Next-generation contracting: Managed Medicaid for individuals with special or supportive care needs
This new framework can help states improve their ability to design and contract for managed Medicaid programs for these individuals—and maximize the programs’ likelihood of success.
The past decade has seen considerable innovation in how specialty services are provided to individuals with special or supportive care needs—those with behavioral health (BH) conditions or intellectual or developmental disabilities (I/DD), as well as those who require long-term services and supports (LTSS) because of medical conditions or physical disabilities.1 Many state Medicaid programs, for example, are increasingly using managed care to provide these services while keeping costs under control (Exhibit 1). We expect this trend to be resilient regardless of other changes to the Medicaid program that may be considered in the coming years.
Our experience suggests that a structured approach to contracting can help states maximize the potential of a managed Medicaid program for one or more of these groups. The first step is basic: a state should determine what its objectives are and how much potential managed care has for achieving those objectives. It should then consider 15 questions related to the program’s scope, market structure, partnership approach, and terms of agreement. There is no single “right” set of answers to these questions. Each state should base its decisions on the objectives it wants to achieve.
In this paper, we describe the structured approach we recommend, highlighting the 15 key questions. We also discuss several related issues states should bear in mind as they begin to define their approach to managed care contracting.
Individuals with special or supportive care needs represent some of the most vulnerable populations in today’s healthcare system. These individuals often require a combination of medical treatment and supportive services, either in an institutional, home-based, or community-based setting, and can require prolonged assistance performing activities of daily living (e.g., bathing, cooking). As a result, they often require intensive care coordination activities in addition to a higher overall volume of services, and can be subject to detrimental gaps in care.
As McKinsey’s recent report2 makes clear, the three groups with special or supportive care needs present unique challenges. Although they constitute only 20% to 25% of the population, they account for 35% of national healthcare expenditures. Each year, the United States spends over $800 billion on care delivery to these individuals, including more than $450 billion for non-medical services. The Medicaid program bears about two-thirds of these costs.3 About 40% of Medicaid funding comes from state budgets. However, the amount spent does not always correlate well with the quality of care delivered, level of care coordination, or ease with which care can be accessed.
Several economic trends have prompted an increasing number of states to consider alternative approaches to managing their Medicaid populations as a whole—not just those with special or supportive care needs. For years, national healthcare expenditures have been increasing at a rate above GDP growth, and spending levels are projected to rise further because of the aging population, the increasing prevalence of chronic conditions, and other factors. In many states, the number of people eligible for Medicaid has risen because of the Affordable Care Act. Cost concerns have prompted states to innovate in how they deliver care to their general pool of Medicaid beneficiaries and, more recently, to individuals with special or supportive care needs.
State Medicaid programs have therefore been introducing new approaches for serving their beneficiaries, including those with special or supportive care needs. One of the approaches being used most often is managed care, in the belief that it can achieve multiple aims:
- Improve care quality, outcomes, and patient experience
- Enhance the overall performance of state health systems, especially in such areas as access to care and population health
- Slow spending growth
In addition, states may be attracted by the increased budget predictability, program flexibility, and accountability that managed care can provide.
Because managed care programs for Medicaid beneficiaries with special or supportive care needs are comparatively new, empirical evidence for their effectiveness is still limited. However, studies have shown that total healthcare costs for Medicaid beneficiaries with BH conditions can be reduced by 5% to 10% within four years through improved integration of behavioral and physical health services.4 Another study has shown that states can achieve cost savings of 10% to 15% by rebalancing their LTSS services toward home and community-based offerings.5 Evidence is also emerging that managed care programs for individuals with special or supportive needs can improve care quality, outcomes, and patient experience.
In 2005, 23 states offered managed care to one or more of the groups requiring special or supportive care through their Medicaid program. Today, 37 states do (Exhibit 2).6 BH programs are the most established; managed care services for individuals with I/DDs are still uncommon. Only seven states currently offer managed care programs to all three populations (Exhibit 3).
