Printer Friendly

Minds and hearts: priorities in mental health services.

To care for people with mental illness is to live in a world of inescapable misfortune and tragic choices. Mental illness strikes with a random yet pandemic quality, producing at its worst nightmarish hallucinations, crippling paranoia, unrelenting depression, a choking sense of panic, and uncontrollable obsession - and even in its milder manifestations it produces many other oppressions. Approximately one in three Americans will experience some form of mental disorder at some point in their lives, and according to one estimate, one in every 6.4 adults is currently suffering from some form of mental illness.[1] This figure comes to 41.2 million people and climbs higher if substance abuse is included. A significant number of these - more than 1.7 million Americans - suffers from a persistent and severely disabling condition, such as schizophrenia. The costs of caring for these patients and those whose illness is episodic, curable, or only mildly debilitating have reached $136.1 billion per year.

As if the tragedy of illness were not pain enough - for individuals with mental illness, their families, and the community - as a society we face additional pain as we are pressured to choose which ill people we will treat and with what services. As health care reform moves forward with the aim of providing affordable, appropriate health care that is available to all, there will be pressure to examine whether any health services are less important or dispensable than others. It is unlikely that mental health will escape scrutiny. Indeed, one of the most serious issues that will have to be faced by the mental health community is how it can participate in the health planning process.

Given the past and present dismal plight of mental health services, both its patients and its providers have good reason to be concerned about reform. Historically, mental health services have not received the same support as physical health.[2] private and public funding permit "carve outs" in mental health coverage that provide fewer benefits than those allowed for physical health. Private insurance customarily restricts mental health benefits more stringently, setting caps on numbers of hospital days or outpatient visits, or imposing annual or lifetime dollar limits for mental health services.[3]

Insurance policies, for example, typically limit hospital care for persons

with mental illness to thirty days a year, according to a 1991 survey of employer-provided coverage conducted by the Health Insurance Association of America. While 99 percent of insured individuals have private coverage for inpatient mental health care, in only 37 percent of these cases is the mental health coverage' equivalent to that for other illnesses.[4] The same is true for outpatient care: although 98 percent of persons with private insurance are covered for outpatient mental health benefits, only 3 percent have coverage equivalent to that for other illnesses. Psychotherapy sessions are generally limited to ninety visits a year with a 50 percent copayment. Health maintenance organizations (HMOs) customarily restrict therapy to thirty sessions a year with a 50 percent copayment.[5] Spending caps (lifetime or annual) limit coverage of disorders such as schizophrenia, bipolar disorder, and autism because of the presumed high cost of treatment and uncertainty of successful outcomes. Unfortunately, mandates generally provide uniform limits on services without regard to the severity of the condition or the effectiveness of the treatment.[5] The public system's provision of care is not much better. With strained, if not reduced budgets, and fragmentation in services among levels of government, those in need of mental health care are often either stricken from the rolls or simply fall through the cracks of a highly fragmented system. In a word, mental health continues to be the poor stepchild of the health care delivery system. As someone once quipped, mental health care is more likely to be a "plea for service" than "fee for service."

The causes of this unequal treatment arise in part from deep-seated convictions, if not biases. Foremost is the belief that mental health and disorder are poorly defined. While few would argue today that mental illness is a myth, or maintain that mental illness is whatever psychiatrists treat, nonetheless there is a bewildering diversity of views about mental illness, ranging from biological accounts to social determinism.[7] The nature of mental illness is often conceived as a dichotomy between mind and body,[8] an unfortunate dualism that tends to minimize the physical suffering and disability associated with mental illness. Oddly enough, even within the mental health field there are those who wish to distinguish biological from nonbiological mental disorders and give priority to the former.

Complicating these matters is the lack of knowledge about the efficacy of mental health treatments;[9] mental illness is perceived as having few fully effective treatments, and the care is typically thought to be lengthy and expensive. In the absence of convincing research, it is difficult to differentiate established interventions from the latest fad. Stereotypes are impossible to break: treatment for mental illness is seen as something patients can do without, since, on the one hand, remedies for the severe and persistently ill are perceived to be almost futile, and on the other, relief for the "worried well" is thought to be discretionary. All people with mental illness are typecast as severely and persistently ill, while in actuality many suffer only single or episodic bouts of illness. The lack of agreement regarding effectiveness even gives rise to conflict among mental health advocates, who argue, for example, over the value of psychotherapy as compared with more medically oriented services, the appropriateness of involuntary hospitalization, and the effectiveness of family-based interventions. Fueling the problem is the variety of ways - depending on the provider, the setting, diagnostic status of the patient, and treatment - that mental health services are defined: should rehabilitation services such as job training or housing be considered part of mental health services? The lack of good data on expenditures and costs fuels the disagreement over effectiveness. Unlike those with a physical illness, persons with mental illness are often perceived to be the moral cause of their own problems and for that reason less entitled to generous benefits. These beliefs have conspired to produce a palpable lack of treatment and funding for persons with mental illness.

If mental health constituencies are not noted for harmonious concord on all issues, on one thing they do agree: mental health must have a higher status in health services. In the present initiatives for health care reform one slogan is prominent: "Health care reform without equitable mental health benefits is no reform at all."[10] Some of those who promote mental health call minimally for nondiscriminatory treatment." Others urge parity, meaning either equal funding or treatment on the same conditions and terms as physical health services. In either case, mental health proponents are calling for the relative place of mental health to be improved and rearranged - for mental health to receive money and attention more on a par with physical health. This call for a higher status does not end at mental health services; state and federal movements for health care reform can be construed as a call for new priorities in general. Initiatives to set priorities range from a casual attempt with no special system in mind, to a more formal attempt that is somewhat systematic but employs no numerical ranking, to a formal system that includes an explicit effort to ascertain citizens' opinions and numerical ranking. The most notable formal attempt has been that of the state of Oregon, which has proposed a highly detailed priority scheme that actually ranks mental and other health services together in a single system of condition-treatment pairs.

Whether one looks at the problem as one of mental health's demanding higher ranking, or health service in general requiring a regrading, extremely difficult questions are raised by the process. Can mental and physical health be compared, and if so, how? If mental and physical health are compared, will mental health services suffer? Should mental health services be ranked at all, and if so, should this be done separately, and by what criteria? Does it harm mental health services to set their priorities separately.? What in mental health services is more or less important, more or less dispensable?

In practical and unfortunate terms, the priority debate is often framed as a choice between alternative treatments or competing populations - for example, should society make available a treatment for schizophrenia, such as Clozapine, or should it concentrate on providing talking therapies for those who suffer from milder and perhaps only episodic forms of mental illness? Or both, and in what proportion? Framing the question this way often assumes, however, that there is agreement about the general terms of the emerging controversy, such as what it means to set priorities, or what criteria should be used, or what counts as care for persons with mental illness. Even if there were agreement that priorities should be set in the first place (and there is not), and the criteria were clear (and they are not), the process itself forces a reconsideration of issues central to our life together as a community. Who shall make the decisions about setting priorities? This question squarely frames another: who shall govern? In a democracy, how shall we balance expert opinion and public preferences? Are consumer preferences coterminous with what is best for society, and if not, which of these should prevail in the decision?

Justifying Priority


This report explores the issues enumerated above and others central to setting priorities in mental health. It examines three components essential to understanding priority setting. First, it surveys the purported reasons both for and against setting priorities in mental health. This must be done in order to answer the question of whether priorities ought to be established at all. Second, the report identifies de facto priorities in the history of mental health policy in the U.S. as well as in the state examples of Oregon and New York. Probing these cases reveals a complex set of social factors - funding mechanisms, intergovernmental relations, and cultural perception of mental illness, to name a few - that propagate a nonlinear, often conflicted development of priorities in mental health. Third, the report appraises the ethical criteria that have been proposed to evaluate priority plans. Finally, the report identifies the present state of affairs in priority setting (where we are and where we need to go) and what mental health policy has to gain from discussions of this kind.

Priorities: Why We

Need Them

At the mere mention of setting priorities, some vehemently object that there is insufficient motivation to do any such thing. Yet the question has surfaced because of the dominance in health policy of three concerns: to contain rapidly rising health care costs, to assure access to health care for millions who are without it, and to give health care that is appropriate.

The rising cost of health care and its crippling effect on budgets, private and public, motivates in large measure a need to set priorities. American corporations find it increasingly difficult to compete internationally, impaired in part by rising employees' and retirees' health insurance costs. Most international competitors have health services paid for by their governments, leaving them lower overhead, lower prices, and a stronger competitive edge. Individual consumers feel the pinch of higher costs as expenses are shifted to them in the form of higher copayments for health services. State and local governments too, faced with rising deficits and tighter budgets, have identified escalating health care costs as an impediment, if not the sole obstacle, to gaining control of their budgets. It is not likely that society can curb the accelerating cost of health services and at the same time offer everyone every service of any possible benefit. Choices will have to be made about what services are more or less vital or expendable.

The national push to insure the 37 million Americans who lack insurance of any kind, and to offer an adequate level of coverage to those who are seriously underinsured (another 30 million) is further impetus to rethink the present distribution of health services. It is generally assumed that a nation as wealthy as ours ought to be able to provide all of its members with some level of health care, but to do so will entail either further increasing a seriously strained health care budget, or rearranging the present distribution of health services in a fairer and more agreeable manner. While access to care in general is already somewhat limited, for individuals with mental illness it is much more restricted." The practice of medical underwriting discriminates against those who have received mental health care in the past, often by denying them coverage altogether, and "mental disorder" is one of the most common conditions for which a medically underwritten group is denied coverage. Even though some twenty-two states have enacted "bare bones" or "basic benefits" legislation to permit insurers to market low-cost, low-coverage benefits to those purchasing insurance, the tangible effect for people with mental illness is often attenuated because of gaps in coverage, very high deductibles, or inability to pay for treatment.

Professionals and consumers demand health care services that are appropriate, which further motivates a need to set priorities. Emerging data suggest that certain health care services are often ineffective and wasteful, and deserve either low priority or none at all. Even if health care dollars grew on trees, simple rationality dictates that useless services be eliminated and marginally beneficial services be given low priority. There is mounting information in the field of mental health to determine what treatments are beneficial for what populations. For example, it is wasteful to provide effective services to people who do not need them, and it is increasingly possible to determine who they are. People who have never fully met the criteria for mental illness should not receive monies apportioned for mental illness, even though this has happened in the past and still occurs in some states. Likewise, it can hardly be a priority to provide treatments to inappropriately diagnosed conditions - for example, treating the physical symptoms of anxiety disorder rather than the disorder itself. Ehminating the useless is the first step toward setting priorities. More difficult will be to make the judgments ranking marginally beneficial treatments, if they should even be included.

Ironically, these concerns are often at odds with each other. If, for example, costs are kept under control, a system is unlikely to be able to assure desired amounts of high-quality, appropriate care to everyone. And if universal access to care is offered, the system could bankrupt itself in the attempt to maintain quality.

