Medicaid: Recent Trends in Beneficiaries and Spending

This report discusses medicaid and recent trend is beneficiaries and spending.

Medicaid: Recent Trends in Beneficiaries and Spending Kathleen King Richard Rirnkunas Specialists in Social Legislation and Dawn Nuschler Technical Information Specialist Education and Public Welfare Division March 27, 1992 CRS MEDICAID: FtECENT TRENDS IN BENEFICIARIES AND SPENDING SUMMARY Total (Federal, State and local) Medicaid spending increased by 18 percent in FY 1990. In FY 1991,it increased an additional 26.9 percent to $92.0 billion. Recently, Medicaid spending has grown faster than both Medicare and all other national health spending. In examining Medicaid spending trends from FY 1987 through FY 1991,two striking features emerge. One is the rapidly accelerating rate of growth in inpatient hospital spending--24 percent in FY 1990 and almost 45 percent more the following year. Second, the rate of growth in Medicaid spending varied tremendously by State and region of the country, with spending growing fastest in southern States. While it is not possible to explain definitively why Medicaid spending has increased so rapidly, a number of contributing factors can be identified. Among those discussed in this report are: inflation; rapid increases in the number of Medicaid beneficiaries; increased reimbursement for selected Medicaid services; and changes in States' sources for Medicaid revenues. In FY 1990,the number of beneficiaries was 5.7percent greater than in FY 1989. When beneficiary data for FY 1991 become available, they are expected to reveal an even greater rate of growth in recipients; continuing growth is expected through 1997. The number of beneficiaries is rising because Medicaid eligibility is partly tied to cash welfare programs (Aid to Families with Dependent Children (AFDC)and Supplemental Security Income (SSI)), which are growing rapidly. Between 1989 and 1991,the number of families enrolled in AFDC rose almost 19 percent and SSI enrollment increased 9.5 percent. In addition, congressionally mandated eligibility expansions for low-income pregnant women and children, low-income Medicare beneficiaries, and others are factors in increasing enrollment. Rising reimbursement for Medicaid services also contributed to growth in expenditures. Medicaid law requires hospital and nursing home payment rates to be "reasonable and adequate" to meet the costs of "efficiently and economically" operated facilities. In a number of States, providers have sued, claiming that the payments failed to meet this test. As a result of these lawsuite, some States have been required to increase payments for inpatient hospital and nursing home services. Payments to hospitals have also risen in recent years because States have increased payments to hospitals serving a disproportionate number of low-income people. The effects of voluntary contributions and provider-specific taxes, new State revenue sources for the program, are less clear. Some maintain that these revenue sources allowed States to increase Medicaid spending far more than they would have otherwise. They contend that these funding sources were primary contributors to rising Medicaid spending. Whether spending would have increased so quickly in the last few years without these funding sources is not easily determined. Although Congress enacted legislation in 1991 phasing-in limits on use of these revenue sources, they are likely to comprise a substantial share of total FY 1992 spending. ACKNOWLEDGEMENTS The authors want to express their appreciation to a number of individuals who provided data and assistance in the preparation of this report. Their thanks go to John Mauck, David Evans, Martin Gilbert, John Klemm, Miles McDermott, and Bill Surine a t Health Care Financing Administration for their invaluable and timely assistance in providing Medicaid data. Several Congressional Budget Office analysts, including Jean Hearn, Jan Peakin, Pat Purcell,.and John Tapogna also provided data, analyeis, and helpful comments. They also thank David Colby and Anne Schwartz of the Physician Payment Review Commission and Carl Volpe of the National Governors' Association for their aseistance. CONTENTS EXECUTIVESUMMARY .................................. 1 CHAPTER 1. RECENT EXPENDITURE AND BENEFICIARY TRENDS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 BACKGROUND ON MEDICAID . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Eligibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Reimbursement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 MEDICAID OUTLAYS: RECENT TRENDS AND PROJECTIONS . . 7 MEDICAID BENEFICIARIES: RECENT TRENDS AND PROJECTIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 CHAPTER 2. RECENT TRENDS: SERVICES AND STATES . . . . . . . . SERVICE SPENDING TRENDS . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acutecarespending . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Long-Term Care Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . STATE SPENDING TRENDS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 13 15 16 17 CHAPTER 3. FACTORS ASSOCIATED WITH INCREASING MEDICAID SPENDING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 HEALTHCAREPRICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 REIMBURSEMENT POLICIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Hospitals and Nursing Facility Services . . . . . . . . . . . . . . . . . . . 22 Effect of the Boren Amendment . . . . . . . . . . . . . . . . . . . . . 23 Effect of Disproportionate Share Payments (DSH) . . . . . . . 27 Quality Assurance in Nursing Homes . . . . . . . . . . . . . . . . . 31 New Funding Sources for Hospital and Nursing Home Spending: Provider Donations and ProviderSpecificTaxes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Physicians' Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Prescription Drugs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Enrollment Increases: Growing Numbers of Cash Welfare Recipients and Medicaid Program Expansions . . . . . . . . . . 40 Aid to Families With Dependent Children (AFDC) Enrollment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Supplemental Security Income (SSI) Enrollment . . . . . . . . 42 Eligibility Expansions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 Pregnant Women and Children . . . . . . . . . . . . . . . . . . 44 Qualified Medicare Beneficiaries (QMBs) . . . . . . . . . . . 46 Aid to Families With Dependent Children-Unemployed Parents (AFDC-UP) . . . . . . . . . . . . . . . . . . . . . . . 47 OtherFactors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 Acquired Immune Deficiency Syndrome (AIDS) . . . . . . . . . 48 MEDICAID: RECENT TRENDS IN BENEFICIARIES AND SPENDING Since the mid-1980s, Congress has enacted a series of measures to broaden accees to Medicaid or increase payment for Medicaid cervices. Examples of these measures include: (1) successive expansions of Medicaid eligibility to poor pregnant women and children; (2) extension of limited Medicaid benefits to poor and near-poor Medicare beneficiaries; and (3) requirements that State Medicaid programs increase payments to hospitals serving a disproportionate number of low-income persons. Congress enacted these and other measures in response to a perceived need to provide poor people with necessary health care services. For example, the primary purpose of eligibility expansions for pregnant women was to reduce unacceptably high infant and neonatal mortality rates. Medicaid expenditures rose more rapidly over the last several years than they did in the mid-1980s. In the 4 years between FY 1987 and FY 1991, total (Federal, State and local) Medicaid spending nearly doubled from $49.3 to $92.0 billion, with most of the growth concentrated in the last 2 years. In FY 1990, Federal Medicaid outlays grew by 18.8 percent, indicating a sharp upturn in Medicaid spending after a period of more moderate growth in the mid-1980s. In FY 1991, the rate of increase in Federal Medicaid spending was even higher--28 percent. These rates of growth have been considerably higher than anticipated, raising questions about why spending is growing so rapidly and whether the current rate of growth will continue. This report examines detailed Medicaid expenditure and beneficiary trends in an attempt to explain this rapid rate of growth. It presents data on Medicaid epending and beneficiary trends from F Y 1987 through FY 1991 and projections for FY 1992 to F Y 1997. I t also discusses epending trends for health services covered by Medicaid and variations in State spending patterns. Although data limitations and the interaction among complex factors contributing to Medicaid spending preclude a definitive explanation, this report provides information and analysis about how various factors have affected Medicaid epending. Major factors include: inflation; Federal or State changes in reimbursement policies; decisions of the courts; and rising numbers of Medicaid beneficiaries. The underlying rate of general inflation in the economy and specific price increases in health services exert considerable pressure on health spending trends. These factors generally account for more than half of the increase in national health spending each year. Their effects on Medicaid are probably lower because Medicaid reimbursement rates are frequently not tied to a cost of living index. However, changes in reimbursement policies, particularly for inpatient hospital services, have contributed to rising Medicaid spending. In 1987, Congrese enacted legislation requiring States to increase payments to hospitals serving a disproportionate number of low-income patients. Theee supplementary payments are provided to hospitals that meet State-defined disproportionate share payments (DSH) criteria. In FY 1992, DSH to hospitals are estimated to equal $14 billion. Medicaid spending for inpatient hospital services increased by 24 percent in FY 1990 and almost 45 percent in FY 1991. In 1991, Congress enacted limita on growth in DSH. The Administration has cited two new sources of State revenues as major contributors to rising Medicaid spending: State revenues derived from voluntary contributions from health care providers and/or taxes on health care providers (referred to as provider-specific taxes.) Since the funding is from provider donations or provider-specific taxes, the Administration maintains that these revenue sources allowed increased Medicaid spending to occur with only small or no increases in spending from State general revenues. By law, the Federal Government is required to pay for a share of all Medicaid spending. It is argued that reliance on provider donations and taxes, along with Federal matching payments places more of the burden of increased payments on the Federal Government. This perspective argues that providers receive larger payments, and the Federal Government pays a larger Medicaid share of spending, while net State spending remains close to what it was before the donation or tax plan. In contrast to the Administration's perspective, the States assert that the source of revenues used for Medicaid spending is irrelevant. Furthermore, the States assert that these are essential resources in helping them cope with burgeoning Medicaid spending, some of which has been required by Federal law. Some of the growth in both inpatient hospital and nursing facility spending also appears to stem from laws enacted in the early 1980s. The "Boren Amendment" permitted States to deviate from cost-based reimbursement policies and required only that payment rates be reasonable and adequate" to meet the costs of "efficiently and economically" operated facilities. The law did not define reasonable and adequate payments or efficiently and economically operated facilities. Most States implemented alternative payment systems in the next several years. Although the Boren Amendment was intended to give States more flexibility to implement reimbursement systems that would contain costs, Medicaid providers of inpatient hospital and nursing facility services have also cited the Boren Amendment in legal challenges to Medicaid reimbursement rates. Lawsuits filed in many States asserted that rates were inadequate to meet the costs of efficiently and economically operated facilities. In many instances, the courts have sided with providers and have ordered States to revise their Medicaid plans and increase payment rates to comply with the Boren