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.
About the Series
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.
Contents
Introduction
Background
Types of Medicaid Managed Care
Medicaid Managed Care Programs for Persons with Disabilities
Enrollment Issues
Mainstream versus Specialized Programs
Behavioral Health
Rate-Setting
Quality Assurance
Conclusions
Notes
About the Economic and Social Research Institute
About the Authors
Introduction
About one of every six persons on Medicaid
1 can be classified as a "younger person with a disability"—that is, a child or an adult under age 65 who qualifies for Medicaid coverage in part because of a disability.
2 Although this population is extremely heterogeneous, these individuals generally are grouped into three broad categories: persons with physical disabilities, mental retardation/developmental disabilities, or mental illness. It is common for individuals with disabilities to fall into more than one of these classifications.
Because persons with disabilities are a costly population to serve, state Medicaid programs have begun to incorporate younger persons with disabilities into their managed care programs. In fact, some states view integrating their disabled population into managed care as a key component in their overall cost management strategy over the next several years. Attempts to rein in spending on disabled groups could yield significant savings to state Medicaid programs.
Persons with disabilities, however, are costly for a reason. On average, their needs for health services are greater than those of persons without disabilities. Although there may well be room for efficiencies in their care, they are particularly vulnerable in managed care programs that restrict access to services to control costs.3
This report describes the various types of Medicaid managed care and then examines how states have made choices on several policy dimensions, including use of capitation arrangements, voluntary versus mandatory participation, enrollment, mainstream versus specialized programs, the role of behavioral health, rate setting, and quality assurance. This analysis is a part of the Urban Institute's Assessing the New Federalism project, a multi-year undertaking designed to track and analyze the decentralization of social programs in the United States. A major component of this project is intensive case studies of policies in 13 states: Alabama, California, Colorado, Florida, Massachusetts, Michigan, Minnesota, Mississippi, New Jersey, New York, Texas, Washington, and Wisconsin. These states were selected because together they represent nearly 50 percent of the nation's population and provide a broad perspective with respect to geography, fiscal capacity, child well-being, and traditions of providing government services. The information in this report was obtained largely from in-person interviews and documents collected at site visits conducted during the second half of 1996 and the first half of 1997, with updates obtained from various tracking services. This information was supplemented with a telephone survey of state Medicaid agencies in January 1998.
Background
Nonelderly individuals qualify for Medicaid benefits based on disability status in one of two ways. They can qualify for Supplemental Security Income (SSI) if they meet federal income and asset tests and a national standard for disability. Alternatively, some disabled persons who do not meet the SSI income and asset tests qualify for Medicaid because their medical expenses cause them to "spend down" to a state's medically needy standard, which can be set up to 133 percent of the state's old Aid to Families with Dependent Children (AFDC) level. Thirty-four states have medically needy programs. Even in states without such programs, severely disabled persons in institutions qualify for Medicaid if their income is no greater than 300 percent of the SSI cash payment (or roughly 200 percent of the federal poverty level).
4
Virtually all measures of spending reveal that younger persons with disabilities are a costly group for the Medicaid program. For example, while
16.6 percent of Medicaid's 35.2 million beneficiaries are classified as blind and disabled, they account for 33.7 percent of total Medicaid expenditures. The Medicaid program spent on average $3,789 per beneficiary in 1995, but spent more than double that amount—$8,784—on health benefits for disabled individuals.5 More than half of these costs (58 percent) are spent on acute care services. Although they do not consume costly long-term care services in the same proportions as elderly Medicaid recipients, younger persons with disabilities use long-term care more frequently than nondisabled beneficiaries.
State spending on disabled individuals varies enormously, as shown in table 1. Spending per beneficiary ranges from a low of $4,423 in Mississippi to a high of $16,605 in New York. In each of the 13 states under review, spending per disabled individual is at least 160 percent of average per beneficiary spending. Nine of the states spend more than twice as much on the disabled beneficiary as on the average beneficiary.
Within the disabled population, there is considerable variation by eligibility category. Medically needy individuals tend to be much more costly than other disabled Medicaid beneficiaries. For example, in 1994, payments for persons on SSI were nearly $7,000 per person, but were over $12,600 for medically needy and other non-SSI disabled beneficiaries.6
Types of Medicaid Managed Care
Forty-eight states and the District of Columbia now either require or allow some or all of their Medicaid population to enroll in some form of managed care program;
7 between 1991 and 1996, the percentage of Medicaid beneficiaries enrolled in managed care plans grew from 9.5 percent to just over 40 percent,
8 with most of this growth occurring among the nondisabled AFDC population. Despite this rapid increase, the variation of Medicaid managed care penetration among those 48 states and the District is extremely wide. Some states claim that more than three-quarters of their Medicaid population is enrolled in some form of managed care, while others have fewer than 5 percent enrolled.