As a practical matter there may seem little point in publicly examining reasons to set priorities. Many proponents of priority setting recite a litany of examples to show that priorities are now and have always been set, but that the process takes place out of the public eye and in no methodical fashion. The present mix of services constitutes de facto priority setting, even though the story of how resources came to be distributed as they currently are is difficult to trace. It seems politically prudent, even if one thinks it wise or even necessary to set priorities, not to be forced to acknowledge a preference for one service or population over another. Yet when, for whatever reason, die fact of existing priorities goes unacknowledged, they are set by default in a constant shell game of cost-shifting. Ignoring underfunded public mental health services, for example, in an effort to avoid drawing attention to the question of priorities at all, may have the net effect of shifting the cost of caring for persons with mental illness to the welfare and prison systems, or to consumers in higher out-of-pocket costs. However ingenious or devious this method of redistributing the burden to public budgets other than mental health services, the cost for care still remains. So does the question of whether and how best to provide it. And if the concern is to stop a potentially or actually violent person with mental illness from harming others, so does the question of which agency is responsible for assuring the public safety in this case.

Priorities: Why We

Resist Them

Setting priorities is unthinkable! The foremost hurdle to setting any health care priorities is that many people vigorously resist the idea that they ought to be set in the first place. Setting them, critics allege, is a pernicious and unthinkable act. In the face of suffering, can anyone in good conscience turn away from the ill? Is there ever any justification for deciding that certain populations are more important than others, or that some services are dispensable and there is no social obligation to offer them? What justification would be sufficient for society's omitting to offer those services, and where can a line be drawn between what society is and is not obligated to provide?

The instant abhorrence to setting priorities stems from equating the enterprise with rationing. The two are not necessarily the same. While priority setting can mean different things to different people, it is broadly construed to mean ranking. The ranking can be used for at least three purposes: to provide a wise distribution of health services in a relatively affluent society; to identify low-preference, marginally beneficial services; and to eliminate beneficial but less needed care. Rationing, only one aspect of setting priorities, can be defined as the denial of services demonstrated to be beneficial.[13] Yet by potent innuendo, the connection of priorities with rationing frequently puts a halt to any reasoned discussion. In our land of plenty, known for its largesse, where people have become accustomed to the postwar expansion in the standard of living, the mere suggestion that we will have to live with scarcity and perhaps be denied goods that most of us perceive to be essential cuts sharply against the grain of our cultural identity.

Rationing aside, however, it must be acknowledged that the process of setting priorities serves other purposes as well. Grading services along a continuum of efficacy or cost, for example, can inform consumers and enable them to select services that comport with their values. Ordered priority lists can, by implication, identify for researchers lacunae in needed health services, or potential areas for basic research. For private venture capitalists and those who control the public purse, decisions can be better informed when strategic planning for basic scientific research, data collection, and the provision of services includes ranked priorities.

As a practical matter, priority setting may result in rationing, but in fairness to both processes, the objections to each must be examined separately. In either process, perhaps the least easy objection to address is that lodged against the means used to accomplish the task. Because there are so many different proposals for setting priorities and no homogeneous means of doing so, objections to the means by which priorities are set usually comprise a myriad of objectionable subissues, including objections to what is being ranked, or what criteria - such as cost or effectiveness - are being used, or whether certain procedures are observed. Some plans sort by populations - for example, by targeting persons who are severely and persistently mentally ill - while others sort by emphasizing broad categories of services such as outpatient care or emergency services. Others sort by conditions and treatments, and still others never sort and mean nothing more by priority setting than that some broad public policy issues need more urgent attention. Objections will vary according to what is sorted and ranked. It is important to remember that a declaration that priority setting is objectionable tout court is likely to be an objection to individual aspects of the process, each of which must be examined separately.

Setting priorities harms mental health! Many would argue that priority setting would harm mental health. If setting formal priorities within the health care system in general is not so unthinkable after all, it is nevertheless perfectly true that to single out mental health alone for priority setting could be harmful.[14] It has been argued that there are no similar attempts to set priorities in intensive care or ophthalmology, and that to set them for mental health services alone would be to countenance existing patterns of discrimination and to exacerbate battles among constituents who, in our most recent history, have been competing with one another for a limited budget.

While the possible harm in setting priorities for mental health only is hypothetical (it has not proved harmful in Alameda County, California) it makes little sense in the long run. Practically speaking it is not easy to disentangle treatment for mental and physical health, given their interrelated symptoms and overlapping places, modes, and providers of treatment. The motivation for setting priorities in general is to achieve a sensible, cost-conscious balance of resources within all health services, not simply one service. To this end it is imperative not to exclude any health service, mental or physical, from the analysis. Those within mental health have good warrant for insisting that priority setting should be an all-or-nothing process.

However, if all health services are prioritized inclusively, examining the priorities within mental health as part of a larger enterprise is a sensible thing to do. These services are traditionally administered for some portion of the population in public, state-operated programs and funded separately, and for that reason mental health requires its own analysis. More important, focusing on mental health places in bold relief certain features of health care services that are all too frequently overlooked in health care reform. In large measure health policy has tended to privilege short-term illness served by general hospitals, where the emphasis is on acute care.[16] Because much (but by no means an) of mental illness is chronic or episodic it does not fit this norm.[17] Because it does not, it can serve as a model for thinking about priority setting in other areas of chronic and long-term care as well - areas such as the care of the elderly, which is sure to require a larger proportion of the health care dollar in the near future. Further, it would be a disaster in the long run to overlook the services that are used to identify, diagnose, and treat mental disorders, including a range of social services offered in the stabilization and rehabilitation of certain groups of persons with mental illness.[18]

Setting priorities is impossible! A question that could promptly stop priority setting is whether it can even be done. History provides strong evidence that priorities are, by and large, difficult to set deliberately. They are established, many would argue, through political processes and organizational power and have little to do with a systematic examination of alternatives and their social and economic significance. Comprehensive plans such as those under discussion belong more to the imagination of idealistic public policy, makers than to pragmatists. All too frequently the best planned course of action is waylaid by a complex set of social factors. Mental health stands as an important case in point, buffeted as it is by a host of interrelated and conflicting factors, including its charismatic leaders, intergovernmental responsibilities and failures, legislation, court cases,[19] external funding, conflicting understandings of the nature of mental illness, composition of the mentally ill population, professional ideology, interprofessional friction, popular attitudes, consumer demands, and the means to deal with disease and dependency. Nevertheless, we believe these social factors merely make priority setting difficult, not impossible. In the face of formidable odds, the alternative is to do nothing; and that has not worked well.

Priorities: Some


Setting priorities is sensible! It is important not to undervalue the need to set explicit, formal priorities. Even in an affluent society, one not constrained by budgetary limits and one in which almost every beneficial service could be offered to every citizen on demand, setting priorities is a sensible way to safeguard health care budget,s, which are always vulnerable to political pressures and market forces.[20] The lack of evidence about how contemporary priorities have been set suggests that the process has been hidden from the public view. Political pressure and public attention often skew informal priorities in the direction of interest-group concerns that may be at odds with what is good for a society as a whole. Market forces left to their own devices are, in part, responsible for the lack of access to health care for some, but by setting priorities and using governmental mechanisms to adjust for supply and demand, the maldistribution caused by the market, can be compensated for.

Most importantly, priorities might bring balance to a lopsided health policy. An examination of general health policy reveals that society usually hews to the "rule of rescue," that is, it cares for individual, identifiable persons, no matter how desperately ill or how ineffective or costly the treatment, even though this may mean that faceless patient populations must do without beneficial care. The present system is geared toward meeting individual patients' health needs, with less concern for what mix of services will have the most impact on the health needs of the population. While it may be difficult to convince people of it, setting formal priorities is a sensible means of expanding the focus of health care policy beyond individual patients' needs to the health needs of the population as a whole.

The inducements for setting priorities - cost, access, and appropriateness of treatment - are the fuel for all health care reform, a fate we will not likely escape. Explicit or not, priorities will be part of the machinery for health reform and objections to setting them will persist. Some objections will be resolved factually. Will people with mental illness be made worse off if priorities are set? Will social forces make it virtually impossible to set sensible priorities? The merits of other objections will stand or fall according to the coherence and adequacy of criteria for setting priorities. These objections notwithstanding, forces acting on health policy such as political influence, the market, and current biases in health policy that do not fully account for population needs provide some incitement to entertain thoughts about priorities. The question we face is not whether we should set priorities, but whether priorities will go. Is the present direction of priorities acceptable, and if not, what is the alternative? What measures are there for evaluating the setting of priorities?

Priorities: Informal

and Formal

Talk about priorities is a permanent conversation in the political forum; however, the contours of the meaning of the term are continually being sharpened. In health policy, priorities have traditionally been informal and meant nothing more than that particular services or populations would receive special attention. More than a century of mental health policy reform bears witness to the setting of priorities in this sense. In a more recent and specific sense, priority setting has come also to mean a formal and methodical attempt to redirect health services, sometimes using numerical rankings, as Oregon has done, or at other times using existing health services strategies to effect change, as has been done in New York State. An examination below of the broad cycles of mental health reform in the U.S. uncovers social factors that have bent policy to the prevailing ideological breezes, while a look at the expeiience of Oregon and New York exposes striking and disputed features of the current situation.

Cycles of Priorities

Priorities, informal but potent, may be glimpsed in the four cycles of mental health reform that took place in the nineteenth and twentieth centuries." With each reform a new set of priorities (normally informal) "established, based on newly developed approaches to treatment and a new or different facility or locus of care. But each set of priorities was profoundly affected by social, political, economic, and demographic factors. Before the nineteenth century, care for lunatics, as they were called at that time, was charitably dispensed by the family and informal community networks, and not the government. A sharp departure in the kind of care occurred in the first reform cycle early in that century, when reformers championed a view of "moral treatment." A belief arose that those newly diagnosed could be cured of mental illness if provided with humane care and instruction in small, pastoral asylums. With a consensus emerging that the severely insane should be committed, public policy made it a priority that a public state asylum system should be created that offered care for acute cases.[23]

During this first cycle of reform, the shared responsibility for the health and welfare of the citizenry resided with the state and local governments and not with the national government. Yet there was no uniformly acknowledged responsibility for persons with mental illness - it varied from community to community. The state usually purchased the land for mental hospitals, built the bldiding, and paid for some employees. In fact, the cost of administering public asylums constituted one of the largest items in many state budgets, suggesting that institutional care had a high priority.[24] The local community retained responsibility for the day-to-day care of those not dangerously insane, but since the community was also financially responsible for those of its citizens who received care in the state institutions, there was an incentive to keep persons with mental illness in the local almshouses, where per capita costs were low. Despite this incentive, an increasing number of patients were admitted to state asylums and failed to recover, thus requiring long-term care.[25]

The increase in the number of institutionalized chronically ill patients was accompanied by early examples of questions related to priority setting. Should the state build additional hospitals? Did the presence of chronically ill patients undermine the therapeutic goals of the state asylums? Should the local community retain patients with chronic or severe mental illness in almshouses, or was it better to put them in other institutions? What level of government - local or state-should provide the greatest level of support?

