The Future of Children, Princeton - Brookings: Providing research and analysis to promote effective policies and programs for children.

Managed care is not new. The first health care organization to employ many features of managed care was the Ross-Loos Clinic, established in Los Angeles in 1929. The concept was subsequently adapted by the Kaiser Permanente Corporation as a way to provide health care to employees during World War II. The original plans were called prepaid group practices, but by the 1970s such plans were renamed health maintenance organizations (HMOs).³ The key idea behind these plans was simple: health insurance coverage and the delivery of medical care are integrated into a single organization to facilitate access to care while at the same time removing financial incentives to provide “extra” care to plan enrollees.

HMOs with organizational structures quite different from the format of the early prepaid group practices have evolved over the years. Especially since the passage of federal legislation to promote the expansion of HMOs in the mid-1970s, managed care has become a growth industry which has spawned a variety of different organizational models, including independent practice associations (IPAs), preferred provider organizations (PPOs), point-of-service (POS) plans, service-specific case management plans, and high-cost case management programs.⁴ These comprehensive managed care models are described in greater detail in Table 1. The intersection of the features of many of these alternative health plans has become smaller over time, leading to confusion in the public discussion of managed care. Some observers even question the utility of the generic term “managed care.”⁵

Recently, the Health Insurance Association of America (HIAA) attempted to clarify the debate by defining a managed care plan as one that integrates the financing and delivery of specified health care services by means of four key elements: (1) arrangements with selected providers to furnish a comprehensive set of health care services to enrollees; (2) explicit standards for the selection of participating health care providers; (3) formal programs of quality assurance and utilization review; and (4) significant incentives for enrollees to use providers associated with the plan. All of these features are part of the original HMO concept which also typically includes the assumption by doctors of some financial risk for the costs of care and, sometimes, the integration of medical facilities which was a feature of the original prepaid group practices.⁶ In this paper, we follow the HIAA definition of managed care which requires incentives for subscribers to use a defined network of health care providers associated with the plan. We exclude from our consideration more traditional indemnity insurance plans which may include some features of utilization management in order to control costs but do not attempt to limit enrollees' choice of providers.¹

Regardless of which specific combination of features individual managed care plans use, they all share the common goal of constraining health care costs and may do so in ways that increasingly intrude into the doctor-patient relationship. There is increasing concern that managed care represents a potential threat to the health of vulnerable populations such as children and pregnant women, especially those who are poor.⁷ To assess whether this is true requires an understanding of the large and more subtle distinctions among the variety of structures comprising managed care.

It is difficult to derive a completely satisfying typology of managed care.⁸,⁹ As recently as the mid-1980s, it would have been possible to completely distinguish the two predominant forms of managed care plans, HMOs and PPOs. However, the differences between what are frequently referred to as four different types of HMOs (staff, group, individual practice association [IPA], and network) and PPOs have become blurred (see Table 1 for specific details). PPOs have incorporated many of the features of IPAs. (For a description of some distinguishing features of different plans, see Table 2.) Otherwise restrictive HMOs with the addition of a point-of-service option which provides for some out-of-network use, become from the enrollee's perspective virtually indistinguishable from many PPOs. There is also considerable uncertainty among industry analysts and managers as to the stability of the current mix of managed care plans. It is possible that health care markets are now in a transitional stage which will facilitate enrollment into managed care plans that will gradually become more restrictive or change in other important ways.¹⁰,¹¹

Growth in the number of managed care plans and in the number of individuals enrolled in them has been dramatic over the past 20 years, with the most rapid growth concentrated in PPOs, IPA HMOs, and most recently point-of-service plans (Figure 1). At midyear 1992, there were more than 575 HMOs in the United States, accounting for an enrollment of almost 42 million people.¹² In 1991, there were almost 1,000 individual PPO plans in operation.¹³ Although precise estimates of PPO enrollment are not available, it appears that approximately 37 million workers were eligible to use PPOs in 1991. Adding eligible dependents to that estimate suggests that as many as 85 million individuals may have been eligible for PPO benefits as part of their employmentbased health coverage in 1991.¹³ Even more recent data indicate that enrollment in the 40 largest individual PPOs increased by 10% in the first six months of 1992.¹³ This growth rate was more than double the growth rate of these plans in the first half of 1991.

Data permitting precise estimates of the number of children and pregnant women covered in managed care arrangements are not available; however, children are well represented in managed care systems because plans are typically marketed through employers, and most provide benefit packages that emphasize ambulatory care services attractive to young families. The Group Health Association of America (GHAA) estimates that about 26.5% of HMO enrollees are less than 15 years old (10.6 million children in 1991).

