By and large, neither bioethicists nor economists have offered a satisfactory account of how managed care organizations should ration health care. Both disciplines would like to guarantee adequate care to all without defining adequacy. But it cannot be done. The more we rely on market forces to

The role of organizations in health care financing and delivery has increased in recent years, in response to technological change and cost containment pressures. These organizations are more likely to be for-profit enterprises than in the past, and whatever their profit orientation, they must cope with radically altered financial incentives and constraints. These changes have attenuated the public's trust in health care organizations, and generated genuine confusion on the part of those who run them about what they should be doing. As a result, we now have an active public debate on organizational ethics in health care.

I want to look at a key topic in this debate, namely, the role of managed care organizations (MCOs) in health care rationing. I will consider two questions: How should managed care organizations determine the standard of care they guarantee? and What role should physicians play in rationing in a managed care environment? I will take up these questions from the contrasting perspectives of bioethics and economics and argue that there are fundamental contradictions between the market framework within which MCOs must operate and the way bioethicists and economists think about the ethical obligations of the managers and physicians who work in them. Either the economic framework or the conceptualization of ethical obligations must be changed. The optimal policy is to change both.

The Nature of the Rationing Problem

The definition of rationing is controversial. For my purposes here, the term means that we limit the beneficial health care an individual receives by any means--price or non-price, direct or indirect, explicit or implicit. (1) The necessity for rationing health care is also controversial. It is obvious that scarcity requires limits on goods in general, but many people would like health care to be an exception. Still, most people realize that the benefits of health care form a continuum, from saving a life to eliminating a minor inconvenience, and they acknowledge that exempting health care from any limits would cost too much--we would have to sacrifice too many of the benefits that could be obtained from alternative ways of using the resources put into health care.

Rationing must occur for reasons of both efficiency and equity. The efficiency argument is based on the economics of insurance. Health insurance is the market response to two facts--that a person's desired expenditure on health care depends on his or her state of health, and that one's state of health is uncertain. In a given year, one person's desired expenditure could range anywhere from a portion of one paycheck to a sum that exceeds annual income.

By buying health insurance, people can "pool" this financial risk; they pay a fixed sum in the form of the insurance premium into the pool and then are able to draw on the pooled funds to pay for health care. The risk-pooling works because the health care expenditures of the entire group are more predictable than those of any individual in the group. Insurance provides subscribers with valuable peace of mind; however, it introduces inefficiency into decisions about the use of care. When insurance pays the bill, people consume health care as if it were free, rather than weighing the benefit of a particular service against its cost (as they do for the other goods and services they buy and consume). The health care is not free, of course, since the premium must be set high enough to cover the total cost of the health care provided to the entire group. (2) To obtain the security of health insurance at an affordable cost, it is prudent for subscribers to agree to reasonable limits on their use of care.

The equity argument is based on societal value judgments about the meaning of equitable access to health care and the locus of responsibility for ensuring that access is equitable. This paper starts from the premise that there is a societal moral obligation to ensure, first, access for all to some basic level of health care (also called "an adequate level" or "a decent minimum") without an excessive burden in terms of money or travel, and second, a fair distribution of the cost of meeting the obligation. Since the obligation does not extend to all care that would be of benefit, limits are appropriate on the care delivered at public expense. (3)

Two different kinds of rationing are therefore at issue: limits desired by prudent participants in voluntary risk pools, and limits on the care provided to fulfill the social obligation. In either case, the limits must vary with health status, which is the major determinant of the benefit a patient receives from a service. As a result, solving the rationing problem--setting the entitlement in private or public insurance or delivery systems--is complex. A good solution is not merely a set of rules about what services are covered or which health conditions will be treated. Rather, it must be a process in which collectively determined rules and constraints and individual patient and physician decisions interact to determine a cost-conscious "standard of care" (SOC), which addresses both the amounts and quality characteristics of the goods and services provided to individuals given their particular circumstances. The process must be one that allows the standard of care provided by a system to evolve over time, in response to changes in technology; resource availability, and individual and societal values and preferences. (4)