As chronic patients became the priority of the public system, attempts were made to shift the responsibility for this priority to the federal government. A leading actor in this drama was Dorothea Dix, who attempted to persuade the federal government to sell ten million acres of land, the proceeds of which were to be used to support the indigent insane. But she was up against a pervasive belief stemming from the colonial period that for most issues local government should retain the authority to look after its citizens. This belief about local autonomy was crystalized in the Tenth Amendment to the Constitution, which holds that powers not delegated by the Constitution are reserved to the state. As responsibility for health and welfare was taken to reside with the state, the federal government played no role whatsoever in mental health policy during the eighteenth and nineteenth centuries. Dix's proposal was nonetheless passed by Congress but vetoed by Franklin Pierce, who justified his actions in these words: "The fountains of charity will be dried up at home, and the several States, instead of bestowing their own means on the social wants of their people, may themselves ... become humble suppliants for the bounty of the Federal Government."[26]

A second reform cycle, that of "mental hygiene," began in the early 1900s with aftercare services, occupational therapy, and outpatient clinics as a priority. A coalition of social reformers and physicians lobbied for an end to dual responsibility between the state and local governments; insane persons were to become wards of the state. This became the general rule until the 1940s, by which time care was being provided to some 425,000 residents. The most critical downstream effect of this shift in responsibility was that, as localities transferred their charges from the almshouses to the state, they also transferred responsibility for care of other residents almshouses, especially the demented elderly. To aid this shift, local officials began to redefine "senility," as it was then called, in psychiatric terms.[27]

A third cycle of reform, that of "community treatment," shifted the priority in the mid-1950s from a hospital-based system of care to a community-based system, through a state-local partnership for the delivery of mental health services. Community treatment became the priority due o shifts in the ideology of psychiatry and newly available interventions. The experiences of World War II supposedly demonstrated that community and outpatient treatment of people who were mentally disturbed was superior and more efficient.[28] Psychodynamic and pshychoanalytical modalities of therapy prevailed, along with a view that pshychiatry, in collaboration with other social and behavioral sciences, could ameliorate the social and environmental conditions that played such an important role in mental illness. Also prevailing was a view that early identification of persons at risk, along with treatment in the community, could stave of course biological treatments, including psychotropic drugs, facilitated the shift to community care.

Most prominent in bringing about the shift was Robert H. Felix, a pschiatrist and chief orchestrator of the National Mental Health Act of 1946. His goal was to employ the prestige and resources of the national government to redirect mental health priorities by weaning the nation from its reliance on mental hospitals and replacing them with a network of community institutions. Arguing that a community-oriented policy would be more efficient, he was able to define mental disorders in public health terms by establishing a new system of outpatient community clinics that would provide preventive and therapeutic services for persons with mental illness.[29] As part of the act, the National Institute of Mental Health (NIMH) came into being. The new institute was to perform a number of functions: to speak to a national constituency, to frame a national agenda by collecting data and providing them to public policymakers, to promote research and demonstration projects, and to expand the number of community clinics.

The shift in priorities culminated with the passage of the Community Mental Health Centers Act of 1963, which established federally funded local centers for mental health care. The regulations had strikingly little reference to state mental hospitals and thus the community centers served quite a different clientele. To guarantee their development, the community centers were organized to be highly responsive to community pressures and reflected local priorities by catering to problems encountered frequently within the community: marital and family difficulties, children's emotional problems and delinquency, and substance abuse.[30] As severe and persistent mental illness plagued only a tiny minority of the local citizenry, and as the centers had little formal connection with the state mental hospitals, the care of people with severe and persistent mental illness diminished in priority. In any case, caring for individuals with mental illness in the community was time-consuming and arduous.

Priorities also shifted in response to funding. The most dramatic change was effected by the Medicare and Medicaid amendments to the Social Security Act. The enactment of Title XVIII (Medicare) provided insurance covering hospital stays and physicians' fees for the aged, while Tide XIX (Medicaid) allotted money to the states for medical assistance programs for the indigent. Medicaid permitted a rapid decline in the number of elderly in mental hospitals. The reason for this was that while Medicare severely limited payments to support die elderly in mental institutions, states could employ Medicaid payments to care for the same elderly patients if they were shifted to nursing homes. This decline in the population of chronic patients in state hospitals generally had the unintended effect of improving the quality of care available there.[31]

The fourth and most recent cycle of reform, that of "community support," began in the late 1970s and took as its priority the correction of the weakness of the preceding cycle. The earlier reform had failed to recognize that many formerly institutionalized patients had needs that were once met by the institution, which provided food, shelter, work, recreation, security, and mental and medical care. Once these patients were outside the institution, the need for these goods remained but there was no one to provide them.[32]

Three important social factors converged in this cycle to transform the context and substance of mental health priorities: the return of persons with severe and persistent mental illness to the community, demographic trends in the population as a whole, and bureaucratic changes within the mental health system.[33] The return to the community of persons with severe and persistent mental illness was hastened by the growth of a series of federal welfare programs, including housing programs, food stamps, the expansion of Social Security Disability Insurance (SSDI), and the creation of Supplemental Security Income (SSI). During the 1980s the federal government, through the Omnibus Budget Reconciliation Act, drew back from the mental health responsibilities that it had assumed in the postwar decades by cutting budgets for mental health services to the states. Demographics made the story worse because the disproportionately large size of the "baby boom generation' meant that the number of persons at risk for developing severe mental illness was high. The mental health system changed too, with the emergence of a decentralized psychiatric system and more and more responsibilities devolving to the states and local governments. The consequences of this "New Federalism" of the Reagan presidency are only now appearing within certain populations of those with mental illness who, while formerly enrolled in services, are now being struck off the rolls or dumped.[34] Given the increasing amount of bold experimentation at the state level, the number of patients "falling through the cracks" and not receiving care appropriate to their need is only likely to grow. The problem of setting priorities in this cycle of reform will be in part to establish clear lines of responsibility among the several levels of government. In particular, the push for mental health reform within local community health alliances puts pressure on state governments to determine their exact role in and responsibility for the care of certain populations, such as those with severe and persistent mental illness. The shifting structural relationship between state and local levels of government is increasingly likely to be a focus of the resource allocation debate.

This brief record of cycles of reform discloses several features that thinking about priority setting must engage. Chief among them is that each cycle has had distinct priorities, sometimes set proactively, other times reactively. To the extent that priorities were proactively set, it can be claimed that deliberate planning is in fact possible. To the extent that priorities were reactively set, often unpredictably and incompatibly with public planning, planners can be alerted to the dangers of simplistic solutions. Comprehensive plans are not merely the fruits of idealistic imaginations, but of pragmatists who can keep clearly in mind the history of social forces.

Lessons from Oregon

The state of Oregon riveted the attention of those interested in priority setting when, through legislation, it enacted the most explicit and formal priority plan to date.[35] At once an object of diverse bitter attacks and high accolades, Oregon's proposal has been condemned as unfair because it discriminates against the poor and the handicapped, and it has also been held up as a model of fairness because it offers access to care for all Oregonians living at or below the poverty level. Most uniquely the Oregon Health Plan straightforwardly addresses the integration of mental and physical health. But the plan, which has been almost five years in the making, has only recently been approved and has yet to be implemented. Once it is implemented a certain amount of time will have to pass before assessments can be made about how well it works. The question then will be whether general conclusions for mental health services can be drawn from the Oregon experience.

In 1989 the Oregon legislature passed several bills that created the Oregon Health Plan.[36] The intent was to achieve three goals: universal access to health services, reasonable provider reimbursement, and cost containment. The plan was to develop a prioritized list of services, the most important of which would be defined as basic coverage for the 400,000 Oregonians currently without medical insurance. This coverage applies both to the Medicaid population (thus expanding coverage to virtually everyone at or below the federal poverty line) and to uninsured workers through legislation requiring employers to "pay or play."

The Oregon Health Plan established a Health Services Commission, whose first task was to develop list of health service priorities based on community values about health care and on technical information about the effectiveness of various health services relative to the needs of the population serviced. The initial proposed legislation contained a glaring omission: mental health care had, once again, been marginalized. This confirmed both for those who were mentally ill and for those who cared for them that they were being victimized, not only by inequalities in coverage and reimbursement rates, but also by the new initiative. Lobbying efforts led to legislation modified to include a Mental Health and Chemical Dependency (MHCD) subcommittee to consider whether and how mental health could be integrated and prioritized.

The process used to develop priorities was anything but simple. By the time an agreed-upon list was developed in 1991, there had been many negotiations and adjustments on the part of the commission and among three of its committees: the MHCD, Health Outcomes, and Social Values. The Health Outcomes Committee initially decided to value and sort on a model derived from the work of Robert Kaplan. Simply put, the model associated health care conditions with an appropriate treatment. For each condition-treatment pair, outcome and cost data for both treating and not treating a condition were to be obtained and entered into a quality of well-being scale (QWB), producing a mathematical value. The QWB scale was modified with information gained from the Social Values Committee, which had collected information about individual and community values. Individuals' values were collected by means of a random-sample telephone survey of the Oregon population and the responses were aggregated and used to weigh health care outcomes. Community values were obtained via forty-seven community meetings held throughout the state.

Once they arrived at a sense of how Oregonians felt about various states of health, the Health Services commissioners ultimately identified and ranked seventeen categories of care as the primary determinants of priorities. Woven into community values was information gathered from expert panels about the outcomes of treatment for individual conditions. The commission's first attempt to combine condition-treatment pairs into a ranked list by using a method of strict cost-benefit ratios proved seriously defective. Practitioners were reluctant to disclose the cost of treatment, and reliable data were so difficult to acquire that the figures were finally dumped except for use as tie-breakers when two net-benefit scores were identical in the same category. A second list using a net-benefit ratio was developed by subtracting the probable outcome if a condition were left untreated from a probable outcome of treating the condition.

Blending mental and physical health was no minor political task. From the outset the MHCD committee wondered whether it could in good conscience even join the prioritizing process at all.[37] its reservations about equitably prioritizing mental and physical health stemmed from the Kaplan methodology, which was biased in favor of acute conditions. In choosing to go along with the process, mental health services stood to lose funding from its separate budget as well as from reduced insurance mandates. On the other hand, if it opted out of the ranking process, mental health services might receive reduced funding and the old patterns of segregation and discrimination would certainly be perpetuated.

The committee agreed to join the process while stressing two goals. The first was to see to it that all those involved with the process understood that mental health deserved parity with physical health. The second goal was continually to critique the commission's methodology, noting where it must be modified to ensure a fair assessment of the value of mental health services. The committee urged the commission to modify the Kaplan QWB public survey instrument to include more symptoms associated with mental illness, while at the same time pressing for all due speed in collecting the data for mental health. Even though they obtained the best data available, the committee retained reservations about its quality. These were somewhat assuaged by comparing notes with other groups and discovering that the outcome data from other specialties was no more reliable than the mental health data.

Once the MHCD committee sorted and ranked mental health conditions, the health list was reviewed by mental health experts within and outside the state. As the committee expected, the list turned out to be plausible and valid, with the major mental conditions falling appropriately into the highest categories and integrating well with other health conditions. Committee members had had to anticipate doubts that such an outcome would be valid, and they did so by adhering wherever possible to the methodology used to prioritize medical-surgical conditions. The list integrates mental health conditions irrespective of their etiologies; conditions and treatments are ranked on the basis of their severity as experienced by the patients, while benefit is measured by outcomes. The list gives a high priority to the treatment of biological illnesses such as schizophrenia and major mood disorders. It also gives priority to conditions that cause widespread suffering and loss of functioning (such as eating disorders) and to illnesses that have a disproportionate impact on children.

