urban institute nonprofit social and economic policy research

Health Policy for Low-Income People in Texas

Read complete document: PDF


PrintPrint this page
Share:
Share on Facebook Share on Twitter Share on LinkedIn Share on Digg Share on Reddit
| Email this pageE-mail
Document date: November 01, 1997
Released online: November 01, 1997

The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.


About the Series

Assessing the New Federalism is a multi-year Urban Institute project designed to analyze the devolution of responsibility from the federal government to the states for health care, income security, employment and training programs, and social services. Researchers monitor program changes and fiscal developments, along with changes in family well-being. The project aims to provide timely nonpartisan information to inform public debate and to help state and local decisionmakers carry out their new responsibilities more effectively.

Key components of the project include a household survey, studies of policies in 13 states, and a database with information on all states and the District of Columbia, available at the Urban Institute's Web site. This paper is one in a series of reports on the case studies conducted in the 13 states, home to half of the nation's population. The 13 states are Alabama, California, Colorado, Florida, Massachusetts, Michigan, Minnesota, Mississippi, New Jersey, New York, Texas, Washington, and Wisconsin. Two case studies were conducted in each state, one focusing on income support and social services, including employment and training programs, and the other on health programs. These 26 reports describe the policies and programs in place in the base year of this project, 1996. A second set of case studies to be prepared in 1998 or 1999 will describe how states reshape programs and policies in response to increased freedom to design social welfare and health programs to fit the needs of their low-income populations.

The income support and social services studies look at three broad areas. Basic income support for low-income families, which includes cash and near-cash programs such as Aid to Families with Dependent Children and Food Stamps, is one. The second area includes programs designed to lessen the dependence of families on government-funded income support, such as education and training programs, child care, and child support enforcement. Finally, the reports describe what might be called the last-recourse safety net, which includes child welfare, homeless programs, and other emergency services.

The health reports describe the entire context of health care provision for the low-income population. They cover Medicaid and similar programs, state policies regarding insurance, and the role of public hospitals and public health programs.

In a study of the effects of shifting responsibilities from the federal to state governments, one must start with an understanding of where states stand. States have made highly varied decisions about how to structure their programs. In addition, each state is working within its own context of private-sector choices and political attitudes toward the role of government. Future components of Assessing the New Federalism will include studies of the variation in policy choices made by different states.


Contents

Highlights of the Report

Thumbnail Sketch of Texas

    Sociodemographic and Economic Overview
    Economic Status
    Political Landscape
    Budgetary Environment
    Roadmap to Rest of Report

Setting the Policy Context

    The Health Care Environment
    State Health Care Indicators

State Health Programs

    Medicaid
    Department of Health and Department of Human Services
    Department of Mental Health and Mental Retardation

Assessing the New Federalism: Possible State Responses to Additional Flexibility and Reduced Funding

    Interstate Competition and the "Race to the Bottom"
    Medicaid and Federal-State Relations

Third-Party Health Coverage for the Low-Income Population

    Medicaid Eligibility
    Other Public Financing Programs
    Insurance Reforms

Financing and Delivery System

    The Health Care Market and Access for Low-Income Populations
    Medicaid Managed Care
    Medicaid Provider Payment and Disproportionate Share Hospital Policy

Delivering Health Care to the Uninsured and Low-Income Populations

    State and Local Health Departments
    Hospital Districts
    Impact of Government Policies and Market Changes on Safety Net Providers: Houston and El Paso

Long-Term Care for the Elderly and Younger Persons with Disabilities

    Supply, Utilization, and Expenditures
    Long-Term Care for Older People in Texas
    Younger People with Disabilities

Continuity and Change in Texas's Health Care Policy for Low-Income Groups

Notes

Appendix: List of People Interviewed

About the Authors


Highlights of the Report 1

Texas is a big, diverse state, with a large low-income population, high levels of uninsurance, and many foreign-born residents, largely from Mexico. Politically, it is very conservative, with fairly minimal health and welfare programs beyond what is required to draw down a federal match. Cash welfare and taxes are highly unpopular, but health care is viewed more favorably, and the state would like to expand Medicaid eligibility if it can do so without spending more money. Most state officials acknowledge that health and welfare benefits in Texas are less than what are provided in many other states, and few observers are concerned about the state ever becoming a "welfare magnet." Texas officials were supportive of a Medicaid block grant because they believed that funding was adequate and would not require reductions in eligibility or services. Currently, there is very little competition for budgetary resources between the elderly/disabled and families with children. Funding for each group is largely driven by the requirements of federal entitlement programs, a situation that would change drastically under a Medicaid block grant.

Medicaid expenditures grew rapidly between 1990 and 1995, although the rate of increase has fallen sharply since then. Medicaid expenditures increased by 40.9 percent a year between 1990 and 1992, and by 12.0 percent between 1992 and 1995. Although the huge increase in 1990-92 is largely explained by state efforts to maximize federal revenues through the disproportionate share hospital (DSH) program, enrollment and expenditures per enrollee also increased sharply during this period. 2 Expenditure increases are now in the single digits, and enrollment is flat.

Medicaid eligibility is quite restrictive, and the state funds only a very small medical program for the low-income population ineligible for Medicaid. The state has followed a middle-of-the road policy on immigrants, seeking to maintain services to current beneficiaries but not attempting to fill the gap created by federal termination of coverage. Policy on Medicaid eligibility has been driven by a desire to limit state spending while expanding the number of people covered. To that end, the state has been pursuing a Medicaid research and demonstration (a so-called "Section 1115") waiver since 1995, with the goals of protecting DSH payments, expanding coverage, and drawing down additional federal funds. Particularly motivating the state in 1995 was the fear that federal changes in the Medicaid DSH program would result in substantial reductions in federal funds. The state's plan would use local funding for indigent care for the state match to draw down federal funds; thus, no additional state spending would be required. Although the initial proposal was never approved by the Health Care Financing Administration (HCFA), more importantly, the proposal lacked political support by local hospital districts that were concerned about loss of control over their funds and the potential use of their money to fund managed care or other nonpublic hospitals. A revised, more modest proposal that focuses on low-income children and that requires newly eligible individuals to receive their care through provider networks organized by "public-funding entities" was submitted to HCFA in November 1996, but rejected in August 1997.

The health care financing and delivery system in Texas is changing rapidly. First, in recent years, the state has substantially increased Medicaid funding for safety net hospitals by aggressively expanding its Medicaid DSH program, using intergovernmental transfers from local hospital districts to cover the state match. DSH spending now accounts for almost a fifth of total Medicaid expenditures, making federal changes in the program of great concern. Second, overall enrollment in managed care is growing rapidly, and responding to demands for cost reduction is a major concern of providers. Increasing enrollment in Medicaid managed care is the centerpiece of the state's cost-containment strategy, although it also expects enrollment to have positive effects on quality and access. At this point, commercial health maintenance organizations (HMOs) are competing actively for the Medicaid business as a way of increasing market share, and the state has not had to rely on Medicaid-only managed care organizations. Savings of about 10 percent compared to the fee-for-service system have been obtained. Third, Texas has a high percentage of for-profit hospitals, raising concerns about the provision of uncompensated care and the use of formerly nonprofit assets. Fourth, almost 24 percent of the nonelderly population is uninsured, which is one of the highest rates of any state in the country.

A large number of public hospitals, funded at the local level by property taxes, play a critical role in providing care for the Medicaid and uninsured populations and partly compensate for the low levels of insurance coverage. Many of these hospitals feel threatened by Medicaid managed care. Although the Medicaid-contracting HMOs are required to contract with traditional safety net providers for three years, enrollment of Medicaid beneficiaries in these plans is likely to reduce utilization of public hospitals, both because HMOs seek to reduce hospital utilization and because patients may be channeled to other hospitals with lower payment rates. Although from the perspective of the Medicaid program this decline in the use of public facilities may result in reduced costs, more consumer choice, and better quality of care, the loss of Medicaid patients would be a major problem for these safety net hospitals because there would be fewer paying patients over which to spread the cost of providing care to uninsured patients. Without Medicaid patients, public hospitals would require far larger local contributions, something not likely in the current political environment. As a result, facilities could close or reduce services, diminishing access to care by the uninsured population. In response, the Texas legislature in 1997 passed a law requiring Medicaid to contract with HMOs formed by public hospital districts. A significant number of community health centers serving a large number of Medicaid and uninsured patients, which have relatively high Medicaid payment rates because of existing federal rules, face similar problems but have not been afforded similar protection by the legislature.

Finally, long-term care for the elderly and people with disabilities is a critical component of the state's involvement in health care and accounts for about 30 percent of Medicaid spending on services. Historically, long-term care has been dominated by institutional care. In an effort to alter this, the state's policy initiatives center on expanding the supply of home and community-based services for both the elderly and younger people with disabilities. At the same time, the number of persons in state mental hospitals and intermediate care facilities for the mentally retarded is declining, and the state has aggressively maximized Medicaid reimbursement for previously state-only funded programs. The state is also planning a major demonstration project in Houston that will integrate acute and long-term care services through managed care. Beyond system reform, the state has relied heavily on traditional methods of controlling long-term care expenditures, including a moratorium on new nursing home construction and low nursing home reimbursement rates.


Thumbnail Sketch of Texas

Texas is a diverse state with a large low-income population and many foreign-born residents. Politically, it is very conservative, with a lot of power residing in the legislature. Aside from wanting savings to fund tax cuts and the threat of reduced federal funding, there is no Medicaid budget crisis, and current program projections are for very modest rates of increase in expenditures. The Balanced Budget Act of 1997 reduces federal disproportionate share hospital (DSH) payments to Texas by $200 million in 2002 compared to 1998, an amount that is significant but unlikely to cause a major crisis.

Sociodemographic and Economic Overview

Texas is a big state, both in terms of population and geography. Almost 19 million people live in its almost 262,000 square miles (table 1). Texas has 254 counties, 196 of which are considered rural; of these, 62 are considered "frontier." Despite this vast amount of territory, 81 percent of the population lives in metropolitan areas. Partly as a result of the large distances between cities, there is a great deal of diversity in health and human service capacity and delivery systems. It is commonly believed that major areas of the state—in and around Austin, Dallas, Houston, El Paso, and Lubbock, for example—are each unique, requiring different policy approaches. While some areas of the state are bustling and prosperous, others face enormous problems, with unemployment rates as high as 50 percent and a lack of basic services, such as running water and sewage disposal.

Immigrants are an important part of the Texas population and have been increasing in number. Not surprisingly, given the state's geographic location, a high proportion of the immigrant population is from Mexico. This is especially true in places on the border, such as El Paso. Although almost a third of the Texas population is Hispanic, the vast majority are American citizens. Some cities, such as Houston, also have large numbers of immigrants from Asia and other areas. Non-Hispanic whites constitute only slightly more than half of the population of Texas.

Economic Status

The economy in Texas has been doing well, with relatively low unemployment. However, the state has a high level of poverty. Unemployment in 1996 was 5.6 percent, slightly higher than the national average of 5.4 percent (table 1). The proportion of the total population and of children below the federal poverty line is considerably higher in Texas than in the country as a whole; per capita income is somewhat lower than the national average but grew faster than that for the country overall between 1990 and 1995.

Political Landscape

Texas is generally fiscally conservative. Republicans control the governorship and the state Senate. In contrast, the Democrats have a comfortable majority in the House. There is a substantial amount of bipartisan cooperation.

Structurally, Texas state government is characterized by a weak governor and a strong legislature. The governor, for example, does not prepare an annual budget. Part-time legislators meet for only 140 days every other year. Legislative committees are assigned topics to study between sessions. In practice, a great deal of power resides with the Legislative Budget Board, which is chaired by the lieutenant governor and consists of members of the leadership of the House and Senate. This group proposes budgets and is empowered to reallocate funds when the legislature is out of session.

Budgetary Environment

There is a very strong antitax environment in Texas, and welfare is viewed with disfavor; health care has more support. There is no income tax, although local property taxes are relatively high. During the 1995-96 biennium, the state enacted major welfare reform and authorized an ambitious Medicaid expansion and managed care demonstration. New spending was very limited.

For the 1997-98 biennium, the budget was also fairly constrained. Governor George W. Bush made local property tax relief and restructuring the tax system major priorities. Approximately a third of the funds for local property tax relief would have come from Medicaid savings. The legislature enacted a $1 billion property tax cut over two years (but did not restructure the tax system). The loss of local revenues was offset by increasing state education funding.