Bioethics and Health Care Rationing

Bioethics was slow to address rationing as a practical problem for a public or private risk pool. The field's original focus was on the issues of individual physician-patient relationships--the protection of research subjects and the conflict between physician paternalism and patient autonomy, for example. Since bioethicists were theologians, philosophers, physicians, and lawyers who knew little about social science (especially economics), they were not inclined to think about "system issues." When they did consider rationing, it was usually in a "lifeboat" context, in which one tries to allocate a specific resource perceived to be in fixed supply, such as donated organs for transplant or time on dialysis machines in the period immediately after effective kidney dialysis became possible. (5) Those bioethicists who considered access to care in a broader context tended to concentrate on whether there was a right to health care rather than on the content of the potential right and on who should pay to make sure everyone got it. (6) The few who tried to translate the abstract concept of a basic level of care into a concrete standard of care found the task maddeningly difficult. (7)

Oregon's Medicaid reform plan, developed in the late 1980s, was an important milestone for bioethics and rationing. The proposal was controversial, but many bioethicists praised the state for raising the rationing issue openly, whatever they thought of the actual methodology, and they viewed the plan as a practical attempt to define a decent minimum. They were fascinated by the community meetings to set priorities, the quality of life survey, and the rankings of condition-treatment pairs, and they wrote extensively about them. (8) Oregon's process for determining the Medicaid standard of care had serious flaws; in the reforms implementation, it did not play the role that many bioethicists expected it to play. (9) Nevertheless, the discussion of the Oregon plan helped to shape bioethicists' perspective as they turned to the ethical issues associated with the increasing share of managed care organizations in the health insurance market.

There is no generally accepted precise definition of an MCO. The glossary of Essentials of Managed Health Care, a leading textbook, says this of managed health care.

   A regrettably nebulous term. At
   the very least, a system of health
   care delivery that tries to manage
   the cost of health care, the quality
   of that health care, and access to
   that care. Common denominators
   include a panel of contracted
   providers that is less than the entire
   universe of available providers,
   some type of limitations on benefits
   to subscribers who use non-contracted
   providers (unless authorized
   to do so), and some type
   of authorization system. Managed
   health care is actually a spectrum
   of systems, ranging from so-called
   managed indemnity through
   PPOs [Preferred Provider Organizations],
   POS [Point of Service]
   plans, open panel HMOs [Health
   Maintenance Organizations], and
   closed panel HMOs. For a better
   definition, the reader is urged to
   read this book and formulate his
   or her own." (10)

Such difficulties notwithstanding, however, managed care organizations generally differ from traditional indemnity insurance organizations in that they take a more active stance toward health care providers. Traditional insurers did not focus much on cost containment, and what they did to control costs was directed primarily toward subscribers; they excluded service categories from coverage and set up cost-sharing in the form of deductibles and co-payments. In contrast, MCOs put direct and indirect pressure on the providers themselves to manage cost, quality, and access. They sought, in other words, to adjust the standard of care provided to patients to take better account of resource scarcity. This cost containment approach is controversial. When is limiting the care people receive a morally acceptable response to scarcity, and when is it morally unacceptable corner-cutting?

Bioethicists have approached the task of establishing the standard of care in an MCO from an equity perspective. They see it as identical to the task of defining the morally obligatory level of care. In this view, the MCO's mission is to distribute a fixed amount of resources fairly over a fixed population. The members of the plan form a mini-society that must collectively decide what its health care obligations are to its members.

The managed care structure most compatible with this view is the closed panel HMO, in which a defined group of physicians provides comprehensive care to a defined population of enrolled members--the group or staff model HMO. Since the bioethics literature seems to tacitly assume that actual MCOs are toward this end of the organizational spectrum, let us focus for the moment on the closed panel HMO.