Oregon's Contribution

If the most dramatic contribution of the Oregon Plan was that of placing formal, public priority setting on the public policy agenda, an equally stunning gain for mental health was the MHCD committee's single-minded approach to assuring parity between mental and physical health. No one could dispute evidence of discrimination and lack of parity toward mental health; what can be and widely is disputed is MHCD'S driving claim that mental health can be compared with physical health. The MHCD committee's strategy took the form of a pragmatic, relentless, several-tiered approach whose dominant characteristic was a refusal to argue about whether there should be parity. Instead, they concentrated on how to assure parity. Not only did the committee make certain that mental health was included in the process, it also ensured at every turn that mental health would not be put at a disadvantage by being underrepresented on the final list of ranked condition-treatment pairs.

The MHCD committee did not advance reasons for rejecting a distinction between mental and physical health in evaluating public policy. They took a more positive approach. From a medical perspective health and illness are inextricably part of mind and body; disorders of the body affect the mind and vice-versa. Mental health is not merely a subset of physical health but an integral part of physical health. Regardless of whether mental health services are medical treatments or other kinds of care, no principled reason exists to discriminate between care provided to those who suffer mental illness and those who are physically ill. Practically speaking, this means that mental health should be considered on the same terms as physical health, but this is not to say that they must receive parity or equal amounts of funding. Differences in funding can be tolerated so long as the same standards for evaluation are applied to both.

While it is difficult to rank which feature of the Oregon plan is most dramatic, high on the list is the criterion for valuing and sorting condition-treatment pairs according to effectiveness. It is controversial because it requires that the most beneficial treatment be given high priority and low-benefit treatments be given low priority, and as a consequence some services might not be offered. One criticism of the effectiveness criterion is that the concept of effectiveness is inherently malleable. Something that is ineffective for a cure might be effective as palliation. Which kind of effectiveness counts? Oregon tried to grapple with this problem democratically, by ranking according to broad areas of effectiveness that Oregonians said they preferred. For example, treatments for acute fatal conditions that effected a full recovery received highest priority, while treatments for both fatal and nonfatal conditions that produced only minimal or no improvement in the quality of life rated the lowest priority.

Another criticism Oregon had to deal with was that mental health services might fare worse if the effectiveness criterion were used. This criticism turned out to be unfounded. Some mental health services in fact ranked in the highest categories of effectiveness. Acute fatal conditions for which treatment prevents death and provides full recovery include major depression, drug-induced delirium, and single-episode, post-traumatic stress disorder. Acute fatal conditions for which treatment prevents death but cannot always deliver a full recovery include alcohol and drug abuse diagnosis. Chronic fatal conditions for which treatment improves life span and quality of life include dysthymia, eating disorders, bipolar disorder, recurrent major depression, and schizophrenia. To the surprise of critics, Oregon's method actually improved the plight of individuals with mental illness over what it had been before formal priority setting was attempted.

Some challenges to the use of the effectiveness criterion remain unanswered. One, for example, is whether certain highly effective treatments might at some point financially squeeze out less effective treatments. This depends on how many effective but costly treatments there are and how much the legislature is willing to fund. Another is the exact rationale for giving effectiveness a privileged position. Why this and not other important goals of society, such as cost, or prevalence? To its credit, the Oregon plan attempts to meet this objection by granting the health commission the authority to hand-adjust the list to reflect these other values after the computations have been made.

Oregon's plan also made headway in that its actions spoke clearly about who should set priorities, although the rationale behind these decisions can only be surmised. In the broader, national discussion on priorities those who should be actively involved in the process include experts drawn from the professions, elected or appointed officials, independent representative health boards, or the public at large. The Oregon Health Services Commission opted to empower representative groups, including physicians, other health providers, and lay members, each group functioning as a check and balance on the others. For example, given the task of ranking the effectiveness of treatment, a plan to set priorities might have deferred to the judgment of professionals alone, on the assumption that they have the proper expertise for the question of setting priorities. Oregon's actions suggest this assumption can be challenged. Those involved in the planning seemed to see clearly that discussion about setting priorities is not purely scientific or value neutral; values impinge on every step of the discussion. Professionals have no special expertise in deciding how the effects of a medical intervention ought to be valued by the society. Oregon avoided this pitfall by filtering the assessment of effectiveness through solicited public preferences.

But the checks and balances offered by public preference are attended by shortcomings of their own. It is to Oregon's credit that the plan is informed by a good inventory of the values of the community who must live with the results, yet if broad societal participation is an important means of achieving a vision of the overall good of society, it is also true that a good statement of community values goes beyond a mere aggregation of public preference and extends to incorporating perspectives of what is good for society. 38 The voices of all parties affected by the distribution of services must be heard, and those who cannot speak for themselves must have effective advocates. That is the first step. However, there is a further task, one not addressed by Oregon's actions. In this most important second step, the task is to blend public preference with society's needs.

The Oregon experiment identifies and resolves some but not all of the practical objections raised against attempts to set formal priorities in mental health. The results not only argue for parity, but in fact demonstrate that mental health services need not be worse off under formal health care priority plans. Moreover, integrating and ranking mental health with physical health produces the extra dividend of allowing planners to appreciate the value of preventive and long-term care services -varieties of care that have too often been neglected in the past. However, consequential issues remain. Do democratic principles require that citizens have as much say as they did in Oregon? Does the merger of mental and physical health i la Oregon present any drawbacks?

Lessons from New York


The state of New York differs from Oregon State in its "size, complexity, and diversity of mental health populations, providers, and service settings,"'[39] and it serves as an important contrast to Oregon's highly formal and well-publicized priority-setting initiative. It is undoubtably difficult to generalize from New York to other states, but we can safely say that compared to Oregon, New York's means of setting priorities are more typical. The state ranks populations and services by formal means (in the sense that they are explicit), but unlike Oregon the practical consequences for redistribution are not immediately identifiable. like many states, New York has experimented with setting priorities by (1) focusing on interventions that reduce inpatient costs; (2) introducing nonmedical interventions into the health care system; and (3) identifying priority populations.40 Like many states, New York's priorities and plans for service are defined by a host of influences, including large-scale implementation of managed care plans, cost control in government entitlement programs, and vestiges from earlier cycles of mental health reform, such as inpatient state hospitals. Until the mid-1980s public mental health systems were largely state-controlled and self-contained. Priorities were operationalized by the state's executive budget process. The bulk of the expenditure went to pay for inpatient care, which still makes up 62 percent of New York State's entire mental health budget. While every state directly operates one inpatient hospital service, only fifteen states (New York is by far one of the largest) operate ambulatory services. Most states fund a network of nonprofit community providers that offer outpatient treatment and support services.

During the 1980s many states formed priorities by responding to Medicaid funding, which increased monies for certain important populations such as children, and which allowed for optional Medicaid services such as targeted case management, clinic services, and rehabilitation.[41] Perhaps exceptional in its ability to expand services dramatically without increasing state appropriation, New York was able to enlarge its public mental health budget between 1986 and 1993 by $1 billion, 85 percent of which was captured from Medicaid funding.[42]

One of the crucial moves toward setting formal priorities in the states occurred in 1986, when the federal government enacted the State Comprehensive Mental Health Planning Act (Public Law 99-660), which created a high-priority obligation to serve targeted populations. The law required that statewide service plans describe the prevalence of persons with serious mental disorders and delineate strategies for delivering services. Prioritizing by populations led planners to identify a preferred set of services unique to specific groups of persons.[43] Prior to this law, die distribution of community services was left to market forces with little intervention from the state; providers determined which populations with mental illness would receive priority. Consequently services were lopsided. During the 1980s in New York the number of licensed outpatient treatment programs doubled, but the modality of treatment in this setting was psychotherapy, which is not necessarily appropriate for persons with severe and persistent mental illness. While the amount of service grew in the aggregate, only certain geographic areas benefitted. Midtown Manhattan, for example, had an oversupply of certain services, while other parts of New York City offered virtually no services of any type.[45] By the late 1980s emergency rooms and psychiatric units in general hospitals were routinely overcrowded, despite the 1,200 new acute psychiatric beds added to these hospitals in die 1980s. In fact, emergency rooms became the staging areas for a variety of social problems-homelessness, substance abuse, and criminal behavior. Studies demonstrated that a relatively small number of individuals was using a disproportionate number of emergency services and inpatient days. Patients were literally "bouncing between emergency rooms, general hospitals, state hospitals, outpatient services, jails, and homeless shelters without ever being effectively linked to appropriate services."[45]

In the late 1980s these problems, along with a faltering economy, growing unemployment, declining tax revenues, spending caps, and setbacks in government social welfare programs, forced states like New York to do what had at one time been unthinkable, namely, set formal priorities. Without increasing state revenues, priorities were set by reallocating resources, principally by closing state inpatient hospital beds and reinvesting the resources in community outpatient treatment. The state accomplished the task by using traditional always been available, including (1) certificates of need (CON) for granting expansion of inpatient and outpatient services; (2) needs assessment to services; (3) licensing to oversee the services provided; (4) rate setting and financial incentives to promote services for targeted populations;and (5) a formal, comprehensive plan to redirect resources. The priorities it set were guided, n the one hand, by a mandate to provide services for a number of very needy patients and, on the other hand, by pressures to reallocate existing resources by offering new, less expensive, and frequently less intrusive services which were expected to respond more reasonably to unique patient needs. The service strategies given priority were (1) to increase access to psychiatric emergency services; (2) to develop case management programs for certain high-risk patients who consume a disproportionate share of services; (3) to develop a broad array of community housing; and (4) to refocus outpatient programs toward outcomes associated with rehabilitation and recovery.[46]

New York's Contribution

New York State's attempt at setting mental health priorities offers an important alternative to Oregon's. It strongly suggests that, despite daunting circumstances that might inhibit comprehensive planning, public policymakers have a clear chance to bring about a fairer and more sensible distribution of services using means already available. Interim reports on this process indicate that the number of inpatient beds is decreasing and that resources have been reallocated to the designated areas without increasing state funding.[47] Data about the overall impact on the states' mental health services remain to be gathered.

Despite these remarkable advances, New York and other states that are using already-available health policy techniques to redirect priorities will most likely fall victim to some of the unresolved conundrums in priority setting. The most obvious of these is associated with the process of decisionmaking. The historic role of the mental health authority to plan for people with mental illness remains firmly entrenched, directed by legislation and implemented by a commissioner and a state-mandated advisory council that reviews the state mental health plan. This shared authority has become somewhat more democratic through community participation. In 1988 the Office of Mental Health launched a three-pronged planning effort with expanded participation from a broad range of constituent groups, including service recipients and their families, advocates, representatives of local and state human services agencies, educators, and service providers. To guarantee expanded public participation, the commissioner's Citizens' Mental Health Advisory Committee was established, working through five parallel structures around the state to recommend strategies and initiatives for change. The committee's deliberations are supported by experts who provide information on technical aspects.[48] The locus of decisionmaking, while informed by community values, ultimately resides with the publicly appointed Commissioner for Mental Health.