Roadmap to Rest of Report

The rest of the report is divided into eight topics. The next two sections briefly describe the policy, budgetary, and administrative context of Medicaid and other health care programs for the low-income population in Texas. This description is followed by an assessment of how Texas views its role in providing health care and other services to the low-income population and how that might change if it were given additional flexibility to run Medicaid and other programs. After this section is an analysis of eligibility for third-party health care coverage for the low-income population, particularly Medicaid. Two sections on financing and delivery of health care to the low-income population follow, after which there is a discussion of long-term care for the elderly and disabled. Finally, there is a discussion of the major health care challenges Texas faces in the future.


Setting the Policy Context

The state's health care agenda is driven by two competing concerns—cost containment and improving access. Although a lower priority than in 1995-96, health care remains important because of its budgetary implications. The rate of increase in Medicaid costs has slowed dramatically, but the state continues to be concerned about controlling Medicaid expenditures. Medicaid is a very substantial portion of Texas's state spending and dominates state health policymaking.

The state's strategy for controlling health care costs depends very heavily on enrolling Medicaid beneficiaries into managed care organizations, a move that the state believes will also improve quality of care and access. At the same time, the state has continued to pursue eligibility expansions through a Medicaid research and demonstration waiver, even though its proposals met with local and federal resistance and failed to be approved.

The Health Care Environment

There are two characteristics of the health care environment in Texas that are crucial to understanding the politics and policy of health care in Texas. First, counties, through hospital districts, play an extremely important role in providing health care for the uninsured and Medicaid beneficiaries. Hospital districts have taxing authority (through the property tax) and run public hospitals and clinics. In addition, the University of Texas, especially the campus in Galveston, plays an extremely important role in providing charity and Medicaid care. These public hospital districts fill some of the gap that is created by low Medicaid eligibility. Local hospital districts are deeply worried that the expansion of Medicaid managed care will result in their losing Medicaid patients, who are crucial for the hospitals' financial stability.

Closely associated with the role of public hospitals, the state has aggressively maximized Medicaid DSH funds, which account for about 19 percent of Medicaid spending. The federal legislative intent of DSH payments was to provide financial relief to hospitals that serve a large number of low-income patients. Intergovernmental transfers from the hospital districts provide most of the state match for DSH payments, and public hospitals receive most, but not all, of the money going to hospitals. Thus, the reductions in federal DSH funds recently passed by Congress could adversely affect these hospitals.

A second characteristic of the health care environment in Texas is that the health care marketplace is in turmoil. Enrollment in health maintenance organizations (HMOs), which served less than 20 percent of the population in 1995, is growing rapidly. Competition among HMOs is fierce, and a shake-out is widely expected. At this point, Medicaid is viewed as a better payer of hospitals than are HMOs. Commercial HMOs are very interested in the Medicaid business as a way of gaining market share. In addition, there are a lot of mergers and conversions of nonprofit facilities to for-profit status.

State Health Care Indicators

Texas is about average among states in the health status of its population, but it has a higher proportion than average of uninsured. In the ReliaStar ranking of the relative healthiness of the populations in all 50 states, Texas ranked 24th in 1996 and 33rd in 1995. 3 Texas has about the same infant mortality, low birthweight, and premature death rates as the national averages (table 1). Despite these rankings, Texas has many areas that are extremely poor and have major health problems. For example, relatively unique to southern Texas are areas called colonias, which have substandard housing, unpaved roads, and limited or no access to potable water, waste water disposal systems, and garbage collection. Many residents of the colonias are undocumented aliens.

The lack of health insurance is a major problem in Texas, with 23.9 percent of the nonelderly population uninsured in 1994-95, compared with 15.5 percent for the nation as a whole. In addition to low Medicaid coverage, this high rate of uninsurance is due to the low proportion of people with employer-sponsored health insurance—only 58.0 percent of the nonelderly population in Texas, compared with 66.1 percent for the country as a whole.

Although many urban areas, such as Houston and Dallas, have a large number of health care providers, the state as a whole has significantly fewer providers than the national average. For example, in 1994, while there were 81 generalist physicians per 100,000 population in the United States, there were only 61 generalist physicians per 100,000 population in Texas. In 1995, 112 entire counties and parts of 22 others were considered health professional shortage areas. 4


State Health Programs

In Texas, the Health and Human Services Commission oversees all acute and long-term care policy. It was created in 1991 in order to consolidate policymaking in one entity, but operational responsibility remains with a variety of departments. The Texas Department of Human Services (TDHS), which previously was responsible for most Medicaid programs, now oversees long-term care and Medicaid eligibility determinations. Its acute care counterpart, the Texas Department of Health (TDH), oversees all acute care Medicaid and core public health functions. Mental health and mental retardation programs, many of which also are funded by Medicaid, are located in the Texas Department of Mental Health and Mental Retardation (TDMHMR). Other health-related departments include the Commission on Alcohol and Drug Abuse, the Department on Aging, the Rehabilitation Commission, and the Council on Early Childhood Intervention Services.

Medicaid

As in other states, Medicaid in Texas is the dominant health care program, accounting for 11.3 percent of state general-fund spending (table 2). State Medicaid expenditures grew by 17.2 percent per year from 1990 to 1995. This is faster than the overall state spending growth of 11.0 percent per year but less than state spending on corrections, which grew 26.5 percent annually. Combining state and federal sources of funds, spending on Medicaid grew 20.2 percent annually, compared with 9.3 percent for the total budget from 1990 to 1995. Growth in Texas Medicaid expenditures outpaced the national averages for 1990-92 and 1992-95 (see table 3); this is true for both acute and long-term care services and across all beneficiary groups. Growth in DSH expenditures from 1990 to 1992 was explosive, with expenditures increasing from $4.8 million to $1,513 million. The rate of increase in total Medicaid spending averaged 12.0 percent per year from 1992-95, compared with 9.9 percent nationally. Expenditures grew rapidly because of a large increase in the number of beneficiaries, state decisions to increase DSH, and efforts at Medicaid maximization. Since 1995, the rate of increase has fallen to about 7.5 percent a year.

As is true nationally, spending for acute care in Texas grew faster than spending for long-term care from 1990 to 1995. In Texas, however, spending was already more skewed toward acute care than long-term care. Texas now spends 70 percent of its Medicaid funds on acute care services, compared with 60 percent nationally. In 1990 the share was 60 percent for acute care, compared with 53 percent nationally.

Despite high Medicaid growth rates in the early 1990s, Texas spent less than the national average per eligible person in 1995 ($2,488 vs. $3,202, or 78 percent of the national average) (see table 4). Only six states had lower non-DSH expenditures per beneficiary than Texas in 1995. In part, this reflects limited benefits; for example, Texas restricts coverage to 30 days for inpatient hospital stays and three prescriptions per month for adults. The per capita spending difference has narrowed since 1990, when Texas spent 73 percent of the national average per enrollee.

Department of Health and Department of Human Services

The Texas Department of Health is responsible for Medicaid acute care operations (e.g., hospital and physician reimbursement and managed care) in addition to core public health activities (e.g., disease control and prevention, health care quality and standards, environmental health, and personal services such as children's health, primary care, and nutrition). In fiscal year (FY) 1998, 85 percent of the $6.3 billion TDH budget is for the Medicaid program; 61 percent of the budget is federally funded (Medicaid and non-Medicaid). 5

The Department of Human Services oversees Medicaid eligibility and numerous long-term care services primarily for elderly beneficiaries, in addition to Food Stamps, Aid to Families with Dependent Children (AFDC)/Temporary Assistance to Needy Families (TANF), domestic violence, and disaster assistance programs. In addition, the department regulates nursing homes and other long-term care facilities. Funding for long-term care derives from Medicaid and social services block grants. Seventy percent of the DHS budget is federally funded. 6 For FY 98, 70 percent of the department's $3.4 billion in expenditures are appropriated for Medicaid programs.

Department of Mental Health and Mental Retardation

The Texas Department of Mental Health and Mental Retardation oversees most of the state's long-term care services for nonelderly populations. It operates 9 state hospitals, 12 state schools, and 5 state centers for mentally ill and mentally retarded persons and funds 36 community mental health/mental retardation centers. Its programs encompass targeted case management, rehabilitative services, diagnostic services, home and community-based services, and institutional care. For FY 98 the department's operating budget is $1.65 billion (37 percent for mental health and 63 percent for mental retardation services). 7 The state has kept funding flat, arguing that the shift from institutional to community-based services has meant savings. In addition, the state has used Medicaid to fund many TDMHMR programs that had previously been financed with all state funds.


Assessing the New Federalism:
Possible State Responses to
Additional Flexibility and
Reduced Funding

Texas takes a minimalist approach to health and social welfare programs. Texas culture values "independence," and higher levels of assistance for welfare recipients are generally thought to be inconsistent with this philosophy. Health care, however, receives more political support.

Interstate Competition and the "Race to the Bottom"

Texas state officials and consumer and provider groups readily acknowledged that Texas ranks at or near the bottom on many measures of health, welfare, and educational expenditures. As a result, modest increases in spending would not materially affect its standings. Thus, there is little fear of becoming a "welfare magnet" for either cash assistance or Medicaid. Texas, and the United States more generally, is believed to be a magnet for Mexicans, but that is largely thought to result from huge differences in wages rather than the generosity of social and health benefits.

In general, Texas officials do not look to other states to establish their welfare and Medicaid benefit levels. To the extent that they do look elsewhere, Texas policymakers look to Florida, another large, fast-growing, low-benefit Southern state. Northern liberal states were irrelevant to their decisionmaking, although they sometimes served as negative examples. There is a strong preference for "homegrown" solutions, and initiatives in other states are often dismissed as unworkable in Texas.

Although Medicaid is associated with welfare, health care is thought to be slightly different from cash assistance. Medicaid is viewed somewhat as an insurance program, rather than a pure welfare program. There is a general acceptance of the notion that people ought to be able to obtain needed health care services; Texans, however, are unwilling to pay enough in taxes to fund a very generous Medicaid program. Texas relies instead on an extensive network of public hospitals, financed through property taxes. Thus, the extensiveness of services depends on local wealth and willingness to pay taxes; there is no intrastate cross-subsidization across local areas. While policies to increase the number of welfare beneficiaries are so politically unpopular that they are never proposed, policies that would increase the number of Medicaid eligibles are politically acceptable if the state does not have to spend more money. For example, the state's Medicaid research and demonstration waiver proposal would substantially increase the number of children receiving Medicaid. Because many elderly Medicaid nursing home and home care beneficiaries were originally middle class, long-term care is very important in maintaining public support for Medicaid.

One reason for the better political support for Medicaid compared to cash assistance is that there are several well-financed health care provider groups that are financially dependent on Medicaid funds and lobby to protect the program, while few groups lobby in support of welfare. In the health arena, the for-profit nursing home association was widely believed to be the most powerful interest group. Physicians and hospitals were also said to have considerable political power, much more than the HMO and home care associations.

Consumer advocacy groups are not very strong but are more important than in some other states. While the American Association of Retired Persons' presence in the state is recent and there are few other advocacy groups to represent the interests of the elderly, older people are believed to be a powerful political force. Groups representing people with disabilities have tried to increase their power by joining together to form a consortium. Advocates for low-income families and children lack political influence. According to one observer, advocacy groups in Texas "tend to want to 'get along' rather than playing to win. They play to not lose."

Medicaid and Federal-State Relations

Like many other states, Texas would like more flexibility in running its Medicaid program, but it has learned to live within the current constraints. The proposed Medicaid block grant was generally viewed positively by state officials because the funding level was thought to be adequate and it would have provided more flexibility. Changing to a block grant would, however, dramatically change the financial incentives for the state, which are now largely driven by strategies to maximize federal Medicaid revenues.

Federal-State Relations under Existing Law

Relations with the federal Health Care Financing Administration (HCFA) were generally good. Some state officials, however, thought the requirements to apply for Medicaid waivers were burdensome and time consuming and limited the ability of the state to "try things out" before implementing them statewide. In general, state officials found existing federal rules concerning the Early and Periodic Screening, Diagnosis, and Treatment (EPSDT) program, nursing home and hospital reimbursement, and managed care to be bothersome, but most respondents did not voice much passion concerning their repeal. Finally, Texas officials were not interested in additional flexibility to reduce Medicaid eligibility.