Many managed care organizations (MCOs) have responded to the opportunity states have created to provide programs for Medicaid beneficiaries with special or supportive care needs and have demonstrated willingness to invest in new services and new markets, sometimes even before a formal solicitation is announced. The long-term nature of these contracts is attractive to MCOs because they can provide financial stability. Furthermore, well-run managed Medicaid programs for individuals with special or supportive care needs can give MCOs exposure, affording opportunities for footprint expansion. A structured approach to contracting can increase the effectiveness of a state’s and MCO’s joint efforts, ensuring that the programs are well run and beneficiaries receive the care they deserve.
Evaluating the potential for managed care
When considering whether to transition to a managed care program for one or more of the groups with special or supportive care needs, a state should begin by identifying its objectives for these groups. It should then evaluate managed care’s ability, compared with alternatives, to meet those objectives.
A managed care program for Medicaid beneficiaries with special or supportive care needs can, potentially, achieve several goals, but managed care may not be the only available path to meeting those goals. Clarifying and prioritizing the state’s objectives through a fact-based performance diagnostic is an important first step in assessing the available options, including managed care (Exhibit 4). The diagnostic can be structured in a variety of ways, but in all cases, it should include analyses of claims-based data (to identify performance gaps and areas of high-cost growth) and the state’s performance compared with that of its peers.
The claims-based analysis should address these questions:
- What is the breakdown of services currently being provided to the individuals with special or supportive care needs?
- For each group, what is the best way to segment beneficiaries, services, and programs?
- What are the trends in core medical and pharmacy spending for each group?
- What providers currently, or could potentially, serve each group?
Exhibit 5 offers a selection of national benchmarks that can be used for state-by-state comparisons.
Evaluating managed care against alternatives
States generally have a number of options for achieving their objectives for Medicaid beneficiaries with special or supportive care needs. A fully capitated, risk-based managed care program is one. Other options include implementing new provider payment methodologies within the current fee-for-service delivery system (e.g., by using case-mix groups) or making wholesale changes to provider reimbursement rates. States can also introduce new technologies or vendors to enhance existing capabilities for managing complex beneficiaries (e.g., through independent assessments). Yet another option is changing the groups’ medical or payment policies, or the application of those policies, within the current system (e.g., by introducing new prior authorization requirements).
Nevertheless, certain elements specific to a capitated, risk-based managed care approach make it an attractive way to achieve a state’s objectives. First, a capitated managed care model can give the state greater predictability for budgeting purposes than is possible in fee-for-service models. Second, a managed care approach can bring in new capabilities, resources, and experience if the contracts are with MCOs that have experience in other states. Third, well-designed MCO contracts can increase the system’s flexibility and accountability. Finally, managed care programs that include multiple vendors can introduce competition between health plans, enhancing client choice and driving innovation.
Design and execution decisions
Should a state decide to pursue managed care for one or more groups with special or supportive care needs, it will have to consider a number of design and execution decisions in four areas: program scope, market structure, partnership approach, and terms of agreement. Although each of these decisions needs to be thought about early in the process, most decisions do not have to be made until contracts are awarded. In fact, it is likely that many of these decisions will evolve during the process.
Exhibit 6 describes all 15 decisions and outlines when they should be made. In the sections that follow, we discuss several of these decisions in detail to illustrate the specificity with which each one needs to be considered.
States first need to define the new program’s scope and determine how well it would fit with existing managed Medicaid programs. Selecting which population(s) to include and deciding how programs will be integrated are among the most important components of program scope.
Choice of population(s). The results of the diagnostic should determine which groups with special or supportive care needs should be prioritized. For example, if a state discovered that the proportion of its LTSS Medicaid beneficiaries being cared for in institutional settings is much higher than in other states, a managed care LTSS program that shifts beneficiaries to home- and community-based settings could provide a cost-reduction opportunity.
When deciding whether to pursue one, two, or all three program areas simultaneously, states should consider their capacity for managing change.