Surface comparisons between the New York and Oregon experiences have a limited instructive value; if pushed too far, the resemblances become inaccurate or misleading. Still, it is useful to compare the process of decisionmaking in both states. Are there any strong moral reasons for preferring a publicly elected commission such as Oregon's over a publicly appointed official? On first reflection it would seem there is little difference between the two approaches. Regardless of who makes the final decision, whether a publicly elected board or a publicly appointed official, the party in question must be held accountable for integrating a good understanding of community values. There would seem to be little to choose between them. While it is true that independent boards can offer an inter-disciplinary perspective not attainable by one official, and that the final decisions of such a board cannot be dismissed as the idiosyncratic thinking of one person, what this arrangement gains in political acceptability it loses at times in a lack of efficiency; indeed, such boards sometimes experience total paralysis. A single official at least has the capability of acting, although the actions will raise questions of whether a sufficient number of perspectives were integrated into the final choice. Although the differences between them are significant, each decisionmaking process has something to recommend it and arguably each is ethically acceptable. Less carefully examined has been the question of whether priority decisions ought ever to be left to a semi-private forum. In a democracy, must everything be open to those who will be affected by a decision? If not everything, then how much?

Another controversial feature common to New York's experience and the broad discussion on priorities is the targeting of special populations and services. Is it justifiable to prefer services aimed at those made worst off by mental illness? The fear, of course, of those who object to targeting the least well off is that their own interests may not be served, or served well. Quite reasonably, the defense the New York State Office of Mental Health has offered for its action is that federal legislation mandates that states target patient populations. An unstated but equally plausible defense of the practice is that state budgetary cutbacks make it impossible to provide every citizen with every beneficial treatment, which means that some services and populations must be targeted. Yet these actions and justifications do not respond to some profound philosophical riddles associated with the issue of the worst off. For example, whom shall we count as the worst off? Perhaps the worst off have diseases relatively immune to treatment, while the better off could be served well by the limited resources. These problems are not the problems of New York State alone, but inhere in any attempt to set priorities with some form of targeting.

Turns in the



The experiences of Oregon and New York demonstrate that answers to some important questions that arise from setting priorities are already available. However, some means of setting priorities will raise equally important questions to which there are as yet no answers. For this reason any responsible attempt to set sensible priorities will crucially involve an ongoing and national debate. If in turn this debate is to be thoughtful and responsible, it will have to be thoughtfully framed. Conceptual resources for such a frame can be acquired by looking to certain turns in the philosophical debate. This report now examines four of these turns.

The Problem of


Most attempts to develop general theories about how to set priorities, both in formulating criteria and in applying them, are plagued with substantive vagueness and indeterminacy.[49] Criteria that are used in public policy debates on how best to rank health services cannot be applied to yield anything like a determinate set of spending priorities or the appropriate selection of services or populations; this report itself bears witness to the nebulousness of such attempts. In practice, claimants competing for scarce medical resources often appeal to the same general norm - for example, "fair equality of opportunity" - to settle the dispute.[50] Opposing sides can use this general norm to lead to wildly different moral conclusions about the proper distribution of mental health services. How can one principle produce a conclusion and its opposite? It may be true that the principles frequently used are too schematic to draw helpful normative priority schemes, but it remains unclear whether the problem of indeterminacy is one associated with the principles, the resulting priority scheme, or both.

Claims that the entire venture of setting priorities is vague and indeterminate can be understood in at least two ways. First, the principles might be thought of as indeterminate simply because there are an infinite number of them. For example, there is little doubt that each person who invokes a theory of priority setting can have a different set of assumptions - including what is to be ranked, what medical criteria should be used to establish priorities, and what ethical theories should be used. Each perspective will yield numerous practical conclusions, and the aggregate of perspectives will generate what might well be an infinite number of conclusions, some contradictory. The deep moral disagreement that is experienced in priority setting is simply that there are too many options and ways of viewing things. However, to call this variety of options or perspectives indeterminate is an odd use of the term. Usually principles are not so much indeterminate, as insufficiently determinate to specify a unique answer. So, for example, a principle of nonabandonment of the sick by itself cannot fully determine a unique solution to the problem of priority setting. What it can do, however, is rule out certain kinds of actions, such as a draconian use of triage.

A claim that priority schemes are indeterminate might, on an alternative interpretation, be taken to mean that priority schemes never yield anything but indeterminate conclusions. In moral reasoning, principles will typically need to be joined with relevant empirical data and clearly defined concepts in order to achieve some determinacy. To know whether an act of killing is justified, for instance, one would have to examine the principle, "Do not kill the innocent," distinguish the concepts of killing from letting die and the innocent from those threatening harm, and then judge how the factual situation fits the conceptually clarified principle. However, conceptual clarification alone does not always create determinacy. In mental health an adherence to the principle, "Everyone is entitled to some care," is at odds with other principles such as providing appropriate care or obligations to exercise stewardship by containing costs. Not all principles can be accommodated at once. In a case of this kind, the attempt to establish priorities might rest on several principles insufficiently determinate to yield a unique answer, but each might singly rule out some actions and all in combination might further determine a just course of action. The point is that the claim that priority setting schemes are indeterminate is not in principle an insurmountable objection. It is more accurate to claim that the principles of priority setting are insufficiently determinate and consequently will yield approaches to priority setting that will be underdetermined, but certainly not indeterminate.

Health Measurement


A central problem in the priority debate is over the diverse empirical health measurement criteria that can be used to set priorities. What health data shall be employed? Are the criteria that were used by Oregon, for example - severity of illness, net benefit, and the like - rationally compelling and defensible? If these are the wrong criteria, then which ones are right?

The implicit norm or standard from which most policy and priority debate begins is some form of cost-benefit measurement.[51] Decision-makers use this measurement against the existing set of resources and select that combination of resources that maximizes overall benefit. Many different, highly sensitive measures have been developed that take this form,[52] among them the quality adjusted life year (QALY) that measures health services in terms of quality and length of life,[53] or the quality of well-being scale (QWB) that evaluates treatments in terms of the degree and duration of health improvement.[54] Prototypical cost-benefit measurements pose some common philosophical problems. For example, why are certain features such as cost and benefit, rather than other features such as severity of illness or prevalence of disease, chosen to be measured or given pride of place in any of the commonly employed scales? Even if there is agreement on what should be measured, what do concepts such as "benefit" mean? Will palliation as well as cure be considered a benefit? For whom - the individual or society? If maximizing the benefit of the society becomes the measure, the policy can be criticized on the grounds that it disregards the distribution of benefits to the individual. Similarly, if maximizing the benefit to the individual becomes the measure, the policy can be criticized on the grounds that an ethically defensible standard must attend to something more than, or distinct from, individual well-being.[55] While all health measurements have problems in common, when taken by themselves, the problems are unique. Any policy maker responsible for priorities would have minimally to examine aspects of the most frequently used measurements: severity and prevalence of illness and dysfunction, and effectiveness and cost of treatment.

How much priority should be

given to effective outcomes?

A common means society employs to decide its priorities and distribute its resources among a number of ill persons, all of whom are equally needy, is to measure and rank according to the effectiveness of the treatment. Is it more important to provide services to a severely and persistently mentally ill person for whom treatment can do little, or should services be given instead to people with problems in living or anxiety disorders for whom treatment is highly effective?[56] An argument from effectiveness would lead one to rank according to treatments that offer more longevity and better quality of life.[57] However, the term |effectiveness' admits of many meanings. Effective for what? For improving function? For restoring it? At times improving function might be more important than restoring it, depending on how greatly function was compromised in the first place. Furthermore, in mental health, effectiveness can be a measure of how well a treatment reduces the risk of physical danger to the patient or to others who might be harmed by the patient's actions. Yet there is often so little agreement about what exactly ought to be made more effective that discussion can seem pointless.

Not all considerations of effectiveness are useless in determining priorities, however. When considering two treatments for one disorder and all else is equal, including cost and invasiveness of treatment, the choice between a marginally effective and a highly effective service is an easy one. Simple rationality and concern for a just use of public funds would privilege the measure of effectiveness for cases of this kind.

More troubling, however, would be to determine the priority between two persons with mental illness with the same disorder, same degree of severity of illness, and same cost of treatment, but where the outcome for one would be more favorable than for the other. Suppose, for example, that two schizophrenic patients, S1 and S2, had the same severity of disease and cost of treatment but that S1 would improve enormously while S2 would show only moderate improvement. Should we favor S1, who stands the best chance of responding to treatment? If preference is given to patients who are likely to sustain highly effective outcomes, rather than treatments that produce such outcomes, then a patient such as S2 who has been refused treatment could complain that she was being asked to forgo any chance at all of benefit. Few would deny that if treatment for S2 provided no benefit to speak of, then S2 should be given a low priority; however, it is less clear how to prioritize treatments when a person receives only a somewhat smaller benefit than someone else with the same disorder. To add precision in these ambiguous choices, some have proposed deciding outcomes by assigning the ill a number of chances proportional to the benefits of treatment - for example, by throwing a ten-sided die whose sides are weighted relative to the benefit of treatment[58] or some other multiplicative weight.[59] Where benefits are nearly equal in outcome, the best chance of benefit is no longer a useful tie-breaker.

We have some reason to think that relying solely on effectiveness can be counterproductive to priority setting. Just as yielding to a public policy of the "rule of rescue" - dramatically responding to meet the needs of an identifiable individual to save life at whatever cost - leads to an impossibly expensive system, so privileging effectiveness when it is accompanied by a very high price tag would distort the priority process. Other treatments of moderate to high benefit would be squeezed out.

How much priority should be

given to the costs associated

with mental illness?

In the more widely known health measurement scales used to set priorities there is typically an assumption, whether explicit or implicit, that costs are relevant, especially those connected with treatment. As with all empirical measurements, a cost criterion has its share of ambiguity, some intuitively agreed-upon uses, and some terrible consequences.

The conceptual problems with cost are straightforward. Which will count - treatment costs or overall costs, economic or noneconomic costs? Most people agree that overall costs, and not merely treatment costs, are essential to setting priorities; however, proponents disagree among themselves over which costs are part of overall costs, and to what degree treatment costs and overall costs should be compared. For mental illness, overall costs will include not only the present and long-term care of an individual, but also treatments needed for maintenance, the economic and social costs of nontreatment, and the cost of diverting resources from other socially desirable endeavors, including education and other social services.

