Editor's Note: In the commentary below, James E. Sabin, M.D., discusses issues raised by the article reprinted on page 1385 from the March 1998 issue of Psychiatric Services. In that article, Jeffrey L. Geller, M.D., M.P.H., and his colleagues examined patterns of service use in a public-sector managed behavioral health care program and found that for some patients the program led to discontinuities in the provision of care. Dr. Sabin makes the case that policy makers and providers can create an ethically and clinically admirable approach to managing care. He explains the ethical necessity of rationing health care and describes factors in the development of the private-sector managed care industry that have influenced public-sector managed care programs. Dr. Sabin describes how such public-sector programs must differ from employer-based insurance to ensure care for persons with severe mental illness. He emphasizes that all managed behavioral health care programs must provide a clear rationale for their limit-setting policies and urges providers and policy makers to learn from what he calls "our extraordinary national experiment."

Managed care in 2000 represents our society's effort to craft a private-sector solution to the public-policy problem of health care limits and fair resource allocation while pretending that the problem does not exist. Put bluntly, managed care is, in my opinion, rationing in drag.

An ethical society cannot and should not avoid rationing health care. The aging population and extraordinary medical advances combined with the increased public zest for health and well-being that comes with a strong economy create a desire for more health care interventions than society is prepared to pay for. As significant as health care is, it is not the only important individual or social good. The mature response to this situation would be to deliberate about priorities, make hard choices, understand the rationale for the limits we impose on ourselves, live with the results, and periodically repeat the priority-setting cycle in the light of what we have learned from experience. As individuals we do this all the time with important goods we pay for ourselves, such as housing, food, education, and health care that is not covered by insurance.

It is easy to understand why politicians avoid acknowledging the need to ration health care. They much prefer to tell the public that their electoral opponent and the opposition party will put limits on health care, but that they themselves would never support such an unnecessary and cruel course of action. A politician who talks about limits looks callous and invites attack. Governor John Kitzhaber of Oregon is a rare example of a political leader who leveled with the public about the ethical necessity of setting health care limits.

When a full history of the managed care era is written a decade or two from now, I predict that it will be one-third comedy, one-third tragedy, and one-third a solid foundation for the system changes that will be unfolding in its aftermath. This article reflects on three questions. Why did the United States opt for its version of managed care? How does managed care affect the care and treatment of persons with severe mental illness in public-sector programs? And what lessons can we derive about health care reform and mental health parity from our national experience with managed care?

Geller and colleagues (1) began to derive some of those lessons in the article reprinted in this issue. Their study exemplifies an empirically based approach to policy and system design guided by clearly articulated values and supported by a managerial strategy.

Because this is a decidedly personal commentary, I need to start with an element of disclosure. The commentary starts from my belief that there is an ethically and clinically admirable approach to managing care that provides the best basis for our health care system (2). This population-based vision draws on 25 years of clinical experience in a capitated, not-for-profit, multispecialty group practice, Harvard Vanguard Medical Associates; leadership of the corporate ethics program in a not-for-profit insurance company, Harvard Pilgrim Health Care; and extensive policy and ethics research at managed care organizations around the country. Although I prefer not-for-profit health systems, I do not regard for-profit health care as unethical per se (3). For me, the phrase "ethical managed care" is not an oxymoron.

Managed care reflects our country's distinctive approach to a universal human challenge. All societies regard some level of health care as a distinctive social good that the society is obliged to provide to its members. However, even in a country as wealthy as the United States, the cost of health care must be contained. Containment requires limits and priorities. Throughout our history the United States has cultivated an optimistic, can-do persona. As Franklin Delano Roosevelt told us, we have nothing to fear but fear itself. Priorities and limits are for the old world, not the new one.