Integration across programs. States with existing managed Medicaid programs need to determine whether to integrate the new effort into an existing plan (Exhibit 7). For example, BH benefits could be “carved in” to an existing managed Medicaid program. Carve-ins can simplify vendor management by reducing the number of MCO relationships and create opportunities for improved care coordination. However, stand-alone programs enable states to select vendors with specialized expertise.
States selecting a stand-alone approach should decide whether to integrate coverage for the different groups into the same program or to administer them separately. A common approach is to integrate coverage for the LTSS and I/DD populations into a unified program for the aged and disabled, but keep the BH program separate.7 When making this decision, states should consider such factors as the overlap of populations and providers, implied contract sizes, and the ability to attract MCOs with the capabilities required to serve multiple populations.
Early on, states should develop a perspective on the market structure(s) they aim to create, because structure heavily influences the opportunity’s attractiveness to MCOs. Two important factors to consider are geographic reach and member choice.
Geographic reach. Which regions are in scope determines how states should structure their contracts geographically. Today, states are taking three approaches (Exhibit 8). Some states (e.g., Indiana and Idaho) have designed programs in which each vendor serves all regions of the state. States that have chosen this statewide approach tend to have low population density and few large metropolitan areas. Other states (e.g., New York and Pennsylvania) have taken a regionalized approach by subdividing the state into regions for contracting purposes, even if the MCO itself operates in all areas of the state. Yet other states (e.g., California and Texas) are limiting programs to specific regions or have adopted a staged rollout.
A clearly defined statewide approach helps ensure consistent messaging and commitments to stakeholders. Nevertheless, states may find it useful to subdivide geographically so they can partner with multiple regional MCOs, or to limit the areas served to a select subpopulation.
Member choice. States also need to determine the level of choice and competition they would like to instill in their managed Medicaid markets. For states that have opted to integrate one or more of the groups with special or supportive care needs into their existing managed Medicaid programs, member choice among health plans is generally required. States that have taken a stand-alone approach to managed care for these groups are more evenly split among three approaches: no member choice, full member choice, and a hybrid model in which MCOs compete for some but not all beneficiaries (Exhibit 9). Many states report they find value in contracting with multiple MCOs; this approach creates competition for beneficiaries and provides greater latitude in managing MCO performance.8 However, the value of MCO choice to the state may exceed its value to beneficiaries, because members typically view other factors—such as ability to retain their physician—as more important than the choice of an MCO.9
States also need to give early attention to the types of relationships they aspire to develop with MCOs. A key decision here is which group(s) within state government will have responsibility for overseeing the program.
Responsible party. The responsible party within state government typically sets the tone for the partnership(s) and manages vendor performance. In many states, responsibility for each of the groups with special or supportive care needs is shared between the Medicaid program and a separate division or agency (e.g., a division of developmental disabilities services). In transitioning to managed care, states have taken a variety of approaches: giving sole responsibility to the Medicaid program, sole responsibility to the relevant division, or a hybrid. When making this decision, states need to consider internal factors—such as where talent and capabilities reside—and the fit with other program design choices.
Terms of agreement
States should also consider the intended terms of agreement, starting with contract length, before they begin the contracting process.
Contract length. The length of contracts is likely to influence the level of investment MCOs will make and set the tone of the partnership(s). Today, many states opt for three- to five-year contracts, with options for extension.10 However, states are increasingly using longer contracts to form long-term partnerships, encourage innovation, and provide attractive terms to MCOs.
States also need to determine who will hold options for extension or exit. In some cases, a state may decide on a short initial contract but give itself the option to extend the contract. In other cases, the state agrees to a longer contract but builds in exit clauses that either side can exercise. It is likely that many states may eventually use both extension and exit options.
Other factors to consider
The design decisions described above give states a range of market-specific options for a managed Medicaid program. In all cases, however, states may want to take four steps before asking MCOs for proposals:
Design the program around desired partnerships. States should begin engaging MCOs early, bringing a range of potential partners to the table to generate ideas. Structuring these conversations to allow for substantive dialogue is important. Throughout the contracting process, states and MCOs should jointly define goals, such as quality improvement or desired changes to the delivery system.