There is also a great deal of variation among state managed care programs, although they generally fall into one of two classifications: primary care case management (PCCM) programs or capitation arrangements. PCCM programs match beneficiaries with primary care physicians who coordinate care on behalf of the enrollee and who serve as gatekeepers to specialty, inpatient, and other services. Primary care providers who participate in PCCM programs generally receive a monthly management fee, with services paid on a fee-for-service basis.9 In this case, states contract with PCCM providers to provide a "medical home"—a stable and regular source of care—to enrollees and to perform their case management/gatekeeping functions.
States also contract with managed care organizations (MCOs) on a capitated basis for comprehensive service delivery and place these MCOs at full risk for most or all services provided to Medicaid enrollees. Some states have developed hybrid versions of these two models, placing individual providers or group practices at risk for only portions of Medicaid services such as primary care or ambulatory care services.
Potential Strengths
Like others on Medicaid, disabled beneficiaries can benefit greatly from a medical home and careful management by a primary care provider who is compensated for this management function. As stated earlier, these beneficiaries consume a sizable share of Medicaid dollars and are frequent users of health services. Any mechanisms that can coordinate care and more appropriately target services to need could improve care delivery for this population.
Comprehensive, capitated health care arrangements also hold promise for persons with disabilities. Providing health plans with a lump sum payment creates incentives to provide care efficiently and to invest in resources that could prevent costly hospitalizations and emergency room use. This arrangement encourages better disease and disability management strategies to maintain the health and functional status of enrollees. It also allows plans to use the funds more creatively and flexibly so that the monies spent on individuals with disabilities are more effectively directed to their individual needs and preferences.
Potential Weaknesses
Managed care, however, closes the door on certain opportunities inherent in the current fee-for-service system and contains incentives to undertreat or inappropriately treat persons with disabilities. All managed care plans limit choice of provider to some degree, although the extent of limitation varies greatly among plans and across states. Since many persons with disabilities receive their "routine" care from medical specialists and various therapists, even PCCM programs restrict choice by placing access to specialists within the purview of a primary care provider/gatekeeper. Some persons with disabilities who are on Medicaid have long-standing relationships with primary care physicians (PCPs) who are not part of the PCCM program. Such individuals may be required to see a PCP who is unfamiliar with their histories. At least in theory, however, PCCM programs refer to any specialists willing to accept Medicaid fees, so choice of specialists is not necessarily restricted by the PCCM model.
10
Unlike PCCM programs, capitated plans define a network of specialty medical and ancillary therapy providers, making it likely that at least some of an individual's providers will not be included in the plan. Persons with disabilities, who may have multiple acute and chronic conditions, may actually have to sever relationships with many providers, creating considerable discontinuity in their treatment.
Health maintenance organizations (HMOs) and other prepaid health plans may provide care based on a medical model to a population of individuals who need habilitative and rehabilitative, supportive, therapeutic, and assistive services, as well as the full range of acute and preventive care that is commonly provided to plan enrollees. HMOs traditionally have placed very strict limits on therapeutic, supportive, and home care services. These limits can shift the balance between medical and nonmedical care even more toward the medical model. Critics of managed care are skeptical about the emphasis on medical management and fear that supportive services that are nonmedical in nature will be undervalued and underreferred. In addition, persons with disabilities may at times need highly specialized services—not merely access to specialists, but access to pediatric and adult specialists who are experienced in treating rare and complex conditions. Some critics doubt that health plan networks can offer the depth and breadth of services that many persons with disabilities require for their health and functional status.
Arguably, the most serious concern about moving persons with disabilities into managed care is the financing structure that pays health plans to care for individuals, regardless of the specific services provided. The potential of managed care to slow or even decrease spending creates the most obvious concern for persons with disabilities—how can we be sure that they will get the services they need within a system that does not compensate health plans for providing more care to certain persons?
Because persons with disabilities are a costly group for the Medicaid program, states may be tempted to try to squeeze significant cost savings from this group without first understanding how to manage their care in a capitated environment. While states currently appear to be moving slowly in this regard, there may be more interest in cutting rates for disabled enrollees as savings from nondisabled groups taper off over the next several years.
Medicaid Managed Care Programs for Persons with Disabilities
Eleven of the thirteen states described in this report include some younger persons with disabilities in at least one managed care program, either statewide or in one or more counties.