There is substantial agreement about some uses of costs, especially when costs are used as a brake on spending. Simple rationality would require health planners to estimate how much the health care system could afford. If a priority-setting plan is to have any ethical grounding, there must be a good fit between its measurements and its stated goals. It follows that priority setting that is motivated by the goal of containing costs must contain criteria for measuring costs, and indeed, accepting a priority plan that lacks a measurement for its stated goals "would be like buying a car without a steering wheel."[60]

An example of counting cost that can be determinative in and of itself is the case of Clozapine.[61] Clozapine, an antipsychotic drug originally costing $9,000 a year, has been demonstrated to be beneficial to a segment of schizophrenic patients who previously were not satisfactorily helped by drugs (39 percent of once-seriously disabled patients reported they were able to regain employment, and 68 percent were able to live in the community).[62] One state, New York, with upward of 16,000 schizophrenic patients, has distributed the drug to a relatively small number of patients because legislative appropriations are fixed and will not allow additional expenditures for making this drug widely available. Legislators put a priority on providing alternative treatments for schizophrenics after weighing the physical risks of treatment, the monetary cost of providing Clozapine, the cost of squeezing out other effective treatments, and costs associated with nontreatment, including a loss of productivity, increase in crime, and heightened drug abuse.

There are of course some troubling consequences to the use of cost to set priorities, especially when it involves canceling expensive interventions altogether. Sometimes arguments in favor of preventive measures in mental health are justified on the basis of money saved through gains in future wages or through avoiding costly inpatient treatment. Yet even if the preventive interventions never helped avoid future costly treatment, there would be a reluctance to abandon prevention strategies merely because they had been shown not to be cost-effective. The problem finds an unusual twist in mental health. Where there is a "moral hazard" - whereby a treatment is made available but is then overused, thus driving up the cost and leaving the system worse off than it was initially - disincentives, such as high copayments or deductibles, are erected for the purpose of attenuating the demand for the treatment.

Factoring in overall costs can also have its dark side. If, for example, the costs used to set priorities included wages lost because of the treatment, then persons who have low-wage jobs (as some mentally ill persons do) might be given a low priority for treatment. This logic might be applied to any group - the elderly, or children, for example - that tends to absorb social resources while contributing relatively little in the way of work productivity.[63]

How much priority should be given to the worst off, if any?

There is a strong intuition that society should give the highest priority to treating the sickest or most disabled patients. This "hierarchy of pain" criterion is seen clearly in the strong human proclivity to respond to the worst off victims of illness or accident. Oddly, the intuition favoring the worst off is sometimes ignored by cost-effectiveness methodologies that are biased toward services maximizing effective treatment outcomes while minimizing costs. Yet in certain cases the intuition seems appropriate. Take for example the case of a minor mental dysfunction that causes little harm, is limited to the individual, and has no real impact on the smooth functioning of the society; our intuition that this sort of case ought not to be a high public priority is probably correct.

Intuitions fail us at other times, however. When, for example, the choice is between treating persons with severe and persistent mental illness and treating a severe anxiety disorder, such as panic disorder, peoples' intuitions will differ. The National Alliance for the Mentally Ill (NAMI) argues that individuals with severe and persistent mental illness are substantially worse off as a result of their disease and so ought to have a higher priority for treatment, while spokespeople at the Anxiety Disorders Association (ADA) argue that people who suffer from panic disorder, although much less seriously ill, ought to receive equal or even higher priority. Rather than directly confronting the severity issue, the ADA'S argument involves other reasons to make the less severely impaired a priority, such as greater efficacy of treatment, lower cost, or the injustice of discriminating against those whose illness is episodic rather than chronic. Services for the less severely ill might well be given higher priority if they are less costly, serve more people, or meet some socially agreed-upon goal.

On the other hand, the sickest often claim priority on the basis that justice requires giving priority to the worst off. The best known example of this argument is John Rawls's "Difference Principle," which requires that "the social and economic inequalities attached to offices and positions are to be adjusted so that, whatever the level of those inequalities, whether great or small, they are to the greatest benefit of the least advantaged members of society."[61]

The Difference Principle has been much criticized. One criticism is that if one gives absolute priority to the least advantaged, only the worst off will get services. A second criticism is that Rawls's theory is based on economic and social abstractions that do not easily apply in any straightforward way to priorities in mental health. What should be included in the definition of "worst office" - the relative harmfulness of a medical condition if left untreated? Economic disadvantage? Danger to others? Amartya Sen, like Rawls, has argued that if a benefit is to be distributed to two individuals, the initially worse off individual should receive an equal or greater share of the benefit so as not to widen the initial inequity.[65] This principle requires giving some, but not absolute, priority to the worst off. On either formulation it seems likely that other principles, such as efficaciousness and cost, will be used to temper this one.

How much priority should be given to providing benefit to the many?

Some hold that "numbers count" in setting priorities; the prevalence of a disorder is thought to be determinative for ranking. The principle of "the greatest good for the greatest number" would set common conditions at a high priority, while rare disorders would be given low priority. Setting priorities that prefer higher prevalence intuitively seems sensible, as treating a common disorder meets publicly defined goals such as sustaining the functioning of society.[66] In mental health it is certainly true that treating a common disorder such as depression substantially reduces dysfunction by reducing medical costs. Yet here it is not the prevalence of the disorder that makes it a high priority treatment but a host of other considerations, including effectiveness of outcomes and costs.

There is a strong reason not to make prevalence determinative when setting priorities, however. If, for example, a condition is rare, but its cost and benefit per case are equal to a common condition, refusing to treat the rare condition displays an "irrational bias against people and problems that come in small groups."[67] Considerations of prevalence are an inappropriate basis for priorities when two disorders are of equal severity and can be treated equally effectively.[68] Prevalence should be a relevant public consideration only when, for example, a decision must be made whether to treat a rare, harmful disorder at such great cost that it stands to squeeze out treatment of a common and less harmful disorder. Society would have to judge whether in this case it was important to ensure that everyone received some health care.

An examination of all the health measurements put forward at one point or another to establish priorities suggests several items that public policymakers should keep clearly in mind. First, as is obvious from the discussion above, the public policy reasons for selecting some health measurements over others are cloaked in a certain amount of complexity and mystery. The examination of the four measurements to which this discussion has been limited provides ample evidence for refusing to give any of them a privileged position. Moreover it is difficult to establish clear meanings for any of them. One way to determine which measurements should be included and how broadly they should be understood is to leave it to the community to decide what it values. A public process could be used to determine, for example, whether the community prefers fewer costs or fewer burdens. Second, health measurements alone do not seem to be determinative of what priorities to set. No matter how effective or cheap a treatment, or how grave or prevalent a disease or dysfunction, there are other considerations that need to be incorporated or adjusted for. For instance, quite aside from the health data about a treatment, the population to whom the treatment is offered might have suffered a history of oppression which society now wishes to rectify. In short, the health data must be balanced against the political, social, and cultural forces that have and have had an effect on the equitable distribution of services.

Who Should Decide?

The question of what empirical information should count in the determination of priorities is merely one of the many issues that remains unresolved. Another question of great importance is who should make these decisions. How should we determine the locus of decisionmaking? It is appealing to rely on democratic principles as a means of achieving a resolution that will guide policy. But the question then becomes, Is a democratic process the best or fairest means of determining what constitutes an acceptable priority-setting scheme, or are there alternatives, such as reliance on experts or public officials? While mental health care at some times and in some places has been allocated in ways highly responsive to community input, this has not been universally the case, nor have all mental health policy professionals been pleased with the results.[69] In fact, not all practitioners or theorists regard public openness about priority-setting or certain complex public policy matters as a good thing. One argument, that of Guido Calabresi and Philip Bobbitt, stipulates that some choices are so tragic that the good order of society requires they be made in the least damaging way. Choices are tragic when they run afoul of society's fundamental commitment to certain absolute ideals - for example, that life is priceless. If society is unable to live up to the ideal that life is priceless, but it has good reason to protect this ideal, then those who must make the hard choices should do so behind closed doors.[70] Another argument from those who propose measures to make priority setting less public is based on a claim that patterns of invidious bias against persons with mental illness are so rampant throughout society that the decisions must be made away from the heat of these biases, by knowledgeable and sympathetic specialists or officials.

It is important to realize that at a practical level, public officials do not consciously hide decisions, yet in their good intentions to better the plight of the ill, the method used for dealing with tragic choices might be to keep the public in ignorance. In health services, for example, eligibility criteria are used to allow some identified persons to receive a service; however, if eligibility criteria are eliminated and quietly replaced with new patterns in the organization of care, then a tragic choice might have been made, even though the policymaker does not admit that it has been made.

However, both the concealment advocated by Calabresi and Bobbitt and the lesser concealment created by measures to protect persons with mental illness from bias pose tradeoffs that a democratic society can ill afford. The strongest rebuttal to the tragic choices argument is that our society is to some extent held together by openness and honesty. Tragic choices that invisibly set priorities cannot but serve to dissolve the glue that secures society.[71] If we are to create a well-ordered society, it must be animated by a notion of "fair terms of cooperation and mutual advantage."[72] Fair cooperation is only possible if all the parties to the cooperative agreement know where they stand, and that is possible only if all the relevant considerations are available to the participants.[73]

As for priority-setting measures designed to avoid the sting of discrimination, it is not clear that we gain much when we leave these decisions to a less than open forum. A priority decision is anything but a mere algorithm, where an the data are clear and complete and need only be entered into an agreed-upon equation by skilled minds. Rather, priority decisions rest firmly on people's values - about disease and disability, about treatment and palliation, about costs - and these are the issues about which individuals will reasonably disagree. Highly skilled technicians and public officials have no more insight about them than anyone else does. Countering movements to keep the decision from hoi polloi, John Rawls stipulates a "publicity condition": when it comes to determining what should guide the allocation of scarce resources, everything must be open to the view of those affected by the decision.[74]

One might too hastily, and inappropriately, conclude that the publicity condition requires direct citizen participation in every decision. Leaving all decisions about priority setting to democratic vote seems impracticable and would mire any progress. The proper lesson to be drawn from the publicity condition is that everything should in principle be open to view, and that the basic framework of the priority-setting structure should be sensitive to community values. However, the exact degree of community involvement beyond these two constraints should be a local decision. If, for reasons of practicality, a community decides to empower a public official or board to make the final decisions, these decisionmakers will be held to the same constraints as the community.

The move to local decisionmaking is not without some important practical moral consequences. Dispersing mental health decisions to allow community boards to participate in policymaking is a case in point. During the era when community mental health centers developed at the local level, for example, citizens - self-appointed in many cases - pushed for expanded ambulatory services for poor, working-class families with a variety of distressing but not disabling problems, rather than supporting intensive services for individuals with severe and persistent mental illness. The priorities of the citizens reflected prevailing needs within the community, but the needs of the minority - persons with severe and persistent mental illness - were neglected. If these priorities represent the authentic preference of the majority, must we accept them at face value, regardless of their implications? Whose preferences should we rely on - those of the public as a whole or only the people who have experienced a condition? Regardless of the persons from whom opinions are sought, fairness would dictate that the views be constrained by moral principles, such as the nonabandonment of persons, that would rule out certain alternatives.

Another unresolved puzzle in decisionmaking pertains to the structure of advocacy. Those who have experienced a condition are often unable fury to participate in decisionmaking. For this reason it is appealing to rely on structures of advocacy. Dozens of groups represent the interests of the persons with mental illness, not to mention their families, providers, and organizations of providers. The moral problems associated with the advocacy alternative are well rehearsed. At a practical level it is important to examine a number of issues. Who is it that elected this group to be advocate? Are the members of the group self-selected or representative? Do they have the interests of the entire movement in view? Where does their authority to advocate come from? At a theoretical level, it is vital to probe the tension between the preferences of consumer advocates and citizens' needs, as these are not necessarily the same. Consumers are principally interested in increasing their access; very seldom are they interested in restraining the quest for new interventions or the search for monies, yet this indifference to restraint runs at odds with society's need for cost containment.