Amendment. Comprehensive data have not been collected that measure the financial impact of Boren Amendment suits in recent years. However, they are expected to have a substantial impact on inpatient hospital and nursing facility expenditures in future years. Some of the growth in Medicaid spending is associated with growing numbers of beneficiaries. These increases stem from two sources: (1)increasing enrollment in cash welfare programs linked to Medicaid; and (2) Federal laws requiring expanded Medicaid eligibility for low-income pregnant women and children and low-income Medicare beneficiaries. Generally, people receiving or eligible to receive Aid to Families with Dependent Children (AFDC)and Supplemental Security Income (SSI) are eligible for Medicaid.' While not all of those eligible for Medicaid actually use eervices, increasing numbers of eligible beneficiaries undoubtedly contribute to rising expenditures. From October 1989 to October 1991, AFDC case loads rose by 18.8 percent, one of most rapid increases in the program's history. Enrollment in SSI is also growing, although not as quickly. From FY 1989 through FY 1991, the number of SSI enrollees grew 9.5 percent. The recent growth rate, which is projected to continue through FY 1997, marks a deviation from a long-standing pattern of very slow growth in SSI enrollment. On a person by person basis, increasing SSI enrollment is likely to have a stronger effect on Medicaid spending. Those enrolled in SSI are aged, blind or disabled and more likely to use the health care system than AFDC enrollees. This is reflected in higher Medicaid expenditures for SSI enrollees. In FY 1990, per capita Medicaid expenditures for SSI enrollees were $4,478, compared to $1,880 for AFDC adults and $736 for AFDC children. Medicaid eligibility expansions mandated by Congress over the last several years have affected Medicaid expenditures. Beginning in 1984, Congress enacted a series of Medicaid eligibility expansions for low-income pregnant women and children. Currently, States are required to extend Medicaid eligibility to pregnant women and children under age 6 below 133 percent of the Federal poverty etandard. They are also required to phase-in Medicaid coverage to children under age 1 9 under 100 percent of the poverty standard during the 1990s. Health Care Financing Administration (HCFA) data indicate t h a t in FY 1990, optional and mandated legislative expansions accounted for about 50 percent of the 1.4 million increase in Medicaid enrollees that year. To a lesser extent, congressionally mandated eligibility expansions for lowincome Medicare beneficiaries also contributed to rising Medicaid expenditures. Under current law, States are required to offer limited Medicaid benefits to Medicare beneficiaries whose incomes are below 100 percent of the Federal 'States may impose some restrictions on the link between SSI eligibility and Medicaid. poverty standard. States are required to pay Medicare premiums, coinsurance and deductibles for these individuals.' From May 1991 to February 1992, the number of Medicare beneficiaries for whom Medicaid paid Medicare Part B premiums rose 41 percent from 762,741 to 1,078,200. Qeginning in 1993, Medicaid will be required to phase-in payment for Medicare premiums only to near-poor Medicare beneficiaries. CHAPTER 1. RECENT EXPENDITURE AND BENEFICIARY TRENDS BACKGROUND ON MEDICAID Medicaid is a Federal-State program that provides medical assistance to certain low-income persons who are aged, blind, disabled, members of families with dependent children, and certain other pregnant women and children. Each State designs and administers its own program within Federal guidelines. As a result, State Medicaid program vary eubstantially in t e r n of the number of people covered, the scope of eervices provided, and reimbursement rates for covered services. The Federal Government shares in the cost of Medicaid through grants to the States. It matches States' payments through a formula tied to States' per capita incomes. The Federal share (called the Federal medical assistance percentage or F W ) may range from 50 to 83 percent, with the highest matching rates in States with the lowest per capita incomes. In FY 1992, Mississippi has the highest FMAP rate, 79.93 percent, while 12 States, the District of Columbia and all the territories receive the minimum 50 percent match.= The Federal match rate for benefits and administration was about 57 percent in FY 1991. States are responsible for the nonfederal share of Medicaid payments. In addition to using State funds to finance Medicaid benefits, some States require local governments to share in the nonfederal part of the costs of the program. However, States are required to pay at least 40 percent of the nonfederal share of Medicaid expenditures. AB of September 1991, 14 States required local governments to pay for a least some portion of Medicaid costs. Medicaid spending results from entitlements established in law. Each State administers the program through a single State agency in accordance with a State plan approved by the Federal Government.' Within Federal guidelines, State plans define eligibility, coverage, reimbursement and administrative policies. Individuals who meet eligibility guidelines are entitled to have States pay for covered services provided to them. In turn, States are entitled to receive matching payments from the Federal Government for the Federal share of Medicaid expenditures. The Federal Government is obligated to make payments to States in accordance with payments made under States' Medicaid plans, so there is no absolute limit on the amount the Federal Government must pay. 'Federal spending in the territories is subject to annual dollar limits. 'Massachusetts is the only State that operates Medicaid through two agencies, one responsible for the general program and one Medicaid coverage of the blind. Historically, Medicaid eligibility has been linked to actual or potential receipt of cash assistance (welfare) under the AFDC program or the Federal SSI program for poor aged, blind or disabled persons. These beneficiaries are referred to as the "categorically eligible." States have considerable flexibility in determining eligibility for AFDC and eome leeway in determining the link between SSI and Medicaid eligibility. Over the past several years, Congress has enacted laws gradually extending eligibility for Medicaid to poor people not eligible for either of these two programs, euch as low4ncome pregnant women and children and low-income Medicare beneficia~ies.~ Some of these eligibility expansions were initially optional, but are now required by law. States are also permitted to extend Medicaid eligibility to "medically needy" persons-those who fall into one of the categories of groups of persons State Medicaid programs are required to cover, but whose incomes or resources are above the cash assistance standards set by the States. The medically needy must meet eeparate income standards. As of October 1, 1991, 36 States and the District of Columbia provided some coverage to medically needy groups. Federal law requires States to offer coverage for some mandatory services to the categorically eligible. Among others, mandatory services include: inpatient and outpatient hospital care; physicians' services; laboratory and X-ray eervices; rural health clinic services; nursing facility services for individuals 21 or older; federally-qualified health center services; early and periodic screening, diagnosis and treatment (EPSDT)services for individuals under age 21; and family planning services. States may also offer coverage for a broad range of optional services, such as prescription drugs; dental services; and eyeglasses. States with medically needy programs are permitted to offer a smaller range of eervices to these individuals. States determine service coverage policies within broad Federal guidelines and are permitted to place limits on coverage of all eervices except EPSDT. For example, States may limit the number of covered inpatient hospital days, physician visits, or prescription drugs. Reimbursement In general, Federal law grants States wide latitude in establishing reimbursement eystems and payment rates for many Medicaid services. However, three broad principles apply to these payments. First, Federal law requires that "methods and procedures" for making payments ensure that payments are "consistent with effkiency, economy and quality of care." Second, in response to concerns that Medicaid payments were too low to attract 'Medicaid coverage for low-income Medicare is limited to payment of Medicare premiums, coinsurance and deductibles. providers, Congress required payment rates to be suficient to attract enough providers so that covered services will be as available to Medicaid beneficiaries ~ special rules apply to rates for as they are to the general p ~ p u l a t i o n .Finally, hospital and nursing facility services. There is a great deal of variability in States' reimbursement systems and in the rates paid for Medicaid services. For many services, Medicaid payment rates are lower than those paid by Medicare or other insurers. Sometimes States respond to budget crises by delaying payment updates or reducing payment rates. MEDICAID OUTLAYS: RECENT TRENDS AND PROJECTIONS Table 1and figure 1show actual Medicaid outlays (including payments for benefits and administration) from F Y 1987 through FY 1991. Total Medicaid spending nearly doubled from $49.3 billion in FY 1987 to $92.0 billion in FY 1991. Growth in total Medicaid spending has been especially rapid in the last 2 years--18.4 percent in FY 1990 and 26.9 percent in FY 1991. Federal Medicaid outlays grew slightly faster during this period than Medicaid spending by State and local governments. In F Y 1991, Federal Medicaid spending grew by 27.8 percent, compared to 25.8 percent for State and local governments. Table 1 and figure 2 also show Congressional Budget Office's (CBO) projections of Federal Medicaid spending from FY 1992-FY 1997, which were released in January 1992. CBO projects an average annual rate of increase in Medicaid spending of 15.7 percent for these 6 years. Their projection that Federal Medicaid spending will grow 30 percent in FY 1992 is striking, particularly since it follows on the heels of a 28 percent increase in FY 1991. The projected rate of growth in FY 1993 is 17 percent, indicating that the surge in Medicaid spending is expected to continue through next year. After FY 1993, however, CBO projects more moderate yet still robust growth--around 12 percent a year-through FY 1997. Assuming that the Federal share of Medicaid spending remains around 57 percent, total Medicaid outlays would grow from a projected $92.0 billion in FY 1991 to $222.4 billion in FY 1997. Federal Medicaid outlays are projected to increase from $52.5 billion in FY 1991 to $126.1 billion in FY 1997. In F Y 1966, Federal Medicaid spending accounted for 0.6 percent of the FederaI budget; by FY 1991, it had grown to 4.0 percent of Federal spending. Most of the increase occurred between 1966 and 1974, during the program's early years. Between 1974 and 1985, Medicaid consistently accounted for about 2.4 percent of Federal spending. Since 1985, Federal Medicaid spending has represented an increasing share of Federal outlays. If Federal Medicaid epending more than doubles between FY 1991 and F Y 1997, as projected, it will account for 7.3 percent of Federal spending in FY 1997. q h i s provision, which had been required by regulation, was enacted into law in the Omnibus Budget Reconciliation Act of 1989 (OBRA 89). Projected rates of growth for Medicaid spending are eubstantially higher than for Medicare. CBO projects that Medicare benefit outlays will increaee an average of 11.1percent a year, from $128.3 billion in FY 1992 to $217.3 billion in FY 1997. Baaed on these projections, combined Federal, State and local Medicaid spending is likely to equal Medicare spending by FY 1994, and surpass it in future years. Projected ratee of growth in Medicaid spending are also substantially higher than for total national health spending (excluding Medicaid). Recently, HCFA projected national health care expenditures to the year 2000. Theee projections assume that the economy will grow at modeet ratea with low inflation, and that current policies and historical spending trenda will continue.? Based on these assumptiom, national health expenditures (excluding Medicaid) are projected to grow an average of 9 percent a year from $704 billion in 1992 to $1,27212 billion in 1997.8 'Sonnefeld, Sally T., D.R. Waldo, Jeffrey A. Lemieux, and D.R. McKusick. Projections of National Health Ekpenditures through the Year 2000. Health Cum Financing Review, v. 13, no. 1,fall 1991. 