11 Table 2 lists state Medicaid managed care programs that require (M = mandatory) or allow (V = voluntary) enrollment for younger persons with disabilities. The enrollment figures for PCCM and capitated plans are from various state-generated enrollment reports or from state Medicaid officials' estimates of the percentage of disabled enrollees in their total managed care population, and are current as of October or November 1997 with the following exceptions: Alabama's enrollment figure for the BAY health plan is from April 1998, California enrollments are current as of June 1997, and Washington's enrollments are current as of January 1998.
Approximately 630,100 younger persons with disabilities are enrolled in some form of managed care in the 13 states under review. While it is difficult to estimate precisely the proportion of Medicaid disabled beneficiaries that this number represents, it appears that roughly 22 percent of nonelderly blind and disabled beneficiaries are currently enrolled in some form of managed care.12
As can be seen from table 2, managed care enrollment is divided evenly between PCCM and capitated programs. Some of the PCCM programs, however, have capitated components, and in some states (Michigan, Texas, and Wisconsin) so-called capitated programs actually are paying plans 100 percent of fee-for-service for some or all service categories, as a precursor to lower capitation rates.
Three states (Colorado, Florida, and Massachusetts) have mandatory programs that require persons with disabilities to enroll in some form of managed care. Individuals must enroll either in the PCCM program or in an HMO. Five states (Mississippi, New Jersey, New York, Texas, and Wisconsin) have completely voluntary programs for persons with disabilities. Alabama has two relatively small managed care programs (one of which has only recently begun enrolling persons with disabilities) that are mandatory for eligible persons with disabilities. California and Michigan require some but not all of their disabled beneficiaries to enroll in managed care arrangements.
As of January 1998, Minnesota and Washington were not enrolling disabled beneficiaries in any form of managed care. Minnesota initially included persons with physical disabilities or mental illness in its Medicaid managed care initiative, the Pre-Paid Medical Assistance Program, when it was implemented in 1985. Individuals in three counties were enrolled on a mandatory basis over the course of a year, but were returned to the fee-for-service system when one of the health plans dropped out of the program. According to state Medicaid officials, this particular plan (a Blue Cross/Blue Shield product) attracted the majority of disabled enrollees because of its very broad provider network. When the plan left the Medicaid managed care market, Minnesota officials decided to exclude individuals with disabilities from managed care arrangements.
Minnesota is now in the process of implementing mandatory managed care for individuals with disabilities in five state regions. Enrollment in two of the regions is scheduled to begin in late 1998 and early 1999. All five sites will attempt to integrate acute and specific long-term care services, as well as comprehensive mental health services, on a fully capitated basis. The precise mix of services has yet to be determined.
Until the end of 1997, Washington had enrolled 8,300 individuals with disabilities in the eastern part of the state in capitated managed care plans. The state also operated a PCCM pilot program in Clark County, with roughly 1,500 disabled enrollees. In January 1998, all of these individuals were disenrolled from managed care and returned to the fee-for-service system. The state also canceled plans for mandatory managed care for its SSI population in the western portion of the state, which had been scheduled to begin in 1998.
Washington state officials believe that access to services for individuals with disabilities was greatly improved by enrolling them in managed care programs. As access increased, however, so did utilization of services and costs to health plans. Some health plans, experiencing utilization and costs as high as 35 percent above the reimbursement rate, became reluctant to participate in the program. In addition, the Medicaid program, restrained by a limited budget and a legislative mandate to operate the capitated managed care program for SSI enrollees at a discount (99 percent of fee-for-service), refused to increase reimbursement rates for disabled enrollees.
Table 3 shows the percentage of younger persons with disabilities enrolled in primary care case management and capitated programs across the 13 states. With the exception of Florida and Massachusetts (which enroll individuals with disabilities primarily in PCCM programs), only a small minority of younger people with disabilities are in any form of managed care. Most of these state programs exclude institutionalized persons and certain other groups from managed care arrangements. For example, persons in nursing homes or intermediate care facilities for the mentally retarded, and certain others such as the medically needy, may be ineligible for managed care. Likewise, several states exclude persons in home and community-based waiver programs from managed care plans, although there are exceptions. Alabama and Colorado include such persons in their managed care plans for acute care services only.
Because of these exclusions, even in states with mandatory statewide programs, enrollment of Medicaid disabled beneficiaries in managed care arrangements is significantly lower than 100 percent. In Massachusetts, for example, all persons eligible for both Medicare and Medicaid are ineligible for managed care.13 Massachusetts representatives have indicated that they may lift this ban in the coming years, which could bring the percentage of younger persons with disabilities in managed care plans to over 90 percent. However, the federal Balanced Budget Act of 1997 prohibits mandatory enrollment of dual eligibles. In Florida, Medicaid representatives indicate that they have enrolled virtually all of their disabled "eligibles" into managed care, with approximately three-quarters of the Medicaid population enrolled in such arrangements.