Viewing rationing in terms of collective decisionmaking leads bioethicists to advocate participatory processes in which managers elicit members' views and work to develop consensus on what coverage and quality should be. A distinction is drawn between "reasonable" and "unreasonable" reasons for refusing to cover a service, and it is argued that managers are ethically required to have reasonable reasons for their decisions and to make them public. An example of a reasonable reason not to cover a service is that "the service has not been proven to be effective"; an example of an unreasonable reason is that "the service is effective and cost-worthy, but the members who attended the meeting on coverage policy don't expect to want it themselves and prefer to allocate the resources to a less cost-worthy service they will want." (11)

Defining the mission of the organization as establishing a morally obligatory SOC leads to ambivalence about economic goals, and sometimes even hostility. Arguments against coverage such as "because the money not spent can be used to raise shareholder returns" do not qualify as reasonable. Bioethicists sometimes maintain that even for-profit MCOs should resist allowing profit considerations to affect decisions about coverage or quality. (12)

The bioethics literature also reflects ambivalence about contractual commitments. The task of MCO members and managers is to identify the morally acceptable standard of care. Yet since individual moral visions differ, decisionmaking at the level of the plan will yield SOCs that vary across plans. Variations are deemed morally acceptable if the processes and reasons used in setting the SOCs are acceptable. Even after the decisions have been made and embodied in contracts, however, they may be subject to reconsideration; some bioethicists argue, for example, that a contractual exclusion of experimental treatment should not automatically lead to a refusal of coverage for an unproven last chance therapy for a dying patient. (13) The justification seems to be that the facts of an individual case may reveal a defect in the translation of the underlying collective moral vision into a business practice, and the underlying moral vision is what counts.

Bioethicists recognize that MCOs must survive in the marketplace if they are to be able to deliver the SOC they have chosen. They tend to consider this an unfortunate side constraint, however, rather than the engine that should drive the decisions. Some have argued, somewhat unconvincingly, that an MCO that defines its mission as stated and pursues it conscientiously will often be rewarded in the market; there is no real guidance for MCO managers about what to do when it is not. (14)

Even for the closed panel HMO, the above model is problematic, for reasons I take up below. The model is becoming less relevant, however, since the recent growth in managed care has been occurring at the other end of the organizational spectrum, in more loosely organized structures in which physicians care for patients from multiple managed care plans, or patients can seek care outside the plan and still be covered for 40 to 80 percent of the cost. Rather than seeking to establish a coherent, cost-conscious standard of care, these plans focus on contracting with physicians who have resource-conserving practice styles (without inquiring too closely into how the resource conservation is done) and obtaining discounted prices on services from physicians, hospitals, and other care suppliers. Such a plan may be able to form a committee to decide whether to cover a controversial discrete new technology, but it is unlikely to have enough of a communal identity to use group decisionmaking strategies such as "deliberative democracy" to shape a coherent overall rationing strategy and ensure that the strategy is implemented by the individual physicians with whom the plan contracts.

Economics and Health Care Rationing

In contrast to bioethics, the basic subject matter of economics is and always has been rationing. Economists take it for granted that health benefits should compete with other benefits for scarce resources. Markets exist to coordinate individual decisions through the price mechanism to yield an acceptable distribution of goods and services given the available resources, technology, and individual preferences.

Unfortunately, rationing by price does not work well in health care. As noted, uncertainty about one's future health state creates a desire for insurance, but in the insurance context, price cannot play its assigned role of leading people to restrict themselves to the care they consider worth the cost. Information is also a problem; people need physicians to help them understand their options and make the decisions that will enable their health goals to be achieved.

The MCO offers a possible solution to this problem by selling a cost-conscious, physician-administered standard of care--a rationing strategy. In direct contrast to the bioethics perspective, economists see this SOC in terms of the rationing desired by a prudent participant in a voluntary risk pool, not the rationing associated with a social obligation to guarantee basic care. Economists do not expect an MCO to view the task of determining its SOC as one of distributing fixed resources over a fixed population according to principles of justice. Neither the MCO's population nor its resources are fixed, except in the very short run, during which (for both contractual and practical reasons) its SOC is also essentially fixed. The enrolled population, the available resources, and the SOC are actually interdependent, because people enter and exit the organization in response to the SOC and the premium.