Medicaid Block Grants

The governor and most state officials interviewed supported the Medicaid block grant legislation that was passed by Congress (but vetoed by President Clinton). They liked the flexibility and believed that they would not have to reduce eligibility or services, partly because Medicaid caseloads have been lower than anticipated. Observers criticized Congressional Budget Office and Office of Management and Budget baseline projections of Medicaid spending as being unrealistically high.

While state officials expressed no desire to substantially cut the program, the higher Medicaid match rate included in the "Medigrant" plan would have meant that Texas would have drawn down its allocated federal funds faster than under the current formula. 8 State funding without a federal match requirement is extremely difficult to obtain; thus, the Medigrant plan might have resulted in reduced state general revenue spending for Medicaid.

In contrast, advocacy groups were opposed to the block grant, fearing the loss of funds and the legal leverage to require the state to do what they believed was right. Although the nursing home industry felt that a block grant would not adversely affect them because of their political influence and because they believed they provide a cost-effective service, other provider groups, such as the hospital industry, were more concerned about possible negative consequences.

Intergroup Competition for Resources

In terms of resource allocation among groups, there is currently very little competition between the elderly and nonelderly populations or between acute and long-term care. There is some competition between nursing homes and home and community-based services, with the for-profit nursing home association arguing that home care is not a cost-effective substitute for nursing home care. Instead of competition, spending for each group is determined by its own set of federal mandates and entitlements, and the legislature does what it has to do to meet those requirements. As such, there is competition for resources between entitlement and nonentitlement programs, with the entitlements generally winning.

Although there is not much competition among groups now, virtually everyone believed that would change dramatically under a Medicaid block grant because resource allocation would become a zero-sum game—one group's gain must come as a result of some other group's losing resources. In general, most observers thought that the elderly would do well under a block grant. Even though the elderly do not have a strong lobbying presence, the for-profit nursing home association would likely be a critical factor in the expected success of the elderly in claiming resources. In contrast, it was widely believed that low-income families with children would not do well.

Medicaid Maximization

There are few state-funded health service programs outside of those that require state matching in order to receive federal funds. The maximization of federal Medicaid funds is viewed as a legitimate approach to obtaining a "fair share" of federal dollars, which is defined as the state receiving at least as much in federal funds as it pays in federal taxes. Programs that require full state funding without a federal match receive intense scrutiny and are much more difficult to fund. Texas, like many other states, has diligently "Medicaidized" many of its services (e.g., maternal and child health and services for children with special health care needs). That is, it has sought to obtain a federal Medicaid match for as many previously all-state funded services as possible. Unlike some other states, it has not taken the additional federal match as an opportunity to greatly increase the overall level of service for the same amount of state funding. Instead, the resources have been moved into other spending priorities. In addition, the state has carefully crafted a very large DSH program to maximize federal dollars, while minimizing state contributions by relying heavily on intergovernmental transfers from county hospital districts.


Third-Party Health Coverage for
the Low-Income Population

Texas has strict financial eligibility criteria for its health and welfare programs. Largely due to federal requirements, the standards have become somewhat less austere for Medicaid but are still very low compared to those in most other states. The state has an ambitious welfare reform demonstration under way that predates the federal welfare reforms of 1996. The state submitted two proposals to HCFA for research and demonstration waiver programs to expand Medicaid coverage to the uninsured, but neither were approved.

Medicaid Eligibility

Policy on Medicaid eligibility has been driven by the desire to limit state spending while expanding the number of people covered. To that end, the state has sought Medicaid research and demonstration waivers and has followed a middle road regarding federally imposed changes for immigrants.

Trends in Coverage

Although Medicaid eligibility in Texas is quite restrictive, enrollment grew rapidly between 1990 and 1995, at a rate somewhat faster than the national average. Enrollment in Texas grew 14.6 percent per year from 1990 through 1992 and 7.7 percent per year from 1992 through 1995, compared with 11.3 percent and 5.2 percent nationally (table 5). In 1995, approximately 2.9 million Texans were Medicaid beneficiaries; 12.6 percent of the nonelderly population were eligible, about the national average. At the same time, 45.3 percent of the state population below 150 percent of the federal poverty line were eligible for Medicaid, which is significantly below the national rate of 50.3 percent. In 1995, 79 percent of the Medicaid population were children or nondisabled adults, a slightly higher percentage than the national average. While AFDC/TANF caseloads have been dropping, Medicaid enrollment has stayed relatively stable and is projected to remain level through FY 99. Medicaid funds have been returned to the state treasury in recent years because caseloads have been smaller than projected.

Regular Medicaid Eligibility

A large proportion of the Medicaid population is composed of women and children who gain eligibility because they meet AFDC criteria. 9 Texas, like many Southern states, provides a low cash welfare benefit. In 1996, the annual payment for a family of three with no other income was $2,256 (17 percent of the federal poverty line), compared with the national median state payment of $4,668 in 1996 (36 percent of the federal poverty line). There is no interest in increasing cash assistance levels, with the state preferring to focus its attention on moving beneficiaries into the workforce. The state constitution limits total cash welfare expenditures to 1 percent of the state budget, a proportion that it does not come close to reaching. Because of falling AFDC/TANF caseloads, the state has extra state money that it is not required to spend on TANF. 10

Texas covers only a limited number of optional groups under Medicaid. Although the Texas Medicaid program does cover the "medically needy," the income and asset limits are so low that very few people qualify. Texas does not provide Medicaid eligibility to children whose family incomes are above the federal requirements (133 percent of the federal poverty line up to age 6 and 100 percent of the federal poverty line up to age 14). However, the state does cover (using a presumptive eligibility process) pregnant women and infants up to 185 percent of the federal poverty line without any resource test as part of an effort to reduce infant mortality.

Eligibility Components of "Section 1115" Demonstration Proposals

The state has been pursuing a Medicaid research and demonstration (a so-called "Section 1115") waiver since 1995, with the goals of protecting federal DSH funds, expanding coverage, and drawing down additional federal funds. Particularly motivating the state in 1995 was the fear that federal changes in the Medicaid DSH program would result in a substantial reduction in federal funds. Thus, the waiver proposal was an effort to convert DSH payments to managed care payments for individuals, which might be less vulnerable to reductions at the federal level. Under the initial proposal, eligibility would have been expanded to 133 percent of the federal poverty line for children ages 6 through 18 and 45 percent of the federal poverty line for adults (with a goal of 75 percent of the federal poverty line). In the 1995 proposal, the Medicaid program would have become a series of per capita payments to so-called "intergovernmental initiatives" (IGIs) at the county level.

Although never formally rejected by HCFA, the waiver proposal was never approved. Ultimately, the proposal ran into problems because of the proposed administrative structure. HCFA was concerned that the IGIs represented a potential conflict of interest since they were dominated by providers and would run the program at the local level. In addition, the proposal never obtained substantial financial or political support from the hospital districts. Local hospital districts, which were expected to put up matching funds for the expansion, were not enthusiastic about the prospect of "their" funds being used to further enrollment in Medicaid managed care and to help fund nonpublic hospitals. While DSH-contributing hospital districts now receive more funds than they contribute in intergovernmental transfers, the demonstration would not have provided the same guarantees. As a result, only one hospital district agreed to put up its money for the waiver program.

In November 1996, the state submitted a revised research and demonstration Medicaid proposal that was less ambitious than the earlier version. The revised plan sought to expand Medicaid eligibility for children ages 6 through 18 with incomes below 133 percent of the federal poverty level, but it did not provide additional coverage for adults. While adding an estimated 384,557 children in 1998, the incremental number of persons covered beyond federal requirements would decline over time as children below the federal poverty line were phased in as mandatory eligibles under current law. 11 By 2002, about 230,599 children would have Medicaid coverage who would not otherwise be covered. The state match would be provided by eight large hospital districts, the city of Austin, and the University of Texas Medical Branch. New state general revenues would not be required.

Under the waiver, "public funding entities" would have assumed responsibility for delivering or arranging care for all newly eligible children. The state would have contracted directly with managed care organizations formed by the funding entities. Newly eligible children would have received their care through these managed care organizations, which had to provide at least the Medicaid benefit package and meet all process and quality requirements of other Medicaid managed care organizations. Waiver beneficiaries would have faced expanded copayment requirements. The state estimated that the total cost of the additional coverage would be $325.7 million, of which $202.5 million would be federal funds in the first year.

In August 1997, HCFA rejected the second Texas waiver proposal on the grounds that the state needed to include more than one managed care organization in each region in order to provide beneficiaries with some choice of health plans. The passage of the children's health insurance initiative as part of the Balanced Budget Act of 1997 may make the waiver request largely moot. Because of its large number of uninsured children, Texas will be entitled to a very substantial amount of federal money under the new State Children's Health Insurance Program—approximately $560 million in 1998 and $2.5 billion from 1998 to 2002. However, the requirement for a state match in the new program may be problematic for Texas, although the state will likely be able to use intergovernmental transfers to finance the state share.

Issues in Medicaid/Welfare Reform 12

Historically, Texas's stringent eligibility requirements for AFDC led to low levels of participation in AFDC by the poverty population. The state has been very active in welfare reform and, as a border state with Mexico, will be potentially greatly affected by federal efforts to restrict health and welfare benefits to immigrants.

Texas welfare reform.

In 1995, Texas enacted a comprehensive welfare reform package that was approved by the U.S. Department of Health and Human Services in March 1996 as a demonstration program, prior to the passage of the federal Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (PRWORA). The Texas welfare reform plan has time-limited benefits of 12, 24, and 36 months based on each person's work experience, education, and literacy; 12 to 18 months of transitional child care and medical assistance for those whose time limits expire; and a requirement that recipients must work, go to school, or participate in a job training program. In some ways, the Texas welfare reform is less restrictive than the federal legislation. 13 One concern of the state regarding the federal legislation is that severing the eligibility link between AFDC/TANF and Medicaid (as required by PRWORA) could add $100 million in administrative costs.

In 1996 Texas proposed the Texas Integrated Enrollment System (TIES), a project to consolidate and streamline eligibility determination for fifteen programs, including AFDC/TANF, Food Stamps, and Medicaid, and to privatize these functions by contracting out to private firms. The value of the contract was estimated to be $2.8 billion over seven years, with savings estimated to be about $120 million annually. In a highly unusual arrangement, public agencies—the Texas Department of Human Services and the Texas Workforce Commission—teamed with different private corporations for purposes of bidding on the contract. In May 1997, the Clinton administration ruled that federal law prevented private employees from determining eligibility for the Food Stamps and Medicaid programs. The state believed that the President was bending to pressure from the public employee unions. Since then, the public agencies have terminated their alliance with the private vendors, and a private company has been awarded a 15-month contract to help the state reengineer eligibility determination and service delivery and secure a new computer system.

As part of welfare reform, the 1996 federal legislation tightened eligibility for Supplemental Security Income (SSI) (and, relatedly, Medicaid) for children. It is estimated that approximately 12,700 children in Texas with Medicaid expenditures of $52 million will lose SSI. Most will retain Medicaid eligibility through the requirement to cover poverty-level children. However, some children, especially those over age 13, will lose Medicaid eligibility. The 1997 federal budget agreement restored Medicaid eligibility for beneficiaries who lost eligibility as a result of the changes in SSI.

Immigrants.

Changes to immigrants' eligibility for public benefits enacted as part of PRWORA, as modified by the Balanced Budget Act of 1997, will have a significant impact in Texas, although somewhat less than might be imagined because a very high percentage of persons of Hispanic (primarily Mexican) origin are American citizens. Under federal law, almost all legal immigrants who enter the country after August 22, 1996, will be excluded from Medicaid for five years, with the principal exception of refugees. County public hospitals, legally required to serve all medically indigent people, are concerned that the loss of Medicaid eligibility by legal immigrants will increase their levels of uncompensated care.

In addition, the law makes it difficult for states and localities to provide any medical care other than emergency services to undocumented aliens by requiring states to pass a new law making them eligible for state or locally funded benefits. (Undocumented immigrants were, and remain, ineligible for full Medicaid coverage.) For FY 95, roughly 100,000 legal immigrants were eligible for Medicaid at a cost of $440 million. Undocumented aliens who received emergency services accounted for less than 1 percent of total Medicaid recipients and expenditures. (Unlike in California, Texas health care providers have not aggressively sought to obtain Medicaid reimbursement for "emergency" services by identifying undocumented aliens and are philosophically opposed to doing so.)