While the proportions of persons in PCCM and capitation programs are distributed evenly with respect to total national enrollments, this is not the case within states. Florida and Massachusetts have a far greater percentage of their beneficiaries with disabilities enrolled in PCCM programs; in contrast, the majority of disabled managed care enrollees in California, New Jersey, and New York are in capitated arrangements.
Nine of the thirteen states currently operate PCCM programs, although some programs are being phased out. Several state officials reported that they prefer capitated programs for their Medicaid recipients, in large part because of the programs' potential for greater cost savings.14 Michigan's Physician Sponsor Plan was discontinued in summer 1997, and its 50,200 disabled enrollees are now being signed on with the state's HMO program. California's PCCM option is also being phased out; only 1,000 disabled beneficiaries are in that program. Other states, such as Colorado and New York, are working to shift enrollment from PCCM to capitated programs but have indicated a willingness to maintain the PCCM option at least over the next few years.
Enrollment Issues
Many of the states staggered enrollment of categories of persons into managed care, beginning with the AFDC and related populations. According to several Medicaid officials and others interviewed for this report, experiences with enrollment of the AFDC population into managed care arrangements provided valuable lessons for states as they moved to enroll disabled beneficiaries into managed care.
In several cases the enrollment process was extremely problematic. For example, several states began with, but discontinued, direct marketing by managed care organizations. Most of the states now rely on enrollment brokers to handle the process of informing, educating, and signing up Medicaid beneficiaries into managed care plans. Despite the problem of abuses in direct client marketing in many states, Wisconsin's I-Care program, which is a voluntary managed care option, follows up its mailings with telephone solicitations to about 10 percent of its eligible, nonenrolled population each month. This strategy has reportedly paid off, with enrollment increasing by about 20 percent during 1997.
Whether the program is voluntary or mandatory, Medicaid beneficiaries with disabilities generally receive information packets that explain the program and any plan choices that are available. In some states, they may also receive information about a plan's panel of providers. (In some states, provider information is sent after the enrollee has signed on with a particular plan.) States generally also make information about managed care plans available through physicians' offices and social service providers.
In each of the programs under review, disabled beneficiaries are given an opportunity to choose a primary care provider or managed care plan, regardless of whether the program is mandatory or voluntary. In some cases, such as the BAY program in Alabama, there is only one managed care organization, so enrollees choose only their primary care physician. In the case of a PCCM program, enrollees are encouraged to choose a primary care provider upon enrollment. Some states have several managed care organizations providing services to disabled individuals, so enrollees must first choose the plan, and then a primary care provider. In these cases, choice of plan is often driven by the desire to remain with a primary care or specialty provider.
Where enrollment is mandatory, persons with disabilities are generally given between two and six months before they are automatically assigned ("autoassigned") to a plan or provider. Ironically, the state that has set one of the longest enrollment periods has one of the highest autoassignment rates among disabled beneficiaries. To encourage choice, Massachusetts allows at least 12 months before autoassignment and sends reminders periodically to encourage individuals to choose a primary care physician. Still, the majority of disabled beneficiaries are autoassigned to providers, in part because new enrollees are likely to be assigned to a current provider, leaving them with less incentive to preselect a provider.
Special provisions are made for autoassigning disabled beneficiaries in Massachusetts. Unlike other Medicaid beneficiaries, who are assigned to the primary care clinician (PCC) program and capitated plans in proportions that mirror those choosing a plan, all disabled beneficiaries eligible for managed care are autoassigned to the PCC program. There are no lock-in periods for persons with disabilities in either managed care arrangement. Therefore, beneficiaries can disenroll at any time if they are dissatisfied.
Most states try to match persons with disabilities with previous providers when assigning them to managed care plans. If previous providers are not part of the managed care network, Medicaid programs assign disabled beneficiaries on the basis of geography and other factors. For example, Massachusetts surveyed its primary care provider population and developed a database that characterizes provider expertise in 92 conditions. Autoassignments not made to previous providers take into account whether a provider has the expertise to care for a particular disability or condition.
Beginning December 1, 1997, Colorado physicians were given the opportunity to decide which of their patients to enroll in the capitated plan and which to retain in the PCCM program.15 This feature, which on its surface appears to have the potential for substantial risk selection, will be watched closely by Colorado Medicaid officials and may be amended if physicians steer less costly patients to HMOs and leave more costly enrollees in fee-for-service care. This arrangement presumes that these physicians have nonexclusive contracts with certain HMOs and are free to determine which patients to retain in the PCCM program.