In the late 1980s and 1990s health care payers clamored for relief. States referred to Medicaid as the "budget buster." The automobile makers discovered that health care for employees cost more than steel for automobiles. However, the U.S. persona of can-do optimism prevented us from acknowledging that we could not have it all in health care and would have to set priorities and limits. Not acknowledging the situation was the first stumbling block. As every clinician knows, naming a problem is the necessary first step in solving it. Treatment plans only make sense in the context of diagnosis and formulation.

Not naming the problem didn't make it go away. As a nation we had to do something about runaway health care costs. One possibility would have been a budgeted national health care insurance system—Medicare for all. The Canadian system, while not free from problems, proves that single-payer insurance can provide coverage for an entire population with relatively high satisfaction and low administrative cost (4). However, in post-Reagan-revolution America, our political discourse precludes the kind of government activism that launched Medicare, Medicaid, and community mental health centers in the 1960s. Although a majority of participants in the Clinton health care reform project favored a single-payer system, it was not treated as a realistic option.

The Clinton proposal sought to build in substantial choice of health plans for each individual and retained fee-for-service ("unmanaged care") as an option. However, it crashed and burned, brought down by a fusillade of antigovernment sentiment shaped by the "Harry and Louise" advertisements combined with intensive lobbying by the economic powerhouses that felt threatened by the plan. What filled the policy vacuum was the relatively unregulated, market-driven, largely for-profit managed care system we have today.

In the mental health sector we have seen market forces at work in an unusually rapid and instructive manner. In the 1980s an entrepreneurial, cost-expanding hospital "industry" engendered explosive growth, especially in adolescent and substance abuse programs. In accord with Newton's third law of motion, this hospital growth triggered an equal and opposite entrepreneurial, cost-containing managed behavioral health care "industry" designed to rein in the surging costs (5).

Markets are designed to satisfy purchasers, not to educate the public about policy questions or to strengthen communal solidarity or the social safety net. Purchasers and politicians wanted to maintain the fiction that market forces could provide all "medically necessary" services, do no rationing, and reduce costs, all at the same time. It would be naive to expect the managed behavioral health care "industry" that purchasers hired and politicians endorsed to blow that cover (6).

At the start of a new century, the U.S. health care system is caught between an immovable object—the unavoidable need for priorities and rationing—and an irresistible force, the market, which has been asked to solve the cost problem without letting on to the need for rationing. As a nation we have developed unrivaled techniques for managing care with very little societal understanding of the reasons for doing so. We apply powerful managerial tools in the absence of guiding vision.

Not surprisingly, the result is a fierce public backlash. The public knows it is being duped. Rationing is happening. But the public has not yet been helped to understand that without rationing we cannot achieve universal access or better address other societal imperatives.

The comedy of managed care comes from listening to the repetitive debates in which free-market hawks and professional guilds conduct a rhetorical jihad against each other. The tragedy of managed care comes from our inability as a society to join forces around a human need as fundamental as health care and the ethical necessity for rationing. Until we overcome our reluctance to lay out a realistic picture of the problems we need to address, we will not be able to tackle them effectively.

Examining both private and public-sector managed care in detail is beyond the scope of this commentary. For its 50-year history, the purpose of this journal has been "to help mental health clinicians and administrators improve the care and treatment of persons with severe mental illness" (7). Because the public sector has traditionally been the primary source of these services, this section focuses on public-sector managed care.

Managed mental health care cut its teeth and achieved its early cost-containment successes in the private sector in the late 1980s and early 1990s. Some of the initial results were relatively easy to achieve. Hospitals were frequently used when a less intensive form of care could have achieved the same or even better results. Some outpatient treatment was unfocused and regressive. Changing these practices offered win-win opportunities to improve care and reduce costs at the same time. At the other extreme, innumerable anecdotal reports described instances of what was seen to be—and often was—overly restrictive care management. To venture a large generalization, private-sector managed behavioral health care earned its initial spurs by cultivating skill at saying "no."

The individuals, employers, and public agencies that pool funds to purchase health insurance specify the scope of the insurance through benefit design (the conditions and treatments that will be included) and definition of medical necessity (the criteria for inclusion). In many private-sector contracts, medical necessity is defined in terms of symptom remission and return to baseline functioning. This definition is adequate for meeting the needs of a well-functioning person with reliable environmental supports who experiences an acute illness. It is not adequate, however, for many persons with severe mental illness who have progressively deviated from the potential they might have achieved without the impact of the illness and who may live in circumstances of socioeconomic deprivation.

Public-sector managed care has to be different from employer-based insurance. Public-sector programs provide rehabilitation and safety-net services, not simply medically necessary health care interventions. Patients who exhaust their private insurance benefits can turn to the public sector. Public-sector patients can turn only to the streets or jail.

According to a Wisconsin study, typical private insurance definitions of medical necessity would cover only 60 percent of the treatment that good public-sector programs provided for severe and persistent illness (8). This makes for obvious conflict if a managed care company brings private-sector criteria and attitudes into the public arena. Many of the early problems in public-sector managed care arose from applying private-sector conceptions of medical necessity to public-sector practice. If we add to that the mismatch between private-sector definitions of medical necessity and public-sector needs, funding that may be marginal or inadequate, and limited public-sector experience in managing a complex contract, we have a recipe for uneven managed care performance.

Michael Hoge and colleagues (9) have provided a template of ten dimensions that are useful for planning and assessing public-sector managed care programs. Four of the ten can be used to illustrate key positive developments in public-sector managed care: objectives, financing, organizational structure and authority, and strategies for managing utilization.

Objectives. With excellent political leadership from John Kitzhaber, former Oregon senate president and now governor, and active participation by Oregon psychiatrists, the state set unusually clear objectives for its public-sector program. The primary objective was to expand access to a wider population, and the primary means for achieving this goal was to set explicit, democratically defined limits on the extent of the benefits package. In an open, participatory process, people with mental health conditions fared very well. Essentially all treatable psychiatric conditions fell above the cutoff line for coverage. Schizophrenia ranked just behind asthma. Attention-deficit disorder ranked just ahead of hypertension. Chemical dependence ranked four places behind closed hip fracture (10).

Financing. Americans are worried about the way managed care organizations are paid. When a recent nationwide survey asked a random sample of adults if they were "worried that your health plan would be more concerned about saving money than about [providing] the best treatment for you if you are sick," 61 percent of those enrolled in "heavy managed care" but only 34 percent of those in "traditional" plans said they were somewhat or very worried (11). Although 72 percent of the sample agreed that managed care savings "helps health insurance companies to earn more profits," only 49 percent believed that these savings also "make health care more affordable for people like you."

In contracting for a statewide Medicaid carve-out program, the state of Massachusetts structured the contract so that the vendor's profit potential came more from meeting performance standards than from not spending funds budgeted for services (12). Massachusetts did not see profit as the enemy of good public-sector care but wanted to tie profit to clinical performance rather than to earning more by doing less.

Organizational structure and authority. Consumers and providers have criticized managed care organizations for not letting them have enough influence on how the organizations conduct themselves. All major advocacy groups call for consumer and family involvement. The National Alliance for the Mentally Ill, acting on its slogan "Nothing about us without us," has as a key criterion in its managed care report card that "consumers and family members must assume an integral role in the governance and oversight functions of the managed care organizations" (13). The Bazelon Center for Mental Health Law and the Legal Action Center have asserted that consumers and families should be "partners in planning" at the level of the contracting process itself (14). The National Mental Health Association makes consumers and families central to each of its nine standards for "consumer-centric managed mental health and substance abuse programs" (15).

Massachusetts has used the core ingredient of managed care—management—as its fulcrum for empowering consumers and families in its Medicaid carve-out (16). The state and the carve-out identified key points in the management process and built in meaningful consumer and family opportunities for influence. This influence was not global input but high-leverage participation in developing clinical guidelines, setting performance standards, and participating in formal and informal quality monitoring.

Strategies for managing utilization. When the state of Iowa encountered significant turmoil and conflict at the inception of its statewide carve-out program in 1995, it diagnosed a disparity between the vendor's initial criteria for medical necessity and what was required for a successful public-sector safety-net program (17). The state and the program moved from "medical necessity" to "psychosocial necessity" in recognition of the need for rehabilitative, supportive, and even short-term "custodial" services as a component of safety-net functions.

Advocates for parity are caught in a form of double-bind with regard to managed care. On whatever grounds they may oppose managed care, they must assimilate the fact that the managed care "industry" has provided the most persuasive evidence from U.S. experience that mental health costs can be controlled without arbitrary caps on benefits. There is no clinical or principled basis for opposing parity—the only argument against it is affordability. A range of managerial approaches, from global budgets to use of selected networks combined with guidelines and utilization review, provide substantial evidence for the insurability of generous mental health benefits (18).

Ironically, managed care may also prove to be a friend to the psychiatric profession with regard to the trend toward limiting psychiatrists to writing prescriptions and performing "med checks" that began in the community health era. The pervasive belief that splitting treatment, with the psychiatrist handling medications and a nonphysician providing psychotherapy, is less costly has been based on faith, not evidence. The large databases available through managed care organizations allow empirical testing of this hypothesis. Early data suggest that at least for depression, integrated treatment by a psychiatrist actually costs less than split treatment (19).

When the definitive history of the managed care era is written, I predict that it will identify our failure to recognize the need for hard choices in health care policy as the era's greatest failing (20). Developing and disseminating a vision of responsible stewardship of resources at the level of the population as the necessary precursor to good care for all Americans will require political leadership. For the foreseeable future this leadership will have to come at the state level (21). But professionals, advocates, and journals like Psychiatric Services can pave the way for the needed leadership by articulating an agenda for the politicians.

My colleague Norman Daniels and I have identified four requirements for health systems that wish to educate the public about hard choices and persuade them that certain limits are legitimate and fair (22,23). First, the rationale for limit-setting policies must be explicit and readily available to the public. Second, the rationale must show how the policy promotes good care for individuals and optimal use of available resources for the larger population. Third, mechanisms must be in place for debate, appeal, and learning from experience. Finally, the first three conditions must be enforced. We call it holding health systems "accountable for the reasonableness" of their limit-setting policies.

The reprinted article by Geller and colleagues (1) shows how accountability for reasonableness can be put into action. First, the authors ask whether there is an understandable rationale for contracting with a selective network. They conclude that there is. The rationale is part of an effort to "simultaneously increase the use of private settings for inpatient care and curb the explosion of Medicaid expenditures that would accrue from such a shift." Then they look at how that clinical policy performs on the ground. They demonstrate negative outcomes for younger Caucasian females with a personality disorder and a history of substance abuse. In the light of their findings, the authors suggest ways in which public purchasers can enforce improvements through the contracting process. The article puts managed care into a framework of recurrent reform efforts, each of which solves certain old problems while at the same time creating new ones.

During two decades of "reform" through market forces, our society has ceded leadership of the reform process to the invisible hand of the market. Many of the changes have been positive. Many have not. The task now is to learn from our extraordinary national experiment so we can provide wiser political guidance for what comes next. Because the public sector is required by statute to be open and explicit about its policies and practices, it sets the pace for learning how health systems can be accountable for the reasonableness of their limit-setting policies. The article by Geller and colleagues shows us how to conduct the necessary learning process. Private-sector managed care organizations would do well to adopt lessons from the best public-sector practices.

The author thanks the Greenwall Foundation and the Robert Wood Johnson Foundation Medicaid Managed Care Program for their support.

Dr. Sabin is clinical professor of psychiatry at Harvard Medical School and director of the ethics program at Harvard Pilgrim Health Care. Send correspondence to him at Teaching Programs, Harvard Pilgrim Health Care, 126 Brookline Avenue, Suite 200, Boston, Massachusetts 02215 (e-mail, jim_sabin@hphc.org).