The political environment is not hostile to immigrants when state budget impacts are minimal. Governor Bush is viewed as moderately pro-immigrant (with a contrast often being drawn with Governor Pete Wilson of California) and lobbied Congress to restore the federal eligibility cuts enacted as part of the 1996 welfare bill. However, for the most part, he has not proposed offsetting the federal cuts with new state programs or using excess state welfare funds to fund services for immigrants, preferring to propose property tax relief instead. In an exception to that general policy, in October 1997, Governor Bush announced that Texas would provide up to $18 million in food assistance to about 28,000 elderly and disabled immigrants who cannot work and whose food stamp benefits have been cut off by the federal government. Perhaps because Texas is a border state, immigrant advocates in the state have traditionally focused on issues involving the Immigration and Naturalization Service and are less familiar with welfare and health issues.

There has been little discussion of or advocacy for providing benefits or services to new immigrants; most of the debate has centered on continuing to provide for current immigrants. Texas has opted to continue eligibility for Medicaid, TANF, and Title XX/Social Services Block Grant services for qualified aliens in the country as of August 22, 1996. The state has, however, decided to extend the ban on TANF for new immigrants until citizenship, rather than limiting it to the five years after entry into the United States.

In addition to restoring cuts in SSI to certain legal immigrants, the Balanced Budget Act of 1997 guarantees continued Medicaid eligibility for immigrants currently receiving Medicaid on the basis of SSI eligibility. Texas is one of only five states where immigrants losing SSI would not otherwise have retained Medicaid through another existing eligibility category. However, prior to passage of the federal act, the 1997 Texas legislature authorized creation of a new Medicaid category for legal immigrants losing SSI. Because benefits for this group are now restored, the state has not implemented this category. One group whose benefits were not restored by the Balanced Budget Act is legal immigrants already in the United States before August 22, 1996, who are elderly but not disabled. This group is likely to be ineligible for both SSI and Medicaid in Texas, unless the state implements a new Medicaid category authorized by the 1997 state legislature. A relatively small number of legal immigrants falls into this category, and it may be difficult to bring the issue to the attention of state policymakers until more immigrants age into this group.

Other Public Financing Programs

Texas has few programs to provide health insurance for the low-income population, except for Medicaid. For example, Texas does not have a general assistance-medical program for low-income people ineligible for Medicaid. In addition, although Texas has had a high-risk insurance pool since 1989, no funding has ever been approved and it has never become operational. Instead, most uninsured receive care through public hospitals operated by the counties. Counties that do not run public hospitals are eligible for funds through the very small County Indigent Health Care Program. For further discussion of public hospitals and the County Indigent Health Care Program, see the section on "Delivering Health Care to the Uninsured and Low-Income Populations."

During 1996 and 1997, the state debated a variety of initiatives to expand health insurance for children not eligible for Medicaid. Few of the proposals included new state spending or new taxes, limiting the potential impact of planned initiatives. In spring 1997, a law was enacted establishing the Healthy Kids Corporation, which will help insurance companies sell health insurance for children whose parents are not poor enough to qualify for Medicaid. The measure will encourage insurers to cover children through tax breaks and fewer benefit mandates. A major theme of the debate was that providing health insurance to children would save money for employers and Medicaid. Over the next few years, policy initiatives will likely focus on implementation of the Balanced Budget Act's State Children's Health Insurance Program.

Insurance Reforms

Although uninsurance rates are high, a substantial portion of the population below 200 percent of the federal poverty line has private health insurance. Like most other states, Texas has adopted insurance reforms involving rules of issue and rating restrictions for the small-group market (between 3 and 50 employees). However, no individual market reforms have been enacted or are under active consideration.

For the small-group market, the waiting period on preexisting condition exclusions is limited to 12 months for conditions diagnosed 6 months prior to obtaining new health insurance coverage. In addition, a portability provision requires insurers to credit waiting periods if the condition was covered under a previous policy, that is, if there was no gap in coverage longer than 60 days. Moreover, Texas businesses cannot be denied a group health policy because one or more employees suffers from a medical condition. Texas has also established rating bands, with adjustments for age, gender, industry, geographic area, family composition, and group size. Among the nine business classes established, variation across classes is limited to 20 percent and variation within classes is limited to 25 percent. In addition, premium increases at renewal are capped at 15 percent.

In 1993, the Texas Insurance Purchasing Alliance was established to provide health insurance coverage to employers with 3 to 50 workers. Benefits offered through the Texas Insurance Purchasing Alliance were reconfigured in 1996 because relatively few employers had taken advantage of the offerings, claiming that the plans were too expensive or offered too few benefits. Other recent insurance reforms include a 1995 law that authorized a Medical Savings Account pilot program for Medicaid recipients, but it has not been implemented.


Financing and
Delivery System

Managed care and the growth of for-profit hospitals is substantially altering health care financing and delivery in Texas. Overall, managed care is growing, and the state is moving steadily to enroll Medicaid beneficiaries into plans. Medicaid hospital reimbursement is fairly generous, especially when combined with DSH payments, but rate setting is increasingly done by the HMOs rather than the state.

The Health Care Market and Access for Low-Income Populations

The health care market is changing rapidly in Texas. Enrollment among preferred provider organizations (PPOs) and HMOs has grown at the expense of indemnity coverage. Simultaneously, for-profit hospitals have made significant inroads in Texas and account for one-third of hospitals in the state. In response, hospitals and physicians have attempted to become more cost competitive and consolidate services. To date, managed care, conversions of nonprofit hospitals to for-profit organizations, and consolidations have not had a discernible impact on access to care for low-income populations.

Managed Care

Growth among PPOs has been rapid, from 80 PPOs in 1995 to 120 in mid-1996. The Texas Department of Insurance estimates that about half of the indemnity market is now in PPOs. HMO enrollment grew from 1.4 million members in 1991 to 2.2 million in 1995. 14 Of the state's 39 HMOs, only 5 were operated by nonprofit organizations, including one public hospital. There has been considerable competition and consolidation among HMOs. The state reports that net income after taxes per HMO member per month declined by about 60 percent from 1993 to 1995. 15 State officials expect a market shake-out, with mergers and consolidations likely.

HMOs have been affected by several trends in Texas, some of which would tend to hinder and others that would tend to promote HMOs. Trends that might promote further entry of HMOs include the expansion of Medicaid managed care, increases in HMO enrollment within self-insured plans, and relatively high Medicare payment rates in some parts of Texas. However, negative publicity about the quality of care in HMOs—combined with the relatively low political clout by managed care organizations with state legislators—has translated into stricter regulation of managed care in Texas compared with other states. Physicians in Texas are strongly opposed to managed care and have fought for rules to make operating an HMO in the state less attractive. Although a "patient protection act" promoted by the state medical society was vetoed by Governor Bush in 1995, the governor did order that many patient and provider protections be instituted by regulation. These regulations address emergency room services, continuity of care, out-of-network referrals, balance billing, and member information. Additional legislation required the state to collect extensive patient care and quality data from HMOs. In 1997, Texas became the first state to enact a law allowing patients to sue HMOs for malpractice. Aetna is suing the state over this legislation.

Further hindering the development of HMOs is a legal provision that bars the "corporate practice of medicine." Under this ban, nonphysicians may not own physician groups. In response, entities have emerged called 501(a) nonprofit medical corporations, with physician-only boards of directors. 16 The 501(a) entity is the only organization in Texas that can contract directly with an HMO for both physician and hospital services on a capitated basis. Otherwise, risk-bearing entities are subject to full HMO licensure and regulation. Consequently, hospitals, physician management companies, and other medical groups have formed 501(a) organizations in order to contract with HMOs.

Physicians and Hospitals

Both physicians and hospitals have been affected by the growth in managed care that has resulted in pressure on provider rates and competition for patients. Although Texas has a strong tradition of solo physician practice, managed care has changed the face of physician practice. According to a 1996 survey of Texas physicians, HMO and PPO patients accounted for 31 percent of all patients seen in physician practices, up from 20 percent in 1994 and 15 percent in 1992. 17 Forty-two percent of surveyed physicians were members of physician-hospital organizations, 41 percent were members of independent practice associations, 11 percent were members of 501(a)'s, and 10 percent were members of physician practice management organizations. Group practice is viewed as necessary in the face of HMOs and employer purchasing groups. From 1990 to 1996 the proportion of solo practitioners has dropped from 50 percent to 40 percent. 18 Network formation is uneven geographically and generally follows managed care trends.

Hospitals are also undergoing change. Texas has roughly one-third each nonprofit, public, and for-profit hospitals. Its share of public and for-profit hospitals is greater than the national average. Two national for-profit chains, Columbia/HCA and Tenet, have a strong presence in Texas. Columbia/HCA has acquired 71 hospitals, some of which were formerly nonprofit hospitals; at least 3 were subsequently closed. (It is unclear what effect the criminal investigation of Columbia/HCA and the subsequent management changes ultimately will have on the company's Texas operations.) Recent high-profile mergers among nonprofits have been proposed between Harris Methodist and Presbyterian Healthcare Resources in North Texas and Hermann Hospital and Memorial Healthcare System in Houston. Blue Cross and Blue Shield of Texas has merged with Blue Cross and Blue Shield of Illinois. Some officials complain that the state receives only "paper boy" notification (i.e., they read about it in the newspaper) of these mergers and conversions. Texas does not require that the acquiring hospital maintain the level of charity care previously provided.

Like physicians, hospitals are responding to managed care pressure in a variety of ways. Specific activity depends on whether an area has excess hospital capacity and on the extent of pressure from for-profit hospitals. Cutting costs, laying off staff, developing integrated networks, shifting care to outpatient settings, creating partnerships with HMOs, or even pursuing HMO licensure are common hospital responses. Consolidations and for-profit conversions are other responses. In some cases, a chain such as Columbia/HCA will build a new specialty hospital rather than acquire an existing one. Columbia/ HCA was planning a children's hospital of its own in Houston to compete with the existing facility but has recently decided against the project.

Medicaid Managed Care

Texas began its move into Medicaid managed care in 1992 with two modest pilot programs. Through these pilots the state was able to work out some implementation issues, such as plan selection and contracting criteria, beneficiary enrollment, plan capacity, and payment. In May 1995 the state passed legislation to vastly expand mandatory managed care as part of a larger Medicaid reform initiative that also would have expanded eligibility. This legislation, Senate Bill 10, was adopted at a time of great uncertainty over Medicaid block grants and federal DSH policy.

While block grants faded from the national policy discussion, the state continues to phase in mandatory managed care enrollment, one geographic area at a time. In 1996, mandatory managed care was expanded to three additional geographic areas, and two more are planned for 1997. Most of the state will be phased in by 2000. State officials emphasized the potential of managed care to improve access and provide a "medical home" to Medicaid beneficiaries. However, in the view of many, the state's emphasis has increasingly been on cost containment and the desire to "privatize" the Medicaid program.

In 1995, the prospect of expanded Medicaid eligibility through the demonstration waiver assuaged concerns among consumer advocates and providers about the state's managed care plans. However, their support quickly waned when the state proceeded to implement mandatory managed care through a freedom-of-choice waiver without eligibility expansions. In fact, two hospital districts, Tarrant County and Lubbock, and an HMO formed by community health centers, sued—unsuccessfully—to stop the mandatory managed care program.

Under its managed care program, Texas enrolls AFDC and related populations (pregnant women and children) in managed care on a mandatory basis. Persons with disabilities may enroll in managed care voluntarily, but dually eligible (Medicaid-Medicare) beneficiaries are barred due to the administrative complexity of coordinating between the two programs. In 1997 the state began developing a separate program combining acute and long-term care coverage for dually eligible beneficiaries in Houston. (See the section on "Long-Term Care for the Elderly and Younger Persons with Disabilities.")

Managed care plans must cover most Medicaid-defined services, including special services for children, but not dental services or long-term care. Plans that offer benefits beyond those normally covered by the state are favored during the selection process. In particular, plans receive extra points for innovative coverage of behavioral health services. Beneficiaries enrolled in HMOs have coverage for unlimited prescriptions, whereas fee-for-service Medicaid has a three-per-month prescription limit for adults. In addition, HMOs must cover a wellness exam and may not limit inpatient coverage to 30 days per spell of illness, as is the case under Medicaid fee-for-service. HMOs are permitted to use their capitation payment to pay for services not covered by Medicaid, for example, home modifications needed because of disabilities. However, these services are not factored into the payment rates. The original pilot program included a provision to extend HMO coverage for six months, even if a beneficiary lost Medicaid eligibility. Because the cost must be borne by the state without any federal Medicaid contribution, this provision was dropped. 19 Managed care organizations must contract with traditional safety net providers (e.g., public hospitals and community health centers) for three years.

The original managed care pilot programs used a freedom-of-choice waiver to implement mandatory enrollment in Travis County (around Austin) and in three contiguous eastern counties, Jefferson, Chambers, and Galveston. In Travis County the state contracted with an HMO and a prepaid health plan that was partially capitated. In the tricounty pilot, the state tested a primary care case management (PCCM) model, in which beneficiaries are linked to a primary care provider who must provide most care and approve referrals to specialists. The state chose a PCCM model in the tricounty area because it is largely rural and beneficiaries were considered to have high, potentially inappropriate, emergency room use.

For a number of administrative reasons and political pressure from providers and advocates, the schedule for the implementation of managed care throughout most of the state has slipped somewhat. The next phase involves mandatory enrollment around Houston, the largest urban area in Texas. Dallas will follow in 1998.

Plan Selection and Payment Rates

Since the Medicaid pilot programs in 1992, commercial managed care has expanded greatly in Texas, allowing the state to contract with more plans in each area. In addition to Austin and the eastern counties, mandatory enrollment has been implemented in and around Tarrant County (Fort Worth), Bexar County (San Antonio), and Lubbock County (Lubbock). Although Texas continues to operate a PCCM program in the eastern counties, it prefers to contract with commercial HMOs. However, the state was forced, due to political pressure from physician groups, to implement a PCCM program in Bexar and Lubbock counties, giving beneficiaries the choice between an HMO and a primary care physician network.

Except for PCCM programs, the state will only contract with state-licensed HMOs. While the state does not completely rule out Medicaid-only plans, it has a strong preference for commercial HMOs as a way of mainstreaming Medicaid beneficiaries. The state has not lacked bidders, who see Medicaid contracts as a way of increasing their market share.

The selection of plans to participate in Medicaid managed care has been controversial in Texas. Some public hospitals and clinics formed HMOs, motivated by a desire to retain Medicaid enrollees. However, these plans have not been very successful in winning contracts. One HMO formed by a network of community health centers bid unsuccessfully for contracts in three areas in 1996. Although the state selected an HMO formed by the hospital district in San Antonio, it did not award a contract to the HMO formed by the Harris County (Houston) hospital district because it did not score well on the selection criteria. The failure of the principal Medicaid provider in Houston to obtain a managed care contract could be financially crippling to the hospital district. 20 This decision precipitated state legislation in 1997 requiring the Medicaid program to contract with hospital district-sponsored HMOs. The downside of this contracting guarantee is that the state may lack leverage in requiring improvements in these public HMOs.

Plans must bid on individual service areas and are selected using a quasicompetitive process. First, applications are assessed with regard to minimum requirements set out in the application request; they also must meet the regulatory requirements of the Department of Insurance and the Department of Health, which apply to all HMOs operating in the state. Medicaid-contracting plans must respond to required time and distance limits between patients and their designated physicians, emergency care definitions, case management and specialty care for people with chronic conditions, requirements with respect to family planning providers, and a requirement that plans make reasonable efforts to contract with community health centers. The state has established a point system by which plans' applications are scored. Plans are given points for increasing quality, timeliness, and scope of services and specific behavioral health initiatives. 21

In the application request the state also sets maximum capitation rates for AFDC adults, AFDC children, pregnant women, newborns, and two categories of non-AFDC children. Plans must bid at or below the maximum. Maximum capitation rates for a given area are determined using fee-for-service data and data from the pilot programs and typically are set at 90 percent of estimated fee-for-service costs. The fee-for-service costs include costs of graduate medical education but not DSH. Rates are not equal across plans because of differences in proposed coverage. There also is a profit-sharing arrangement between the state and the HMOs. HMOs may keep a percentage of the profits, but the remainder must be shared equally with the state. On the other hand, there are no Medicaid-specific stop-loss or other financial protections for plans; reinsurance is not offered. Limits on risk-sharing within the plan are set by laws and regulations that apply to all plans. In addition to the capitation payments, plans can receive up to $2 per member per month if the plans achieve certain performance targets established in the contract. The performance objectives address EPSDT screens, immunizations, adult annual visits, pregnancy visits, and health education. For enrollees who are blind or disabled, plans are reimbursed on a fee-for-service basis, plus a $14 per month case management fee.

In the PCCM program, primary care providers are paid the state's Medicaid fee-for-service rates plus a $3 per month case management fee. In the future Texas hopes to conduct more aggressive referral management in order to lower costs and perhaps develop a specialty provider network. In addition, the PCCM program is used as an opportunity to renegotiate hospital reimbursement rates. The state negotiates rates with hospitals in the PCCM area on a facility-by-facility basis, rather than paying the state's usual diagnosis-related group (DRG)-based rates (which are discounted through selective contracting in some areas).

Enrollment and Implementation Issues

In some areas the state has experienced start-up difficulties because many beneficiaries did not select an HMO or PCCM provider as required. If beneficiaries do not select a plan, they are automatically assigned to one. The auto-assignment procedure takes into account a client's claims history and makes an assignment based on prior primary care provider, geographic location, the health plans of other family members, and special needs. Auto-enrollment is on the order of 40 to 50 percent in some areas, including Austin (despite the area's experience with Medicaid managed care). This raised concern in the state that some beneficiaries are not sufficiently aware of their rights and obligations under the mandatory managed care program.

To protect beneficiaries from aggressive marketing pressure, HMOs may not conduct "direct contact" marketing, although they may distribute approved written materials. 22 Instead, the state has contracted with a broker to conduct education and enrollment. Both state officials and advocates alike felt that the education and enrollment materials were inadequate.

Oversight of HMOs occurs through the departments of health and insurance. Overall quality oversight conducted by the health department was transferred in 1996 to the insurance department, which is also responsible for licensure. Only a small staff is assigned the numerous tasks of overseeing a Medicaid managed care program, for example, quality monitoring, contract monitoring, and patient satisfaction surveys. The state feels pressure to build up its managed care administrative capacity, but it faces very tight administrative budgets. In some cases, particularly capable state staff have been hired away by managed care organizations. Recognizing this problem, the state is considering contracting with an outside organization, possibly an accreditation agency, to monitor quality of care.

Medicaid Provider Payment and Disproportionate Share Hospital Policy

Medicaid is a relatively generous payer of hospital care; physician care is less well reimbursed. For hospital care, Medicaid is thought to be a better payer than many managed care organizations. In addition, the state has very aggressively maximized Medicaid DSH payments by using intergovernmental transfers from local hospital districts. Thus, safety net providers are reimbursed quite well.

Medicaid Hospital and Physician Reimbursement

In 1991 the Texas Hospital Association sued the state Medicaid program under the federal Boren amendment, arguing that Medicaid reimbursement rates were too low. 23 After a settlement in 1993, payment rates improved. According to data from the American Hospital Association, Medicaid paid hospitals (including DSH payments) 76 percent of costs in 1991 and 117 percent in 1993. 24 Medicaid rates, in fact, are considered quite favorable by hospitals compared to private payers, where competition and managed care is growing. In many local areas, competition for Medicaid patients, particularly in obstetrics and pediatrics, is intense. Protection of hospitals through the Boren amendment is viewed as less and less relevant because of Medicaid managed care, hospital selective contracting, and DSH. 25 In contrast to the repeal of the Boren amendment, federal policy changes that might eliminate or reduce DSH funds were considered potentially very destabilizing to the hospital industry.

Most hospitals, except children's specialty hospitals and psychiatric hospitals, are paid by Texas Medicaid on the basis of DRGs. The state makes a flat payment based on a patient's diagnosis category, regardless of the actual length of a patient's stay in the hospital. In an effort to save funds under Medicaid, the state in 1995 introduced selective contracting in 24 metropolitan areas for nonspecialty hospitals that were not state owned. Under selective contracting, hospitals were asked to offer the state a discount on their standard payment rates. The state intended to contract with the lowest-bidding hospitals (i.e., those offering the greatest discounts) and exclude the others from participating in Medicaid. In fact, the state was very lenient in its contracting, selecting 220, or all but 20, Medicaid-participating hospitals. Ironically, hospitals that offered substantial discounts in anticipation of gaining additional Medicaid volume were disadvantaged because those hospitals did not see more patients.

Physician payment is not as favorable as hospital payment in Texas, and rates have been frozen since 1993. Physicians were paid 88 percent of Medicare payment rates, based on an index of 28 services, in 1993. 26

Disproportionate Share Hospital Payments

Beginning as a small program in 1986, the Medicaid DSH payment program in Texas is now very large, totaling $1.5 billion, or 17 percent of program expenditures in 1995. In 1996, 25 public hospitals contributed DSH funds through intergovernmental transfers for the state share of the Medicaid match, while approximately 160 hospitals qualified for DSH payments. The contributors include 10 local public hospitals (9 hospital districts and 1 city hospital) and 15 state-owned institutions (10 mental health/mental retardation facilities, 2 hospitals operated by the health department, and 3 university hospitals). Hospital districts are looked upon by the state as a favored source of funds to expand Medicaid.

Unlike some DSH programs in other states, where federal funds have been diverted to the state treasury, the DSH program in Texas has been used to provide financial support to local hospitals serving a large number of Medicaid and uninsured patients. All hospitals receiving DSH funds must formulate a local health needs assessment and address how DSH funds will be used to meet those needs. About two-thirds of the hospitals receiving funds are located in rural areas. Changes in federal law in 1993 that limit DSH payments to any individual hospital to what it spends on uncompensated care have shifted the distribution of funds away from state-owned hospitals toward other institutions. Consequently, the amount that the state facilities can contribute for federal matching funds has declined. Some contributing hospital districts believe it is unfair that "their" money—money that they have imposed taxes to raise—is used to fund hospitals in other areas of the state and, in some cases, competing private hospitals.


Delivering Health Care
to the Uninsured and
Low-Income Populations

Efforts to provide health care to the uninsured and low-income populations in Texas face enormous difficulties. Because the Medicaid program covers a relatively small proportion of the low-income population and private insurance coverage rates are low, uninsured rates are high. Moreover, aside from having many uninsured persons, some parts of the state lack even basic public health services. The state health department plays a minimal role in health care for low-income populations outside of the Medicaid program. To fill this void, Texas depends largely on local health departments and hospital districts to provide health care for the uninsured. Houston and El Paso provide examples of how local communities cope with this responsibility and how changing market conditions threaten traditional arrangements.

State and Local Health Departments

Outside of administering the acute care portion of Medicaid, the state health department does not see its role as funding additional service delivery programs for low-income people. The state appropriates few funds outside of matching funds required in order to receive specific federal grants. Non-Medicaid health programs—including disease prevention and control, environmental health, vital statistics, family planning, services for children with special needs, and maternal and child health services, but excluding the Supplemental Food Program for Women, Infants, and Children (WIC)—are allocated about one-tenth the funds that Medicaid receives.

Efforts were made over the past few years to maximize Medicaid funds and obtain federal matching payments for health care that otherwise would be entirely state funded. However, frequently the additional federal match is used to reduce the state budget rather than to expand existing programs. As a result, over the past few years, funding for non-Medicaid programs has been flat or reduced. Moreover, most of the public health initiatives, such as increasing the number of children receiving health screens, involve Medicaid rather than traditional state-only funding.

The state health department operates through eight regional offices or through contracts with local health departments. Regional offices provide clinical and regulatory services only in areas that choose not to have local health departments, mostly sparsely populated areas. In some areas the regional office complements services provided by local health departments. The state estimates that approximately four-fifths of clients receive state-supported services through local health departments, one-fifth through state regional offices.

County commitment to public health varies greatly. Only counties receiving funds from the state are subject to state requirements. Others elect not to pursue state funding, and the state has very little information and policy control over them. Only 69 local health departments receive funding from the state and are termed "participating local health departments." Local entities enter into program contracts with the Department of Health to provide various services, for example, maternity care, WIC, and screening for sexually transmitted diseases. Populations served with these funds must be low income; immigration status is not used as an eligibility criterion and is rarely checked. In order to receive primary care and now maternal and child health funds, local health departments and other agencies must compete for grants. Awardees must offer 24-hour services and a regular provider for patients—a requirement that most local health departments cannot currently meet. Previously, local health departments were allotted funds on a population basis. The change has caused great concern among health departments providing important, albeit minimal, services such as immunizations and health screens. Many local health departments worry that they will no longer qualify for state grants to continue those services.

Local health departments also are expected to undergo change as a result of Medicaid's shift to managed care. Most health departments had become Medicaid providers during relatively recent efforts to maximize Medicaid reimbursement. Now health departments are expected to move away from direct service delivery and Medicaid dependence and perform system planning and coordination functions. Some may form relationships with managed care organizations. In Austin, for example, the city health department is a subcontractor to a commercial HMO. In Galveston the health department is a PCCM provider. Others may withdraw from patient services and focus only on community and environmental health.

Hospital Districts

Counties must have a program to serve the medically indigent. Most counties fulfill this requirement by forming hospital districts, which have property taxing authority and use the funds they collect to operate public hospitals and, in some cases, clinics. In 1995 there were 102 hospital districts in Texas. Counties without hospital districts are eligible for funds through the County Indigent Health Care Program. 27 However, the program is very small—less than $6 million is budgeted for 1998—and operates in only 17 counties. In addition, the state operates a few university-based hospitals, which serve significant numbers of Medicaid and uninsured patients. Of particular note is the University of Texas Medical Branch in Galveston, which is viewed as the absolute provider of last resort and which draws patients from all over the state.

Most public hospitals are at a distinct competitive disadvantage. In fact, their ability to serve commercial patients is restricted, and many county officials believe that the public sector should not compete with private organizations. More recently, some counties have sought to limit their role in supporting indigent care through public hospitals. Three hospitals have been shifted to private management (in Austin, Corpus Christi, and Amarillo), one hospital district facility was contemplating a merger with a nonprofit hospital (in Fort Worth), and another public hospital was "for sale" (in Lubbock).

State government does not have an explicit policy to shore up these safety net hospitals, aside from Medicaid DSH funds. State officials see their major responsibility as being the Medicaid program and its beneficiaries, even when it is at the expense of public hospitals. For example, although Medicaid-contracting HMOs are required to enter into agreements with traditional safety net providers for three years, enrollment of Medicaid beneficiaries in these plans is likely to reduce utilization of these hospitals, both because HMOs seek to reduce hospital utilization and because patients are channeled to other facilities because of lower payment rates. Public hospitals tend to have higher costs because they are teaching institutions, may see sicker patients, serve the uninsured, and may be inefficient because of constraints imposed by public management. If public hospitals see fewer Medicaid patients, that will significantly lower third-party revenue to these institutions. With the exception of state-operated hospitals, the burden of lost revenues is then shifted to counties. Counties may have to raise taxes, which is difficult, or reduce services.

Impact of Government Policies and Market Changes on
Safety Net Providers: Houston and El Paso

In order to assess the impact of government policies and market changes on safety net providers, public hospitals, teaching hospitals, community health centers, and other providers were interviewed in Houston and El Paso. Houston, located in Harris County, is the fourth-largest city in the country, with substantial black, Hispanic, and immigrant (both Hispanic and non-Hispanic) populations. Although there is a large low-income population, Houston/Harris County is generally prosperous. In contrast, El Paso is a medium-size, very poor city with high unemployment. Situated on the border, it is overwhelmingly Hispanic and is dominated by its relationship with Mexico. While Houston is very much at the center of state concerns, El Paso is distant geographically and very marginal politically to Austin.

In both areas, the hospital district plays a central role in delivering care to Medicaid and low-income populations, but other hospitals also have large uncompensated care loads. Competition has increased for the Medicaid population, and more Medicaid patients are being seen in the private hospitals. In both cities, safety net providers expect managed care to dramatically alter the system of care.

Health care in Houston is characterized by a large number of tertiary teaching hospitals (including three private academic medical centers, a large tertiary-care public hospital system, a children's hospital, and a state-owned cancer institute), with an oversupply of hospital beds and physicians. Many of these institutions are located adjacent to one another in the center of the city. They are all affiliated with two medical schools, Baylor College of Medicine and the University of Texas at Houston, and are committed to the three-fold mission of patient care, education, and research. Houston has a high proportion of for-profit hospitals, including 19 owned by Columbia/HCA, which have a reputation for being tightly managed and aggressively competing on price. Managed care came rapidly to Houston and, by 1996, reached 16 percent of the population. 28 Access to care is generally considered good in Houston; however, access in certain low-income pockets of the city and to basic preventive care is poor. 29

In contrast, El Paso has poor or limited access to care in many areas, the most extreme of which are the colonias, which have limited or no access to potable water, waste water disposal, and paved roads, and minimally acceptable housing. Managed care has not reached El Paso to any significant degree, although it was anticipated to enter the market in the near future. For-profit hospitals are extremely important in El Paso; as a result of recent mergers, El Paso has one public hospital and four for-profit hospitals. Two national chains, Columbia/HCA and Tenet, each operate two hospitals.

Hospitals and clinics in Houston and El Paso demonstrated great diversity in financial status, in competitive position and bargaining power, and in the sophistication of their responses to market changes. Medicaid is viewed as a desirable payer in both areas, and increased competition for Medicaid patients has drawn these patients away from public and traditional safety net providers to other hospitals.

Both areas anticipated changes due to Medicaid managed care, although Houston will be affected earlier. Hospitals emphasized the importance of capturing Medicaid managed care enrollment early on, to avoid losing existing patient bases and facing barriers to entry later. Hospitals in Houston were especially active, applying for HMO licenses, forming networks with other hospitals and physician groups, expanding primary care capacity and clinics, cutting costs, and preparing to accept risk-based contracts.

Although currently in relatively good financial shape, the public hospital system in Houston is vulnerable in regard to Medicaid managed care. Although it has expanded its clinic network and obtained an HMO license, it has no commercial managed care contracts and is heavily dependent on Medicaid for its revenue. In 1997, the public hospital HMO failed to win a Medicaid contract, the financial consequences of which could have been catastrophic for the institution. In response, the state passed a law requiring Medicaid to contract with licensed HMOs formed by hospital districts.

In El Paso, the hospital district has benefited from large DSH payments. However, it is even less well prepared for managed care than Houston's hospital district. El Paso's public hospital has almost no outpatient capacity and has strained relationships with Texas Tech Medical School, which provides almost all physician services, and with the county commissioners, who approve tax revenues. In contrast, private hospitals in El Paso have made concerted efforts to form integrated care systems and cooperative relationships with local physicians.

Adding to the uncertainty in the health care environment of El Paso and Houston is that Columbia/HCA is the target of a nationally publicized federal criminal investigation, which began in El Paso. Although the investigation is very wide-ranging, the principal focus has been on possible fraudulent Medicare billing and on the financial relationship of physicians with Columbia/ HCA. As a result, several top managers at Columbia/HCA have resigned and others have been indicted.

In Houston, the local hospital district operates a very large network of clinics, leaving a smaller role in personal health services for private health centers and the city and county health departments, the latter which have focused on preventive services, maternal and child health, dental services, sexually transmitted diseases, and AIDS. A relatively small proportion of revenues for the county and city health departments is derived from Medicaid. The local health departments are grappling with their new role in a managed care environment and are investigating the possibility of subcontracts with the local hospital district and commercial HMOs.

In contrast, private community health centers play a more important role in El Paso. Clinics at the medical school have long waiting lists for appointments, and the public hospital has little outpatient capacity. At this point, the community health centers have no experience with managed care and are concerned about the potential "culture clash" between bottom line-oriented, for-profit HMOs and the "serve-people-in-need" philosophy of the health centers. The combined city and county health department has sustained major budget cuts in recent years that have forced clinic closures. The department cannot meet some of the state's requirements for funding, such as providing 24-hour care. With so few resources available, the potential loss of state funds to the El Paso public health department was viewed as of more immediate concern than changes that might ensue due to Medicaid managed care.


Long-Term Care for the
Elderly and Younger Persons
with Disabilities

Long-term care for the elderly and younger people with disabilities is a critical component of the state's involvement in health care and plays a major role in the Medicaid program. While delivery has historically been dominated by a large number of relatively low-cost institutions, Texas is attempting to alter the balance between institutional and noninstitutional services. Quality of care has been a consistent concern.

Supply, Utilization, and Expenditures

Given its large population and huge territory, it is not surprising that Texas has a very large number of long-term care providers and spends a lot, at least in absolute terms, on long-term care. As shown in table 6, Texas spent $2.2 billion on Medicaid long-term care for the elderly and younger persons with disabilities in 1995. The vast majority of expenditures are for institutional care, although expenditure growth in recent years has been much faster for home and community-based services.

The state has a large supply of nursing homes and home health agencies and a slowly shrinking supply of institutional mental health and mental retardation beds. In 1995, Texas had approximately 130,000 nursing facility beds (68 beds per 1,000 elderly compared with the national average of 53 beds per 1,000 elderly), which is the second largest nominal number of beds of any state in the country. In 1995, the state had 2,399 licensed home health care agencies. The state runs 9 state mental hospitals, 12 state schools for persons with mental retardation/developmental disabilities, 5 state centers (two serve only persons with mental retardation; three also serve persons with mental illnesses), and 36 community mental health and mental retardation centers.

Long-Term Care for Older People in Texas

Long-term care for the elderly in Texas has been dominated historically by institutional care. In an effort to alter this, the state's policy initiatives center on financing and delivery system reform that would increase the supply of home and community-based services. It is also planning a major demonstration that will integrate acute and long-term care services through managed care. At the same time, it has relied heavily on traditional methods of controlling long-term care expenditures: a moratorium on new construction of nursing homes and low nursing home reimbursement rates. Partly as a result, quality of care in nursing facilities is a major issue that has recently been addressed in the legislature.

Medicare Maximization and Private-Sector Initiatives

Unlike some states, such as New York, Texas does not have major initiatives to reduce Medicaid spending by maximizing Medicare nursing home and home care revenues or by promoting private-sector initiatives. For example, there are no major initiatives to encourage the purchase of private long-term care insurance. Despite the requirement of the Omnibus Budget Reconciliation Act of 1993 that the state operate an estate recovery program, the state does not, because to do so would mean certain "political death." (Earlier proposals by one state legislator to establish estate recovery led to his electoral defeat.) The state believes that the transfer of assets in order to qualify for Medicaid nursing home benefits is a problem, but there are no major efforts under way to prevent it.

Home and Community-Based Services

Although not nationally known as a model for its provision of noninstitutional services, Texas, in fact, finances a substantial amount of home and community-based services. Texas Medicaid covers personal care, has a large home and community-based care waiver, and is one of the few states to participate in the Medicaid "frail elderly" option. A majority of long-term care beneficiaries receive home and community-based care rather than institutional care. In a recent study by Ladd et al., Texas ranked eighth among the states in 1992 in "commitment" to home and community-based services. 30 While officially supportive of widening the continuum of care, the nursing home industry contends that home and community-based services are not cost effective.

Demand for Medicaid home and community-based care waiver services far exceeds the number of funded slots, and there are long waiting lists. The number of persons budgeted to receive services is considerably below the number of slots approved by HCFA. Given excess demand, the state wants to ensure that persons receiving services are at high risk of institutionalization, a strategy which has been controversial among consumer and home care advocacy groups.

Managed Care and Long-Term Care

In the Medicaid managed care waivers that have been implemented in Texas to date, the elderly population needing long-term care has been excluded. However, the state is developing a model program—known as the STAR+Plus Integrated Care Project—that will enroll 50,000 aged and disabled Medicaid beneficiaries in Houston in managed care plans that are responsible for both acute and long-term care. While the state cannot require dually eligible (Medicare-Medicaid) beneficiaries to receive their Medicare benefits through the HMOs, the state is seeking to provide incentives for the aged and disabled to enroll. Specifically, building on the fact that many Medicare HMOs in Texas offer extra benefits with no additional premium, Medicaid beneficiaries enrolling in the program will obtain additional prescription drug and eyeglass benefits beyond what is currently covered under Medicaid. HMOs appear to be interested in enrolling dual eligibles in the pilot project, according to some observers, because long-term care clients are typically more middle-class than other Medicaid beneficiaries. The high costs of long-term care necessitate that many middle-class elderly rely on Medicaid to finance their nursing home care. Given the lack of managed care organizations' experience with long-term care, long-term care providers are "nervous" about the reimbursement rates that HMOs will pay, particularly for nursing home care.

Nursing Home Care

While the state has kept Medicaid nursing home rates fairly low and has a moratorium on new construction, it has not tried to restrict eligibility. Quality-of-care deficiencies, according to some, are a significant problem. Politically, the nursing home industry is a major force in the state, especially on Medicaid issues, and its views must be taken into account when policies are proposed.

Nursing home eligibility.

The state has not attempted to tighten admission criteria for nursing homes as a way to control nursing home costs, despite home care advocates' contention that approximately 25,000 nursing homes residents have low levels of disability and do not need to be institutionalized. In response, the nursing home industry argues that lower-disability nursing facility residents still need care, and savings are not possible by shifting to home-based services. In 1994, the state tried to implement preadmission screening for nursing homes, but it was an administrative failure because no additional funds were allocated for staff.

Nursing home reimbursement and the Boren amendment.

Texas nursing homes are reimbursed by Medicaid on a case-mix adjusted, flat-rate basis. Each resident is classified into one of 11 categories, each with a separate statewide rate. In 1995, the average reimbursement rate was $56.17, well below the national average of $85.05. Indeed, the state has the third-lowest average rates in the country, although one expert contended that the average rates were depressed by the large number of low-disability (and, therefore, low-cost) nursing home residents.

A key issue is that facilities make profits in flat-rate reimbursement systems by spending less than they receive in reimbursements. This creates an incentive for low levels of service, which may result in poorer quality of care. The position of the Texas nursing home industry is that the state gets what it pays for: Facilities have low staffing ratios because the state is not willing to pay for higher staffing levels. At the same time, the state asserts that nursing homes do not always use the money properly when Medicaid rates are increased to improve staffing.

The state has been sued twice under the Boren amendment, and there has been one major threat of an additional suit. Most recently, in November 1996, the Texas Health Care Association sued the state over its nursing home reimbursement system, claiming that rates were inadequate to cover the federally mandated increase in the minimum wage. In February 1997 a settlement was reached that provided Texas nursing homes with about half of their requested rate increase.

According to state officials interviewed before the most recent lawsuit, federal nursing home reimbursement rules put the burden on the state to prove the adequacy of the rates, and the state cannot win in court. (In addition, the limited resources of the attorney general's office are no match for the law firms hired by the nursing home industry.) As a result, the threat of a lawsuit is just as effective as an actual lawsuit. Many state respondents felt that it is inequitable that nursing homes receive special protection whereas other providers do not. Speculating about what might happen if the Boren amendment were repealed (as it has been by the Balanced Budget Act of 1997), the nursing home industry believed that there are some general provisions in state law requiring adequate payment that it could use to prevent rate reductions.

Controlling the nursing home supply.

Nursing homes in Texas are significantly overbuilt, resulting in very low occupancy rates (only 76 percent in 1995); this situation has existed for many years. The state has had a moratorium on nursing home construction since 1985, but no certificate-of-need requirement. The nursing home industry, which supports the moratorium, would like to see the moratorium broadened to include hospitals that are trying to convert empty beds to nursing home beds.

Quality of care in nursing homes.

Quality of care in nursing homes has been a major focus of legislative activity in the last two years. Passed in the closing hours of the 1995 legislative session without much public scrutiny, HB 2644 prohibited the state from establishing nursing home standards that are different from those of the federal government for purposes of certifying facilities to participate in Medicare and Medicaid. The nursing home industry, which proposed the bill, contends that the state rules do not add much to meaningful quality-of-care regulation and state rules never figure into the federal certification decisions. Opponents say that it repeals a host of quality standards and makes adding onto the requirements much more difficult.

In September 1996, a committee of the Texas Board of Nursing Facility Administrators developed new, much stronger draft regulations. These rules were proposed after it was reported that the board had failed to discipline any nursing home administrators since 1993, including the 23 administrators working at homes where conditions were so bad that they were put under state control. However, opposition to the proposed regulations by nursing home administrators caused them to be scrapped.

Reacting to the controversy over HB 2644 and the publicity concerning the Texas Board of Nursing Facility Administrators, the legislature passed several new laws in 1997 that strengthened nursing home regulation. The prohibition against state standards being more stringent than federal law for certification for participation in Medicare and Medicaid was repealed, and many new standards were imposed, including a detailed listing of patients' rights. To aid consumers, a nursing home consumers' guide is to be created. Soon after these bills were signed by the governor, the state auditor released a report sharply criticizing the state's enforcement of nursing home quality standards.

Younger People with Disabilities

As with the rest of its health and welfare programs, Texas, historically, has not spent very much for mental health and mental retardation services. For example, in 1994, Texas ranked 48th in the nation in state mental health agency expenditures per capita, even after substantial expenditure increases between 1990 and 1993. State initiatives concerning long-term care for the younger population with disabilities reflect developments on four policy issues—fragmentation of responsibilities for services, the relationship between institutional and noninstitutional services, Medicaid maximization, and managed care.

Fragmentation of Services

Coordination among agencies responsible for serving younger persons with disabilities is difficult. Long-term care for younger persons with disabilities in Texas is funded by a variety of state agencies, including the Department of Human Services, the Department of Mental Health and Mental Retardation, the Department of Health, and the Texas Rehabilitation Commission. Each operates different home and community-based waiver programs for younger disabled persons. Moreover, because Texas is a big state, there is a lot of intrastate variation in how state policies are implemented. Advocates and some in state government believe that the Health and Human Services Commission is not providing the coordination that is needed. Although the Health and Human Services Commission is supposed to have policymaking authority, the line between policy and operations is often difficult to establish, and a great deal of authority remains in the operating departments.

Institutional and Noninstitutional Services

Like virtually all states, Texas is moving away from institutional care for its mental health and mental retardation/developmental disabilities (MR/DD) populations and toward more community-based options. A major impetus for this movement was two lawsuits, a class-action suit involving the state psychiatric hospitals (RAJ v. Jones) and a second class-action suit requiring improvements in services to individuals with mental retardation (Lelsz v. Kavanagh).

Community-based service funding as a percentage of total program funding of both mental health and mental retardation has increased steadily. 31 In 1983, just under 30 percent of mental health funding was for community programs; this will approach 62 percent in 1997. For mental retardation programs, community funding was slightly more than 10 percent of expenditures in 1983; in 1995, community funding represented about 50 percent of spending, which will continue to rise, reaching nearly 68 percent by 1997. Medicaid home and community-based waivers figure heavily in providing funds for services for persons with mental retardation. While services have expanded, there were 30,000 people on waiting lists for community-based MR/DD and mental health services. First priority is given to people who are or were in institutions, making access very limited for those people who have always received services in the community.

The census of state schools (which serve those with mental retardation or developmental disabilities) is down compared with historical figures, as is the number of persons in psychiatric facilities. In 1995, there were 6,156 individuals in state schools, which is projected to decline to around 5,300 by 1999; but remaining residents of the state schools are much more disabled and costly to serve. Although the trend is toward less institutionalization, there are vocal parent groups that are very strongly opposed to closing the state schools. Moreover, the state has more than 8,100 persons in community-based intermediate care facilities for the mentally retarded (ICFs/MR), utilization of which is not projected to decline. The number of persons in state psychiatric facilities is also projected to stay virtually flat in the coming years. Compared with the rest of the country, Texas residents are much more likely to reside in large facilities.

Despite Texas's movement toward community care, the state lags behind other states. For example, the state ranked 30th in Medicaid home and community-based care waiver outlays per participant in 1994. In terms of fiscal effort, which is spending compared with state wealth, Texas ranked 36th overall.

Medicaid Maximization

The state has a very conscious policy of maximizing Medicaid revenue for mental health and mental retardation services. In FY 95, Medicaid accounted for about 27 percent of the operating budget for TDMHMR. About half of the funds for MR/DD services come from Medicaid and other federal sources, while less than 10 percent of mental health spending is Medicaid or otherwise federal. TDMHMR's FY 95 Annual Report states that the legislature "expects TDMHMR to earn more Medicaid dollars, and it reduced the agency's FY 1996-97 general revenue appropriations accordingly." 32 Expansion funds are expected to come from billing Medicaid more aggressively.

Managed Care

There is a great deal of interest in managed care for people with disabilities. Beyond just capitation, the state is interested in learning from the private sector about utilization review, quality management, uniform assessment systems, and focusing on outcomes. Four new pilot programs are being implemented to provide mental health care for those on Medicaid. Under these pilots, a seriously mentally ill person will receive acute behavioral services from an HMO and long-term rehabilitation from the local mental health authority. Advocates are concerned that managed care plans will be unable and unwilling to adequately treat persons with chronic mental illness. Younger people with disabilities will also be enrolled in the STAR+Plus Integrated Care Project.


Continuity and Change in
Texas's Health Care Policy for
Low-Income Groups

Texas is a large, high-poverty state with a substantial number of foreign-born residents that has traditionally provided relatively minimal health and welfare benefits beyond what is required to receive federal funds. Texas culture values independence and private-market solutions to problems. Compared with those of other states, cash welfare payment levels are especially low. From the state's perspective, its modest benefit structure makes it unconcerned about becoming a "welfare magnet." Politically, there is no interest in raising state taxes to fund a richer set of services for the low-income population. Instead, in 1997 the governor and legislature agreed to increase state funds for education in order to reduce property taxes at the local level.

To partly compensate for the lack of health coverage at the state level, Texas has a large number of public hospitals and clinics funded at the local level by the property tax. Counties are constitutionally responsible for providing health care for the indigent. Unlike localities in California and New York, however, the counties are not involved in running the Medicaid or cash welfare programs. Texas hospital districts are major providers of health care for the indigent and Medicaid populations. As a result, access to care is better than one might imagine from looking solely at Texas's Medicaid eligibility levels.

This duality between the state and local governments creates a number of health policy dilemmas. For example, providing health care to the low-income population is viewed more favorably than cash welfare benefits, and the state is pursuing initiatives that would increase Medicaid eligibility. However, while the state is interested in raising Medicaid eligibility levels, it is unwilling to pay any more in state funds. As a result, Texas's strategy depends entirely on using intergovernmental transfers from local entities to draw down additional federal Medicaid dollars. For their part, local hospital districts are concerned about losing control of their funds and having the state use their money to pay providers not supported by the hospital district. Thus, in some cases they have been reluctant to provide the funds necessary to obtain the additional federal Medicaid match. Some at the federal level may be concerned that these initiatives may result in merely a substitution of federal for local funds and little actual expansion of access to or improvement in quality of care. This is especially true if newly entitled beneficiaries are required to use mostly hospital district providers that they previously used.

A second, closely related dilemma involves the relationship between Medicaid managed care and the public hospitals. Expansion of managed care is the centerpiece of the state's Medicaid strategy to control costs and improve quality and access. Texas is moving steadily to enroll all of its Medicaid beneficiaries into managed care organizations and is contracting almost exclusively with commercial HMOs, an approach that is viewed as a way to mainstream Medicaid beneficiaries. Competition among HMOs is intense and the state has had little difficulty obtaining bidders. Although these HMOs are required to contract with traditional safety net providers for three years, enrollment of Medicaid beneficiaries in these plans is likely to reduce utilization of these hospitals, both because HMOs seek to reduce hospital utilization and because patients are channeled to other facilities with lower payment rates. Although from the perspective of the Medicaid program this reduction in the use of public facilities may result in reduced costs, more consumer choice, and better quality of care, the loss of Medicaid patients would financially strain these safety net hospitals because there would be fewer paying patients over which to spread the cost of providing care to indigent patients. Without Medicaid patients, public hospitals would require far larger local contributions, which is not likely given the current political environment. As a result, facilities could close or reduce services, endangering access to care by the uninsured population.

The recent failure by the Harris County Hospital District HMO to obtain a Medicaid managed care contract and the efforts of the legislature to ensure that public hospitals will obtain contracts is indicative of this problem. Without the contract, the Harris County Hospital District will face financial disaster. However, guaranteeing any provider a contract, as the legislature did for public HMOs, could limit the leverage of the state in requiring that standards be met.

The state faces at least two other major policy challenges. One involves the relationship between welfare reform and Medicaid. The Personal Responsibility and Work Opportunity Reconciliation Act of 1996 separated eligibility for cash assistance and Medicaid. As people lose welfare benefits because they have reached the time limits or are sanctioned for failure to meet work and other requirements, Texas will have to develop the systems to make sure that people understand that they may still be eligible for Medicaid. Even more daunting is that Texas must educate people to understand that they may be eligible for Medicaid even if they are unwilling to accept the various requirements of cash assistance. In any event, given the low levels of employer health insurance coverage in Texas, people working their way off welfare may find themselves without health insurance coverage, raising the number of uninsured and placing additional strains on public hospitals. Although the Balanced Budget Act of 1997 restored some Medicaid and cash benefits to legal immigrants that were to be ended as a result of changes enacted in 1996, most new legal immigrants will not be able to obtain Medicaid for five years, again possibly placing additional strains on the public hospital system.

The other challenge relates to long-term care for the elderly and disabled. Like the rest of the country, the elderly and disabled populations in need of long-term care are increasing. The state is pursuing a strategy of changing the balance of care in the direction of additional home and community-based services; but institutions still dominate funding, and the nursing home industry is particularly influential in determining state policy. From the state's perspective, it would like to extend managed care to include both acute and long-term care services, but it recognizes that there are formidable practical, political, policy, and legal barriers in pursuing this policy objective.

In sum, Texas is a big, bustling state with enormous contrasts of wealth and poverty. Its system of restrictive state benefits combined with an extensive locally funded health care safety net is both a strength and a problem for the state as it tries to expand coverage and move toward managed care.


Notes

1. This research is based largely on interviews conducted in Austin and Houston, Texas, in July 1996 and in El Paso, Texas, in February 1997.

2. The "disproportionate share hospital" portion of the Medicaid program requires states to make additional Medicaid payments to hospitals that serve a "disproportionate share" of Medicaid and uninsured patients.

3. ReliaStar, The ReliaStar State Health Rankings, 1996 Edition (St. Paul, MN: ReliaStar Financial Corporation, 1996).

4. Texas Department of Health and Texas Higher Education Coordinating Board, Physician Work Force Strategy for Texas: A Report to the Subcommittee on Health and Human Services and Education of the House Appropriations Committee (Austin: Texas Department of Health, 1995).

5. General Appropriations Act, Texas 75th Legislature, 1997-98 Biennium, House Bill 1, June 1997.

6. General Appropriations Act, Texas 75th Legislature, 1997-98 Biennium, House Bill 1, June 1997.

7. General Appropriations Act, Texas 75th Legislature, 1997-98 Biennium, House Bill 1, June 1997.

8. Texas's Federal Medical Assistance Percentage was 64.44 in 1996 and 64.18 in 1997.

9. AFDC has been replaced by Temporary Assistance to Needy Families, but the old AFDC criteria are used to determine eligibility for Medicaid.

10. Because the welfare block grant was based on spending in FY 94 (when AFDC caseloads were higher), Texas, along with most other states, received more federal money than it would have under the previous program. Thus, even after meeting the maintenance-of-effort requirement specified in the federal law, the state has funds left over from what it was spending previously.

11. Current federal law requires state Medicaid programs to phase in, one year at a time, coverage of all children younger than age 19 who have incomes below 100 percent of the federal poverty line.

12. For a more detailed discussion of welfare reform issues, see: Nancy Pindus et al., Income Support and Social Services for Low-Income People in Texas (Washington, D.C.: The Urban Institute, forthcoming, 1997).

13. Only the adults would lose eligibility, not children (as in the federal law). Extensions are available in counties with high unemployment. Recipients are exempt from this provision until the youngest child at home is four years old. Time limits (12, 24, and 36 months) apply only after employment services begin for the client.

14. Texas Department of Insurance, data as of December 31, 1995.

15. Texas Department of Insurance, "Basic Service Industry: Financial Data Graphs as of December 31, 1995" (Austin: Texas Department of Insurance, no date).

16. David W. Hilgers and Joe H. Cunningham, "Integrated Delivery Systems: What Does it Mean?" (materials for the 2nd Annual Rural Health Conference, "Community Oriented Care: Community Survival in the New Environment"), Texas Tech University Health Sciences Center, Lubbock, Texas, April 19-20, 1996.

17. Texas Medical Association and MTI Research, Inc., 1996 Survey of Texas Physicians: Research Summary Report (Austin: Texas Medical Association, April 1996).

18. Ibid.

19. The Balanced Budget Act of 1997 now provides for a federal match when a state guarantees Medicaid eligibility for six months to beneficiaries who enroll in managed care organizations.

20. "Hospital District Not Selected by State as Medicaid Provider," Houston Chronicle, May 13, 1997, 1A, 8A.

21. Texas Register, "Texas Department of Health, Medicaid Managed Care, Standards for the State of Texas Access Reform (STAR)," Proposed Rule, 21 TexReg 7324, August 6, 1996.

22. Ibid.

23. This provision in the Medicaid law requires states to pay hospitals and nursing homes enough to cover the costs of "economically and efficiently" operated facilities that meet the quality standards. Congress repealed that requirement as part of the Balanced Budget Act of 1997.

24. Prospective Payment Assessment Commission, "Medicare and the American Health Care System: Report to Congress" (Washington, D.C.: ProPAC, June 1995).

25. The Health Care Financing Administration contends that the Boren amendment does not apply to rates negotiated between hospitals and HMOs.

26. Stephen Norton, "Medicaid Fees and the Medicare Fee Schedule: An Update," Health Care Financing Review (fall 1995), vol. 17, 167-181.

27. Counties seeking funding under this state program must first spend 10 percent of "general revenue tax levy" on indigent care. State funds are then available for subsequent services with a 20 percent county match.

28. Hardy Loe et al., "The Houston-Galveston-Brazoria Consolidated Metropolitan Statistical Area Health Needs Assessment" (Houston: University of Texas, School of Public Health, 1994).

29. InterStudy, The Competitive Edge 6.2, Part III: Regional Market Analysis (St. Paul, MN: InterStudy, December 1996).

30. Richard C. Ladd, Robert L. Kane, Rosalie A. Kane, and Wendy J. Nielsen, "State LTC Profiles Report" (Minneapolis: University of Minnesota, 1995).

31. These figures must be viewed with caution because the state defines as "community" any service that is not located on the grounds of state institutions.

32. Texas Department of Mental Health and Mental Retardation, "Setting the Stage for Change," FY 95 Annual Report, p. 15.


APPENDIX
List of
People Interviewed

Most of the information in this report was collected through interviews conducted in Austin and Houston during July 1996 and in El Paso during February 1997. People interviewed included the following:

State Government Officials

Office of the Governor
Ronald Lindsey

Health and Human Services Commission
Michael D. McKinney
Cathy Rossberg
Linda Wertz

Department of Health
Lynn Hudson
Demetria Montgomery
Pattie Patterson
Debra Stabeno
Steve Svadlenak

Department of Human Services
Anita Anderson
Ron Borg
Dee Church
Amy Orum
Kay Priest
Susan Syler

Department of Mental Health and Mental Retardation
Nancy Gravley
Rose Ann Rossman
Sam Shore
James Williamson

Department of Insurance
Tyrette Hamilton
Leah Rummel

Department on Aging
Mary Sapp
Peggy Seely
Karl Urban
John F. Willis

Attorney General's Office
Jan Soifer
Mark Tobey

Texas Comptroller of Public Accounts
Bee Moorhead

Legislative Staff

Offices of State Legislators
Marianne Dwight, Senate Health and Human Services Committee
Erin Florence, Speaker's Office
Laura Lawlor, House Public Health Committee
Kathy Spurgin, Rep. Harvey Hildebran
Chris Taylor, Rep. Harvey Hildebran

Legislative Budget Board
Maureen McCormack
Anita Zinnecker

Provider Associations

Texas Association for Home Care, Inc.
Anita Bradberry
Heather Vasek

Texas Association of Homes and Services for the Aging
David Latimer
Sandy Derrow

Texas Health Care Association
Tim Graves
Tom Seuhs

Texas Hospital Association
James Houdek
Joe Dasilva
Richard Shirmer
Camille Miller

Texas Medical Association
Connie Barron
Helen Davis

Other Provider Associations
Jose Camacho, Texas Association of Community Health Centers
Brian Sperry, Children's Hospital Association of Texas
Geoffrey Wurzel, Texas HMO Association

Consumer Advocacy Groups

Robert Almanzan, Mexican American Legal Defense and Education Fund
Candice Carter, American Association of Retired Persons
Lynne Coyle, Lawyers Commission on Civil Rights
Ann Dunkelberg, Center for Public Policy Priorities
Beth Ferris, Texas Advocates for Nursing Home Residents
Penny Kendall, Disability Policy Consortium
Suzan Kern, Border Rights Coalition
Lisa McGiffert, Consumers' Union
Maureen O'Connell, Advocacy, Inc.

Experts

Paso Del Norte Health Foundation
W. Blake Fry
Geri Kaugh
Ann G. Pauli

University of Texas at El Paso
Fernando Galan, Social Work Program
Amy K. Liebman, Center for Environmental Resource Management
Dennis Soden, Public Policy Research Center

Others
Charles Begley, School of Public Health, University of Texas at Houston
Cliff Dasco, Baylor College of Medicine
DeAnn Friedholm, Independent Consultant
Richard Ladd, Ladd & Associates

Local Health Departments and Government Officials

City of El Paso
Raymond Telles, Mayor Pro-Tempore and City Councilman

El Paso City-County Health and Environmental District
Jorge Magaña
Raul V. Muñoz, Jr.

Harris County/City of Houston
M. des Vignes-Kendrick
Thomas Hyslop
Carol Pierson
Antonia Stewart

Providers—El Paso

Thomasson Hospital
Pete Duarte
Kent Giles

Others
Salvador Balcorta, Centro De Salud Familiar La Fe
Cindy Hague, Maternidad La Luz
Amy Lacera, Project Vida
Randall H. Rolfe, Columbia West Texas Division
Mario Sanchez, Centro Medico de Valle

Providers—Houston

Mel Bishop, Columbia North Houston and Columbia Doctors
Jeannie Fraser, Hermann Hospital
Lois Moore, HarrisCounty Hospital District
Mark Wallace, Texas Children's Hospital


Tables & Figures

Table 1
Table 1

Source of Area Shaded Blue: Current Population Survey (CPS) three-year average
(March 1996-March 1998, where 1996 is the center year) edited by
the Urban Institute to correct misreporting of citzenship.


About the Authors

Joshua M. Wiener is a principal research associate at the Urban Institute's Health Policy Center, where he specializes in research on Medicaid, long-term care, and health care for the elderly. Prior to coming to the Urban Institute he did research and policy analysis for the Brookings Institution, the Health Care Financing Administration, the Commonwealth of Massachusetts, the State of New York, and the City of New York.

Alison Evans is currently in the Ph.D. program at the University of California, Berkeley, School of Public Health. At the time of the research for this report, she was a research associate in the Urban Institute's Health Policy Center, where she specialized in research on Medicaid, managed care, and health care for the elderly.

Crystal Kuntz is an associate director for business at National Association for Home Care. At the time of the research for this report, she was a research associate in the Urban Institute's Health Policy Center, where she specialized in research on Medicaid, long-term care, and health care for the elderly.

Margaret Sulvetta is director of computer services at the Urban Institute. At the time of the research for this report, she was a senior research associate in the Urban Institute's Health Policy Center, where she specialized in research on hospitals, reimbursement systems, and Medicare.

This report is part of The Urban Institute's Assessing the New Federalism project, a multi-year effort to monitor and assess the devolution of social programs from the federal to the state and local levels. Project codirectors are Anna Kondratas and Alan Weil.

The project has received funding from the Annie E. Casey Foundation, the Henry J. Kaiser Family Foundation, the W. K. Kellogg Foundation, the John D. and Catherine T. MacArthur Foundation, the Charles Stewart Mott Foundation, the Commonwealth Fund, the Robert Wood Johnson Foundation, the Weingart Foundation, the McKnight Foundation, and the Fund for New Jersey. Additional funding is provided by the Joyce Foundation and the Lynde and Harry Bradley Foundation through a subcontract with the University of Wisconsin at Madison.

The authors would like to thank the many state, county, and local officials and others who participated in interviews and provided information. The authors would also like to thank Alan Weil, John Holahan, Nancy Pindus, Anita Zinnecker, and Maureen McCormack for useful comments on an earlier draft. The views expressed are those of the authors and should not be attributed to The Urban Institute, its trustees, or its funders.



Topics/Tags: | Health/Healthcare


Usage and reprints: Most publications may be downloaded free of charge from the web site and may be used and copies made for research, academic, policy or other non-commercial purposes. Proper attribution is required. Posting UI research papers on other websites is permitted subject to prior approval from the Urban Institute—contact publicaffairs@urban.org.

If you are unable to access or print the PDF document please contact us or call the Publications Office at (202) 261-5687.

Disclaimer: The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. Copyright of the written materials contained within the Urban Institute website is owned or controlled by the Urban Institute.

Email this Page