Mainstream versus Specialized Programs
Another important issue for states to consider when planning to move persons with disabilities into managed care programs is whether a "mainstream" or specialty program will best serve the special health needs of this population. Mainstream programs integrate the disabled into managed care arrangements that serve the general Medicaid population. Because they combine all recipients into one programmatic structure, they may be administratively less complex while also providing certain economies of scale in the provision of care. Some advocates believe that mainstream programs can also reduce discrimination against persons with disabilities.
16
Critics of this approach, however, fear that mainstream programs cannot accommodate the heterogeneity of conditions among the disabled because they lack experience in individual care planning and service provision, are not equipped to address the full spectrum of health-related and social service needs of the disabled, and may find it difficult to determine capitation rates for this population. Without reliable risk adjusters, mainstreaming the disabled could put patients and providers at risk.
An alternate approach is to allow disabled beneficiaries to enroll in specialized programs that cater to their specific health and provider needs. Programs that accommodate the unique needs of this population can be designed to address the distinctions among different groups of people with disabilities and allow specialty providers to develop expertise in caring for the disabled, especially with respect to rare or complex health conditions. Even proponents of specialized programs, however, question their financial viability. The answer to this question may turn on developing adequate risk-adjustment mechanisms for determining payment rates.
Most states seem to favor mainstreaming disabled beneficiaries into managed care programs that care for their nondisabled Medicaid enrollees rather than developing or maintaining specialized programs. Alabama, Massachusetts, and Texas have mainstreamed portions of their disabled populations into managed care programs that were already enrolling nondisabled Medicaid beneficiaries.
In addition to their mainstream programs for persons with disabilities, Michigan and Wisconsin offer managed care services through specialized programs for disabled adults or children. Children in Michigan who are dually eligible for Maternal and Child Health Block Grant (Title V) services and Medicaid can enroll in a comprehensive plan for all their health needs. Approximately 24,000 children are enrolled in the program, half of whom are dually eligible (and half of whom are eligible for Title V only and pay sliding scale copayments for the services). Wisconsin's I-Care program is a joint venture between Humana and the Milwaukee Center for Independence that exclusively serves a portion of the eligible SSI population in the state. (The program is voluntary, and about 10 percent of the eligible population is enrolled.)
New York is planning an ambitious program that would offer comprehensive services through capitated special needs plans (SNPs) for HIV/AIDS to individuals and their families. These SNPs would provide acute and long-term care services and would include providers with expertise in HIV and AIDS. New York plans to require individuals in a given geographic area to join these plans if and when they become available. State Medicaid officials estimate that up to 100,000 people could eventually enroll in the HIV/AIDS SNPs.
Behavioral Health
States are using a mix of approaches to provide behavioral health services to their disabled Medicaid populations, as is illustrated in
table 4. Many states continue to rely on the fee-for-service system for mental health and substance abuse services. In Alabama, California, Florida, Michigan, Minnesota, Mississippi, New Jersey, New York, and Texas, most mental health services for persons with disabilities already enrolled in some form of managed care are provided on a fee-for-service basis. This is true for both PCCM and capitated program enrollees.
Massachusetts provides capitated behavioral health services to all of its enrollees in managed care plans. The state requires all enrollees, including those with disabilities, to join either the PCC program or the state HMO program. In both cases, services are fully capitated—in the case of the PCC program, they are carved out and provided by one behavioral health firm that serves the entire state. As a consequence, this firm provides services for most of the Medicaid disabled beneficiaries in the state, who chose the PCC program at a rate of 10 to 1 over the HMO option.
Several states are dividing responsibility for mental health services among HMOs, traditional fee-for-service providers, and in some cases, county-based mental health authorities. California, Florida, and Michigan provide some mental health services on a capitated basis and either exclude all community-based care from capitated programs (Florida) or set explicit utilization limits on inpatient and outpatient services. This situation can create concerns about care coordination, especially if different sets of providers are unclear about who carries the ultimate responsibility for mental health services.
Despite attempts by several states to carve behavioral health partially or fully out of their "somatic" managed care programs,17 HMOs and other managed care organizations still often provide some basic behavioral health services, causing discontinuities and inefficiencies in service delivery. Physicians in HMOs may retain medication management responsibilities, for example, while referring patients outside the HMO for therapy and other services. In other cases, patients may begin therapy with the HMO and find themselves referred out to other care when their needs exhaust the HMO's service levels.