This interdependence seriously undermines the bioethics view of the MCO as a stable mini-society establishing a stable morally obligatory standard by democratic deliberation. Instead, it supports the economics view that MCOs sell standards of care that are determined by consumer preferences expressed through market behavior. Since people differ in their preferences and their ability to pay, economists assume that they will prefer to buy different rationing strategies, just as they prefer to buy different kinds of shoes and automobiles. MCOs will survive in the market only if they offer a combination of SOC and premium that enough people want. Variation in the rationing strategies available is to be expected; in fact, many consider this a major advantage of a private, market-oriented health care system. In this context, explicitly pursuing profit is desirable, since it makes a supplier more attentive to the consumer's wishes. MCOs are free to have other goals. If they devote a significant share of their resources to pursuing them, however, competition on the tradeoff between the quality and the comprehensiveness of the SOC and the premium will drive them into financial difficulties. In sum, market forces determine the SOC, and many economists consider this desirable. (15)

This picture is oversimplified, in that it omits the issues raised by so-called "market imperfections." (16) Some of these are related to inherent features of health care. For example, given the complexity of medicine, a buyer has difficulty learning enough about the competing SOCs/rationing strategies to make a good choice. Also, the MCO must be aware that the cost to produce its product varies with the buyer; someone already chronically ill is likely to be more expensive to care for than someone in good health. Other market imperfections are related to historical accidents of American insurance markets. For example, because of income tax provisions, most health insurance is group coverage purchased by employers on behalf of their employees. (17) Market imperfections mean that in practice, the rationing strategies available in the market have only an indirect connection to consumer preferences. To make market competition work better, economists recommend public policies establishing "rules of the game" that help consumers get their preferred quality-cost tradeoffs--for example, policies to improve the information available to potential members. (18)

How (if at all) does this economic perspective bear on MCO managers' ethical concerns? Economists certainly believe that business people have ethical obligations. A market system cannot function without a commitment to common moral standards. (This is readily apparent to anyone who has spent time in a country in which common ethical standards have broken down.) Market participants should not lie, cheat, steal, violate contract provisions, mistreat employees, evade taxes, and so on. There are public sanctions, from social disapproval to a jail sentence, to make the standards clear and to penalize violations. But the choice of product characteristics is not considered an ethical issue, provided the seller is honest about them. It would be unethical to sell a product so dangerous that only uninformed buyers would buy it, and such a product may even be banned, but economists usually prefer to inform buyers and let them decide whether to acquire. (19)

What about the justice issues seen by bioethicists as central to an MCO's rationing decisions? Economists recognize distributive justice as a vital element in health policy, and most acknowledge the existence of a societal moral obligation to ensure access to basic care. Since health care financing in the United States is currently based on private insurance, many economists believe that the societal obligation might best be met by using redistributive taxation to subsidize the poor and people with high health risks on a sliding scale, allowing them to participate fully in the insurance market.

The literature on "managed competition" offers versions of this strategy. (20) The basic premise is: everyone is required to have coverage for at least a "basic package" defined in terms of broad categories of services. People choose health insurance plans in a context in which the publicly administered rules of the game are designed both to encourage competition on quality and cost and to ensure that the minimum SOC prevailing in the market has been freely chosen by people of at least average income. Managed competition advocates implicitly assume that the market behavior of those who are not poor will set a SOC that is at least adequate (probably more than adequate). (21) If consumers of average income choose not to be insured against the cost of some treatment, what better indication can there be that it is not morally required?

Unfortunately, the market result is the answer to the wrong question. It identifies the standard of care for which people choose to insure given their own risk profiles, not their opinion of the standard that must be provided to all as a matter of moral obligation. The market-determined standard of care will inevitably offer biased answers to the question: What level of care does the average American think is morally required for all? And it will be biased in both directions, sometimes toward too little care, sometimes toward too much, depending on the health condition. For example, one might believe that treatment for schizophrenia is morally required, but also believe that one's own risk of the disease is negligible during the contract period and choose a health plan that excludes it. Conversely, a woman who is having difficulty becoming pregnant might believe that in vitro fertilization treatment is not morally obligatory but choose a plan that includes it.

