Number A-9 in Series, Assessing New Federalism: Issues and Options for States
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.
Services to younger persons with disabilities (children and adults under 65 years of age) form a substantial share of overall state health spending and are an important part of the Medicaid program. This population accounted for 16 percent of all Medicaid beneficiaries in 1995 (about 5.7 million persons), but 32 percent of Medicaid expenditures (almost $50 billion). Younger persons with disabilities also accounted for nearly one-third of the extraordinary growth in Medicaid spending over the 1988–1995 period. 1 Thus, this population is critical to state efforts to control Medicaid expenditures and to reform health care generally.
This brief discusses a variety of delivery and financing issues that states are facing as they rethink Medicaid and other health programs. A brief overview of the population and the costs associated with health services provided to this group helps set the stage for the discussion.
The Population and Its Service Costs
Younger people with disabilities are a very heterogeneous group, consisting of individuals with physical disabilities, mental illnesses, and mental retardation/developmental disabilities. The number of younger people with substantial disabilities living in the community ranged from 1.0 to 14.1 million people in 1995, depending on the definition used.
2 In addition, about 500,000 younger people with disabilities were in institutions: 135,000 in intermediate care facilities for the mentally retarded (ICF/MRs), 220,000 inpatients in psychiatric institutions, and 150,000 in nursing homes.
3 There is an ongoing trend away from institutional use.
As noted above, the Medicaid program served about 5.7 million younger persons with disabilities in 1995. Spending on blind and disabled Medicaid beneficiaries averaged $8,685 per person in 1995, compared to $1,728 for younger beneficiaries who are not disabled. 4 For the very severely disabled, the cost can be much higher. For example, the average cost of a year’s institutional care in an ICF/MR was almost $71,000 in 1995. 5 These institutional costs are particularly high because of extensive quality standards, use of (relatively expensive) unionized state employees in public facilities, and low occupancy rates resulting from the deinstitutionalization movement. The average Medicaid cost of serving younger people with mental retardation/developmental disabilities in home and community-based settings under Medicaid waivers was just over $24,000 in 1996.
Medicaid expenditures for younger persons with disabilities historically have been about evenly split between acute and long-term care services, but in recent years the balance has shifted toward acute care expenditures. In 1994, 57 percent of Medicaid expenditures for blind and disabled beneficiaries were for acute care, while 43 percent were for long-term care (figure 1).
Service Delivery
States play a much larger role in the direct provision of services for younger people with disabilities than they do for the elderly. They are especially important providers of institutional services for younger persons with mental retardation/developmental disabilities and with chronic mental illnesses. Two major issues concern states as they review their delivery of services to younger people with disabilities: the balance between institutional and home and community-based care and the integration of acute and long-term care services within a managed care system. Both are important features of ongoing debates about cost and appropriateness of care with respect to this population.
Balance Between Institutional and Noninstitutional Services
There is an extremely widespread, although not unchallenged, policy consensus among state policymakers and disability advocates that institutions should play a far smaller role in providing services to younger persons with disabilities than has traditionally been the case. Many advocates go further than policymakers and are unwilling to grant even a residual role for institutional care (which sharply distinguishes the disability movement from advocates for the elderly). There have been numerous lawsuits forcing deinstitutionalization at many state mental hospitals and ICF/MRs. As a result of transferring less disabled individuals to other settings, the remaining residents of state institutions are very severely disabled; moving them to less structured environments may be more difficult and expensive than past shifts. Although the institutionalized population has declined, funds and services have not always followed individuals from the institution to the community.