New York's plan to establish SNPs for mental health services is an example of a multi-tiered approach to mental health services. New York intends to create separately identifiable mental health SNPs, similar to its plans for HIV/AIDS care, that would offer an extremely comprehensive set of services to adults with serious and persistent mental illness and children with serious emotional disorders.18 Unlike the HIV/AIDS SNPs, mental health SNPs would provide mental health services only, with individuals referred back to HMOs or alternative care arrangements for their other needs.
Individuals with mental illness would come to the SNP in one of two ways: They would either automatically qualify for SNP services by virtue of a history of serious mental illness, or they would exhaust the mental health benefits available in the HMO program and be referred to the SNP. At this point it is unclear how persons needing mental health services would move between the basic mental health benefits provided by the managed care plan and the enhanced benefit package available through the SNP. For instance, if SNP patients showed improvement and required less intensive services, it is uncertain whether they could retain certain benefits available from the SNP or be referred back to the HMO for a less comprehensive set of services. Advocates for persons with mental illness are concerned that the mental health SNP could create a two-tiered system in which patients bounce back and forth between levels of care.
Some states include substance abuse services in their managed behavioral health programs. Florida, Massachusetts, and Michigan, for example, offer substance abuse services as part of their behavioral health programs, which are managed along with mental health services. New York includes substance abuse in its managed care programs in some cases and not in others; in the planned mental health SNPs, health plans are required to develop care plans for persons with substance abuse problems but not to provide or arrange for these services.
Rate-Setting
Most Medicaid managed care payment systems rely on very crude measures of risk, thereby creating financial incentives for plans to avoid enrolling persons with disabilities altogether or to enroll only persons with disabilities that are likely to require less costly care. Managed care organizations do not advertise that they have developed innovative systems of care for persons with diabetes, quadriplegia, AIDS, and other chronic illness, in part because they fear that the costs of treating these individuals will exceed the reimbursement they receive.
19 Most states that include persons with disabilities in their Medicaid managed care programs do not have sophisticated risk adjusters in place, but merely differentiate payments according to a person's categorical status (e.g., SSI, medically needy), age, and perhaps gender. With a few states beginning to experiment with diagnosis-based risk adjusters that should more closely match an individual's likely needs, it may be possible to evaluate the clinical and administrative value of such adjusters within the next few years. But until that time, there may not be a very close match between the cost of serving people with disabilities and the Medicaid payment.
States rely heavily on historical fee-for-service costs to set capitation rates for persons with disabilities. All of the states with capitated plans have separate rates for persons with disabilities, generally broken into age groups. For example, Mississippi, with less than 1 percent of its disabled population in capitated care, has developed seven age-based rates for disabled individuals enrolled in its capitated pilot program: ages 1-2 months, 3-12 months, 1-5 years, 6-14 years, 15-20 years, 21-44 years, and 45-64 years. Florida and Michigan (and Washington, in its former capitated program) use similar categories to reflect age, gender, and disability status. None of these states, however, has a risk adjustment for type of disability or associated functional limitation(s).
Alabama's capitated rates for enrollees in the BAY program will be, on average, about 97 percent of what the fee-for-service system pays for the same population who are not enrolled in Medicaid managed care. Texas pays managed care plans 100 percent of fee-for-service costs as a capitated rate for persons with disabilities enrolled in the plans (capitated rates for nondisabled persons are discounted on average about 10 percent). California's Medicaid program relies on historic fee-for-service costs but develops assumptions about trends in future spending to create "shadow" fee-for-service rates that become the basis for the capitation rates.
Michigan's program for Title V/Medicaid children also sets capitation rates at 100 percent of anticipated fee-for-service costs. The state originally intended to provide a fully capitated rate at less than full cost; however, advocates for the disability community persuaded the state not to place providers at risk, at least for the first few years.
New York is attempting to adjust payment rates to reflect higher-than-average enrollments of certain high-cost individuals. New York adjusts payments to health plans that enroll disproportionately high numbers of persons with HIV/AIDS. If a plan's enrollment of persons with HIV/AIDS reaches 20 percent, its capitation rate for all enrollees increases by 1 percent.
A few states are beginning to experiment with risk-adjustment mechanisms to reflect the greater utilization and costs of persons with disabilities who enroll in managed care plans. Colorado's capitated program relied on discounted (about 5 percent) fee-for-service rates for several years, but has since changed to a diagnosis-based system of risk-adjustment for its AFDC and disabled populations. The system, implemented in October 1997, uses 2,400 International Classification of Diseases-based diagnoses, which are grouped into 40 diagnostic categories, to set capitation rates. Wisconsin is planning to move away from total reliance on historical fee-for-service costs and to incorporate diagnostic categories, such as Ambulatory Care Groups, as part of its rate-setting process. It is interesting to note that Wisconsin recently adjusted its I-Care rates in response to criticisms that the capitation rate did not accurately reflect the high costs of caring for disabled enrollees. Now, instead of basing capitation rates on fee-for-service costs for disabled persons statewide, it will tie rates to its enrollees' actual prior use of services. This change has already resulted in a small increase in I-Care's capitation rates.
