Medicaid Expenditures and State Responses

Abstract

This overview summarizes issues addressed in this issue of the Health Care Financing Review, entitled “Medicaid and State Health Reform.” Articles cover the following topics: growth in the level of expenditures for Medicaid and creative financing strategies by States to manage these increases; section 1115 demonstration waivers; States' experiences with implementing approved section 1115 demonstrations; how section 1115 demonstration waivers fit into larger State health reform efforts; and other reform efforts in two States.

Introduction

The Medicaid program experienced extraordinary growth during the late 1980s. This growth was affected by a number of factors, including an increase in the number of enrollees, higher provider payments, and changes in Medicaid financing arrangements. States responded to these changes in a variety of ways, including taking advantage of special financing arrangements, modifying the service delivery and provider payment approaches through innovations to the Medicaid program, and enacting more extensive changes to the service delivery system through broader health reform efforts. This article addresses factors affecting increases in Medicaid expenditures as well as State responses.

Medicaid Spending

During the 1980s, the Medicaid program experienced only gradual annual increases in spending averaging about 10 percent. Medicaid State and Federal spending totaled about $25.5 billion in fiscal year (FY) 1980, rising to $41.3 billion in 1985 and $72.2 billion in 1990. In the early 1990s, however, spending began to grow at an accelerated rate. Over a 1-year period (FY 1990 to FY 1991), spending increased by about 31 percent, to a total of $94.3 billion. In FY 1992, spending continued to rise, totaling $120.2 billion (an increase of about 27 percent). By FY 1993, spending increases for Medicaid had slowed, rising by only about 9 percent. Medicaid spending in FY 1994 is expected to reach $143.8 billion, representing a 10-percent increase over FY 93 spending.¹

The Federal and State governments expressed concern not only about the magnitude of increases in Medicaid spending, but also about the growth level compared with other payers. During most of the 1980s, Medicaid spending was comparable to that of other major payers. For example, from 1983 to 1987, the average annual expenditure growth rate was about 10 percent for Medicaid, 9 percent for Medicare, and 8 percent for private payers. However, between 1990 and 1992, Medicaid's annual expenditure growth rate jumped to about 28 percent, while growth rates for Medicare and private insurance remained at rates similar to those of the 1980s (i.e., 11 percent and 6 percent, respectively) (Coughlin, Ku, and Holahan, 1994).

A second fundamental problem, and one critical to States, was the disparity between the growth in Medicaid and that of State budgets. During most of the 1980s, the growth in State budgets and Medicaid State-only expenditures averaged around 9 percent However, beginning in 1988, the growth in State revenues slowed to around 7 percent annually, while State-level Medicaid spending grew at a much higher rate (about 19 percent per year). As a result, Medicaid became a much larger component of State budgets (Coughlin, Ku, and Holahan, 1994; National Association of State Budget Officers, 1993).

Factors Affecting Medicaid Spending

A number of factors at the Federal and State levels are hypothesized to have affected Medicaid expenditures. Such factors include growth in the number of persons eligible for Medicaid, increases in provider payments, and special financing arrangements used by States to increase Federal matching payments. These factors are briefly discussed in this section.

Eligibility for Medicaid is linked to eligibility for two Federal cash assistance programs: Aid to Families with Dependent Children (AFDC) and Supplemental Security Income (SSI).² Because of this link, the number of Medicaid enrollees typically rises and falls with the number of persons eligible for AFDC and SSI. Although AFDC eligibility criteria remained restrictive through the 1980s (with an average annual growth of only 2 percent from 1987-90), growth in that population averaged 9 percent per year in the early 1990s (Coughlin, Ku, and Holahan, 1994). These increases were most likely the result of the economic recession. The number of persons eligible for Medicaid through SSI grew similarly during the same period, increasing from an average annual growth of 3 percent from 1987-90 to 8 percent from 1990-92. This growth was due primarily to increases in the number of blind and disabled persons (Coughlin, Ku, and Holahan, 1994; Gurny, Baugh, and Reilly, 1993).