Build in competition. States should communicate the planned market structure early on. MCOs typically value knowing the number of likely vendors and regions so they can develop a competitive strategy. Allowing sufficient time for new entrants to prepare a bid is important for widening the set of potential MCOs. States should develop contract terms that encourage innovation, such as member auto-enrollment based on achievement of quality and cost goals, and contract extensions based on performance.
Encourage continuous innovation. States should define the particular areas in which ongoing innovation will be needed and, early in the process, seek partners with relevant expertise in those areas. Interaction with MCOs can be tailored to encourage continuous innovation through incentive programs and shared savings.
Adhere to an ambitious yet realistic time frame. A sample of recent state procurements suggests that the process, including implementation, can take anywhere from 12 to 28 months (Exhibit 10). Exhibit 11 outlines the steps states need to take, which generally happen over a two-year period, to move from initial consideration of managed care for one or more of the groups with special or supportive care needs to program launch.
Implementation time is especially important to consider. Although a few states have been able to launch managed Medicaid programs within three or four months of the contract award, most states require more time. Among the factors that most strongly influence the implementation timeline are the state’s level of experience with managed Medicaid, the infrastructure and experience of the MCOs already present in the state, and the degree to which the Centers for Medicare and Medicaid Services and other important stakeholders have been actively engaged throughout the process. An overly ambitious timeline can be counterproductive if it impedes the transparency and engagement required for a successful launch.
* * *
Managed care programs present an opportunity for states to serve populations with complex needs in a new and effective way. MCOs, in turn, can benefit from the opportunity for strategic expansion. In designing these programs, each state should carefully consider a number of specific factors that will ensure the delivery of sustainable value for MCOs and the state, while improving quality and outcomes of care delivered for beneficiaries.
The authors would like to extend special thanks to Pavi Anand, Jason Barell, Kara Carter, Patricia Freeland, Tom Latkovic, Razili Lewis, Rocio Garcia Villaverde, and Tim Ward for their help with the preparation of this article.
- Behavioral health issues include mental health and substance abuse conditions, which can range from mild disorders to severe illnesses (e.g., schizophrenia). Individuals with intellectual or developmental disabilities (I/DDs) require help performing activities of daily living (e.g., bathing, cooking) for a prolonged period and thus need longterm services and supports (LTSS). The other individuals needing LTSS have chronic, complex medical conditions or physical disabilities, and thus require extended care in home, community, or institutional settings. Because the services needed by individuals with I/DDs are often more highly specialized than those required by other people needing LTSS, we have categorized the two groups separately in this paper.
- Carter K, Lewis R, Ward T. Improving care delivery to individuals with special or supportive care needs. McKinsey white paper. August 2016.
- Kaiser Family Foundation. Federal and state share of Medicaid spending.
- Economic Impact of Integrated Medical-Behavioral Healthcare: Implications for Psychiatry. Milliman American Psychiatric Association Report. April 2014.
- Kaye HS. Gradual rebalancing of Medicaid long-term services and supports saves money and serves more people, statistical model shows. Health Affairs. 2012;31(6):1195-1203.
- Includes Washington, DC.
- Arizona, Illinois, and Wisconsin are examples of states that have taken this approach. Arizona’s Regional Behavioral Health Authority program is stand-alone, but the Arizona Long-term Care System runs an integrated LTSS and I/DD program. Illinois’ Integrated Care Program serves the state’s aged and disabled populations, but BH coverage is provided through a separate, managed care program. Wisconsin’s BadgerCare includes BH services, but its Family Care program covers all three areas. Some states, like Wisconsin, have multiple programs for one or more of the groups because of legacy effects or differences in the subpopulations being served.
- Based on interviews with state Medicaid leaders.
- Based on interviews with state Medicaid leaders and MCO executives.
- Based on interviews with state Medicaid leaders and MCO executives.