Home and community-based services are expanding rapidly and embrace an increasingly wide range of services, including home health, personal care, homemaker services, assisted living, adult foster care, day habilitation, prevocational services, supported employment, supported living, chore service, homemaker services, meals-on-wheels, respite care, family training, modifications to the home, and personal emergency response systems. Many of these services are beyond the traditional definition of "medical care," but are important supports for people with disabilities. As a sign of the shifting balance of care, in 1994 for the first time the number of Medicaid beneficiaries with mental retardation or developmental disabilities receiving home and community-based services exceeded the number of persons receiving care in ICF/MRs. 6
For home and community-based services, two key issues are the flexibility and scope of services and the use of "personal assistance services." Advocates 
argue that because each person is different, people with disabilities should be able to tailor services to their own needs by choosing from a very broad array of services. In the most far-reaching formulation of this argument, some advocates have proposed "cashing out" services and giving individual consumers total freedom to spend the money as they see fit. Although supporting increased flexibility, federal and state officials worry that, with an ever-widening array of services on the list of offerings, more and more persons with disabilities will come forward to claim services. In an open-ended program like Medicaid, according to this view, people cannot be entitled to "everything" without expenditures increasing greatly.
Advocates reject this argument, contending that flexibility would lead clients to choose lower-cost, less intensive services than are now forced upon them by the narrower set of services traditionally available.
Closely related to this issue of service flexibility is how much control the client should have over the service provider. This choice is crystallized in the debate over personal assistance services and agency-directed services. Under the traditional agency-directed service approach, an organization hires and directs the personnel who deliver the services. In contrast, in the personal assistance service model, the individual client hires, fires, and directs the service provider. Some states, such as California, have been drawn to personal assistance services because of the potential to lower costs while at the same time giving persons with disabilities greater control over their lives. Other states, however, have been wary of this approach because of potential problems of quality assurance and the administrative complexities. Can persons with severe disabilities enforce acceptable service standards? And can they be counted on to handle income tax, unemployment, Social Security taxes, and workers’ compensation contributions for their employees?
Integration of Acute and Long-Term Care Services through Managed Care
Managed care is increasingly being seen as a way for states to control their Medicaid costs. For younger people with disabilities, as with the elderly population, there is increasing state policy interest in integrating acute and long-term care services through the use of health maintenance organizations (HMOs) and other managed care organizations. Initiatives are either under way or being developed in several states (including Wisconsin, Ohio, and Massachusetts). Because costs are so high for persons with disabilities, the potential savings of more efficient service management are large. But debate rages over whether managed care is appropriate for this group.
People with disabilities currently receive their care in a fragmented and uncoordinated system. Proponents argue that applying the principles of managed care can greatly improve the quality of and access to care at the same time that it controls spending. Opponents, including advocates for younger people with disabilities, point out that managed care organizations have very little experience in serving younger people with disabilities and may not be sensitive to their needs or capable of meeting them. The medical necessity criteria often used by HMOs for home care, rehabilitation, durable medical equipment, and therapies are typically much narrower than commonly exist under Medicaid, raising questions of whether these Medicaid-covered services will be provided as required.
In addition, most managed care organizations depend on primary care physicians to reduce use of specialist care. But persons with disabilities often prefer to rely, even for routine care, on specialists who are knowledgeable about their conditions. This factor is potentially aggravated by managed care’s restrictions on provider choice, which may mean losing access to the providers best qualified to deal with the problems connected with a specific disability.
Finally, managed care organizations are typically dominated by "the medical profession," i.e., physicians and hospitals. This carries two dangers. Long-term care may become overmedicalized and less consumer directed. And managed care organizations may end up shifting Medicaid resources from long-term care to acute care services.
Financing
States are involved in financing a substantial portion of health care for younger people with disabilities. Medicaid accounts for the lion’s share of state health care spending on this population, but a variety of state-only programs provide significant additional funds. As policymakers at the federal and state levels continue to debate ways to restructure the Medicaid program, four financing issues stand out among the many that confront them with respect to younger people with disabilities: the great variation among states in Medicaid spending for this population; the link between Medicaid eligibility and the definition of disability used to qualify for Supplemental Security Income (SSI); the use of Medicaid home and community-based waivers; and the incentives currently facing states to put as many service programs for younger people with disabilities (as well as for other eligible groups) under the Medicaid umbrella as possible in order to partially shift state-only costs to the federal government.
