Florida that want to use managed care on a statewide basis, but see few prospects for capitated arrangements in rural areas in the near future. Moreover, the PCCM model has allowed community health centers and local health department clinics to participate without having to make major operational changes or risk loss of favorable compensa-
tion methods.16
The limitations of PCCM, including minimal care management, lack of incentives for integration, and the fact that financial risk remains with the Medicaid agency, have led states such as California and Colorado to decide to move to full-risk arrangements as soon as possible. States that operate parallel programs, such as Massachusetts, Colorado, and Michigan, are attempting to shift more beneficiaries into HMOs by autoassignment and other mechanisms. Florida and Texas are essentially giving beneficiaries equal choice between models and defaulting only to the PCCM. The problem with this approach is that it may impede growth of HMOs while contributing to adverse selection for the Medicaid agency in its residual fee-for-service PCCM program. Few states (Massachusetts and Michigan may be exceptions) appear to have sufficient data or analytical capacity to determine if this, in fact, is happening.
Whether building from a PCCM program or from a limited capitation pilot project, the movement to full capitation is widely evident across the ANF states. Such an approach provides the sought-after risk-shifting to MCOs and the guarantee of savings that only mandatory full capitation can provide. It also conforms Medicaid managed care more with commercial managed care initiatives, which comports with the avowed goal of trying to capture for Medicaid the managed care gains of private-sector purchasers. Another important consideration is that a full-capitation approach can permit states to offload administrative functions like provider payment and customer and provider relations to prepaid plans in order to facilitate downsizing of state government and privatization of some previously public-sector functions.
Special Population Accommodation
The limitations of simply adopting commercial managed care models into Medicaid are perhaps most apparent when comparing the Medicaid population with privately insured individuals. A substantial portion (approximately
70 percent) of Medicaid beneficiaries are young, healthy individuals who are generally similar in medical needs to the employed persons and their dependents found in private groups. But a sizable part of the 35+ million Medicaid beneficiaries have far more substantial health and other human service needs than are commonly found in commercial groups, due to serious chronic disease or disability. These beneficiaries represent nearly 70 percent of Medicaid expenditures.
Most Medicaid beneficiaries with disabilities and serious chronic illnesses have remained in the fee-for-service system, where they have enjoyed a
high degree of freedom to move among providers and, in many cases, self- or family-designed care systems that may include clinical, social, and other supportive services. Resistance to enrollment in managed care models, even low-intensity models like PCCM, has been high among beneficiaries, their families, advocates, and specialized providers who have traditionally delivered much of their care. In addition, considerable heterogeneity among these individuals makes developing sizable programs infeasible for commercial MCOs. Commercial managed care organizations enrolling such members have
typically dealt with their needs with individualized high-cost or catastrophic case-management approaches.
A high percentage of the most needy of these individuals have dual eligibility for Medicaid and Medicare, creating substantial, if not insurmountable, obstacles to designing authentic capitated managed care models. Finally, in some instances persons in institutional settings (nursing homes, intermediate care facilities for the mentally retarded, or foster care) are excluded because of the unsuitability of managed care models for their situations, or because the logistics of managed care for them are impractical. A rare exception to this can be found in some of the "county operated health system" programs in California, where nursing home services are part of the capitated program, and in Minnesota, which is in the process of developing an experimental program.
Consequently, most states have exempted beneficiaries with disabilities or serious chronic illness from mandatory programs and have merely allowed them to enroll on a voluntary basis if they so choose, even in California and Minnesota, which have the highest HMO penetration rates in the country. Massachusetts and Wisconsin have operated small pilot capitated programs to gain experience in contracting, performance monitoring, care coordination, and, perhaps most significantly, risk-adjustment for capitated rates for these beneficiaries. Similarly, New York plans capitated programs for certain targeted subgroups of the SSI population. New Jersey is deferring enrollment of persons with disabilities or serious chronic illness until full implementation of capitated care for AFDC/TANF (Temporary Assistance for Needy Families, AFDC's successor) beneficiaries is completed. Still other states seem content to make PCCM programs mandatory for Supplemental Security Income (SSI) beneficiaries, at least until more evidence emerges that HMOs can ably serve them and states can find ways to pay appropriately for their care.