In an effort to increase access to Medicaid services, Congress enacted a series of mandates and State options that greatly enhanced States' ability to extend coverage to specific Medicaid populations. These congressional actions were motivated by concerns over health care coverage for low-income children, as well as a rise in infant mortality. Beginning with the Omnibus Budget Reconciliation Act (OBRA) of 1986, eligibility for pregnant women, infants, and children was expanded. That legislation, and subsequent congressional actions, expanded coverage for these groups regardless of their eligibility for a State's AFDC program, breaking the tie of Medicaid and AFDC eligibility (Buck and Klemm, 1993). Findings suggest that eligibility expansions affecting low-income pregnant women, children, and infants may have accounted for 45 percent of the growth in the number of Medicaid recipients³ from 1988-92 (Coughlin, Ku, and Holahan, 1994). In addition, Congress mandated expanded coverage for the elderly and disabled. The Medicare Catastrophic Coverage Act of 1988 required States to pay Medicare premiums, coinsurance, and deductibles for qualified Medicare beneficiaries⁴ and mandated an increase in the assets and maintenance allowance for a person when his/her spouse was institutionalized (Buck and Klemm, 1993).

Increases in payment levels also affected Medicaid expenditures, including changes brought about by the Boren amendment and by OBRA 1987. The Boren amendment requires States to set “reasonable” payment rates for nursing facilities (NFs) and hospitals. This amendment appears to have slowed institutional rate increases during the 1980s. However, the amendment seemed to provide a standard against which institutions could challenge their rates. In instances where providers prevailed, States were usually required to raise the level of Medicaid payments and, in some cases, make retroactive payments to providers (Buck and Klemm, 1993; Anderson and Scanlon, 1993). OBRA 1987 established a single category of NFs and required that intermediate care facilities (ICFs) meet the higher standards of Medicare skilled nursing facilities (SNFs). Moreover, it required States to account for the cost of institutional compliance with the regulations in the payments made to NFs (Liu, Taghavi, and Cornelius, 1993).

Special Financing Programs

Although the Medicaid financing system has always provided States with incentives for maximizing Federal support,⁵ State efforts to increase Medicaid funding intensified in the early 1990s. States took advantage of a number of these special financing arrangements to effectively increase Federal support for their Medicaid programs. Such arrangements include disproportionate share hospital (DSH) payments, provider-specific tax and voluntary donation (T&D) programs, and intergovernmental transfers (IGTs). In general, under each type of arrangement, the Federal Government provides matching payments to the State for funds collected through the financing arrangement and disbursed to providers, in addition to the Federal contribution to the State's regular Medicaid program payments. These special financing programs are briefly described in this section.

In the early 1980s, Congress required that States consider the needs of hospitals that served a disproportionate number of low-income or uncompensated care patients. The DSH program was originally designed to help public and non-profit hospitals that serve low-income patients. Since these hospitals have high Medicaid and uncompensated care patient caseloads, they are less able than other hospitals to shift the costs of uncompensated care to private-pay patients (Buck and Klemm, 1993; Holahan et al., 1993; Coughlin, Ku, and Holahan, 1994). Although States were initially reluctant to implement DSH programs, legislation passed during the mid-1980s stimulated their development. Since that time, DSH program expenditures have risen from $4.7 billion in FY 1991 to a high of $17.4 billion in FY 1992 (Coughlin, Ku, and Holahan, 1994). The Medicaid Voluntary Contribution and Provider-Specific Tax Amendments of 1991 (the Voluntary Contribution Act) implemented spending caps on DSH payments, limiting them to a 12-percent national target. As a result, DSH payments began to decrease in FY 1993 and will total approximately $16.9 billion in FY 1994.⁶