'In projecting national health expenditures to 2000, the Health Care Financing Administration (HCFA)also projected Medicaid expenditures. The Medicaid projections use different data than those cited in this report. These projections assume that Medicaid will grow a t an average rate of 12.5 percent a year from calendar 1992 through 1997. TABLE 1. Federal and State and Local Medicaid Spending, Actual Spending F Y 1987-FY1991, Projected Spending FY 1992-FY1997' (dollars in billions) Fiscal year Federal outlays Annual increase State/ local outlays Annual Increase Total Annual increase 'Projected Medicaid spending is baeed on CBO projections of Federal Medicaid spending as of Jan. 1992. These projections will be updated in the summer of 1992. CBO does not project State and local government Medicaid spending. bFiscal year 1991 Federal outlay information as reported in the Final Monthly Treasury statement for FY 1991. The Treasury statement does not provide information on State and local spending. State and local spending figures reported are preliminary HCFA estimates which assume that the overall Federal matching percentage is 57.07 percent. 'The CBO does not provide projections for State and local epending. Total and State and local projections are estimated by Congressional Research Service (CRS)on the assumption that the Federal share of Medicaid outlays remains at 57.07 percent. NOTE: Spending figures reflect Federal outlays and use a slightly different accounting method than expenditure information which is reported in eubsequent tables. Source: Health Care Financing Administration data for 1987-1990.Data for 1991 from Department of Treasury. Projections for 1992-1997based on CBO projections as of Jan. 1992. FIGURE 1. Federal and Total Medicaid Spending FY 1 9 8 7 t o F Y 1991 Billions of Dollars 100 Total Program 80 60 Federal Outlays 40 20 0 Flscal Year FIGURE 2. Projected Trend in Medicaid Spending FY 1991 to FY 1997 Billions of Dollars I Total Program 220.0 Federal Outbys 68.2 - 52.5 Fiscal Year N&: All figwm .np m j ~ ~ud W s u b j ~@ t Ih. Nthodr and dat. mod in thoir ulculabonr. CBO projrcb F d w J spending only.7ot.l rpond~ngurumos Fodual matching porcontlgb Of 57% Souco: Prmpud by CRS. B u o d on dmla obtained from Un CongrrrmnJ Budgot Oflro. JuLIODP CRS-11 MEDICAID BENEFICIARIES: RECENT TRENDS AND PROJECTIONS Table 2 shows the number of Medicaid beneficiaries for FY 1987 through FY 1991 and CBO's staff estimates of beneficiaries for FY 1992 through FY 1997. A beneficiary is defined as a person enrolled in Medicaid who receives a medical service paid for by Medicaid during that fiscal year. Between 1989 and 1990,Medicaid beneficiaries increased by 5.7 percent. The last time Medicaid experienced such large increases in beneficiaries was the early 1970s,following the enactment of major eligibility expansions. Figure 2 displays the historical trend in beneficiaries, while figure 3 provides the program's projected trend. TABLE 2. Actual and Projected Number of Medicaid Beneficiaries, FY 1987-FY1897 (in millions) Fiscal year Beneficiaries' Annual increase .Beneficiary is defined as a Medicaid enrollee who receives a medical semce paid for Medicaid during the fiscal year. bEstimated and projected figures were obtained from the CBO in Jan. 1992. These projections are unpublished staff estimates and are likely to change as new information becomes available. '1991 rate of increase is calculated from 1990 actual enrollment. Source: Health Care Financing Administration Form 25g for actual beneficiary enrollment for FY 1987-FY 1990, Unpublished staff estimates of CBO for FY 1991-FY1997,Jan. 1992. FIGURE 3. Number and Rate of Growth of Medicaid Beneficiaries, N 1987 to P/ 1991 Mllllon8 of Benmflclarler 40 Percent Change Numbor of Boneflclerles -me: ncm Fonn a%. FIGURE 4. Projected Number and Rate of Growth of Medicald Beneficlarier, N 1991 to PI 1997 --. 40 Mllllonm of Bonmflclarlmr 1 I CHAPTER 2. RECENT TRENDS: SERVICES AND STATES SERVICE SPENDING TREXDS This section describes eervice expenditure trends from FY 1987 through F Y 1991. Table 3 divides total Federal, State and local) Medicaid expenditures into two large categories: spending for acute care eervices and spending for longterm care services. As ueed in this report, long term care services include all covered nursing facility eervices, services provided in intermediate care facilities for the mentally retarded (ICF/MRB), nursing and other personal care services provided in an individual'e home, home and community-based waiver (HCBS) services, and inpatient mental health services. The acute care eervice category is dominated by three eervices: hospital eervices, physician eervicee, and prescription druga. All other remaining Medicaid covered services are included in the "other" acute care services category. Some services in this category include: other practitioner services; dental services; clinic eervices; laboratory and radiological services; EPSDT eervices; and rural health clinic services. Some Medicaid service spending has grown much more quickly than other service spending. Table 3 highlights a number of important aspects about this trend. First, payments for acute care services have been growing much more rapidly than spending for long-term care eervices. In 1987, acute care spending was 10 percent larger than long-term care spending, but, by 1991, spending for acute care was 45 percent larger. This reverses a trend during the mid-1980s, when long term care spending increased a t a faster rate.' "Chang, Deborah and John Hollahan. Medicaid Spending in the 1980s: The Access-Cost Containment Trade-off Revisited. Urban Institute Report, m2. Washington D.C., Urban Institute Press, 1990. TABLE 3. Spending for Selected Medicaid Services, FY 1987-FY1991 (All payment amounts are in thousands) Service category 1987 1988 1989 1990 1991 Percent change 1987-1991 Inpatient hoepital Outpatient hompital Phyeician Preecription druge Other eervicee SubtotalAcute care eerviees Nureing homes ICFMR Home health HCBS waivers Peraonal care servicee Mental health services SubtotalLong term care services $22,342 $24,439 $27,164 $31,379 $35,752 60.0% Totel-Medical aaaietance payments $46,956 $51,646 $58,646 $69,754 $87,554 86.6'16 NOTE: All figures are cetimatea and eubject ta data limitatione and methods of calculation. Service epending amounte do not include adjustments for overpayment or underpayments, or other adjuetments to pnyment amounte. FY 1991 information ie preliminary and subject to change. Payments are for medial eervicea and exclude any adminietrative coets. Source: Table prepared by CRS baaed on data submitted by the Statee to IICFA. HCFA form 64--Quarterly Medicaid Statement of Expndituns fir the Medical Assistance Pmgmm. Acute Care Spending Much of the growth in acute care spending is attributable to very rapid growth in inpatient hospital spending. Between FY 1987 and FY 1991, it increased almost 120 percent, from $11.5 billion to $25.2 billion. For the last 2 of those years, spending for inpatient care grew faster than any other Medicaid service. In FY 1990, inpatient hospital spending was 24 percent higher than in FY 1989. In FY 1991, the rate of increase was nearly twice as fast-45 percent. That year, for the first time, Medicaid spending for inpatient hospital services exceeded spending for nursing facility services. This dramatic and rapid rise in inpatient hospital spending can be contrasted with utilization and reimbursement statistics. American Hospital Association (AHA) data on Medicaid hospital use suggests that the rise in inpatient hospital spending coincides with greater inpatient use, but not increased lengths of stays in hospitals. From 1987 through 1990, the latest year for which AHA data are available, Medicaid hospital discharges increased at an average annual rate of 5.9 percent. The total number of Medicaid hospital days increased at a somewhat lower average annual rate of 4.2 percent.'' The result is that average length of stay for Medicaid patients dropped from 8.6 days in 1987 to 8.2 days in 1990. This suggests that there is some increase in service use, but not intensity of hospital services over this time period. In addition, there are indications that Medicaid payment rates have increasingly fallen below hospitals' costs. A recent Prospective Payment Assessment Commission (ProPAC) study reported that Medicaid paid 92 percent of hospitals' costs in treating Medicaid patients in 1980, but that the percentage of costs paid by Medicaid fell to 72 percent by 1989." However, it should be noted that the ProPAC study covers a time period that does not include the most rapid increase in inpatient payments. More timely data on reimbursement rates and utilization are needed to determine whether the most recent increase in inpatient hospital payments has been associated with increased access, use and or increased reimbursement rates. The trend in spending for outpatient hospital services is similar to inpatient care, but less pronounced. Spending for outpatient hospital care doubled from $2,122 billion in FY 1987 to $4,244 billion in FY 1991. It grew 22 percent in FY 1990, nearly a t the same rate as inpatient spending. In FY 1991, however, - - - -- - - '%published American Hospital Association (AHA) data obtained through personal communication. These data reflect hospital cost-reporting years 1987 to 1990, which do not necessarily coincide with Federal fiscal year data, so they should not be directly compared to other hospital expenditure data cited in this report. "U.S. Prospective Payment Assessment Commission. Medicaid Hospital Payment. Congressional Report, C-91-02. Oct. 1, 1991. (Hereafter cited as Prospective Payment Assessment Commission, Medicaid Hospital Payment) outpatient spending grew at a much slower rate, 28 percent, than inpatient spending. Medicaid spending for physicians' services rose 81 percent from $2.9 billion in FY 1987 to $5.5 billion in FY 1991 and consistently accounted for 6 percent of Medicaid expenditures. The rate of increase in physician spending was more modest in FY 1988--6.8 percent, but accelerated each year thereafter, reaching 26 percent in FY 1991. The "other" service category represents spending for all remaining acute care senrices, as well as payments made by Medicaid for other health insurance, such as Medicare. In fact, more than 43 percent of the increase in the other payment category is associated with increased Medicaid payments for Medicare's part A and part B premiums, coinsurance amounts and deductibles, and payments for other health insurance.I2 As will be discussed below, over the last few years Congress has expanded the number of individuals eligible for these payments. Among others, some additional spending items in this category include: dental services, other practitioner services, clinic services, laboratory and radiological services, hospice benefits, transportation semces, physical therapy and other services. Long-Term Care Semcee Spending on long term care services reveals two distinct patterns. First, although nursing home spending increased from FY 1987 to FY 1991, it grew more slowly than overall Medicaid spending. It increased from $13.6 billion in FY 1987 to $20.8 billion in FY 1991, or 54 percent over 4 years. Because nursing facility spending grew more slowly than other eemces, it accounted for a smaller share of total Medicaid spending by 1991. However, spending for nursing facility services grew faster in FY 1991 than in other years--by about 16 percent, suggesting that the rate of nursing home spending may be accelerating again. Still, the rate of growth in FY 1991 was well below the rate of increase for both inpatient and outpatient hospital services. The aecond trend in long-term care spending reflects rapid increases in spending for community-based aervices. Spending on home and community based waivers, personal care, and home health services increased more than twice aa fast as institutional long-term care services.'' '*A State may enroll individuals eligible for Medicaid in a group health plan if it is cost-effective. 'Wnder the home and community-based services waiver process established in section 1915(c) of the Social Security Act, States are permitted to cover services that go beyond medical and medically-related benefits covered under Medicaid. These services are intended to prevent or postpone the institutionalization of persons who could otherwise use services in the community. They include a variety of nonmedical, social and supportive services (continued ...) STATE SPENDING TRENDS State Medicaid programs are highly variable in many respects. Some of the major differences in State programs include: the numbers of persons served, eligibility criteria, covered services, and reimbursement rates. Overall Medicaid spending has increased steadily in recent years, but the rate of increase varies widely across States. For instance, preliminary data from Illinois indicate that total Medicaid expenditures increased by about only 1.5 percent in FY 1991. In contrast, Missouri Medicaid spending rose 73 percent the same year. Table 4 provides State-by-State annual rates of change in Medicaid paymenta from 1987 to 1991. There are two noteworthy aspects about States' Medicaid spending. Historically, a few populous States have accounted for a large share of the total Medicaid spending. That trend continued from FY 1987 to FY 1991;spending in New York, California, Pennsylvania, and Texas accounted for more than 36 percent of all M 1991 Medicaid spending. However, the share of total program spending represented by these States shrunk between FY 1987 and FY 1991, because a number of other States experienced very rapid Medicaid spending growth. Figure 5 highlights States that doubled their Medicaid spending between 1987 and 1991. These States include: Alabama, Arizona, Florida, Kansas, Louisiana, Massachusetts, Missouri, New Hampshire, North Carolina, South Carolina, and Wyoming." Figure 6 highlights those States whose annual growth rate exceeded the U.S. average rate of growth in Medicaid spending each of the last 4 years. These States include: Alabama, Arizona, Florida, Kansas, Kentucky, Massachusetts, Mississippi, North Carolina, Tennessee, and Wyoming. - - - - - - lS(...continued) such as case management, homemakerhome health aide services, personal care, adult day health, habilitation services and respite care. "Arizona is the only State that does not operate a traditional Medicaid program. Its Arizona Health Care Cost Containment System (AHCCCS) operates as a Medicaid demonstration. It was implemented in 1982 for acute care only. During the last several years, the State has been phasing-in a longterm care program. TABLE 4. Annual Rates of Change for Total Medicaid Spending, by State,FY 1987to FY 1991 FY 1991 total State Alabama Alaska Arizona Arkansas California Colorado Connecticut Delaware District of Columbia Florida Georgia Hawaii Idaho Illinois Indiana Iowa Kaneas Kentucky Louisiana Maine Maryland Massachusetts Michigan Minnesota Miesieeippi Mieeouri Montana Nebraska Nevada N e w Hampshire New Jersey New Mexico 1987 1988 1989 1990 to 1988 to 1989 to 1990 to 1991 Medicaid spending (in millione) TABLE 4. Annual Rates of Change for Total Medicaid Spending, by State,FY 1987 to FY 1991-Continued FY 1991 total 1987 1988 1989 1990 to 1988 to 1989 to 1990 to 1991 State Medicaid spending (in millions) New York North Carolina North Dakota Ohio Oklahoma Oregon Pennsylvania Rhode Island South Carolina South Dakota Tennessee Texas Utah Vermont Virginia Washington West Virginia Wisconsin Wyoming U.S. Total 8.6% 12.2% 18.6% 26.9% $91,530 NOTE: All rates of change are based on total Medicaid spending (i.e., Federal, State and local spending) after all adjustments for overpayments, collections, etc. The U.S. total includes spending for benefits and administration in the 50 States and the District of Columbia and excludes spending for all outlying territories. Source: expenditures. Based on HCFA form 64--Quarterly report of Medicaid FY 1991 figures are preliminary and subject to change. FIGURE 6. States Where Medicaid Spending More Than Doubled Between 1987-1991 Source: Map prepared by CRS based on data from HCFA Form 64. FIGURE 6. Sates Where Annual Growth Was Greater Than U.S. Avemtge Each of the Last Four Years Source: Map prepared by CRS based on data from HCFA Form 64. CHAPTER 3. FACTORS ASSOCIATED WITH INCREASING MEDICAID SPENDING Increases in Medicaid spending can be explained by a number of factors, such as: health care price increases; increases in the number of beneficiaries; Federal or State changes in eligibility and reimbursement policy; and judicial decisions Setting reimbursement and eligibility. The remainder of this report is devoted to a discussion of these factors and what is known about their effecta on Medicaid spending. HEALTB CARE PRJCES The underlying rate of general inflation in the economy and specific price increases in health services (exceeding general inflation) exert considerable pressure on national health spending trends. For example, analyses of overall health spending trends show that general inflation and changes in the prices of health services account for more than half of the increase in health spending each year during the eighties. Beginning in the mid-1970s and continuing into the early 19808, both general inflation and changes in the prices of health services were very high and contributed to record rates of growth in national health expenditures.16 By 1982, general inflation subsided considerably, coinciding with more moderate rates of growth in health epending in the next few years.16 In the late 19806, the relatively low rate of general inflation continued, but the rate of increase in national health spending accelerated. The long term effecta of health care inflation on Medicaid were analyzed last year by the Actuarial Research Corporation. It examined Medicaid spending trends from 1980-1990and estimated that "medicalprice inflation" accounted for 59 percent of the increase in Federal Medicaid spending over this 10 year period." lbNational health expenditures include all public and private spending on health care, services and supplies related to that care, and funds spent for construction of health care facilities, as well as public and private noncommercial research spending. 16For more information about trends in national health spending, see: U.S. Library of Congress. Congressional Research Service. National Health Expenditures: Trends from 1960-1989. CRS Report for Congress No. 91-588 EPW, by Kathleen M. King and R.V. Rimkunas. Washington, 1991. 17Department of Health and Human Services--0f'fice of Management and Budget Medicaid Management Review, Team #4 Report, Part B: Independent Consultant's Report, Actuarial Research Corporation, p. vii. Because the Actuarial Research Corporation did not estimate general price inflation separately, it is assumed to be included in the estimate of medical price inflation. However, the change in health care prices typically affects Medicaid less than other payers because Medicaid reimbursement rates are frequently not tied to an index of the cost of living or eervices. Generally, Medicaid reimbursement rates are more likely to be affected by States' budgetary considerations than inflationary pressures. In many States, Medicaid reimbursement rates remain unchanged until State legislatures authorize payment updates or political pressure builds for regulatory updates to payment rates. Moreover, it is not uncommon for States to reduce reimbursement rates, cancel or delay payment updates during fiscal crises.I8 REIMBURSEMENT POLICIES States design Medicaid reimbursement eystems and set payment rates for services within broad Federal guidelines. The next section of this report discusses how recent changes in reimbursement policies have affected payments for hospital and nursing home care; physicians' services; and prescription drugs and how these payment policies translated into greater program payments. Hospitals and Nursing Facility Semces Before 1980, States were required to use Medicare reimbursement principles for hospital and nursing facility services. Under these cost-baaed principles, institutional providers were reimbursed the actual costs of providing care to Medicaid beneficiaries. In response to criticisms that cost-based reimbursement principles provided few incentives for providers to perform efficiently and were inflationary, Congress enacted the "Boren Amendmentmfor nursing facility services in the Omnibus Reconciliation Act of 1980. The Boren Amendment freed States from cost-based reimbursement requirements for nursing homes and directed only that payment rates must be "reasonable and adequate" to meet the costs of "efficiently and economically operated facilities" in providing care meeting Federal and State quality and safety standards. The law did not define reasonable and adequate payments or efficiently and economically operated facilities. In the Omnibus Budget Reconciliation Act of 1981 (OBRA 81), Congress applied the Boren Amendment to hospitals aa well. Hospital inpatient rates must be sufficient to ensure reasonable access to services of adequate quality. Nearly all States eliminated cost-based reimbursement principles and established alternative eystems designed to control costs and encourage e f f i c i e n following enactment of the Boren Amendment. In October of 1981,16 States were using some alternative to a retrospective cost-based system for hospital payments. In July 1991, only I8For a current discussion of States' proposed Medicaid reductions see: Battle of the (Medicaid) Bulge: States Gird for Sizeable Cutbacks. Intergovernmental Health Policy Project, State Health Notes, no. 122, Dec. 16, 1991. four States continued to use a retrospective cost-based system for hospital payments .Ig The OBRA 81 also required States to "take into account the situation of hospitals which serve a disproportionate number of low-income patients with special needs." Some States responded by adopting Medicaid State plan amendments to make additional "DSH" payments to these hospitals. These payments were supplementary Medicaid payments for services rendered to beneficiaries by facilities meeting State established criteria for designation as a DSH. In some cases, these payments created a potential conflict with a regulation limiting aggregate Medicaid payments to the amount allowed by Medicare. Congress responded by enacting a provision in the Consolidated Omnibus Budget Reconciliation Act (COBRA) of 1985 prohibiting the Secretary from limiting DSH. This provision is significant because DSH were the only Medicaid payments the Secretary was not allowed to limit. Some felt that States' responses to OBRA 81 were insufficient. In some States, no hospitals qualified for DSH. In other States, additional DSH were not set as high as some thought necessary. In the Omnibus Budget Reconciliation Act of 1987, Congress established minimum criteria for DSH. A hospital must receive additional DSH if: (1)its Medicaid utilization rate is more than one standard deviation above the average Medicaid utilization rate for all Medicaidparticipating hospitals in the State; or (2) its low-income utilization rate is at least 25 percent.20 However, this is a minimum criterion. States can use more liberal definitions, as long as the State's plan is approved by HCFA. Eflect of the Boren Amendment While the Boren Amendment permitted States to develop alternative reimbursement systems, it also established a standard against which to measure those ~yeterns: States must provide assurances satisfactory to the Secretary of Health and Human Services that their Medicaid rates are reasonable and adequate. Over time, Medicaid providers began to sue State Medicaid agencies, arguing that States had not met the Boren Amendment standards. In 1990, the U.S. Supreme Court confirmed providers' right to seek judicial review of States' aseurances of the adequacy of Medicaid rates or adequacy of the rates themselves under the Boren Amendment in Wilder us. Virginia Hospital Association. 'gProspective Payment Assessment Commission, Medicaid Hospital Payment, Oct. 1,1991, figure 2-1. 30The low income utilization rate is defined as the sum of two percentages: (1) Medicaid payment and State and local patient care subsidies as a percentage of the hospital's total patient revenues; and (2) inpatient charity care charges (excluding contractual allowances or discounts other than those for indigent patients ineligible for Medicaid) as a percentage of a hospital's total inpatient charges. In August 1991, the AHA surveyed State hospital associations and found that Boren Amendment law euits on hospital payment rates had been filed in 21 States.*' Similarly, the American Health Care Association (AHCA) has compiled information regarding Boren Amendment suits filed on behalf of nursing homes. Ae of summer 1991, euits had been filed in 21 States. Table 5 shows the States in which Boren Amendment suits are pending and the States in which lawsuits have been resolved. (Cases listed as "resolved"have either gone to trial or have been settled out of court.) The AHA reported that Boren Amendment lawsuits had been resolved in 10 States ae of August 1 9 9 1 . ~An AHCA summary of Boren Amendment caaes obtained in November 1991 reported that suits had been resolved in eight States.= Many resolutions have favored providers. The thrust of theee decisions is that States did not generally identify objective standards as to what constitutes an eficiently and economically operated facility or establish findings that their rate structures met these standards. In most cases, courts have held Medicaid State plans invalid and ordered States to revise their State plans to demonstrate compliance with the Boren Amendment. Typically, the revised State plans increase payment rates. 21Theseinclude suite filed by State hospital associations, groups of hospitals, or individual hospitals. %rice the time of the AHA survey, cases have been settled in Illinois and New York. In both cases (Illinois Health Care Association us. Bradley and Rye Psychiatric Hospital Center us. Surles), the Medicaid agency was found in violation of the Boren Amendment. The Ohio Supreme Court (Ohio Hospital Association us. Ohio Department of H u m Services) recently held that Medicaid had reduced ite outpatient hospital reimbursement rates solely for budgetary reasons, but did not explicitly hold the State in violation of the Boren Amendment. The State may appeal this decision on the grounds that the Boren Amendment does not apply to hospital outpatient services. q i n c e that time, a case has been settled in Massachusetts (Massachusetts Fedemtion ofNursing Homes, Inc. us. Commonwealth of Massachusetts). In that case, the court found that the State did not violate the requirement of providing assurances that the rates satisfy the Boren Amendment. TABLE 5. States in Which Boren Amendment Suits Have Been Filed by Hospitals and Nursing Homes Hospitals' State Alabama Alaska Arizona Arkansas California Colorado Connecticut Delaware District of Columbia Florida Georgia Hawaii Idaho Illinois Indiana Iowa Kansas Kentucky Louieiana Maine Maryland Massachusetts Michigan Minnesota Missiesippi Missouri Montana Nebraska Nevada New Hampshire New Jereey New Mexico New York North Carolina North Dakota Ohio Oklahoma Oregon Pennsylvania Pending Resolved' Nursing homesb Pending Resolved' X (2 cases) X (2 cases) Xg TABLE 5. States in Which Boren Amendment Suits Have Been Filed by Hospitals and Nursing Homes--Continued Hospitals' State Rhode Island Sout). Carolina Souti; Dakota Tennessee Texas Pending Nursing homesb Resolvede Pending Resolvedc Xd Utah Vermont Virginia Washington West Virginia Wisconsin ukoming X Xd X@ X 'Medicaid lawsuit summary as of Aug. 27,1991,obtained by CRS from the AHA. "Summary of Boren Amendment cases known to the AHCA obtained by CRS in Nov. 1991. Ceeea listed as "resolved" have either gone to trial or have been eettled out of court. d S ~ i tfiled s by State hospital associations. S u i t s filed by individual hospitals. 'Suits filed by groups of hospitals. C a s e s that have gone to trial. bHospitals in North Dakota sued South Dakota Medicaid Department. Source: Table prepared by CRS based on information from the AHA and the AHCA. Systematic information about the fiscal effects of Boren Amendment lawsuits is not readily available. Therefore, it is not possible to quantify the current or future fiscal effects of Boren Amendment challenges. However, the greatest impact of these lawsuits will probably occur in future years. Some States have issued preliminary fiscal estimates of costs they will incur to meet Boren Amendment standards, but these estimates are not all for the same fiscal year.% Most are also subject to revision. Nevertheless, they shed some light on the magnitude of Boren Amendment suits on Medicaid expenditures. In Washington State, Medicaid recently settled a suit brought by hospitals (MultiCare Medical Center, et.al. us. State of Washington, et.al.)after a judge ruled that the State's assurances that "its rates were substantively adequate had no factual basis." The State estimates that it will spend $62 million over the next 2 years ($28 million in State funds and $34 million in Federal funds) to increase average Medicaid hospital reimbursement rates by 10 percent. Oregon (Oregon Association of Hospitals us. State of Oregon) recently settled a similar hospital suit. I t agreed to pay hospitals an additional $65 million between July 1, 1991 and June 30,1993. In Tioga Pines Living Center, Inc. vs. Indiana State Board of Public Welfare, an Indiana circuit court judge found that the State's Medicaid nursing home rates had been unreasonable and inadequate since mid-1987. The State, which has appealed, estimates that it could pay between $100 million and $150 million to comply with the court's order, which would have constituted between 21 and 27 percent of the State's nursing home spending in F Y 1990. In Virginia, (Wilder us. Virginia Hospital Association), the State estimates payments of an additional $120 million from State FY 1993-FY 1996. Michigan, (Michigan Hospital Association vs. Babcock) estimates costs of $70 million in FY 1991. Its rough estimate of FY 1992 costs is $30 million, with costa in succeeding years as yet undetermined. Also in Michigan, the costs of a nursing home suit (Health Care Association of Michigan us. Babcock) are estimated a t $20-$25 million in FY 1991. The specter of a Boren Amendment lawsuit, especially in the wake of recent decisions, may affect States' policy decisions. They may raise reimbursement rates to avoid a perceived Boren Amendment challenge or may settle lawsuits to avoid judicial review, which could result in even higher costs to States. Effkct of Disproportionate Share Payments (DSH) The National Association of Public Hospitals (NAPH) surveyed States to determine how they implemented the disproportionate share requirements of OBRA 87.% Between February 1989 and the summer of 1990, 31 of 47 States %All Boren Amendment fiscal estimates reported include both Federal and State funds. %National Association of Public Hospitals. Revised State Medicaid Policies for Dispmportionute Share Hospitals: An Updated Status Report. Washington, D.C.,1990. that reported information had an increased number of hospitals qualifying for DSH. In 10 States, the number of hospitals qualifying for DSH increased by more than 50 percent.26 Forty-one States reported DSH totalling $569 million in FY 1989. NAPH reported projected spending of $831 million in FY 1990 and $1.1billion in FY 1991.27 Based on information provided by States, HCFA recently projected DSH in FY 1992 a t $14.3 b i l l i ~ n . ~ Table 6 provides State projections of disproportionate share spending in FY 1992. a6Theae 10 States are: Alabama, Florida, Georgia, Iowa, Kentucky, Massachuaetts, Missouri, Ohio, South Carolina and Tennessee. nBecauee these projections do not include data from several States, including Michigan (for FY 1991), New York and West Virginia, they underestimate total disproportionate share payments (DSH). Bunpublished data obtained by personal communication in Dec. 1991 from HCFA'e Medicaid Bureau. TABLE 6. Eetimated Disproportionate Share Payments, by State,FY 1992 State Alabama Alaeka Arizona Arkanem California Colorado Connecticut Delaware District of Columbia Florida Georgia Hawaii Idaho Illinois Indiana Iowa Kansas Kentucky Louisiana Maine Maryland Massachusetts Michigan Minnesota Miesissippi Miseouri Montana Nebraska Nevada New Hampshire New Jersey New Mexico New York North Carolina North Dakota Disproportionate share payments Total medical assistance payments Disproportionate share payments as percent of total TABLE 6. Estimated Disproportionate Share Payments, by State,F'Y 1992--Continued State Ohio Oklahoma Oregon Penneylvania Rhode Island South Carolina South Dakota Tennessee Texas Utah Vermont Virginia Washington West Virginia Wisconsin Wyoming Total Total medical assistance payments Disproportionate share paymente as percent of total 112,887 17,870 11,000 784,200 36,500 4,337,262 952,873 776,300 6,865,586 577,328 2.60% 1.88 1.42 13.37 6.32 448,220 50 444,025 1,390,900 4,004 1,485,006 243,195 2,413,143 6,087,562 390,691 30.18 0.02 18.40 22.85 1.02 9,613 30,000 186,900 34,234 7,273 140 242,948 1,502,429 1,877,288 736,327 1,936,596 117,243 3.96 2.00 9.96 4.65 0.38 0.12 $14,393,676 $114,351,942 Disproportionate ehare paymente 12.59% NOTE: All figures are estimates of FY 1992 payments based on information supplied by State Medicaid programs in Nov. of 1991. These estimates are subject to the limits of the data and methods used in their calculation. Source: Unpublished HCFA estimates. Quality Assurance in Nursing Home8 The Omnibus Budget Reconciliation Act of 1987 (OBRA 87) included comprehensive provisions designed to improve the quality of care in nursing home^.^ Most provisions were implemented by regulation on October 1, 1990; a number of them were expected to substantially increase Medicaid nursing home expenditures. Among the requirements imposed on nursing homes are the following: (1)they must complete a comprehensive assessment of residents' phyeical and mental abilities shortly after their admission to the facility and update these aseessmenta regularly; (2) they must meet Medicare's higher level of licensed nurse staff~ng;and (3) they must ensure that nurse aides complete a training program or competency evaluation program.s0 The OBRA 87 required States to increase Medicaid nursing home rates to reflect costa incurred by nursing facilities in meeting these requirements. As of October 30,1990, HCFA had approved increases in Medicaid nursing home rates ranging from a low of $0.10 per patient day in New York to a high of $4.64 in New H a m p ~ h i r e .The ~ ~ median rate increase approved was $1.16 per patient day. No further information is available concerning the aggregate effects of the OBRA 87 requirements on nursing home expenditures. New Funding Soume8 for Ho8pitaZ and Nurdng Home Spending: Provider Donations and Provider-Specific Taxes In recent years, some States have sought alternative financing mechanisms to help them cope with rapidly rising Medicaid expenditures. Financing burgeoning Medicaid spending is especially problematic for States because, unlike the Federal Government, every State but one is required by law to balance its budget every year. States have sought to alleviate this increased fiscal stress, without altering the types of covered services, number of beneficiaries or reallocating State general funds. Some States have turned to funds donated by health care providers (called voluntary contributions) or taxes %ome provisions of Omnibus Budget Reconciliation Act of 1987 (OBRA 87) were amended by subsequent legislation. For a complete description of all the law's provisions see: U S . Library of Congress. Congressional Research Service. Medicam and Medicaid Nursing Home Reform Provisions in the Omnibus Budget Reconciliation Act of 1987. CRS Report for Congress No. 90-80 EPW, by Richard J. Price (revised). Washington, 1990. %oat nursing homes either provide the required training on site or subsidize its cost. "Unpublished data obtained from the American Association of Homes for the Aging. providers (called voluntary contributions) or taxes impoeed on providers (referred to as provider-specific taxes) to finance increased Medicaid spending.s2 Prior to 1985, Federal regulations did not permit States to uee donated funds except for training State personnel to administer Medicaid. In 1985, however, a new regulation permitted States to uee donated fun& for Medicaid under the following conditions: (1) the funds had to be transferred to the Medicaid agency and be under its adminietrative control; and (2) the fun& could not revert to the donor unlees the donor wae a nonprofit organization and the Medicaid agency decided independently to use the donor'e f a ~ i l i t y . ~ Following issuance of this rule, some States received voluntary contributions from Medicaid providers, chiefly hospitals and nursing homes. These contributione are ueed ae part of the State's share of spending for covered eervices and matched with Federal funds. Around the time States began using voluntary contributions to help finance their Medicaid programs, eome States imposed provider-specific taxes on health care providers. These revenue sources have become controversial. The Administration contends that there is a connection between these revenues and increasing reimbursement, A eimple example will illustrate this point. Assume that a State has a Federal matching rate of 60 percent. In this example, the State receives a donation of $40 from a provider. Medicaid pays the provider $100 for services rendered to Medicaid patients. The Federal ehare is $60 and the State ehare is $40. Net State spending (i.e., State spending after the donation is taken into account) is $0. The net payment to the provider-the provider's Medicaid payment aRer the donation is eubtracted-is $60. Beginning in 1987, HCFA made repeated efforts to disallow Federal matchingpayments for both voluntary contributions and provider-specifictaxes. In HCFA's view, States were unfairly increasing Medicaid expenditures without meeting the etatutory requirement that States actually pay a t least 60 percent of the State ehare of Medicaid funds from State funds. In turn, States maintained that HCFA bad no legal right to scrutinize the eource of State matching funda. In their estimation, all funda and revenues obtained by the States belong to them, regardless of the ultimate source from which they were derived. Furthermore, the States maintain that euch funds are essential to coping with rapidly rieing Medicaid expenditures. The reason why voluntary contributions and provider-specific taxes have been so controversial etems from the Administration's concern that these revenue sources are contributing disproportionately to rising Federal Medicaid T o r more information about this iesue, eee: U.S. Library of Congress. Congreesional Reeearch Service. Medicaid Provider Dorurtions and ProviderSpecific T w s . CRS Report for Congress No. 91-722 EPW,by Merlis, Mark. Washington, 1991. %0 Fedend Register 46662, Nov. 12,1985. expenditures. In the Mid-Session Review of the Budget, the Office of Management and Budget (OMB) reported that voluntary contributions and provider tax initiatives accounted for only a small portion of the increase in Medicaid expenditures from 1980 to 1990. However, OMB stated that they appeared to constitute a substantial portion of spending increases for FY 1991 and 1992 in the States using such programs and that recent evidence suggested dramatically increased reliance on themaU The HCFA has estimated the FY 1992 fiscal impact a t $11.4 billion in total outlays, of which $6.9 billion are Federal funds and $4.5 billion are from voluntary contributions and provider-specific taxes.= Table 7 provides estimates of the size of these donations and taxes. However, two observations should be made. First, it is impossible to predict whether States would have incurred the same Medicaid expenditures in the absence of these revenues, or whether these revenues made additional expenditures possible. Second, the impact of these revenue sources was projected to grow substantially over time. A number of States began using these funding mechanisms in FY 1992. From 1988 to 1990, Congress enacted a series of measures prohibiting HCFA from implementing regulations banning the use of funds from voluntary contributions and provider-specific taxes except in certain cases. The last moratorium, enacted in the Omnibus Budget Reconciliation Act of 1990 (OBRA 901, was scheduled to expire on December 31, 1991. On November 27, 1991, Congress passed the Medicaid Voluntary Contribution and Provider-Specific Tax Arnendmenta of 1991, which the President signed into law on December 12, 1991.96 Effective January 1, 1992, States are prohibited from using most voluntary contributions to claim Federal matching funds. States with voluntary contribution programs in effect or reported for States' FY 1992 are permitted to continue them. Effective January 1,1992, the law prohibits use of Federal funds to match revenues derived from provider-specific taxes unless these taxes are broad-based and apply uniformly to all providers of a given type and all business of the providers within a class of services. Examples of a qualifylng tax include a tax based on all inpatient days or a head tax on all patients. States with nonbroadbased taxes in effect or approved as of November 22, 1991 are permitted to continue them temporarily, but the taxes may not be increased. The law also applies a general limit on using revenues derived from contributions and taxes to obtain Federal matching funds. It limits the use of voluntary contributions and revenue from both provider-specific taxes and broad-based taxes during FY - - - aU.S. Dept. of' Health and Human Services. Office of Management and Budget. Mid-Session Review of the Budget, Improving Medicaid Estimates: Report of the HHS-OMBTask Force. July 15, 1991. Washington, 1991. p. 17. aHCFA, Medicaid Bureau, unpublished data, Dec. 1991 provided to Congressional Research Service (CRS) through personal communication. =H.R. 3595, P.L. 102-234. 1993-FY 1995 to the greater of 25 percent of the State share of Medicaid funding or the amount of donations and taxes collected in State FY 1992.37 Congress also included a provision ultimately limiting DSH in the law on donations and taxes because the Administration contended that States could conceivably use these revenue sources to increase DSH. Beginning in Federal FY 1993, a national limit is imposed on DSH. States' aggregate DSH can not exceed 12 percent of national (Federal, State and local) Medicaid expenditures. States whose DSH already exceed the 12 percent limit are permitted to continue them and are allowed to increase such paymenta as long as they do not exceed the same percentage of the State's Medicaid expenditures as they accounted for in F Y 1992. States below the 12 percent cap are permitted to increase DSH under a formula that ensures total DSH will not exceed the national 12 percent limit. According to HCFA'e calculations, 17 States will be subject to the 12 percent cap in FY 1993. Although passage of the recent law will not reduce FY 1992 Medicaid expenditures, it will constrain States' attempts to expand use of these funding sourcee in future years. ''See disproportionate share hospitals, below, for a discussion of the provisions of the Medicaid Voluntary Contributions and Provider-Specific Tax Amendments of 1991 concerning payments to disproportionate share hospitals. TABLE 7. Provider Taxes and Donations as a Share of Total State Medicaid Expenditures, FY 1992 State Provider taxes Alabama Alaska Arizona Arkansas California $172,251 0 0 27,255 0 Colorado Connecticut Delaware District of Columbia Florida 0 0 188,200 Georgia Hawaii Idaho Illinois Indiana 0 0 0 634,080 71,195 Iowa Kansas Kentucky Louisiana Maine 0 5,750 143,344 0 30,882 Maryland Massachusetts Michigan Minnesota Mississippi 41,663 200,000 0 13,352 21,500 Missouri Montana Nebraska Nevada New Hampshire 0 1,243 0 44,166 227,695 New Jersey New Mexico New York North Carolina North Dakota 0 0 276,200 0 0 14,512 - Taxes & donations Provider Taxes & as a share of State donations donations Medicaid spending TABLE 7. Provider Taxes and Donations as a Share of Total State Medicaid Expenditures, FY 1992--Continued State Provider taxes Taxes & donatione Provider Taxes & as a share of State donations donation8 Medicaid spending Ohio Oklahoma Oregon Pennsylvania Rhode Ialand South Carolina South Dakota Tenneseee Texas Utah Vermont Virginia Washington West Virginia Wisconsin Wyoming Total $3,002,915 $1,516,150 $4,519,065 9.28 NOTE: All figures are estimates and subject to limitations of data and methods employed in their calculationa. Estimates are dependent on estimates of State spending levels in FY 1992. States' estimates are likely to change as the fiscal year continues. Total amount excludes any spending for Puerto Rico and other outlying territories. Information for Connecticut is not available. Source: Based on unpublished HCFA data. Physicians' Services Medicaid payment rates for physicians are subject to the general requirement that payments be sufXcient to attract enough providers to ensure that covered services will be as available to Medicaid beneficiaries as they are to the general population. This requirement, previously established by regulation, was enacted into law in the Omnibus Budget Reconciliation Act of 1989 (OBRA 89). Payments for physicians' services are not subject to the Boren Amendment. Medicaid payments to physicians are typically the lesser of the provider's actual charge or a maximum allowable charge established by the State. Maximum allowable charges are determined either by historical reasonable charges or a fee schedule. Most States now use fee schedules; in 1990,42 States paid physicians on the basis of fee s ~ h e d u l e s . ~ Payments for physicians' services vary widely among the States. In general, Medicaid programs pay physicians considerably less than either Medicare or private insurers. In 1989, Medicaid maximum payments averaged 64 percent of the maximum allowed by M e d i ~ a r e .No ~ ~ comprehensive analyses of the ratio of Medicaid fees to private insurance fees have been conducted to date, but the Physician Payment Review Commission (PPRC) has reported that Medicaid fees for five common services averaged 55 percent of private insurance fees in 1989.'O Many Medicaid programs do not update physicians' fees every year. The most current information on fee increases indicates that 35 States increased fees between 1987 and 1989." However, only 21 States adopted across-the-board increases, while most of the remaining States implemented targeted increases for maternity and pediatric services. These increases were designed to address concerns that low payments contributed to unacceptable infant mortality rates and inadequate access to primary care for children. Congress also responded to these concerns in OBRA 89 by establishing specific reporting requirements for obstetric and pediatric services so the Secretary can determine the adequacy of State payments for these services. -- - --- -- -- - -- - - - =Physician Payment Review Commission, Annual Report to Congress, Washington, D.C., Apr., 1991. (Hereafter cited as Physician Payment Review Commission, Annual Report to Congress, 1991) 3@Ph~ician Payment Review Commission. Physician Payment Under Medicaid. No. 91-4, July 1991. Washington, D.C.,1991. 40These five services are: total obstetric care (vaginal delivery); vaginal delivery; cesarean delivery; tonsillectomy and adenoidectomy (under 12 years); and repair inguinal hernia (under 5 years). Physician Payment Review Commission, Annual Report to Congress. The HCFA instructed States that compliance with the OBRA 89 requirement could be demonstrated by meeting one of three standards: (1)a t least 50 percent of obstetric and pediatric practitioners are full Medicaid participants or Medicaid participation is a t the same rate ae Blue Shield participation; (2) Medicaid payment rates equal a t least 90 percent of the average payment by private insurers; or (3) other d o c ~ m e n t a t i o n . ~ Ae of November 1991, HCFA had approved 34 Medicaid State plan amendments and rejected 17. Most States have demonstrated compliance by certifyng that a t least 50 percent of obstetrical practitioners and pediatric practitioners in the State are full Medicaid participants, rather than raising fees to cloee the gap between Medicaid and private sector paymente?s However, information gathered by PPRC shows that some States substantially increased fees for medical visite and obstetric services after 1989." Explaining why Medicaid physician expenditures are growing is not a simple task. Unlike hospitals and nursing facility services, physician payments are not affected by the same pressures influencing hospitals and nursing homes discussed above. In addition, some of the data needed to analyze growth in Medicaid physician expenditures are not available. Although more definitive explanations of the reasons fueling growth in spending for physicians' services must await better data, some observations about physician expenditures can be made. Increases in the number of beneficiaries may contribute to rising physician expenditures. In FY 1990, the latest year for which data about beneficiaries are available, the number of beneficiaries increased by 5.7 percent. The number of beneficiaries receiving physician services increased by 8.9 percent. Medicaid payments for physicians' services per beneficiary using such services also increased almost 30 percent from $181 in FY 1987 to $235 in FY 1990.46 Some part of the increase is probably due to increased fee levels, but it can not be quantified. While most research on Medicaid physicians' fees has focused on whether fees are too low to ensure access to care, it is also possible that the volume and intensity of physician services rendered to Medicaid beneficiaries are increasing. Some physicians do not treat Medicaid patients, but those who do may increase the number of services they render, provide higher '3Fu11 Medicaid participation means that providers are enrolled in Medicaid. According to the PPRC, few States have attempted to determine whether physicians enrolled in Medicaid accept all Medicaid patients or what percentage of their patients are on Medicaid. "According to the PPRC survey, Medicaid fees for vaginal deliveries equalled those paid by private insurers in 1989 in only two States--New Hampshire and South Carolina. "Unpublished PPRC data obtained through personal -ommunication. 46Congressional Research Service (CRS)analysis of HCFA Form 2082. levels of service than they did previously, or shift the site of service from oflice settings to outpatient hospital departments, where payments to physicians may be higher. Limited data are available concerning the number of times Medicaid beneficiaries see a physician in a given year. In 1986, a sample of Medicaid beneficiaries reported seeing a physician 8.3 times that year?6 By 1989, the number of reported physician encounters increased very slightly to 8.5. This increase is not etatietically significant, lruggesting that increasing numbers of physician encounters is not a contributor to rising physician expenditures. Regarding changee in the eite of eervice, research has documented that Medicaid beneficiaries are more likely to receive care in nonoffice based settings in States with low payment levels for office services. However, wider scale studies of Medicaid physician practice patterns are needed for a more complete assessment of volume and intensity of Medicaid physician services. Prescription Drugs Medicaid payment for prescription drugs furnished on an outpatient basis has two components: an amount for drug ingredients and a dispensing fee to the pharmacies for filling prescriptions. Medicaid regulations establish limits on payment for acquisition costs, but do not limit dispensing fees, which must only be "reasonable." Two separate limits apply for drug ingredients-one for multiple source drugs and one for all other drugs. These limits are intended to encourage the use of lower cost generic drugs. From FY 1987 to FY 1990 Medicaid payments for prescription drugs increaeed by 48 percent. Concern over this rate of increase, Congress added another dimension to Medicaid drug reimbursement-prescription drug rebates. The OBRA 90 required drug manufacturers to pay rebates to State Medicaid programs for drugs dispensed and paid for on or after January 1,1991. In return, Medicaid programs were required to cover all drugs marketed by that manufacturer, with certain exceptions. Rebate requirements may also apply to certain nonprescription items such as aspirin if they are covered in a State's Medicaid plan. Rebates do not apply to products dispensed aa part of a service provided in a hospital, a physician's or dentist's office. These requirements apply differently to multiple source drugs and other drugs. It is too soon to assess the effects of the drug rebate law on prescription drug expenditures, but some preliminary observations can be made. From January 1, 1991, when the law took effect, to the end of FY 1991, Medicaid collected $95.3 million in rebates.'? After the rebate amount is subtracted fiom FY 1991 drug expenditures, Medicaid drug spending still rose 20.5 percent 46Unpublished data from the Health Interview Survey obtained through personal communication with the National Center for Health Statistics. "These data may not include complete information on rebates from all States. between FY 1990 and FY 1991. This rate of increase is higher than any year from FY 1987-FY 1990, indicating that savings from rebates may have been more than offset by required coverage for new drugs, price increases and increased drug use. Enrollment Increaseu: Growing Numbers of Cash Welfare Rsdpiente and Medicaid Program Erpansions Increases in Medicaid spending are also affected by increases in the numbers of program beneficiaries. Recently, the number of beneficiaries has risen because of increasing enrollment in caeh welfare program linked to Medicaid. Generally, people receiving AFDC are automatically eligible for Medicaid. In addition, individuals receiving SSI in most States are also eligible for Medicaid. Aid to Familiee With Dependent Children (AFDC) Enrollment The AFDC is a Federal-State program that provides cash assistance payments to needy children in families with only one able-bodied parent and to other members in the households of such children." States determine "need" for these benefits and establish their own income and resource eligibility standards within Federal limitations. As a result of broad Federal guidelines, States' income eligibility thresholds and AFDC payments vary widely. In January 1991, monthly AFDC payments for a family of three ranged from a low of $120 in Missiesippi to a high of $891 in Alaska, with median State payments a t $357.'' From October 1989 to October 1991, AFDC caseloads rose by 18.8 p e r ~ e n t . This ~ is one of the most rapid increases in the program's recent history. While this increase predates the economic downturn of the early nineties, the recession may exacerbate the increase in the program's ~aseload.~' But AFDC caseload is not cyclical. Periods of economic growth have not resulted in eignificant decreases in caseload. Clearly, other factors besides the - - '%ti1 Oct. 1, 1990, States were not required to provide benefits to twoparent families who are needy because of unemployment of one of the parents. See discueeion of Aid to Families with Dependent Children-Unemployed Parents (AFDC-UP) program below. 'Qata collected by the CRS through a telephone survey of the States. US. Congress. House. Committee on Ways and Meane. 1991 Gmen Book: Overview of Entitlement Pmgmms. May 7 , 1991. Washington, GPO, 1991. p. 600-601. bORates of growth in this paragraph are for the basic AFDC program. AFDC- UP program which provides cash welfare payments for families with an unemployed parent are discussed below. 'lU.S. Congressional Budget Office. A Preliminary Analysis of Gmwing b e l o a d s in AFDC. Staff Memorandum. Dec. 1991. recession are contributing to increasing AFDC caseloads. Among others, these factors include an increased number of female headed households, expanded Medicaid outreach efforts, and other AFDC policy changes. The CBO projects continuing increases in AFDC enrollment are likely through FY 1997. Although not all people who are eligible for Medicaid, particularly through AFDC eligibility, use Medicaid services, increases in AFDC enrollment will continue to exert preeeure on Medicaid epending. Prior to 1988, States had the option of providing AFDC coverage to families who are needy because the principal wage earner waa unemployed. The Family Support Act of 1988 made this coverage mandatory for all Statee, effective October 1, 1990. This program is known as AFDC-UP. In order to be eligible for AFDC-UP, the principal wage earner must have had a recent attachment to the labor force or have been recently eligible for unemployment compensation. Prior to the Family Support Act mandate, 31 jurisdictions provided optional AFDC-UP benefits. Families receiving AFDC-UP cash benefits are automatically eligible for Medicaid for the duration of their AFDC-UP eligibility. The typical AFDC family adds three members (one adult and two children) to Medicaid, while AFDC-UP families usually add four Medicaid enrollees (because they are two-parent households). While enrollment in AFDC-UP is much lower than AFDC, it rose a t a much faster rate recently because AFDC-UP enrollment is more closely tied to economic downturns. In FY 1991, AFDC-UP enrollment was 31 percent higher than in F Y 1990. Table 8 provides historical trends in AFDC caseload from 1987 through 1991 and projections through 1997. TABLE 8. Historical and Projected AFDC Caseload, Average Number of Families, FY 1987-FY1987 Average monthly number of families: W1 fimrres are estimates in the thousands) Fiscal year Basic AFDC AFDG UP Total TABLE 8. Historical and Projected AFDC Caseload, Average Number of Families, FY 1987-F'Y 1997--Continued Average monthly number of families: (All figures are estimates in the thousands) Fiscal Ye= Basic AFDC AFDC- UP Total NOTE: The AFDC-UP caseload estimates include estimates of mandated UP programs. All projected numbers are subject to limitations of the data and methods employed in their calculations. Projections are subject to change. Source: Unpublished staff estimates of the CBO. Prepared Jan. 1992. Suppkmental Security Income (SSZ) EnroLLment The SSI program is a means-tested federally funded and administered program that provides monthly cash payments to eligible needy aged, blind and disabled persons. Unlike AFDC, SSI has uniform, national eligibility guidelines and uniform payment levels. In 1992, the maximum monthly SSI payment is $422 for an individual and $633 for a couple.b2 States are generally required to offer Medicaid eligibility to those receiving SSI benefits. However, States may use more restrictive eligibility standards for Medicaid than for SSI if they were using those standards in 1972, prior to the implementation of SSI. States that have chosen to apply more restrictive eligibility standards are known as "section 209(b)" States. In 1992, 12 209b) States use standards related to income, allowable resources, definition of disability or inclusion of children under age 18 that may be more restrictive than SSI's." - - p p p - %oat States supplement Federal Supplemental Security Income (SSI) benefits. The result is a combined Federal SSVState supplemented benefit against which countable income is compared in determining eligibility and benefit amounts. Thus, both eligibility and payment amounts vary by State. v h e s e States are: Connecticut, Hawaii, Illinois, Indiana, Minnesota, Missouri, New Hampshire, North Carolina, North Dakota, Ohio, Oklahoma, and Virginia. TABLE 9. Supplemental Security Income Recipients, 1987-1995 (All figures are estimates in the thousands) Year Aged Disabled Total NOTE: All projected numbers are subject to limitations of the data and methods employed in their calculations. Projections are subject to change. Source: Unpublished staff estimates of the CBO,Jan. 1992. After many years of very slow growth, the number of SSI enrollees began a more rapid increase a t the end of the 1980s. From FY 1989 through FY 1991, the number of enrollees grew 9.5 percent; enrollment is expected to grow at about 4.5 percent a year through FY 1997. There are a number of reasons for the recent rapid growth in the program and the expected rates of growth in the future. Some of the growth in the number of SSI beneficiaries can probably be attributed to disabled individuals who lose their labor force attachment during economic downturns. However, the short-term effect of the recession on SSI enrollment is not the major reason for the enrollment increases. At least for disabled applicants, there has been both an increase in the number of people applying for SSI and the number of approved applications. These increases have been observed for both disabled adults and children. Enrollment increases for disabled children are contemporaneous and subsequent to a recent Supreme Court decision that requires retroactive determinations of eligibility back to 1980 for disabled children." There has been a significant increase in the number of SSI "On Feb. 20,1990 the U S . Supreme Court (Zebley vs. Sullivan) affirmed the Court of Appeals (Third Circuit) decision which required the Social Security Administration to reevaluate childhood claims for disability because the child's functional status was not considered in determining the severity of impairment. New regulations on childhood disability have been issued in response to this decision. Wedem1 Register, Dec. 12, 1990 and Feb. 11,1991.) childhood disability decisions and an increase in the number of children found eligible for SSI by virtue of disability. An additional 125,000 children are expected to become SSI eligible as a result of this decision. Although the projected 4.5 percent rate of growth in SSI enrollment through 1997 is fairly moderate, it could have a significant impact on Medicaid expenditures because SSI beneficiaries are generally more expensive to serve than AFDC enrollees. In FY 1990, per capita expenditures were $4,478 for SSI enrollees. By comparison, FY 1990 per capita expenditures for AFDC a d d & were $1,880 and $736 for AFDC childreneU Eligibility Expanawns Since the mid-19806, Congress has enacted a number of Medicaid eligibility expansions. Most have been geared toward increasing Medicaid access for pregnant women and children, but there have been a number of other significant eligibility expansions. These include: requiring States to pay Medicare premiums, coinsurance and deductibles for low-income beneficiaries; requiring provision of cash assistance and Medicaid eligibility to two-parent families where the principal wage earner is unemployed; requiring coverage for special needs children regardless of the income and resources of foster or adoptive parents; requiring provision of emergency services, pregnancy-related services, and eervices to children under age 18 for aliens who are otherwise eligible. The next section of this report discusses available data concerning the largest Medicaid eligibility expansions. Pregnant Women and Children. Since the mid-1980s, Congress has enacted a series of measures increasing Medicaid access for poor pregnant women and children. It has also permitted streamlined eligibility determinations for pregnant women so they gain access to prenatal care as early in their pregnancies as possible. The most significant feature of these expansions was that they broke the historical link between Medicaid eligibility and cash welfare payments and instead tied Medicaid eligibility to Federal poverty standards for pregnant women and children. States are currently required to provide Medicaid to pregnant women and children under age 6 below 133 percent of the poverty standard. They are also required to phase-in Medicaid coverage to children under age 19, born after September 31,1983, below 100 percent of the poverty standard during the 1990s. Reliable data are not available to determine how many pregnant women and children have enrolled in Medicaid as a result of these continuing eligibility expansions. The paucity of data results both from the incremental nature of the expansions and the fact that HCFA did not adjust data collection 6y8te.m to capture the new eligibility categories until FY 1989. States reported that 349,566 children and 324,559 pregnant women enrolled in FY 1990 as a result -- - "Congressional Research Service calculations from HCFA Form 2082 data for FY 1990. Data on expenditures for SSI enrollees are weighted to reflect the proportion of aged, blind and disabled enrollees. of the legislative expansions that took effect after January 1, 1988.& These numbers include both pregnant women and children under 100 percent of the poverty threshold (the standard that States were required to use in FY 1990) as well as those enrolled in States offering optional eligibility to those whose incomes exceeded 100 percent of poverty. A number of States have taken advantage of optional e x p a n ~ i o n s . ~For ~ example, as of January 1990, States were required to cover pregnant women and children up to 75 percent of the Federal poverty standard. Forty six States exceeded that level; the national average was 127 percent of the poverty standard. In addition, 41 States exceeded the minimum age for coverage of children (which was 1year then), and the average State covered them up to 3.7 years. Most States also took advantage of the flexibility granted them in Federal law to streamline Medicaid eligibility applications for pregnant women. It is difficult to assess the extent to which pregnant women and children have enrolled in Medicaid as a result of the expansions and how accurately HCFA enrollment data capture the numbers of new Medicaid beneficiaries. In 1988, the National Governors' Association and the Alan Guttmacher Institute projected the number of pregnant women and children who would be Medicaid eligible a t different poverty thresholds." Although both organizations noted limitations in their estimating methods, their projections are useful in gauging how enrollment data compare with projections. At 100 percent of the poverty standard (the standard used in FY 19901, the National Governors' Association estimated that 327,737 pregnant women and an equal number of children (assuming a live birth for each pregnancy) who did not already have Medicaid would be eligible. The Alan Guttmacher Institute estimated that 361,000 additional pregnant women not already covered by Medicaid would be eligible a t 100 percent of poverty. The General Accounting Ofice (GAO) conducted a survey between 1989 and 1990 to determine how many of newly eligible pregnant women enrolled in 9 o u r c e : HCFA Form 2082. "This discussion is derived from General Accounting Office's (GAO) report: U.S. General Accounting Office. Medicaid Expansions: Coverage Improves but State Fiscal Problems Jeopardize Continued Progmss. GAO/HRD 91-78, June 1991. Washington, 1991. (Hereafter cited as GAO, Medicaid Expansions, 1991) See this report for a complete discussion of State actions regarding optional Medicaid expansions. uNewacheck, Paul W. Estimating Medicaid-Eligible Pregnant Women and Childmn Living Below 185 Percent of Poverty. Washington, National Governors' Association, 1988; and Torres, A d a , and A S . Kenney. Expanding Medicaid Coverage for Pregnant Women: Estimates of the Impact and Cost. Family Planning Perspectives, v. 21, no. 1, Jan./Feb. 1989. Medi~aid.~'In the 10 States-surveyed, GAO found that between two-thirds and three quarters of potentially eligible women enrolled in Medicaid within 2 years of the expansions. However, there was a great deal of variability in enrollment data across States. Some States enrolled only a little more than a third of the eligible population, while others enrolled nearly all the women estimated to be eligible. Enrollment growth was greatest in States that took advantage of streamlined eligibility processes and did not require beneficiaries to meet asset testa. Pregnant women and children enrolling in Medicaid in FY 1990 as a result of the eligibility expansions accounted for about 50 percent of the 1.4 million increase in beneficiaries that year.w Judging how they contributed to rising expenditures is a more difficult proposition. GAO examined this issue and reported that it was unable to isolate the separate influence of individual mandates on Medicaid outlays from FY 1984-FY 1989.6' However, GAO also noted that the States did not perceive expansions targeting women and children as the primary factor in rising Medicaid expenditures. GAO further reported that States perceived Federal mandates implemented after 1989 (including the qualified Medicare beneficiaries (QM3s) mandate discussed below) to be more costly than earlier expansions. Moreover, eome States were able to replace existing State and local spending with Federal Medicaid funds, thus generating a savings in State spending. For example, another GAO study found that one third of the States were able to substitute Federal for State funds for maternal and child health Medicaid expansions because the States were already providing these servi~es.~' Qualified Medicare Beneficiaries (QMBs). Prior to the enactment of the Medicare Catastrophic Coverage Act (MCCA) of 1988, States had the option of offering Medicaid coverage to aged and disabled persons with family income up to 100 percent of the Federal poverty level: States choosing this option could provide basic Medicaid coverage or could cover only Medicare premiums, coinsurance and deductibles. The MCCA required States to pay Medicare premiums (both Parts A and B),coinsurance and deductibles for Medicare beneficiaries whose incomes are 69U.S. General Accounting Office. Pmrrcrtal Care: Early Success in Enrolling Women Made Eligible by Medicaid Expansions. GAOPEMD 91-10. Washington, D.C.,1991. comparison assumes that all the 674,125 enrollees reported on the HCFA 2082 for FY 1990 were new enrollees. While this assumption is plausible for pregnant women, some of the children may have been Medicaid enrollees in previous years. 61GA0,Medicaid Expansions, 1991, p. 26 and 31. %AO, Medicaid Expansions, 1991. below 100 percent of the Federal poverty standard and whose resources are at or below twice the resource standard used for the SSI p r ~ g r a m .Beneficiaries ~ eligible for these expanded benefits are known as QMBs. This expanded coverage was to be phased-in between 1989 and 1992. Medicare beneficiaries with incomes below 85 percent of the poverty level were covered in 1989, with coverage of beneficiaries below 100 percent of poverty slated for implementation in 1992. The OBRA 90 accelerated phased-in coverage for those under 100 percent of the poverty etandard. It required coverage of those under the poverty standard by 1991, instead of 1992. It also provided limited Medicaid coverage for the first time to near-poor Medicare beneficiaries. Coverage of Medicare premiums only will be required for those below 110 percent of the poverty standard in 1993 and 1994, and to those below 120 percent of the poverty standard in 1995 and thereafter.a Data are not available to trace the number of QMBs from the time the mandate was implemented. The earliest available data, from FY 1990, indicate that 132,131aged Medicare enrollees became Medicaid eligible through the QMB provision during FY 1990. However, this is an underestimate because not all States reported QMB eligibles ~ e p a r a t e l y .The ~ most current data regarding QMBs are Medicare Part B premium data from May 1991 to February 1992. In May 1991, States reported payment of Medicare Part B premiums for 762,741 people. By February 1992, that number had grown by 41 percent to 1,078,200.66 These data also understate the actual number of QMBs because they do not include QMB eligibles from four States (Florida, Michigan, Mississippi, and Nebraska) and, according to HCFA, may not reflect accurate State reporting of the total number of QMBs. Aid to Families With Dependent Children-UnemployedParents (AFDC-UP). The number of new AFDC-ZIT beneficiaries resulting from the mandate is relatively small. During F Y 1991, an additional 22,000 families became eligible for AFDC-UP in States implementing the mandate. Partly as a result of the recent recession, an additional 55,000 families are expected to be program beneficiaries in FY 1992. aA 209(b) State which used an income level lower than the SSI level on J a n u q 1,1987, was allowed an extra year to phase-in coverage of persons up to 100 percent of poverty. These States were required to cover persons a t or below 95 percent of poverty in 1991 and at or below 100 percent of poverty in 1992. % 1992, the poverty standard for a family of three was $11,750. Vource: HCFA Form 2082, FY 1990. ssUnpublished data provided through personal communication from Medicare premium billing data provided by the Bureau of Data Management and Strategy, HCFA. Other Factors Acquired Immune Defkiency Syndrvme (MDS) Although not a major contributor to Medicaid spending trende, Medicaid expenditures for AIDS have risen sharply in the last several yeare and are projected to increaee rapidly for the foreseeable future. Medicaid hae emerged as the moat important eingle source of coverage for penone with AIDS and may play a growing role in funding treatment for other pensone who are infected with the human immunodeficiency virus 0but who have not been diagnoeed as having AIDS. Recently, the Office of the Actuary in HCFA estimated that total (Federal and State) Medicaid expenditures for beneficiaries with AIDS rose from $390 million in FY 1987 to $2.1 billion in FY 1991. By FY 1997, npending is projected to nearly double to $3.8 b i l l i ~ n . ~ g6Unpublishedestimates of the Officeof the Actuary, Health Care Financing Administration, Feb. 6, 1992.