Although estimates of enrollment in PPOs vary, the age distribution of PPO enrollees is probably similar to that in HMOs, which suggests that more than 20 million children may have been eligible for PPO benefits in 1991.¹⁴

More precise figures are available, however, for the number of Medicaid beneficiaries enrolled in various types of managed care arrangements. The passage of the Omnibus Budget Reconciliation Act of 1981 made it legal for states to apply to the Health Care Financing Administration (HCFA) for permission to enroll Medicaid eligibles in managed care. Since that time, when approximately 200,000 Medicaid beneficiaries joined HMOs, enrollment has soared such that, by mid-1992, 3.6 million Medicaid-eligible individuals were enrolled not only in staff, group, and network type HMOs of all varieties, but also in health insuring organizations (similar to county-sponsored IPAs), and in capitated and fee-for-service case management plans (see Figure 2).¹⁵ Some 36 states now have managed care plans as part of their Medicaid programs.¹⁶

The number of Medicaid enrollees in managed care should increase substantially again in the next three to five years following the recent announcements that California and New York will each phase 50% of their Medicaid populations into managed care over that period.¹⁶,¹⁷ These are the states with the largest Medicaid programs, and their actions, if successful, will add about three million enrollees to the ranks of Medicaid managed care. Many states, such as Arizona, Kentucky, Ohio, and Wisconsin, have mandatory enrollment programs which require Medicaid beneficiaries to join managed care arrangements either throughout the state or in several urban areas. Other states under voluntary enrollment programs contract with managed care providers so that Medicaid beneficiaries may choose managed care plans.¹⁸

Far and away the largest fraction of managed care–enrolled Medicaid beneficiaries are women and children who qualify for Medicaid by receiving Aid to Families with Dependent Children (AFDC). In some states, the only group for whom enrollment is mandatory or even possible is AFDC recipients. Many women qualify for Medicaid only when they become pregnant and then join an HMO or other type of managed care arrangement. Indeed the bulk of the data available on how children and pregnant women fare in managed care comes from Medicaid AFDC managed care programs.

Many claims, positive and negative, are made about managed care; however, which of these hypothesized effects are, in fact, realized remains to be determined empirically. Among the advantages of managed care, purchaser savings are regarded as a primary benefit.¹⁹ Such cost savings are supposed to derive in large measure from reductions in unnecessary inpatient hospital and emergency room use, and reductions in the use of ancillary medical services.²⁰ Increased quality of care, greater receipt of preventive care, and improved continuity of care also are often cited as potential benefits resulting from the introduction of managed care.¹⁸ It has been suggested in the literature that capitation and other types of payment incentives may bring about these desirable changes.¹⁹ Organizations that participate in capitated arrangements have a strong incentive to minimize costs and may do so by delivering only care that is truly necessary. For the same reason, it is hypothesized that utilization of those preventive services which reduce the likelihood of expensive debilitating illness should be greater in managed care plans. The incentives to increase preventive care may, however, be counterbalanced by the recognition that a plan's investment in prevention will pay off only if enrollees remain in the plan for many years. High disenrollment rates and frequent plan switching will dampen the returns to and hence the incentives for preventive services. Table 3 presents a summary of the hypothesized effects of different types of managed care plans on health services utilization and costs.

Because physicians manage access to the health care system, physician behavior is key to the success of managed care plans. There are, however, so many different ways in which physicians are compensated in managed care organizations that it is difficult to make generalizations about the effects of payment incentives on their behavior.²¹ As a result, there are few if any studies which permit a full understanding of whether there are differences in the extent of cost savings or changes in utilization according to the ways physicians are remunerated.²²

Managed care plans also differ in other domains, including factors that can affect costs, utilization, and health outcomes. For example, increases in preventive care, an aspect of health care thought to be particularly important for children, may result because physicians in managed care plans are encouraged to prescribe such care more often. In addition, managed care plans also frequently reduce patients' out-of-pocket payments for these services in comparison to fee-for-service arrangements, and this change in the incentives patients face should, in and of itself, also lead to increased use of preventive services. Changes in utilization patterns may also reflect greater access to primary care in some managed care plans. Increased care may also result from the treatment of unmet needs by primary care physicians or from greater recognition of these health problems in some managed care plans.

Negative effects of managed care are also quite possible. These are of particular concern for vulnerable populations. For example, plans with strong incentives to reduce utilization might limit services needed by children and pregnant women whose medical care may be difficult or costly to deliver. Special concern is raised because primary care physicians who are at risk financially may face disincentives to refer their chronically ill patients to specialists, to rehabilitation, or to extended psychiatric care and the like. The restrictive features of some managed care plans may make it more difficult to find the appropriate services and may necessitate breaking existing provider relationships. Because only a small proportion of children enrolled in a typical managed care plan are likely to need a substantial amount of expensive specialty services, it is possible for managed care plans to satisfy most enrollees while underserving those whose needs are greatest.