T&D programs became popular in the late 1980s. Provider-specific taxes include assessments on certain groups of providers, including hospitals, NFs, and physicians. Under this arrangement, providers pay a tax to the State; the State, in turn, repays the provider and uses these dollars as the State share for generating Federal matching support Estimates are that between 29 and 33 States used T&D programs in 1992 to fund some part of their Medicaid program (Holahan et al., 1993; Iglehart, 1993). The Voluntary Contribution Act eliminated Federal matching funds for all voluntary provider donations and limited provider-specific tax revenues to generally not more than 25 percent of the State's share of Federal Medicaid expenditures. Prior to the Voluntary Contribution Act, providers were promised that they would receive DSH payments that at least equalled their contribution. The new law eliminated the ability of States to return the tax to providers.

IGTs were another form of special financing arrangement used by States. Under this arrangement, tax dollars from local governments or State/county hospitals are transferred to the Medicaid agency. Then Medicaid payments are made to the contributing providers, often as a DSH payment, with the State collecting Federal match dollars on the transferred amount. The Voluntary Contribution Act prohibited HCFA from issuing regulations that changed the allowability of IGTs. Because the Voluntary Contribution Act had closed many loopholes in earlier laws, thereby limiting provider taxes and eliminating provider donations, many States have reportedly increased their use of IGTs (Lipson, 1993; Coughlin, Ku, and Holahan, 1994; Buck and Klemm, 1993).

Research on Medicaid Expenditure Growth

The first four articles in this issue discuss approaches to identifying the reasons for and explanations of the increase in Medicaid expenditures during the early 1990s. One article summarizes the results of a multivariate analysis of factors that affect Medicaid expenditures and includes the issues previously discussed as variables in the models tested in the study. A second article focuses on T&D programs and reports on a survey designed to estimate the effect of these programs on States. The effects of changes in Medicaid coverage from 1984 to 1991 on Medicaid enrollees and State taxpayers are examined in a third article. The final article expands on this theme and looks at trends in the breadth and depth of eligibility.

The article by Wade and Berg analyzes the reasons underlying Medicaid expenditure growth. Using multivariate analysis, the authors assess the impact of a number of factors on Medicaid, including the following: mandated expansions in Medicaid eligibility; T&D programs; the Boren amendment; the nursing home provisions of OBRA 1987; the extension of Medicaid eligibility to low-income Medicare beneficiaries; the Zebley decision, which revised eligibility criteria for disabled children; and other factors (i.e., the economy, the prevalence of acquired immunodeficiency syndrome, State policy variables). Models are analyzed for four enrollment groups (adults, children, the blind and disabled, and the aged), as well as different categories of expenditures (e.g., ambulatory, inpatient, prescription drugs, home health care, ICFs/SNFs, etc.) within enrollment group. As expected, the results suggest that total Medicaid expenditures for each enrollment group and for almost all categories of expenditures significantly increased as enrollment increased. However, the growth in enrollment of low-income pregnant women, children, and infants accounted for only 9 percent of expenditure growth from 1988 to 1992. This finding, then, does not support the contention that the eligibility expansions for low-income pregnant women were the predominant factor behind the increases in Medicaid expenditures. Other results suggested that Federal Medicaid policies that directly affect reimbursement, specifically OBRA 1987 and the Boren amendment, were related to expenditure growth for specific types of services. For example, the findings suggest that the cost of maintaining an ICF bed at the SNF level is approximately $5,000 per year. Other factors, including State policies (e.g., tax price and tax capacity) and T&D programs, are found to have a significant effect on growth only for certain enrollment groups and expenditure categories.