Medicaid Variation among States
State Medicaid programs vary widely in the size of their younger population with disabilities, in total Medicaid expenditures for this population, in the proportions of Medicaid program funds spent on this population, and in per-beneficiary expenditures on this group (Table 1). New York spends considerably more Medicaid dollars on this group than any other state ($7.6 billion in 1994), with California second ($4.3 billion) and Wyoming at the low end ($61 million). Idaho spends the largest share of its Medicaid dollars on younger persons with disabilities (45.6 percent) and Arizona the smallest (16.6 percent). 
In terms of expenditures per younger disabled beneficiary, Connecticut spends the most ($16,262 in 1994) and Tennessee the least ($3,136).
Changing Medicaid into a block grant (as was proposed by Congress in 1995) or imposing per-beneficiary caps on spending (as proposed by President Clinton in 1995 and 1997) could lock into place existing interstate differences in spending if allocations or caps are based on historical spending patterns. Without adjustments to loosen the grip of historical spending patterns on federal allocations, there would be no way for low-spending states to come up to the national average, let alone to the levels of high-spending states—and still obtain a federal match. Conversely, high-spending states will be allowed to retain their more generous programs (with the federal match) if they like.
The SSI-Medicaid Eligibility Link
To a large extent, Medicaid eligibility policy drives Medicaid spending. Medicaid eligibility for younger persons with disabilities is tightly linked to the definition of disability used by the SSI program. The SSI rolls have been increasing for both children and nonelderly adults. Between 1988 and 1995, the number of blind and disabled Medicaid beneficiaries increased by over two-thirds (68 percent), partly reflecting the 1990 Supreme Court decision in Sullivan v. Zebley, which greatly broadened SSI eligibility for children. 7 The Personal Responsibility and Work Opportunity Reconciliation Act of 1996 tightened SSI eligibility for children. Under current law, however, most children would continue to qualify for Medicaid as low-income children even if they lose eligibility for SSI. 8
For adults, the SSI definition of disability is based on an inability to work. Many people on SSI have little work experience, reflecting in part a greater likelihood of congenital problems (such as mental retardation). Only about a quarter of disabled Medicaid beneficiaries have enough quarters of work and earnings to be eligible for Medicare before age 65. As a result of the lack of work experience and their disabilities, transition from public assistance to work is difficult and relatively rare. In fact, the linkage of Medicaid eligibility to an inability to work creates a Catch-22 for younger persons with disabilities who might want to work (and the social service agency personnel who want to find them jobs). If Medicaid-funded services help an individual to find and keep a job, then the person is deemed to be no longer disabled, and the very services that enabled him or her to function in the labor market (such as personal assistance services or prescription drugs) are withdrawn after a transitional period because the person will no longer be covered by Medicaid. Without those services, the person may no longer be able to work and will again qualify for SSI and therefore Medicaid. Both the employment and associated Medicaid savings will have been temporary.
Even if younger persons with disabilities are able to find jobs that provide comprehensive private health insurance, this problem is not likely to disappear. Such insurance is unlikely to include the long-term care services covered by Medicaid and may well exclude coverage for preexisting conditions, which this population has by definition. Existing Medicaid provisions that attempt to alleviate these problems have not succeeded in moving many persons with disabilities into the labor force.
Medicaid Home and Community-Based Waivers
All states have Medicaid home and community-based waiver programs for younger persons with disabilities, most commonly for people with mental retardation or developmental disabilities. (Arizona operates a similar program through an 1115 research and demonstration waiver.) Under these programs, states offer a wide range of home and community-based services to a population that is at high risk of institutionalization without these services. In order to obtain federal approval for its waiver, a state must demonstrate that its program will be "cost effective," that is, that the state’s average per capita Medicaid costs with the waiver will not exceed its average per capita costs without the waiver. Unlike the rest of the Medicaid program, states may explicitly limit participation in these programs to a predetermined number of people. To help make this possible, the Health Care Financing Administration waives requirements for "comparability" (i.e., the requirement that services be provided to all groups equally) and "statewideness" (i.e., the requirement that all benefits be covered in all parts of the state).