Moreover, the issue is not simply whether a condition is covered, but whether the quality of the treatment is appropriate. If a plan provides morally inadequate care for a chronic disease and uses the saved resources to provide better than adequate care for sports injuries, it is likely to be rewarded in the market even if everyone would agree that the chronic disease care should have greater moral importance. The plan becomes more attractive to the young and healthy (who are, on average, cheaper to care for) and less attractive to the chronically ill.

Thus, even in a world of closed panel HMOs, in which managers strive to develop a cost-conscious SOC that appeals to consumers and consumers are well-informed about the various SOCs available, the prevailing SOCs may not meet ethical criteria for adequacy. In a world dominated by more loosely organized managed care plans, an outcome that is both equitable and efficient is even less likely. Not only do consumers have difficulty comparing standards of care across plans--plan managers themselves may have incomplete knowledge of the standards of care they are selling, since they rely heavily on indirect pressure on providers to hold down plan expenditures.

The Role of Physicians in MCO Rationing

Physicians play a key role in health care by advising patients on what care to have and granting access to it. They have always had to cope with conflicting messages about their ethical duties in this role. A physician is told to be the patient's advocate and do everything of benefit without regard to cost, yet in practice, he or she must also weigh the needs of patients competing for the same resources--intensive care beds, dialysis machines, or simply the physician's time and energy--and decide whose need is greatest. In neither case is the physician supposed to allow personal gain to influence clinical decisionmaking, yet if he does not receive full payment for the care he provides, from the patient or a third-party payer, his personal goals and obligations are endangered.

There is no easy resolution to this conflict, but before the rise of managed care, most physicians were able to handle it well enough on an ad hoc basis. Since private insurers defined medical necessity as everything of benefit and paid on a fee-for-service basis, physicians could do everything beneficial and still do well financially by taking care to select most of their patients from the ranks of the well-insured. They could make triage decisions in the emergency room or the ICU when forced to but also push for more resources to minimize the need for such hard decisions. They could simply refuse to accept broader responsibility as stewards of societal resources by emphasizing their ethical duty to the individual patients for whom they had chosen to accept responsibility.

The reluctance of physicians to accept any responsibility for limiting beneficial care contributed to the cost escalation that led to the rise of managed care, and managed care has now brought the contradictions in societal expectations of physicians out into the open. In an MCO, physicians are inevitably the administrators of the MCO's standard of care and are pressured to limit beneficial care, even though most still believe it would be unethical to do so.

Bioethicists on Physicians' Ethical Obligations

In the early days of bioethics, most people in the field subscribed to the pure patient advocate view, according to which physicians are ethically required to do everything beneficial for their patients without regard to cost. (22) As they focused more on the societal problem of fair distribution of health care resources and began to recognize the need to limit care received by individual patients to achieve it, they began to appreciate the genuine conflict between the role of pure patient advocate and steward of collective resources.

One approach favored by some was to separate rationing into two levels--limit-setting at the "population" or "macro" level and limit-setting at the individual patient or "micro" level--and argue that a physician can help establish limits at the macro level but should seek everything of benefit for the individual patient at the micro level. The lawyer's traditional obligation to be an advocate for his or her client is often invoked as an analogy to this advocacy obligation of the physician. (23)

This approach is fundamentally contradictory, however. It implies, for example, that a physician who participates in a national consensus conference to establish prudent limits is morally required to work to subvert those limits for the individual patients he treats. In the context of an MCO, it implies that the MCO'S physicians should be excluded altogether from the participatory process of setting the standard of care, or that having participated, they should then ignore the group's decisions when they are treating the MCO's patients. Of the many problems with the analogy between lawyers and physicians, perhaps the most important is the fact that delivering health care is a cooperative enterprise. As physician-lawyer William Sage puts it: (24)

   If physicians are to oversee the institutional
   provision of care, serving
   as advocate is an isolating and
   potentially counterproductive
   role, specifically because the opposing
   party will often be the institution
   whose skills and compassion
   the physician is charged with
   furthering. There is no role for
   nonaligned institutions in an adversarial
   model. Institutional resources
   are either controlled by the
   physician-advocate or opposed to
   her. Given the overall need for
   cost-consciousness and the mix of
   insurance and provider organizations
   in today's health care system,
   this seems an unproductive use of
   valuable assets.