The article by Ku and Coughlin focuses on the extent to which T&D and DSH programs affected uncompensated care and worked to reduce State expenditures on Medicaid. Using a survey completed by 39 States and case studies of 6 of those States, the authors identify the amount of revenues that were collected and how they were distributed to county hospitals, private hospitals, and State hospitals. The results show that the States in the sample received $4.9 billion in extra funds in 1993 through special financing programs. While this amount is modest compared with the overall size of many State budgets, special financing arrangements were more important in some States than in others. For example, in New Hampshire, the total State gain was 25 percent of the State's general fund; in Louisiana, 17 percent, and in South Carolina, 10 percent. The authors find that, of the total revenue collected by the survey States under T&D and DSH programs for 1993, about one-half was used to pay back providers for their contributions to the State, one-sixth was used to help private and county providers pay for uncompensated care in those facilities, and one-third was retained by States to finance other expenditures, including but not restricted to health care costs. Only a small share of these funds was available for uncompensated care provided by State hospitals. In fact, almost one-half of the survey States kept more than 50 percent of the extra funds they generated from providers and the Federal Government.

The authors conclude that T&D programs had a number of advantages and disadvantages. Among the advantages were that they helped many needy hospitals provide care to low-income populations, assisted States during a period of fiscal stress, and helped to underwrite the costs of federally mandated expansions. Disadvantages included: the additional costs incurred by the Federal Government; the uncertainty over the equity of the program, as its use by States was not necessarily related to the State's needs or to the breadth of the Medicaid program; and the difficulty of determining the proportions of funds generated through the Medicaid program that are used for health care and for other purposes.

The article by Adams examines whether changes in eligibility affected the equity of the Medicaid program for beneficiaries and State taxpayers from 1984, 1991, and 1992. According to the author, equity for beneficiaries means that similarly disadvantaged persons are treated comparably across States in terms of benefits and eligibility, while equity for taxpayers means that persons with comparable incomes bear similar tax burdens. Findings indicate that federally mandated expansions significantly increased equity across States in the coverage of the poor, while inequality in real expenditures per enrollee remained substantial. That is, the ratio of persons enrolled in Medicaid to persons with incomes below the Federal poverty level increased from 71 percent in 1984 to 96 percent by 1992, with the coefficient of variation⁷ decreasing from 38 percent to 22 percent over the same time period. Real expenditures per poor person increased from $989 in 1984 to $1,791 in 1992, with a similar decrease in the coefficient of variation from 53 percent to 48 percent However, while expenditures per enrollee (after including provider taxes and donations and adjusting for the Consumer Price Index) increased from $1,381 to $1,927, the coefficient of variation remained constant from 1984 to 1991 (31 percent) and jumped to 47 percent in 1992. This change is indicative of increased disparity across States and may reflect, in part, differential use of T&D programs. In terms of expenditures by major enrollment group, the findings suggest that disparities remained across States. Slight improvements in equity for the aged and blind and disabled were observed by 1991, but these were offset by a small drop in the equity of spending for adults and no change in spending for children.

The article by Cromwell, Adamache, Ammering, Bartosch, and Boutis continues the analysis of equity of Medicaid coverage across States. The authors assess the impact of various mandates and document trends in the breadth of Medicaid eligibility, as measured by enrollment per person in poverty and depth of coverage (in terms of optional services, utilization limits, and real spending per enrollee by State). The results show that States such as California, West Virginia, Illinois, and Tennessee, as well as the District of Columbia, showed an enrollment bias; that is, they enroll a greater percentage of the poor than the U.S. average. Other States (e.g., Wisconsin, Massachusetts, South Dakota, and New Hampshire) showed a greater depth of coverage to persons deemed eligible for Medicaid than the U.S. average.

In the same article, the authors assess the relationship between States' ability to pay for Medicaid services, or taxpayer burden, and the generosity of a State's Medicaid program. The findings show that the overall relationship between taxpayer burden and generosity to the poor is positive, suggesting that States that offer a more generous Medicaid program to their low-income populations do so by bearing a larger Medicaid tax burden. New York, for example, spends 1.7 times in real terms what the average State spent per person in poverty ($3,115 versus $1,806), with a much higher tax burden. In contrast, West Virginia, Nevada, New Mexico, and Alabama all spent less than 60 percent of the U.S. average on the poor, while bearing a relatively low tax burden for Medicaid.