After a reasonably slow start in the early 1980s, home and community-based waiver programs (for both the elderly and disabled) have grown extremely rapidly in recent years, increasing from $735 million in state and federal spending in 1988 to $4,631 million in 1995. Most of the expenditure growth in recent years has been due to the increase in the number of people participating in programs for people with mental retardation/developmental disabilities.
President Clinton has proposed to allow states to implement Medicaid home and community-based service programs on a cost-neutral basis without having to obtain a federal waiver. Advocates of increased state flexibility argue that many of the waiver requirements are needlessly bureaucratic and do not address quality of care. Moreover, states have substantial experience with long-term care services for people with disabilities and, according to this view, can be trusted with more latitude. Although conflict between the federal government and the states over approval of waivers was substantial and bitter during the Reagan and Bush Administrations, regulatory changes implemented by the Clinton Administration have made obtaining waivers fairly routine. Thus, according to this line of argument, little is to be gained by requiring states to go through the federal waiver process.
Opponents of greater state flexibility argue that the current ease of obtaining waivers is precisely the problem. In this view, the Health Care Financing Administration has not been tough enough in requiring that services be cost effective. And because of lax standards, services are often provided to people—especially the elderly—who are not at a high risk of institutionalization. The net result is an inappropriate increase in federal Medicaid spending.
Cost Shifting and Medicaid Maximization
Because states have a substantial commitment to the financing and direct provision of services for younger persons with disabilities, they have a strong incentive to shift the cost of such services to the Medicaid program (where a federal match is available). For example, in 1993, states spent almost as much for non-Medicaid long-term care services for nonelderly people with physical disabilities and nonelderly people with mental retardation/developmental disabilities as they did for Medicaid-financed services. 9 Particularly for services for persons with mental retardation/developmental disabilities, the increased expenditures for Medicaid home and community-based waivers in recent years represented, in part, refinancing of existing state programs. 10 The line between health care and social services, vocational training, and education is fuzzy, especially for long-term care services for the mentally retarded/developmentally disabled and the chronically mentally ill. Thus, vocational training, education, and social services are typically state funded, but if categorized as a Medicaid service will be eligible for a federal match.
Another example of Medicaid maximization is the use of the Medicaid disproportionate share hospital (DSH) program. By federal law Medicaid does not provide coverage for persons between the ages of 22 and 64 who are in "institutions for mental disease" (that is, mental hospitals). Under the 
DSH provisions of the statute, however, state Medicaid agencies can make extra payments to hospitals that serve a disproportionate share of people who are Medicaid-eligible or uninsured. Some states have made extremely large payments under this provision to their state mental hospitals, using federal Medicaid dollars for the very purpose that is disallowed at the individual beneficiary level under federal law.
How much further states can shift additional expenses for younger persons with disabilities to Medicaid is unclear. The potential may be limited, at least in the case of services for the mentally retarded/developmentally disabled. Some observers believe that Medicaid home and community-based waiver expenditures will grow more slowly in the future because "further increases are more and more dependent on hard-to-obtain new state matching dollars." 11 In addition, the Omnibus Budget Reconciliation Acts of 1990 and 1993 effectively capped increases in spending on DSH payments.
Conclusion
Younger persons with disabilities account for a substantial portion of state health spending. It is probably not possible to control Medicaid expenditures over the long run without addressing services and costs for this population. As states consider their health programs for younger persons with disabilities, there are at least five points to keep in mind:
- Given the high Medicaid costs of persons with disabilities, it is likely that states will enroll greater numbers of persons with disabilities into managed care. To date, however, managed care organizations have not had much experience with low-income persons with mental retardation, mental illness, or serious physical impairments. Efforts to quickly enroll large numbers of people with disabilities into managed care run the risk of either not producing the expected savings or reducing the quality of care that enrollees receive.