Is There a Basic Package of Services?

In a common-sense evaluation of the fairness of a priority proposal, the litmus test is often whether a given service is so critical that it must always be provided. Yet most attempts to use ethical criteria to construct a "basic package" - services that are almost a nonnegotiable right - fail to offer any guidance for specific spending recommendations. The problem of underdeterminacy persists.

Some ethical considerations and constraints can offer a certain amount of determinacy to the problem of the basic package. First, it is important to remember that setting priorities serves at least two functions. The first is that even if society has ample funds (but not infinite funds), priority setting is valuable as a way of using the available money wisely and rationally-it is just a way of establishing what is comparatively more or less important. Second, if funds are tight, then priority setting makes it possible to allocate (or ration) those funds in the most prudent way - it facilitates orderly, reasonable rationing. There is virtually no likelihood in the present financial climate that any government can guarantee it will always make a certain level of care available, regardless of the funds. Consequently, a basic package must always be conceived as contingent on available funds. It makes no real sense to talk about a basic package in the abstract. Designing a basic package in light of available funds is a much easier task with a priority list: the more money one has the lower one can go in the list; the less money one has the higher one must go. It is not inconceivable that a state could make (temporary) guarantees of a basic package, saying that unless there is some unforeseen crisis it will take care of items 1 through n on the list.

A rough ethical constraint of nonabandonment adds determinacy to the basic package in that it establishes who should have access to it. The question whether we should prefer the worst off, considered earlier but left unresolved, can on this principle be answered: the worst off should receive a high but not absolute preference. There is a widespread but lamentable assumption that favoring the worst off implies that those who are better off should get nothing. The implication does not follow. From the time of Hippocrates onward there has been a strong moral requirement forbidding the abandonment of patients. Such abandonment is at odds with respect for persons, a fundamental moral principle. Nonabandonment further shows solidarity with everyone in the community, no matter who they are. If some portion of the population of the mentally disabled received nothing after society has examined empirical data, balanced that data with social and moral considerations, and proposed priorities, then there would be good cause for a discretionary rearrangement of priorities to remedy this failing. That is, a criterion of nonabandonment would require a basic package to ensure that everyone who is mentally disabled receive something. The constraint would also require an ongoing analysis of the provision of services to ensure that no person with mental illness falls through the cracks and receives no treatment.

A third consideration for the basic package emerges from a combination of health measurements, including effectiveness and cost of treatment, and prevalence and severity of illness. It can be argued that any decent society owes its members the kind of services that have had the most effect on improving the health of the population as a whole, which historically has included nutrition, sanitation, general living conditions, vaccinations, and antibiotics. There is also a strong human sentiment favoring care for those whose lives are strikingly and decisively harmed by a particular disease, and ranking lower the care of those whose lives are crippled but not devastated. We know the latter can get by, even if not well, while the former are vulnerable and will not get by. Even though the worst off should not get absolute priority, they do deserve some priority. For those most afflicted by mental illness, social services that have been shown to be effective,[75] such as financial assistance, congregate housing, and sheltered workshops to augment job skills - services intended to minimize the negative consequences of a disability (such as reinstitutionalization) - should be provided if at all humanly possible. This social support care for people with mental disorders comprises services appropriate for anyone suffering from a disability, with or without mental illness.

Thinking about Mental Health Services

Has this investigation of the priorities that ought to be established for mental health generated more heat than light? The number of issues that must be examined have seemingly expanded exponentially, suggesting that questions not ordinarily addressed in priority setting have been coaxed to the surface. Even where the project revisited longstanding dilemmas, some issues have become clearer, if not resolved altogether. The single most important contribution our investigation has made is in its application to mental health services, where, at a minimum, it contributes a language and structure for thinking about certain of its festering, perhaps intractable problems.

Theoretical indeterminacy is the single most dominating and ongoing challenge to this and any other attempt to set priorities. The problem is symptomatic of a relatively deep social conflict over what moral concepts we ought to cherish most dearly - a conflict that will vex not only mental health policy but any other reflection on priorities. From the outset it has been clear that the shape, force, and number of criteria for setting priorities are potentially infinite when viewed through the kaleidoscope of readily available ethical theories. Even if one pattern of ethical reasoning is used consistently, limitless problems in achieving conceptual clarity for each criterion can result in both insufficiently determinate principles and underdetermined normative schemes to set priorities.

What is less obvious and perhaps more fundamental is that the moral disagreement that surfaces in discussions over priorities mirrors a profound disparity among world views. If liberalism offers little more than procedural guides for setting priorities, the many challenges to liberalism that offer substantive content hold out little hope for consensus.[76] Yet as our investigation amply attests, the present standoff must not create paralysis. The amount of determinacy that can be attained creates serious philosophical puzzles, but they must be balanced with the fact that de facto, informal priority setting will continue. It must be acknowledged that the principles of priority setting are insufficiently determinate to offer perfect guidance, but they ride out certain courses of action. At the same time, we must recognize that priority-setting schemes are underdetermined and that further efforts will be required, including moral argumentation about the facts, concepts, and principles that apply.

From its inception, the growing conversation about setting priorities has, although not always, squarely addressed whether it should take place at all; it must confront and finally put to rest the claim that this is unthinkable or impossible. Priority setting is unlikely to face sustained allegations of perniciousness if the process is understood as an attempt to promote a more reasoned and participatory public discussion culminating in policy about issues that intimately affect our lives individually and as a nation and about which people reasonably disagree. Likewise, when priority setting is an attempt to excise utterly useless health services or eliminate marginally beneficial ones, the outcry will not be very loud. Most people see the merits of proposals that identify and eliminate health care services that have been demonstrated to be useless, safely forgone, or only marginally beneficial. There will then be little dispute regarding the necessity of an ongoing mechanism, such as a health outcomes bureau, to accomplish this end.

However, when priority setting is used to help sort services and delete beneficial but less needed services, accusations of impropriety will be more pronounced. And to the extent that the stated goals - cost containment, greater access, and more appropriate treatment - are not met by the plan, the critics will be right to complain; criteria must match well the stated purposes for establishing priorities and must further their attainment. Perplexing but not insurmountable is the charge that setting priorities within a particular part of the system is unseemly. There can be little disagreement that, while it may be important to highlight the long-term, chronic features that are representative of a portion of mental health, it would be terribly misguided to rank mental health services apart from other services, either by failing to integrate mental health, or worse, by leaving physical health untouched and unranked. Mental and physical health should be fully integrated in any priority-setting plan. It is important to emphasize once again that the evidence from Oregon indicates that when mental health services are considered together with physical health care, mental health needs rank satisfyingly high.

The claim that priorities, once set, are impossible to implement should also be resisted. Difficult, yes - impossible, no! The history of de facto priorities reveals the convergence of a number of social elements that policymakers and the public must appraise. Any strategy must be based on a keen awareness that social forces of varying degrees shape priorities in unexpected and even counterproductive ways. Any plan that relied on empirical data alone would be insufficient. Priority setting should take place in a way that best ensures a tension among political, social, and moral considerations on the one hand, and scientific evidence on the other. This means that the method for setting priorities must accommodate some degree of discretion on the part of those ultimately responsible for the process, who can adjust for biases and predilections in the empirical data and balance them against a realistic appraisal of what is politically feasible.

Determining what empirical data should count, and in what balance, can seem to be an almost hopeless task. A priority system's first task should be to identify, conceptually clarify, and fully evaluate common empirical health measurements, including the seriousness and prevalence of a condition and the effectiveness and cost of treating it. Care must be taken to include all relevant factors. One that is often overlooked, yet crucial to a segment of the experience of mental illness, is its chronic or sometimes episodic quality, which requires attention over time and perhaps long-term care. Beyond that, if a priority system privileges one empirical health measurement over others, steps should be taken to ensure that the prominence of the measure is warranted by principles of fairness and rationality or authorized by public choice. Additionally, a priority plan must consider whether constraints external to the health measurement invest the measurement with further determinacy. When considering a condition's severity, for example, where there is philosophical agreement and strong public sentiment that those who are worst off should receive some but not absolute priority, the external constraint of nonabandonment should be brought to bear so that no one with mental illness is forced to go without any care at all.

Setting formal priorities requires a continuous mechanism for identification and evaluation of health measurement and outcome data. While this might appear to be no more than old-fashioned common sense, in some early congressional proposals for health care, arrangements for some such mechanism were conspicuously absent. Because of the nature of empirical data, an appeals process must be developed to consider challenges to decisions to exclude certain services that have been pronounced unnecessary or uncertain in outcome.

The problems associated with democratizing the priority process also have been advanced. From the outset, there was little disagreement that priorities should be set by a mechanism that is accountable, enhances objectivity, and allows for a public, participatory process. However, what remained to be considered is whether the entire decisionmaking process is one that requires direct citizen participation in every decision or whether decisions can instead be made by public officials. As messy as it seems, there is strong moral reason for decisionmakers to place decisions in clear view of those who will be affected by them, and the decisionmaker must be sure that the basic structure of decisionmaking is sensitive to community values. The moral requirement does not, however, require direct citizen participation in every decision.

Our consideration of the basic package serves as an example of how insufficiently determinate principles can in combination, and given sufficient conceptual clarity, provide direction for setting priorities. Intuitively it seems correct that any priority system should ideally be prepared to find and note some minimally adequate level of care below which no system should fall. Of course, this ideal cannot always be achieved because it is constrained by resource availability, but it remains an approachable ideal in that any community ought to be willing to mark what it thinks it can give and retreat only in the face of unforeseen catastrophe. Also, the insufficiently determined principle of preferring the worst off provides guidance: priority systems should give a higher place to caring for those who cannot readily be cured (such as chronic schizophrenics) and a lower place to the traditional effort of curing. Another insufficiently determined principle, that of nonabandonment, insists that though a preference for the worst off may be legitimate, those suffering with milder mental illness cannot simply be neglected.

It is unlikely in the short run that all constituents in the field of mental health will warmly embrace attempts to set priorities. No doubt the unresolved questions featured in this investigation will serve as a defense for shunning the process. Yet, regardless of the objections, it is likely that priorities Will continue to be set, formally or informally. The choice will be ours. If we indeed explicitly pursue priorities, further efforts will be needed to deal with the serious puzzles raised by the process, but the benefit will be an improvement over the present state of affairs.