Quality Assurance
The states under review have not yet developed meaningful quality assurance programs for persons with disabilities in managed care. While most states indicate that quality assurance mechanisms are built into their managed care contracts, they are severely constrained in their ability to evaluate quality because plans are just beginning to report information about delivery of services to Medicaid enrollees. Without encounter data, it will be very difficult to identify and track the types of services provided to persons with disabilities and to determine how well Medicaid enrollees are cared for under managed care arrangements.
Some states indicate that two factors act as proxies for quality control. Since many programs in the states under review are voluntary for disabled recipients, and those that are mandatory generally have liberal disenrollment policies, these exit points can serve as indicators of quality. As long as disenrollments remain low, states assume that plans are providing high-quality care. Few states, however, do more than count monthly disenrollments to see whether there are patterns in the health status of persons or episodes of care that result in a disenrollment.
Some states also collaborate with various advocacy groups as part of their quality assurance program. Many of the states under review worked very closely with groups that represent the interests of persons with disabilities during their early plans for Medicaid managed care. In quite a few cases, advocacy groups were instrumental in shaping the managed care program and how it has been implemented. Advocates can also bring to the public's attention anecdotal accounts of problems with service delivery. These accounts, however, are not a substitute for a meaningful program designed to ensure that the quality of service delivery is within the standards determined through the Medicaid contracting process.
Conclusions
The following conclusions can be drawn from the descriptions of managed care programs for persons with disabilities in the 13 states under review:
- Persons with disabilities are a diverse group of individuals who, collectively and individually, present challenges to Medicaid managed care programs. States have responded in various ways, from keeping all of their disabled beneficiaries in fee-for-service arrangements to moving the majority into managed care plans. States usually stagger enrollment of Medicaid beneficiaries into managed care, with persons with disabilities among the last to join the programs.
- When given a choice, persons with disabilities often (although not always) prefer the PCCM option over capitated care.
- States often modify their enrollment procedures to accommodate the needs of persons with disabilities.
- A small but growing group of persons with disabilities is voluntarily enrolling in Medicaid managed care arrangements. Many others who are in states with mandatory enrollment perhaps would have chosen managed care even without the mandate.
- States have taken very different approaches to the delivery of mental health services. Some states run mental health services through their HMO programs, some carve them out to either fee-for-service or capitated arrangements, and others divide responsibility for care among fee-for-service, capitated plans, and local or regional mental health authorities.
- Although progress is slow, states are trying to develop risk-adjustment mechanisms to appropriately compensate providers who are at risk for some or all Medicaid services for persons with disabilities.
- Little is known about how persons with disabilities actually fare in managed care. This is especially true in capitated arrangements, where encounter data are extremely limited. States frequently measure quality by the number of disenrollments, although the disenrollments are not always collected separately by disability status. Disenrollments, however, are only one indicator of quality, and they can be heavily influenced by the availability or movement of primary care physicians.
Tables
Notes
1. Unless indicated otherwise, Medicaid statistics are for 1995 and come from D. Liska,
B. Bruen, A. Salganicoff, P. Long, and B. Kessler. Medicaid Expenditures and Beneficiaries: National and State Profiles and Trends, 1990-1995. 3rd ed. Kaiser Commission on the Future of Medicaid, 1997.
2. This classification understates the true number of Medicaid beneficiaries with disabilities, because there are children and elderly beneficiaries who are also disabled but who are categorically counted as "child" or "aged" beneficiaries.
3. It is important to note that although state Medicaid programs play an important role in service provision for disabled populations, responsibility for care for persons with disabilities historically has resided with many different state-run or federally funded programs, including mentally retarded/developmentally disabled (MR/DD), special education, state mental health, vocational rehabilitation, and Maternal and Child Health Block Grants (Title V) programs. Funding for these programs comes from state general revenues, county contributions, state block grants, and other federal funds, in addition to Medicaid. This discussion is not meant to imply that Medicaid managed care will completely limit access to health-related services. While many of these programs are heavily supplemented by Medicaid funds, efforts to include the disabled in Medicaid managed care models most likely will not affect the funding streams for these other programs.