- As with the rest of the Medicaid program, spending on younger persons with disabilities varies tremendously across states, both in terms of total dollars and spending per beneficiary. 12 Without substantial adjustments, Medicaid block grants or caps on the rate of growth in per-beneficiary spending will lock into place existing variations, making it impossible for low-expenditure states to reach the levels of states that currently spend more.
- There is a tension between covering a broader range of services and the open-ended entitlement structure of Medicaid. While disability advocates argue that, with a very broad menu of services, persons with disabilities will use cheaper and fewer services than professionals would choose, budget officials worry that a broader range of services will lead to higher utilization and substantially greater expenditures. The Medicaid home and community-based waivers put fiscal constraints on an otherwise very large potential demand by limiting the number of people who can receive services. At its extreme, a Medicaid block grant without an individual entitlement to coverage would give states maximum flexibility to provide flexible benefits without having to worry about entitlement-driven increases in demand driving up expenditures. The danger in this scenario is that funding will be inadequate to provide services to all the persons with disabilities who meet current eligibility criteria.
- Efforts to move SSI beneficiaries into the workforce face special difficulties. Persons with disabilities are particularly dependent on Medicaid and other services that come with being "unable to work." Movement into jobs can mean the loss of the very benefits they need.
- Because the costs of serving persons with disabilities are so high, there is a strong incentive on the part of the states to shift costs to the federal government. Many states have refinanced their state-funded programs for younger people with disabilities by moving them into the Medicaid program, thus gaining a federal match for those state (and sometimes local) expenditures. Some states have used the federal match to expand services; others have taken the opportunity to cut back on their own spending. How much more state and local spending on younger persons with disabilities can be shifted to the federal government could have a strong influence on the rate of increase in Medicaid spending in the years ahead.
Tables, Figures, Charts and Graphs
Notes
1. See John Holahan and David Liska, "Reassessing the Outlook for Medicaid Spending Growth," New Federalism: Issues and Options for States, Series A, no. A-6 (Washington, D.C.: Urban Institute, March 1997).
2. Joshua M. Wiener and Catherine M. Sullivan, "Long-Term Care for the Younger Population: A Policy Synthesis," in Joshua M. Wiener, Steven B. Clauser, and David L. Kennell, Persons with Disabilities: Issues in Health Care Financing and Service Delivery (Washington, D.C.: Brookings Institution, 1995), pp. 291-324.
3. Wiener and Sullivan, "Long-Term Care for the Younger Population"; Gary A. Smith and Robert M. Gettings, The Medicaid Home and Community-Based Waiver Program: Recent and Emerging Trends in Serving People with Developmental Disabilities (Alexandria, Va.: National Association of State Directors of Developmental Disabilities, 1996).
4. Holahan and Liska (1997).
5. Smith and Gettings (1996).
6. Ibid.
7. Holahan and Liska (1997).
8. See Leighton Ku and Teresa A. Coughlin, "How the New Welfare Reform Law Affects Medicaid," New Federalism: Issues and Options for States, Series A, no. A-5 (Washington, D.C.: Urban Institute, February 1997).
9. Office of the Assistant Secretary for Planning and Evaluation, Cost Estimates for the Long-Term Care Provisions under the Health Security Act (Washington, D.C.: U.S. Department of Health and Human Services, 1994).
10. Smith and Gettings (1996).
11. Ibid., p. 19.
12. For further variation in Medicaid spending generally, see John Holahan and David Liska, "Variations in Medicaid Spending among States," New Federalism: Issues and Options for States, Series A, no. A-3 (Washington, D.C.: Urban Institute, January 1997).
About the Author
Joshua M. Wiener is a principal research associate at the Urban Institute’s Health Policy Center. His specialties are Medicaid, health care for the elderly, and long-term care. He came to the Urban Institute in 1996 from The Brookings Institution. His most recent book is Persons with Disabilities: Issues in Health Care Financing and Service Delivery (Washington, D.C.: Brookings Institution, 1995).