In addition to this population exemption strategy, states are using benefit or service carve-outs, most commonly for mental health services, to avoid conflicts with special needs populations. Although there are many variants of this approach, these arrangements essentially separate coverage for mental health services from the physical health benefit package and from the purview of plans or providers responsible for physical health services. Several states, including Florida, Michigan, and New York, have or are devising carve-out programs for managed mental health services. Massachusetts has one of the most widely known programs, though it is only for beneficiaries in the PCCM model. Colorado is one of the few states seeking to integrate its mental health carve-out into fully capitated plans.
Decisions surrounding mental health carve-outs often have administrative and interagency rationales that may supersede clinical ones.17 Purchasing mental health benefits is often a complex, contentious, and politically charged issue because so many services to the mentally ill have traditionally been publicly financed and provided. This is a particularly good example of how Medicaid managed care differs from commercial managed care, where behavioral benefits are typically sharply limited in part because the public mental health safety net plays the role of a catastrophic coverage plan. State Medicaid officials must be acutely concerned about their potential to dismantle or undermine this mental health safety net. The decision to carve out mental health is often seen as the least disruptive option.
What kinds of accommodations are states making in selecting MCOs to avoid conflicts with the multiple roles medicaid agencies play, including protectors of safety net providers?
The transformation from passive payer to aggressive purchaser positions large buyers, both private and public, to contribute more directly to shaping health care markets and the configuration of delivery systems. Purchasers wield considerable power over the financial viability of providers. Most private purchasers and their managed care agents, however, remain intent on getting the best value for those they cover, with less regard for how providers fare in the process. Substantial excess capacity in many provider markets, especially in metropolitan areas, facilitates aggressive price negotiation.
Medicaid's position in this regard is clearly more difficult. Provider participation is limited to begin with, and many high-volume Medicaid providers are in underserved markets and often in precarious financial positions. They are also less likely to have already adapted to the demands of managed care because of low commercial payer exposure. Consequently, infrastructure and other system upgrades may not yet have been made. They are also far more likely to be rendering a sizable amount of indigent care given their location, reputation, and mission. As a result, many of these providers have been targeted for special subsidy programs to aid them in meeting both their Medicaid and indigent care responsibilities, including DSH payments and cost-based reimbursement for federally qualifed health centers (FQHCs). These integral components of the financing system of the safety net raise the stakes surrounding Medicaid managed care substantially.
Most states have taken care to ensure a role for high-volume Medicaid providers in both risk- and non-risk-based managed care models. FQHCs have participated in PCCM programs for many years, and most have been able to join HMO networks. In several states such as New York, Massachusetts, Florida, and California, they have formed alliances that have ultimately created their own health plans. Alabama, Michigan, New York, and Texas have provided special opportunities for high-volume hospitals and academic health centers to develop their own plans to facilitate retention of Medicaid revenues or to use Medicaid managed care to foster conversion to integrated delivery systems. Likewise, bids from commercial HMOs often require, or reward in scoring, the incorporation of high-volume Medicaid providers, including hospitals, in their networks.
More overt attempts to preserve and protect these providers include extracting DSH dollars from capitation payments and passing them directly to providers or to plans formed by or for them, as is the case in Minnesota. Many states have a policy of making FQHCs "whole" by paying them the difference between reasonable costs and the rates paid to them by MCOs, as is done in Wisconsin. Washington enabled FQHCs to participate as prepaid plans by providing them with access to stop loss insurance. Michigan allows newly formed prepaid entities, many of which consist of traditional Medicaid providers, up to 18 months before they must obtain a license as a bona fide risk-bearing organization.
In what is probably the boldest and most controversial effort to preserve public-sector providers, California devised its two-plan model for use in several counties that have their own health care delivery systems. Under this model, in which a county-sponsored plan competes for Medicaid enrollees with a private MCO, larger payments are made to the county-sponsored plan than to the private or "mainstream" plan because the former must incorporate all safety net providers. The very protracted period to make this model operational was intended to give these new systems time to develop the wherewithal to be successful cost and care managers. The model was motivated in part because of the distinctive problems faced by California's county health systems, which are the providers of last resort for a substantial undocumented alien population. Texas is another state that is currently wrestling with equally complicated dynamics vis-?-vis public hospitals. State and local officials recognize that capitated managed care may threaten some of the elaborate and complex methods that use local hospital district taxes to maximize Medicaid matching funds and to allocate funds from DSH pools. Development and implementation of Medicaid managed care are being modified to address these concerns.