4. Severely disabled persons who do not otherwise qualify for Medicaid may qualify for home and community-based waiver programs under the same income eligibility standards.
5. This amount does not include disproportionate share hospital payments, administrative costs, and accounting adjustments.
6. C.E. Bishop and K.C. Skwara. Medicaid Spending for Persons with Disabilities: Dimensions and Growth. Waltham, MA: Center for Vulnerable Populations, 1997.
7. Much of this activity requires the states to submit requests to the Health Care Financing Administration for approval to waive various restrictions within the Medicaid law. The waivers that apply to the 13 states under review in this report are discussed in J. Holahan, S. Zuckerman, A. Evans, and S. Rangarajan. "Medicaid Managed Care: Variation in State Approaches." Health Affairs, 17(3), May/June 1998.
8. Holahan et al., 1998.
9. Primary care clinicians in the Massachusetts PCCM program receive enhanced (by $10) fees for primary care visits instead of monthly management fees.
10. Some disability advocates argue that the model itself may not limit choice, but in practice, choice of specialists may be limited by the professional judgment of the PCP. A PCP who is not experienced in caring for a particular type of disability or condition may be less inclined to refer to specialists or other important providers such as physical therapists. This argument, however, is not supported by at least one study of physician referrals, which found that rates of referrals for children with cerebral palsy to physical therapists by general or developmental pediatricians and other physicians did not vary by training or experience of the physician, but were significantly related to the perceived severity of the condition and the belief in the efficacy of physical therapy. See S.K. Campbell, H.G. Gardner, and V. Ramakrishnan. "Correlates of Physicians' Decisions to Refer Children with CP for Physical Therapy." Developmental Medicine and Child Neurology, 1995, 37(12): 1062-74.
11. Unless otherwise noted, program descriptions and enrollment information apply to the nonelderly disabled population and exclude dually eligible, Medicaid/Medicare participants.
12. In 1995, there were approximately 2,912,000 nonelderly blind and disabled Medicaid beneficiaries.
13. According to state Medicaid officials, Massachusetts is likely to begin including dual eligibles in its managed care programs over the next few years. Some states indicated a reluctance to move dual eligibles into managed care because of the administrative and programmatic complications associated with providing "joint" Medicaid and Medicare managed care.
14. Despite this contention, PCCM programs have reduced costs for certain states. Massachusetts commissioned an analysis of the impact of the state's managed care programs on costs and utilization. (See the Lewin Group. Report on the Impact of Massachusetts' Medicaid Managed Care Initiatives. Fairfax, VA: The Lewin Group, 1997.) The report documented a dramatic decrease in inpatient, emergency room, and mental health service utilization among its SSI population that is estimated to have saved between $41 million and $81 million in 1995 alone. Most of these savings came from the 80,000 SSI enrollees in the state's primary care clinician program.
15. Disabled beneficiaries can choose to remain with the PCCM program or, if it is unavailable in their area, to opt for fee-for-service care.
16. R.E. Hurley and D.A. Draper. "Special Plans for Special Persons: The Elusive Pursuit of Customized Managed Care." In S. Somers and S. Davidson, eds. Remaking Medicaid: Managed Care for the Public Good. San Francisco, CA: Jossey-Bass, forthcoming.
17. Somatic managed care programs differentiate care for the "body" from care for the "mind." Some states refer to these programs as their "medical/surgical" component of managed care, in contrast to their mental health component.
18. New York Medicaid officials have not yet determined how substance abuse services will be provided and whether they will be part of the SNPs' responsibilities.
19. Richard Kronick, Zhiyuan Zhou, and Tony Dreyfus. "Making Risk Adjustment Work for Everyone." Inquiry, 32 (Spring 1995): 41-55.
About the Economic and Social Research Institute
The Economic and Social Research Institute (ESRI) is a nonprofit, non-partisan organization that conducts research and policy analysis in health care and social services. ESRI specializes in studies aimed at enhancing the effectiveness of social programs, improving the way health care services are organized and delivered, and making quality health care accessible and affordable.
About the Authors
Marsha Regenstein, vice president of the Economic and Social Research Institute, specializes in analyzing how changes to the health care system affect vulnerable populations, especially children with disabilities and the frail elderly. She has written on Medicaid managed care, long-term care, welfare reform, and early childhood education.
Stephanie E. Anthony is a senior research associate with the Economic and Social Research Institute. Ms. Anthony has conducted policy research and analysis on federal and state programs that improve the health of women, children, and families. Her writing has focused on health issues concerning uninsured children and on youth violence.
Latest Research
Browse by Topic:
PDF
Printer-friendly summary