State Responses to Medicaid Expenditure Increases

The themes in Medicaid financing research are reflected in the States' responses to Medicaid expenditure increases. The first three articles examine the section 1115 demonstration waiver process, the types of demonstrations that are approvable under that authority, the problems encountered by States during the implementation of reform, and how the section 1115 process fits into wider health reform efforts. Two descriptions of more comprehensive efforts at the State level are also provided. These themes suggest a common set of implementation and operational issues faced by States in health care reform efforts.

Under section 1115 waiver authority, States are proposing innovative approaches to service delivery and expanding eligibility to Medicaid for low-income persons. The article by Rotwein, Boulmetis, Boben, Fingold, Hadley, Rama, and Van Hoven provides background information about the section 1115 demonstration process, including a description of the types of innovations that States are either currently operating or planning to implement. The article provides a brief description of the programs in the four States that have implemented statewide section 1115 demonstrations (Oregon, Tennessee, Hawaii, and Rhode Island), as well as two other States that have received demonstration approval but have not yet implemented their programs (Kentucky and Florida). During the section 1115 approval process, a number of areas of each State's application are reviewed, including the following: the public notice process; program eligibility and benefits; Federal financial participation; treatment of essential community providers, including Federally Qualified Health Centers (FQHCs); cost control; quality assurance; data collection; and reporting requirements.

One of the most difficult issues facing HCFA and States during the review is that of budget neutrality. Budget neutrality exists if Federal Medicaid expenditures are no greater than they otherwise would be in the absence of the demonstration. The challenge to States is to demonstrate sufficient cost savings through changes to the delivery system, for example, to offset the cost of expanded coverage. The authors suggest some early lessons that have been learned from section 1115 efforts. Because the administrative and operational efforts for statewide reform programs are often difficult, implementation dates may be delayed. Second, efforts to transition a primarily fee-for-service population into managed-care networks involve careful consideration of the organizational and administrative structures that are needed.

In the article by Thorne, Bianchi, Bonnyman, Greene, and Leddy, representatives from each of four States that are currently conducting statewide section 1115 demonstrations present their thoughts on crucial problems in getting their programs implemented.

Thorne highlights a number of issues that affected Oregon, including unexpected numbers of telephone inquiries from beneficiaries and program enrollment figures that exceeded projected levels. Lessons learned from Oregon's demonstration include the importance of time and resources to be devoted to the reform effort and the involvement of interested parties in the planning and implementation phases.

Next, Bianchi addresses the information systems used by Hawaii in implementing the reform program, including processes for eligibility determination, enrolling and tracking clients, billing, monitoring utilization and quality, processing fee-for-service claims, and generating program management reports. This system is crucial for the effective and efficient administration of Hawaii's program. Payment to FQHCs, which has been a controversial issue for a number of States, is also addressed. In Hawaii, FQHCs have contracts with at least one health plan. However, FQHCs feel that the capitation rate they receive does not provide them with the same level of revenue they received under cost-based reimbursement. In response to this problem, Hawaii set aside a $1 million fund to be used as payment to those FQHCs that provide care to uninsured persons.

Bonnyman discusses a number of issues faced by Tennessee as that State moved from a primarily fee-for-service system to a managed-care system. For example, the State required that any provider who serves State employees must also serve TennCare patients. This highly controversial requirement was critical to ensuring access to primary care for TennCare patients. Bonnyman addresses the importance of beneficiary outreach and education in implementing a successful reform program. In reaching beneficiaries, Tennessee required that all materials be written at a sixth grade level and be field tested. Use of electronic media also ensured that more beneficiaries could be contacted. Problem resolution was enabled through the use of advocacy groups, using telephone help lines and consumers to address grievances.