These special accommodations for traditional Medicaid providers have fostered a new generation of predominantly Medicaid prepaid MCOs. Ironically, many states (Michigan, New York, Texas, California, New Jersey, Wisconsin, and Minnesota) have explicitly asserted their desire to avoid undue reliance on Medicaid-only plans, even though several used time-limited 75/25 waivers, when they were required, to allow start-up Medicaid-only plans to participate. These new plans seem to reflect a kind of de facto acceptance that predominantly Medicaid plans may be inevitable, or at best a necessary compromise to ensure that providers with substantial Medicaid and indigent care commitments are not unduly jeopardized. Criticisms about the lack of a level playing field have been raised by commercial plans, whick seem increasingly skeptical about states' interest in doing business with them. But surprisingly little criticism has been forthcoming from beneficiaries and advocates. For the most part, they have not yet taken umbrage with the probability that this tilt may distance Medicaid beneficiaries from mainstream plans and providers.
In what ways are states approaching the rate-setting and contracting processes, given their tradition of historically low payment rates in Medicaid and the skepticism of plans and providers about the adequacy of payment and the reliability of public purchasers as business partners?
Determining the appropriate methods and amounts to pay managed care plans and providers is very challenging for Medicaid agencies. In part this is because they have already achieved below-market prices for many services, albeit while sacrificing broad provider participation and, some would suggest, access to high-quality providers. Additionally, some payment arrangements are a function of federal requirements or legislative or judicial mandates and thus non-negotiable. Practically speaking, contracting with capitated managed care plans appeals to Medicaid agencies because it liberates them from some of the payment strictures they operated under in the traditional Medicaid program, such as the recently repealed Boren amendment. In the case of PCCM programs, the payment of a case management fee on top of fee-for-service has been viewed as a long-overdue fee increase.
In developing capitation payments for HMOs, states have had to set rates that do not exceed the upper payment limit (UPL), based on fee-for-service experience, while still trying to attract plans to enter into contracts. In the early days of capitated Medicaid managed care, rates were very crude, typically with minimal adjustments even for age, gender, eligibility category, or region, and may have systematically overpaid plans because they tended to attract lower cost beneficiaries, especially when direct marketing was permitted. States are candid about their willingness to have accepted such outcomes in the interest of attracting prepaid organizations, since without them further market development or competition was not possible. Many of these same states acknowledge, however, that now that there are many participants and rate-setting sophistication is far less rudimentary, they have begun to tighten rates, with Florida and New York instituting double-digit rate rollbacks. In addition, in many states there are explicit plans to move to competitive bidding as soon as possible in the hope that market-determined rates can be achieved.
It is striking how pervasive the desire to institute competitive bidding is among the ANF states. It seems grounded in a belief that this is where Medicaid managed care should be going, a viewpoint that has been nurtured by actuarial and other consultants. New Jersey, Washington, Michigan, Colorado, Florida, New York, and California are all in the midst of trying to convert from an administered rate to competitive bidding in which price is one of the prominent criteria, though some states, such as Florida, have had serious problems and legal challenges. States appear to believe that competition can and will intensify under this approach and that state budgets and beneficiaries could gain. There is also a general realization that the dwindling fee-for-service base makes reliance on administered rates less meaningful and reliable. Some states seem reluctant to proceed down this road, in part because they worry that if the rates to plans are too low, provider payment will suffer accordingly, setting in motion a number of the issues raised in the earlier discussion of safety net providers.
Health plan trade associations that participated in some of the case study visits raised concerns about the wisdom of competitive bidding. They commonly pointed out that the UPL is much lower than would be paid for commercial populations (California's rates are a dramatic example of this) because most states historically have paid providers at well below market rates. In addition, high-use rates among some beneficiaries have been declining, and in some instances managed care may improve access for persons needing additional care. Both of these developments indicate that savings possible through aggressive utilization management are limited. They also note that contractual requirements in the Medicaid program exceed those in commercial contracts. Other administrative functions are more burdensome because of high beneficiary turnover and the need for substantial educational and supportive services not commonly rendered under commercial contracts. So despite lower payments, the Medicaid product proves more costly.
Beyond rates, a broader concern is that Medicaid agencies and health plans have had difficulties in establishing long-term business partnerships (an issue that Colorado and New Jersey have explicitly attempted to address). Many MCOs find doing business in the public sector is marked by volatility and arbitrariness. For example, in competitive bidding states may award bids based on rates that are simply inadequate due to bidders' naivet? or lack of candor, and more credible plans may, as a result, be excluded. In addition, previously described accommodations made to allow newly developed, provider-
sponsored, Medicaid-only plans to bid for and win contracts without any track record of managing risk could ultimately drive out established plans. Only Minnesota currently requires HMOs to participate in Medicaid as a condition of licensure. A few others have used financial incentives/sanctions (New York) or tied access to other state government lines of business to Medicaid participation (Washington).
Unlike private purchasers, who are more likely to buy standardized managed care products, Medicaid agencies are more extensively engaged in specifying products to be acquired. This is because of public procurement and contracting policies and because agencies believe it is necessary to ensure beneficiary needs are being met. Most states think they have made progress in devising more sophisticated contracts over time, detailing more completely the structure and operation of the delivery systems from which they wish to purchase. Capacity, access, and service standards are becoming more explicit, data and reporting requirements more rigorous, and consumer protection features more demanding. Counterbalancing this progress, at least in the minds of some observers interviewed in the case studies, is a question about whether states possess the staff and other resources to monitor and enforce these contracts effectively.
How are states performing and adapting to their expanded responsibilities of MCO oversight and consumer protection brought on by the rapid expansion of medicaid managed care?
Compared with private-sector managed care, public-sector managed care programs encounter more oversight and consumer-protection responsibilities due to public accountability and the vulnerabilities of the beneficiary populations. Two areas in which states have experienced problems that have called out for better oversight and consumer protection are marketing and enrollment practices and the assignment of persons who do not choose to plans or providers in mandatory programs. Approaches to both of these problems are promising and are instructive about a third and somewhat broader issue of how ably states are mounting comprehensive program monitoring efforts.
In the early phases of voluntary HMO enrollment, states permitted plans considerable latitude in terms of marketing practices, materials, and inducements since the plans faced a challenging task of drawing beneficiaries out of familiar fee-for-service into unfamiliar managed care. But this laissez-faire approach had some notable negative consequences. It invited and, in effect, rewarded aggressive and sometimes misleading and unethical marketing behavior. The full extent of such behavior is unclear, but it seemed to reach its culmination when a series of newspaper expos?s in Florida documented some of its most unsavory features. In the wake of this series, many states sharply curbed direct marketing, eliminated plan-based enrollment processes, instituted review and approval processes for marketing materials, and more aggressively policed and sanctioned plans when practices became suspect. Not only were these marketing practices confusing and exploitive of beneficiaries, but many were also intended to foster favorable selection into plans that, in turn, resulted in financial losses for Medicaid agencies.
Medicaid agencies across the country have adopted a major innovation in recent years: the use of independent enrollment brokers or benefits counselors to provide information, education, and choice support for beneficiaries. These contractors have appeal both because of their independence and because they can provide states with a substantial cadre of personnel to assist with rapid buildup (California and New York are good examples of this) without adding new state agency employees. Brokers are frequently engaged by state agencies as a response to criticism from legislators, advocates, and providers who fear that accelerated implementation of prepaid managed care invites both exploitation and massive confusion. Only a few states, including Wisconsin and Minnesota, have not joined in this movement, though Minnesota did use a broker when its initial programs began in the mid-1980s. Although their track records have been uneven and many contracts are very recent, a number of states, such as New Jersey and Massachusetts, see enrollment brokers as an integral feature of their successful conversion to managed care. Controversy has arisen in a few states (California, for example) around broker performance and the value brokers add. This is an area, therefore, that will merit future attention.
Brokers have also been valuable because in mandatory programs, unlike private-sector managed care, there is the need to ensure that all persons are enrolled in a plan or with a physician in PCCM models. There are wide variations in the level of default or autoassignment, as this allocation of nonselectors to plans or providers is called, though this may be due to differences in definitions and process differences among states rather than a true measure of success in assigning beneficiaries.18 Autoassignment has become a policy instrument used by states in various ways, in part because some evidence suggests that autoassignees enjoy better health and thus represent good risks to prepaid plans. Some states, such as Michigan, award assignees based on best bid scores; others, such as New York and California, make a disproportionate assignment to plans that are sponsored by or include safety net providers. Still others, such as Colorado, autoassign to HMOs rather than PCCM programs when they are side-by-side options to increase HMO enrollment. Some states, such as Florida and Texas, default to PCCM programs rather than HMOs because they want to encourage HMOs to compete effectively to attract beneficiaries, or because political pressure is being brought to bear by medical associations.
Adding to the challenges of contracting, rate development, marketing, and enrollment are those of ongoing program monitoring and oversight of quality of care. Nearly all states are increasing their resource investment, including personnel and contracted services, to bolster their program administration. This is not surprising, as nearly 40 percent of the Medicaid population is now enrolled in managed care arrangements. But there are problems with meeting some of the skill needs of a value-purchasing organization with largely bill-paying experience. The slow pace at which states have enhanced their quality monitoring activities-a pattern that has been evident in Medicaid managed care for more than a decade-illustrates how difficult it is to gear up at a time when state government is being downsized or many attractive opportunities in the private sector (often with MCOs) beckon. As in the case with enrollment brokers, many states are increasing their reliance on external review organizations to perform core quality-monitoring functions.
As part of the larger backlash against managed care, beneficiary advocates and other critics are questioning whether enough resources are being invested in plan oversight and whether data that have authentic value in assessing plan performance are being collected and analyzed. Nearly all of the states studied reported moderate to severe difficulties in collecting encounter data from capitated plans. Ultimately, because such data may be necessary for both program monitoring and future rate development, most states have interest in resolving these difficulties.
Discussion and Implications
The review of the experiences of the 13 ANF states reveals many differences between Medicaid managed care and private-sector managed care, as well as some important similarities. Clearly the emphasis for Medicaid has been on adapting the features and techniques of managed care to suit its particular needs. The need for customization is hardly surprising given the continuing evolution of managed care, as MCOs attempt to satisfy the demands of aggressive purchasers. As managed care continues to expand in Medicaid, and there seems little reason to believe it will not, it will be important to determine the extent to which Medicaid's versions of managed care become more like commercial models or the extent to which differences in fact grow. Customization of private-sector managed care that Medicaid has undertaken to date merits recapitulation.
First, Medicaid agencies have had to moderate their managed care initiatives to allow private-sector managed care to create both acceptance and momentum in local markets. It is very difficult for Medicaid to lead the market into managed care because its provider constituency is often limited in numbers and financially needy. Until recently, few of these providers have had substantial managed care exposure. They commonly lack the wherewithal to readily develop their own managed care products, and may only be motivated to do so if their current Medicaid revenues are threatened. Moreover, it is difficult to attract plans into Medicaid managed care, especially commercial HMOs, unless they have concurrent opportunities to develop or expand on a private-sector business. Reliance solely or predominantly on a Medicaid line of business may simply be too daunting for plans to accept. One implication of this is that Medicaid agencies' efforts to promote capitated managed care in rural areas where most commercial plans have not ventured and most providers have had little experience may be more difficult than imagined, and the yield from doing so may be very limited.
Private-sector experience may be less instructive in creating Medicaid initiatives to bring managed care to special-needs populations. In this instance, Medicaid may not be able to await private-market "softening" of resistance, simply because private employers are not covering many of these persons and certainly not on the same scale as Medicaid (and to a lesser extent Medicare). Costs to the Medicaid program of these beneficiaries are so high that states will likely not continue their policies of exemption or population and service carve-outs. Therefore, Medicaid agencies will likely have to emerge as the market leaders and promoters of innovation for these persons. It will be challenging, but critical, for them to foster and nurture experimentation and innovation in this area.
Second, most of the states reviewed in this paper have developed strategic plans of varying degrees of specificity to guide their development and implementation of managed care for Medicaid beneficiaries and have attempted to remain faithful to them. But this is very challenging when the entire constellation of forces and factors needed for success are difficult to assemble, manage, and maintain. Managed care programs are complicated, multi-faceted initiatives that depend on coordinating model design, RFP or contract development, rate setting, contractor selection, beneficiary education, marketing and enrollment, and ongoing monitoring capacity-each of which has its own complex features. The cumulative experience of states seems to have increased awareness of the scope of activities to be undertaken and the skill sets needed to carry them out. There are emergent industries of enrollment brokers, marketing and consumer education specialists, rate consultants, and quality monitoring agencies to fill these needs.
Strategic planning also serves the function of galvanizing thinking among all affected parties in advance of implementation. In some instances, this means broadly based input on design and development features that improve the initiative and achieve buy-in and support from those who might have been resistors. Texas, for example, held more than 20 public hearings around the state to introduce its plan. An open process may lead to a major redesign when resistance mobilizes and may push the program structure in a different direction, such as the two-plan model experience in California. On the other hand, if the planning is not done openly or the rollout is seen as heavy-handed, then opposition may coalesce, potentially leading to protracted negotiations. New York's initial Section 1115 waiver-based plan is cited by critics as an example of too little community participation. The counter-argument to such deliberateness is to move ahead expeditiously to prevent resistance from developing a united front. But the program may then expend enormous resources in remedial redesign correction and relationship repair.
Third, the variety of models in use in Medicaid managed care is hardly surprising, given the diversity of beneficiaries, goals, resources, and constraints with which state agencies work. Moreover, early efforts to devise models that addressed pressing concerns, such as using PCCM to achieve guaranteed access to care, may reflect that this was as much managed care as Medicaid could provide at that time, when private managed care was not yet flourishing. Private purchasers have been successful in encouraging managed care organizations to contract aggressively with health care providers for price and service concessions that could be translated into lower premium increases. As discussed in detail in this paper, Medicaid faces significant disadvantages in using its lives for leverage to negotiate still lower rates or to threaten to deselect providers who will not accept terms when provider participation is limited to begin with. And the overall inadequacy of fee-for-service rates makes Medicaid capitation rates based on them unattractive to many MCOs.
Most states now seem on track to move a substantial portion, if not all, of their AFDC/TANF and related beneficiaries into capitated managed care, though the PCCM model still has significant support in Florida, Texas, Alabama, and Massachusetts. The models for SSI beneficiaries are likely to be less standardized and could even cause a resurgence in PCCM models, particularly if risk-adjustment methods for these persons remain unresolved, or if beneficiaries and advocates remain skeptical of the degree to which capitated MCOs can meet their needs. In general, states will probably continue with multiple models, especially given within-state geographic diversity. The array of products and customization mirrors the experience of many large, multi-site private employers, which may have multiple competing HMOs in markets where they have large workforce concentration; PPOs or other looser forms of managed care in areas where small groups of employees are clustered; and even indemnity coverage for widely dispersed workers, such as members of the salesforce.
Fourth, perhaps the most notable and, in some ways, most difficult adaptations to managed care found in the Medicaid program revolve around selecting and qualifying plans and providers. The difficulty lies in part with the fact that some providers on whom Medicaid has greatest dependence have a critical dependence on Medicaid. Thus efforts to reduce or redirect Medicaid spending can have ominous consequences for providers with long-standing commitments to Medicaid beneficiaries. Geography, mission, and tradition also mean that many of these institutions provide a substantial amount of care to persons without insurance, including Medicaid beneficiaries when they lose coverage. Medicaid has been the conduit through which public policymakers
in many states have transferred dollars to these providers for their indigent care. As Medicaid alters its patterns of purchasing, or delegates purchasing to private MCOs, there will be cascading effects for persons with and without Medicaid coverage, and for the providers to whom they have traditionally turned for care.
The contortions in Medicaid managed care because of safety net and related concerns are both impressive and daunting. The creativity and diligence displayed to keep safety net providers "in the game" are notable and necessary, when viewed from the vantage point of these providers who contend they need time and assistance to adapt to a managed care world. Many of them assert correctly that they labor under considerable disadvantage because of their indigent care commitments. But the danger that lurks about this strategy to accommodate safety net providers is that it may put Medicaid beneficiaries at risk of having to remain permanently in delivery systems determined by where they have had to get their care in the past. This may deny beneficiaries the opportunity to use managed care as a means to get improved access to mainstream plans and providers. It is far from clear to what extent managed care could deliver on this promise of mainstreaming. But if state policies give substantial, sustained advantages to plans built around traditional Medicaid providers, then commercial plans could leave this market and mainstreaming will become unlikely.
Fifth, capitated managed care in Medicaid will likely continue to be plagued by concerns about rate adequacy because capitation rates cannot by law exceed the fee-for-service base that has been so discredited in the past. Part of the difficulty for capitation rates seems to lie in the fact that they accumulate low payment rates for an entire package of services and bundle them into a single payment amount that underscores its inadequacy, as seen in the negative reaction to capitation rates in California. Providers in the past may have been willing and able to cost-shift among different lines of business to subsidize at least some Medicaid care. But it is far less likely that prepaid health plans will do this over an extended period of time, particularly as they experience declining profitability in commercial accounts. Moreover, plans that seek to limit their Medicaid membership to minimize financial exposure may conclude that the cost of compliance with increasing demands of Medicaid contracts exceeds any expected gains.
It remains to be seen what effects overt efforts to allow high-volume Medicaid and other safety net providers to become capitated providers will have on rate setting and cost-savings over time. Many of these providers may be willing to accept whatever rates are proffered by Medicaid to maintain revenue flows to institutional providers or to allow them to build membership rapidly. Because many providers are so inexperienced at risk, cost, and care management, it is hard to be sanguine about their ability to live within these rates or meet other contractual performance terms such as detailed reporting requirements. States may be stymied, therefore, in extracting contract compliance and future savings, and, in some instances, may have to increase rates to protect providers from financial crisis. Such concerns are not nearly so apparent in private-sector purchased managed care.
Sixth, Medicaid agencies have made real progress in tackling problems of enrolling large numbers of beneficiaries, promoting informed choice, and devising methods to allocate to managed care arrangements persons who do not make affirmative choices. While progress in each of these areas has been uneven across state programs, significant collateral learning among states is evident. The most pernicious marketing practices have been effectively eradicated in most states, and states have proceeded to tackle more subtle and insidious challenges, including the need to develop effective consumer education and information-sharing mechanisms. Enrollment broker contracting and competence seem to be growing as this embryonic industry meets both manpower and expertise needs for states at a time when state workforces are being reduced. In addition, states have determined several thoughtful ways both to execute autoassignment and to use it as an instrument for public policy purposes.
States freely admit to lagging in the development of quality oversight and other monitoring functions, though it is difficult to conclude that they differ much from their private-sector counterparts in these domains. What may be different is that Medicaid agencies feel compelled to do more than private purchasers because of their beneficiaries' vulnerability and the extent to which their choices may be more restricted either by program design or by their limited personal resources. So the burden to pursue an ambitious set of oversight functions weighs more heavily on public policymakers and program managers than on private-sector purchasers. Whether Medicaid agencies can muster the resources needed to do this in a period of waning public-sector employment remains to be seen.
Conclusion
State Medicaid agencies have achieved considerable success in adopting and adapting managed care for their beneficiaries, particularly over the last half decade. This is evident not only in the aggregate number of enrollees, which has increased more than fivefold over this time period; agencies have also displayed creativity in program development and design and reasonable caution, in most cases, in the scale and pace of implementation. The models employed have both dovetailed with and departed from those in private-sector managed care for reasons related to market conditions, administrative limitations, and beneficiary needs. Perhaps most notably, Medicaid agencies have been able to engineer major managed care arrangements for their beneficiaries, while, by and large, leaving in place most of the indigent care safety net. But continued expansion of managed care for persons with public and private coverage remains on a potential collision course with the cost-shifting that finances much of the uncompensated care in this country. It will be important to continue to monitor Medicaid's experience on this path.
Notes
1. Health Care Financing Administration. 1996. Medicaid Managed Care Enrollment Report. Baltimore, MD.
2. Hurley, R., L. Kirschner, and T. Bone. 1997. "Medicaid Managed Care." In P. Kongstvedt (editor), Essentials of Managed Health Care 2nd Edition. Gaithersburg, MD: Aspen Publications, Inc.
3. Health Care Financing Administration. 1996. Medicaid Managed Care Enrollment Report. Baltimore, MD.
4. Gold, M., M. Sparer, and K. Chu. 1996. "Medicaid Managed Care: Lessons from Five States." Health Affairs 15(3): 153-166.
5. Hurley, R., D. Freund, and J. Paul. 1993. Managed Care in Medicaid: Lessons for Policy and Program Design. Ann Arbor, MI: Health Administration Press.
6. Buchanan, J., A. Leibowitz, J. Keesey, J. Mann, and C. Damberg. 1992. Cost and Use of Capitated Medical Services. Evaluation of the Program for Prepaid Managed Health Care. Santa Monica, CA: RAND.
7. Gold, M., M. Sparer, and K. Chu. 1996. "Medicaid Managed Care: Lessons from Five States." Health Affairs 15(3): 153-166.
8. Meyer, H. 1997. "Medicaid: States Serve Up a Real Turkey." Hospitals and Health Networks 71(22): 22-24, 26, 28.
9. Lipson, D., and N. Naierman. 1996. "Effects of Health System Changes on Safety Net Providers." Health Affairs 15(2): 33-48.
10. Bachman, S., and B. Burwell. 1997. Medicaid Carve-Outs: Policy and Programmatic Considerations. Princeton, NJ: Center for Health Care Strategies, Inc.
11. Hurley, R., and D. Draper. 1998. "Managed Care for Medicaid Beneficiaries: An Overview." In P. Halverson, G. Kaluzny, and C. McLaughlin (editors), Managed Care and Public Health. Gaithersburg, MD: Aspen Publications, Inc.
12. Iglehart, J. 1994. "Rapid Changes for Academic Medical Centers." Part 1, New England Journal of Medicine 331(20): 1391-1395, and Part 2, New England Journal of Medicine 332(6): 407-411.
13. Medicare Payment Advisory Commission (MedPac). 1998. Report to the Congress: Medicare Payment Policy. Washington, D.C.
14. Holahan, J., S. Zuckerman, A. Evans, and S. Rangarajan. In press. Medicaid Managed Care: Variation in State Approaches. Washington, D.C.: Urban Institute.
15. Hurley, R., D. Freund, and J. Paul. 1993. Managed Care in Medicaid: Lessons for Policy and Program Design. Ann Arbor, MI: Health Administration Press.
16. Hurley, R., and D. Draper. 1998. "Managed Care for Medicaid Beneficiaries: An Overview." In P. Halverson, G. Kaluzny, and C. McLaughlin (editors), Managed Care and Public Health. Gaithersburg, MD: Aspen Publications, Inc.
17. Bachman, S., and B. Burwell. 1997. Medicaid Carve-Outs: Policy and Programmatic Considerations. Princeton, NJ: Center for Health Care Strategies, Inc.
18. Horvarth, J., and N. Kaye. 1996. Enrollment and Disenrollment in Medicaid Managed Care Program Management. Portland, ME: National Academy for State Health Policy.
About the Authors
Robert E. Hurley, Ph.D., is a faculty member in the Department of Health Administration at the Medical College of Virginia of Virginia Commonwealth University. He has been conducting research in managed care for more than a decade, with a particular interest in public-sector managed care. He has written a book and more than two dozen articles on various features of Medicaid managed care. His research has been sponsored by several federal agencies, state governments, and private foundations. Dr. Hurley is a member of the National Academy for State Health Policy and a fellow of the Association for Health Services Research.
Susan Wallin is a research associate in the Urban Institute's Health Policy Center. Previously she served as an analyst for the Physician Payment Review Commission. Her research has centered on access to care for low-income populations, including issues of health professional maldistribution, Medicaid managed care, and public health departments.