[1.] Karen Bourdon, "Estimating the Prevalence of Mental Disorders in US Adults from the ECA Survey," Public Health Reports 107, no. 6 (November-December 1992): 663-68. [2.] Gerald Grob, From Asylum to Community: Mental Health Policy in Modern America (Princeton, NJ.: Princeton University Press, 1991). [3.] Miriam Cotler and Brian Gould, "Priorities in Mental Illness: Priorities in the Private Sector," prepared for the Hastings Center Project on Priorities in Mental Health, 1993. [4.] Bureau of Labor Statistics, Employer Benefits in Medium and Large Firms, 1988 (Washington, D.C.: Department of Labor, 1988). [5.] Cotler and Gould, "Priorities in Mental Illness." [6.] Steven Scharfstein, Anne Stoline, and Howard Goldman, "Psychiatric Care and Health Insurance Reform," American Journal of Psychiatry 150, no. 1 (1993): 7-18; Richard Frank, Health D. Salkever, and Steven Scharfstein, "A New Look at Rising Mental Health Costs," Health Affairs 10, no. 2 (Summer 1991): 116-23. [7.] Jerome Wakefield, "The Concept of Mental Disorder: On the Boundary between Biological Fact and Social Value," American Psychologist 47, no. 3 (1991): 373-88. [8.] Kenneth Kendler et al., "The Prediction of Major Depression in Women: Towards an Integrated Etiological Model," American Journal of Psychiatry 150, no. 8 (1993): 1139-48. [9.] Kenneth Wells and Robert Brooks, "The Quality of Mental Health Services: Past, Present, and Future," in The Future of Mental Health Services Research, ed. C. Taube, D. Mechanic, and A. Homann (Washington, D.C.: NIMH, 1991). [10.] Roll Call advertisement, sponsored by a number of mental health activists. [11.] "Mental Health Services in Health Care Reform: Specifications for Implementing the Proposal: Principles for Mental Health Care in Health Care Reform," presented in Washington, D.C., February 1993. [12.] Grayson Norquist and Kenneth Well, "Mental Health Needs of the Uninsured," Archives of General Psychiatry 48 (1991): 475-78. [13.] David Hadorn and Robert Brook, "The Health Resource Allocation Debate: Defining Our Terms," JAMA 266, no. 23 (18 December 1991): 3328-31. [14.] Robert Michels, "Priority Setting in Mental Health," prepared for the Hastings Center Project on Priorities in Mental Health, 1993. [15.] David Kears, "Setting Priorities for Mental Health Services: The Case of Alameda County, California," prepared for the Hastings Center Project on Priorities in Mental Health, 1993. [16.] Rosemary Stevens, In Sickness and in Wealth: American Hospitals in the Twentieth Century (New York: Basic Books, 1989). [17.] C. Keisler, "U.S. Mental Health Policy: Doomed to Fail," American Psychologist 47, no. 9 (1992): 1077-82. [18.] L. K. George, "Definition, Classification, and Measurement of Mental Health Services," in The Future of Mental Health Services Research. [19.] Len Rubenstein, "Law and Priority Setting," prepared for the Hastings Center Project on Priorities in Mental Health, 1993. [20.] Daniel Callahan, "Setting Priorities: Problems and Possibilities," prepared for the Hastings Center Project on Priorities in Mental Health, 1993. [21.] David Hadorn, "Setting Health Care Priorities in Oregon: Cost-Effectiveness Meets the Rule of Rescue," JAMA 265, no. 17 (1 May 1991): 2218-25. [22.] Joseph Morrissey and Howard Goldman, "Cycles of Reform in the Care of the Chronically Mentally Ill," Hospital and Community Psychiatry 35 (1984): 785-93. [23.] Gerald Grob, Mental Institutions in America: Social Policy to 1975 (New York: Free Press, 1973), pp. 84-131. [24.] Gerald Grob, "Government and Mental Health Policy: A Structural Analysis," prepared for the Hastings Center Project on Priorities in Mental Health, 1993. [25.] Grob, Mental Institutions. [26.] Grob, Mental Institutions, Grob, "Government and Mental Health Policy." [27.] Gerald Grob, Mental Illness and American Society: 1875-1940 (Princeton, N.J.: Princeton University Press, 1983), pp. 144-78. [28.] Grob, From Asylum to Community, pp. 5-23. [29.] Grob, From Asylum to Community, pp. 49-50. [30.] Grob, "Government and Mental Health Policy." [31.] Grob, From Asylum to Community. [32.] David Mechanic, "Recent Developments in Mental Health: Perspectives and Services," Annual Review of Public Health 12 (1991):1-15. [33.] Gerald Grob, The Mad among Us: A History of the Care of America's Mentally Ill (New York: Free Press, January 1994), chapter 11. [34.] David Mechanic and Richard Surles, "Challenges in State Mental Health Policy and Administration," Health Affairs 11, no. 3 (Fall 1992): 34-50; Howard Goldman and Richard Frank, "What Level of Government? Balancing Interests of the State and Local Community," prepared for the Hastings Center Project on Priorities in Mental Health, 1993. [35.] Martin Strosberg et al., Rationing American's Medical Care: The Oregon Plan and Beyond (Washington, D.C.: The Brookings Institute, 1992). [36.] Oregon Health Services Commission, Prioritization of Health Services: Report to the Governor and the Legislature (Salem: Oregon Health Services Commission 1991). [37.] David Pollack, Bentson McFarland, Robert George, and Richard Angell, "Prioritization of Mental Health Services in Oregon," prepared for the Hastings Center Project on Priorities in Mental Health, 1993. [38.] Bruce Jennings, "Health Policy in a New Key. Setting Democratic Priorities," Journal of Social Issues 49, no. 2 (1993): 169-84. [39.] Personal communication with Richard Surles, New York State Commissioner for Mental Health. [40.] Richard Surles and Cynthia Feiden-Warsh, "Mental Health Coverage in Health Care Reform," prepared for the Hastings Center Project on Priorities in Mental Health, 1993; Office of Mental Health, Challenges for the 1990s: State Comprehensive Plan for Mental Health Services, 1990-1995 (Albany, N.Y: Office of Mental Health, 1990). [41.] Chris Koyanagi and Howard Goldman, Inching Forward (Washington, D.C. National Mental Health Association, 1991). [42.] Surles and Feiden-Warsh, "Mental Health Coverage in Health Care Reform"; Mechanic and Surles, "Challenges in State Mental Health Policy and Administration." [43.] D. Shern, Richard Surles, and J. Waizer, "Designing Community Treatment Systems of the Most Seriously Ill: A State Administrative Perspective," Journal of Social Science Issues 53 (1989): 105-17. [44.] Surles and Feiden-Warsh, "Mental Health Coverage in Health Care Reform." [45.] L. Meyers et al., New York City's Hospital Occupancy Crisis: Caring for a Changing Population (New York: United Hospital Fund, 1988). [46.] Surles and Feiden-Warsh, "Mental Health Coverage in Health Care Reform." [47.] Office of Mental Health, 1993 Interim Report: Statewide Comprehensive Plan for Mental Health Services, 1993-1997 (Albany, N.Y.: Office of Mental Health, (1993), p. 4. [48.] Office of Mental Health, Challenges for the 1990s. [49.] Norman Daniels, "Rationing Fairly: Programmatic Considerations," Bioethics 7, nos. 2, 3 (1993): 224-33. [50.] Norman Daniels, Just Health Care (Cambridge: Cambridge University Press, 1985). [51.] Dan Brock, "Quality of Life Measures in Health Care and Medical Ethics," reprinted in D. Brock, Life and Death: Philosophical Essays on Biomedical Ethics (Cambridge: Cambridge University Press, 1993). [52.] Ian McDowell and Claire Newell, Measuring Health: A Guide to Rating Scales and Questionnaires (New York: Oxford University Press, 1987). [53.] G. Loomes and G. McKenzie, "The Use of QALYs in Health Care Decision Making," Social Science and Medicine 28 (1989): 299-308. [54.] Robert Kaplan and John Anderson, "A General Health Policy Model: Update and Applications," HRS: Health Services Research 23 (1988): 203-35. [55.] Dan Brock, "Some Unresolved Ethical Issues in Priority Setting of Mental Health Services," prepared for the Hastings Center Project on Priorities in Mental Health, 1993. [56.] Daniels, "Rationing Fairly." [57.] Hadorn and Brook, "The Health Resource Allocation Debate." [58.] Dan Brock, "Ethical Issues in Recipient Selection for Organ Transplantation," in Organ Substitution Technology: Ethical, Legal, and Public Policy Issues, ed. D. Mathieu (Boulder: Westview, 1988). [59.] Frances Kamm, "The Report of the U.S. Task Force on Organ Transplantation: Criticism and Alternatives," Mount Sinai Journal of Medicine 56, no. 3 (1989): 207-20; see also her Morality and Mortality, vol. 1 (New York: Oxford University Press, 1993), in press. [60.] David Eddy, "Oregon's Methods: Did Cost-Effectiveness Fail?" JAMA 266, no. 15 (16 October 1991): 2135-41. [61.] William Reid, "Access to Care: Clozapine in the Public Sector," Hospital and Community Psychiatry 41 (1990): 870-74. [62.] Kenneth Terkelsen and Rene Grosser, "Estimating Clozapine's Cost to the Nation," Hospital and Community Psychiatry 41 (1990): 863-69; Eugene Arnold, "Quantifying the Value of Human Life for Cost Accounting of Safeguards: Clarifying Formulas Applied to the Clozapine Controversy," The Journal of Clinical Ethics 3, no. 2 (1992): 102-13. [63.] Brock, "Some Unresolved Ethical Issues." [64.] John Rawls, Political Liberalism (New York: Columbia University Press, 1993), p. 6. [65.] Amartya Sen, On Economic Inequality (New York: Oxford University Press, 1973). [66.] Daniel Callahan, What Kind Of Life? (New York: Simon and Schuster, 1989). [67.] Paul Menzel, "Oregon's Denial: Disability and Quality of Life," Hastings Center Report 22, no. 6 (1992): 21-25, at 20. [68.] Elizabeth Anscombe, "Who Is Wronged?" Oxford Review 5, no. 1 (1967): 16-17. [69.] Goldman and Frank, "What Level of Government?" [70.] Guido Calabresi and Philip Bobbitt, Tragic Choices (New York: W. W. Norton, 1978), p. 69. [71.] James L. Nelson, "Tragedy, Prejudice, and Publicity in Setting Priorities," prepared for the Hastings Center Project on Priorities in Mental Health, 1993. [72.] John Rawls, A Theory of Justice (Cambridge, Mass.: Harvard University Press, 1971). [73.] Leonard Fleck, "Justice, HMOs, and the Invisible Rationing of Health Care Resources," Bioethics 4 (1990): 97-120. [74.] John Rawls, "Kantian Constructivism in Moral Theory," The Journal of Philosophy 77, no. 9 (1980): 515-72. [75.] George, "Definition, Classification and Measurement." [76.] Ezekiel Emanuel, The Ends of Human Life: Medical Ethics in a Liberal Polity (Cambridge, Mass.: Harvard University Press, 1991; Norman Daniels, "Liberalism and Medical Ethics," Hastings Center Report 22, no. 6 (1992): 41-43.
COPYRIGHT 1993 Hastings Center
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Boyle, Philip J.; Callahan, Daniel
Publication:The Hastings Center Report
Date:Sep 1, 1993
Previous Article:"Ethics and clinical research" revisited: a tribute of Henry K. Beecher.
Next Article:For the benefit of all.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters