Medicare and Medicaid managed care: issues and evidence.

Today in the United States, national health expenditures total more than $700 billion and make up over 13 percent of the gross domestic product (GDP). By 1995, they are projected to rise above $1 trillion and to comprise 15.6 percent of the GDP (Burner, Waldo, and McKusick, 1992). In spite of this level of expenditure, 15 percent of Americans are still without any form of health insurance and an additional 10 percent are inadequately insured.

The President's proposed approach to health care reform and virtually all other national health care reform proposals under consideration rely on health maintenance organizations (HMOs) and other forms of managed care to reduce the rate of health care cost increases while expanding coverage. The focus of this issue of the Health Care Financing Review is on recent empirical evidence from studies of Medicare and Medicaid managed care programs, including an examination of some new methods for setting payment rates to maximize the impact of managed care. As an introduction to these articles, it may be useful to briefly review the history of managed care, with an emphasis on programs financed by the Federal Government.

Health care plans incorporating both a limited provider panel and prepayment, the two primary features of a managed care plan, have been in existence in the United States for over 150 years. While the very early plans were established by large employers and provided services exclusively to their employees, modern prepaid group practice is generally considered to have begun in Elk City, Oklahoma in 1929 with the establishment of the Community Hospital Association. The Community Hospital Association provided heavily discounted medical care to area residents who enrolled in the plan and provided financing through their prepaid memberships. Other prepaid health plans followed, sponsored not only by rural cooperatives but also by unions, municipalities, and industry (MacColl, 1966). Prior to the early 1970s, however, these plans were generally small, few in number, and, with some important exceptions (e.g., Group Health Association, Kaiser Permanente, and the Health Insurance Plan of New York), poorly accepted by both the general public and the medical community.

In 1973, the Federal Government lent significant support to the managed care concept through the National Health Maintenance Organization Act, which required employers within a certain distance of a federally qualified health plan to offer the plan to their employees and provided grants and loans to encourage the establishment and growth of HMOs (Gruber, Shadle, and Polich, 1988). However, despite the support which the Federal Government provided HMOs, most of the growth in managed care during the 1970s took place in the private sector, rather than in the federally financed Medicare and Medicaid programs.

Medicare

In 1979, only 64 HMOs had Medicare contracts. Thirty-two of these organizations were group practice prepayment plans whose contracts did not include hospital care. Another 31 HMOs contracted for the full range of Medicare services on a cost basis. Only one HMO had a contract that placed the plan at risk. Total enrollment for all 64 plans was 508,253 (less than 2 percent of Medicare beneficiaries). Most HMOs were reluctant to participate in the Medicare program, since payment was retrospectively cost based. Cost-based contracting clashed with the prepaid capitation philosophy of HMOs and was perceived by most HMOs as administratively burdensome. In addition, HMOs were accustomed to marketing to groups of enrollees (through employers) and were not willing to develop strategies for attracting individual enrollees, given the other disincentives they perceived in the Medicare contracting system.

In an effort to test contract arrangements and payment methodologies that would provide HMOs with more incentives to enroll Medicare beneficiaries, the Health Care Financing Administration (HCFA) conducted two series of demonstration projects from 1978 to 1985: the Medicare Capitation Demonstrations and the Medicare Competition Demonstrations. These demonstrations, in which 32 plans participated, represented the first attempts by the Federal Government at true prepaid risk contracting. Under these demonstrations, plans were paid, for each enrollee, a monthly rate that was 95 percent of Medicare's adjusted average per capita cost (AAPCC) for beneficiaries with similar demographic characteristics. The AAPCC is a set of payment rates derived from the U.S. per capita cost for Medicare beneficiaries, adjusted for an individual beneficiary's county of residence, age, gender, institutional status, and Medicaid eligibility status. The results from the initial demonstration were encouraging and, by the end of 1984, the second series of demonstrations had enrolled 117,000 beneficiaries (Langwell and Hadley, 1986).

In 1985, the Tax Equity and Fiscal Responsibility Act (TEFRA) of 1982 (Public Law 97–248) was implemented. Utilizing the same payment methodology tested during the Medicare Capitation and Medicare Competition Demonstrations, this legislation allowed the Federal Government to develop risk contracts with HMOs and other organizations on a program-wide basis. The goal of the TEFRA program was to encourage HMO participation in Medicare in order to capitalize on the perceived cost-efficiencies of managed care and to give Medicare beneficiaries a coordinated care alternative to the fee-for-service (FFS) system. In this issue of the Review, Brown, Clement, Hill, Retchin, and Bergeron provide a summary of the findings from a comprehensive 4-year evaluation of the TEFRA risk program, examining the programs' impact on the use and cost of services, beneficiaries' access to care and the quality of the care they receive, their satisfaction with care, and the impact of the program on HMOs' profitability and willingness to continue contracting with HCFA. Their evaluation indicates that, while the risk program provides Medicare beneficiaries with an alternative to FFS that offers expanded benefits at a lower cost to the beneficiary, the program as it is currently structured does not appear to be cost-effective for Medicare. The authors suggest changes to the payment methodology and regulations governing HMOs that could maximize the potential of the program. In a related article, McMillan examines Medicare risk-contracting trends from 1986 to 1993. In addition to describing changes in the organizational characteristics of the plans, the range of benefits offered, and the premiums charged, she provides further insights into the dynamics of the program by examining the entry and exit of plans and beneficiaries from the program.

Not all work on Medicare prepayment systems to promote the efficient management of care involves HMOs. The article by Averill, Goldfield, Wynn, McGuire, Mullin, Gregg, and Bender presents a prospective payment classification system for ambulatory care that is designed to complement the prospective payment system (PPS) that Medicare has been using since 1983 to pay for hospital inpatient treatment for beneficiaries. While the Medicare inpatient PPS has been successful in controlling the growth in Medicare expenditures for inpatient care, there has been a rapid increase in Medicare payments to hospital outpatient departments (same-day surgery units, emergency rooms, outpatient clinics, etc.). Through the Omnibus Budget Reconciliation Act of 1986 (Public Law 99-509), Congress directed HCFA to develop a PPS for hospital outpatient services. The ambulatory patients groups patient classification system presented in this article will be used in establishing an outpatient PPS and, like the diagnosis-related groups used to pay hospitals, it is designed to provide additional incentives for hospitals to manage patient care episodes in a more efficient and cost effective manner.

Medicaid

While HMOs have had legislative authority to enter into contracts with State Medicaid agencies since 1967, there was little initial interest. Then, from 1971 to 1973, 12 States signed 66 contracts with organizations to serve 371,000 Medicaid recipients on a prepaid basis. However, there was little legislated Federal oversight of these contracts and serious complaints were raised about the cost, quality, access, marketing practices and corporate accountability of the plans. These concerns led to stricter regulations concerning the Federal qualification and oversight of Medicaid prepaid plans. As a result, the number of Medicaid HMOs declined and, by 1980, there were only 53 contracts enrolling 270,000 Medicaid recipients out of approximately 20 million eligible individuals (Trieger, Galblum, and Riley, 1981).

In 1981, new legislation relaxed some of the regulations that were inhibiting HMO contracting with Medicaid yet retained enough control to prevent a recurrence of the problems that surfaced in the early plans. In addition, this legislation authorized forms of managed care such as primary care case management (PCCM)—that is, an appointed physician or other primary care provider serving as a gatekeeper to specialist and inpatient services—that provided an alternative to traditional HMOs.

Three articles in this issue of the Review examine a number of issues related to Medicaid managed care. Miller and Gengler examine the impact of the Kentucky Patient Access and Care Program, a PCCM program, on the use of services by Medicaid recipients. Their results support the findings of previous Medicaid managed care studies—managed care often results in uneven reductions in utilization. In this study, the authors find that Kentucky's PCCM program was successful in reducing the use of independent laboratory, physician, emergency, and outpatient services, while not affecting the utilization of inpatient hospital services or prescription drugs.

Fox, Wicks, and Newacheck present the results of a survey of State Medicaid agencies concerning their policies for enrolling and serving special-needs children in HMOs. Their findings suggest that while many States have implemented strategies to protect special-needs children enrolled in HMOs, these strategies are often too limited in scope to ensure appropriate access to specialty services for all children with special health needs. The authors also discuss reasons why HMOs may be reluctant to assume the risk of enrolling such children.

Newhouse, Sloss, Manning, and Keeler also address the issue of HMOs providing adequate access children with special needs from a slightly different perspective. Their article examines the use of health status risk adjusters to determine the capitation rates for enrolling children in HMOs as a way of reducing the incentive for HMOs to discriminate against children with chronic diseases and predictably high costs. They argue against divorcing payment from past and current use and recommend experimentation with a mixed reimbursement system that incorporates both FFS and capitation features.

In addition to the theme-related articles previously discussed, this issue of the Review contains two other articles that, while not directly related to managed care, are quite relevant to the current debate over health care reform.

The first of these articles, by Moffit and Wolfe, examines the relative influences that Medicaid and private insurance exert on an individual receiving welfare when the individual is confronted with opportunity for employment. The authors argue that there is a link between welfare and Medicaid availability and develop econometric models showing that increasing the availability and coverage provided by private insurance can lead to significant decreases in welfare dependency.

Holahan and Zuckerman's article examines the extent to which beneficiaries travel across geographic borders to receive physician services and the types of services for which this most frequently occurs. The authors find that there is substantial geographic variation among both States and rural areas in border crossing to seek services, with more activity taking place within than between States and rural areas being more likely to import services and large metropolitan areas to export services. They discuss these and other findings in terms of defining appropriate geographic areas for a subnational application of Medicare volume performance standards (MVPS) and the need to link MVPS with beneficiary utilization in the providers' service area. Their findings on geographic border crossing may have implications beyond the refinement of MVPS. For example, the accuracy of other payment methodologies such as the AAPCC may be sensitive to border crossing behavior. In addition, these issues will be relevant for the State Health Alliances which would be established under the President's proposed health reform, as the Health Alliances would need to take into consideration both between and within State border crossing in defining their service areas, developing contracts with providers, and setting payment rates.