Finally, Greene and Leddy address implementation issues related to Rhode Island's more targeted program, RIte Care, directed at pregnant women and children. Issues related to FQHC reimbursement, client education, and community relations are addressed. These perspectives highlight a number of different issues that States must face as they implement their demonstrations. However, cross-cutting issues suggest a common link, including beneficiary outreach and education, difficulties in implementation, and establishing adequate operational and managerial systems.

The article by Riley discusses the section 1115 demonstration process in the overall context of State health reform. Riley lays out five approaches to health reform used by States, including insurance reform, State purchasing alliances, universal coverage, expanded coverage for children, and section 1115 demonstration waivers. Two fundamental problems with comprehensive State reform are discussed. First, absent congressional action to create changes in the Employee Retirement Income Security Act (ERISA),⁸ States have insufficient authority to generate revenues and enact reform programs that cover all persons. Second, States have an insufficient financing base to fully achieve reform. The author addresses the relationship between Medicaid program requirements (e.g., ties to AFDC and SSI requirements, coverage restrictions, and “loopholes” in financing) and how these relate to States' proposed section 1115 demonstration programs. That is, States are proposing approaches that expand eligibility to low-income persons, encourage the development of managed-care systems, and maximize Federal economic support. Riley suggests several effects of the section 1115 demonstrations on the Medicaid program, including: Medicaid becoming the base for coverage of the uninsured; administrative difficulties as States move from claims processors to managed-care contractors; concerns over managed-care capacity, capitation rates, budget neutrality, and the needs of special populations; and administrative changes that State Medicaid agencies must undertake. Finally, both the States and the Federal Government must be concerned about the equality, equity, and effectiveness of the demonstration program for individuals who are served by the program.

The article by Fraser describes the New York Prospective Hospital Reimbursement Methodology (NYPHRM). The NYPHRM was designed to meet three different objectives: to contain costs, support financially stressed hospitals, and improve access to care for the uninsured. Beginning in 1970, the State began a per diem prospective payment system for Medicaid and Blue Cross and “coupled” their rates, using the same methodology in calculating the payment rates for each of them. Because rates for commercial payers were set at a fixed amount above the Medicaid rate, reducing Medicaid reimbursements cut payment levels throughout the health care system and decreased the inflow of both local and Federal dollars, since each State dollar is matched by one local dollar and two Federal ones. The introduction of NYPHRM in 1983 was intended to stop this growing inequity.

Fraser addresses seven factors critical to adoption and successful implementation of the approach: (1) strong leadership continuing over time; (2) support by key players; (3) congruence with the State's political culture (i.e., perception of rate regulation as a legitimate State activity); (4) technical capacity and administrative expertise in ratesetting; (5) data capabilities; (6) avoidance of legal challenges; and (7) cross-subsidies of hospitals from Medicare. The New York experience provides some useful lessons for policymakers considering the regulatory approach as part of a health reform strategy. One important lesson from this approach is that the competing goals of improving access to care, containing costs, and supporting financially distressed hospitals can be achieved, but may be difficult to sustain over a period of time.

Jacobson reports on a case study of the implementation of the Health Services Act (HSA) of 1993 in Washington State. The HSA guarantees universal access to health care for all residents through an employer mandate, with caps on premiums as the primary cost-control mechanism. Washington State, in enacting this legislation, was responding to conditions similar to those of States that have implemented reform through Medicaid—unacceptable increases in Medicaid costs, unsuccessful attempts at hospital rate regulation, and heightened concern over the number of uninsured persons. The author highlights many issues that must be addressed before Washington State can operate its system, including: an amendment to ERISA requirements; the design of the uniform benefits package; the setting of the premium cap; the definition of community rating; and the development of working relationships between community health plans and health insurance purchasing cooperatives. Finally, Jacobson discusses other administrative and management issues that must be addressed, including expansion of public education, involvement of stakeholders in the process, and the need for congressional action to enable more comprehensive State health reform.