In the 1980s, some bioethicists began to see that there might be a way out by thinking about rationing from the point of view of the prudent participant in the voluntary risk pool. It is in each member's long-term interest to have fair rationing rules in place, to have physicians agree to administer them, and to have physicians retain some discretion to adjust the rules to an individual patient's circumstances. If so, then it is ethical for physicians to limit beneficial care. (25) This position is now gaining ground in medicine and ethics. (26) Of course, this analysis resolves the physician's ethical problem but may still leave him with a difficult physician-patient relationship, since the patient's short-term and long-term interests conflict. The physician also has the problem of knowing the rules, and the boundaries of his discretion within them.

Economists on Physicians' Ethical Obligations

Economists take it for granted that being an agent for the patient should mean doing what benefits the patient, given both the medical benefit and the cost the patient bears. To an economist, it would be unethical to provide a patient with care that is not worth what it costs the patient. When the costs are covered by insurance, things get more complicated. Then, if the patient does not bear any out-of-pocket cost, it would seem that the doctor has an ethical obligation to do everything of benefit. However, it is in an insured person's best interest to agree as a matter of contract to be subject to non-price rationing, and there is no reason why a physician should consider it unethical to be part of such a system. Of course, as before, the physician still has to know the rules and manage the interpersonal conflict.

Unfortunately, at this point, economists drop the ball. They do not provide a convincing explanation of how this will occur, but implicitly assume that through market competition, there will emerge a mix of organizational and financial constraints on physicians and patients that brings their goals into reasonable harmony, with only the limited government intervention that establishes the overall rules of the game.

In the managed care world, the group or staff model HMO in which the physician delivers the same cost-conscious standard of care to all of his patients is increasingly rare. Many physicians contract with multiple MCOs that explicitly or implicitly require them to administer different standards of care and to face different incentive structures depending on the plan to which a patient belongs. Yet there is little recognition among health economists that American physicians find it morally troubling to ration at all, let alone to administer an array of rationing strategies. There is no explicit discussion of how physicians should resolve conflicts among their duty to their own patients, their duty to the patients of others, and their duty to society. Many economists do not even seem to realize that such conflicts will occur.

Essential Changes

The ethical confusion among MCO managers and physicians is not surprising. Managers are told that they are ethically obliged to set a justice-based standard of care, but the economic pressures they face require them to provide the standard demanded by the market. These are not the same standards in practice, and they would not be the same even in an optimally regulated, universal access, managed competition world of closed panel HMOs.

Physicians are still told by some bioethicists and physician colleagues that they are ethically obliged to be pure patient advocates and ignore cost in their clinical advice to patients, yet this is incompatible with their assigned managed care role of gatekeeper. To make things worse, because both the public and physicians are so resistant to the idea of rationing, MCOs currently rely heavily on indirect economic pressures to induce physicians to limit care. The result is a system in which the details about how care is rationed are set by physicians responding individually to personal financial incentives--even though both bioethicists and economists find such a system ethically troubling.

Both disciplines would like to find a way to guarantee adequate care to all without the need for a national effort to define adequacy. This is understandable, given the difficulty of the task, but in the current world of competing MCOs, it cannot be done. Ironically, the more we rely on market forces to distribute health care, the more we need a national minimum standard of care. With such a standard, there is some hope that competition will take place not on establishing the leanest rationing strategy the market will bear (however ethically problematic it may be), but on delivering the agreed upon minimum standard efficiently. With such a standard, the physician's ethical obligation to an individual patient is to interpret it in the light of the patient's circumstances and to make sure that he or she is offered it.

To do this, there must be a mechanism to establish, at the national level, at least the general outlines of the content of the adequate level, the role of MCOs within it, (27) and measures by which MCO managers and physicians can be held accountable for their performance. There would then be an established threshold standard of care that every MCO would have to provide. Consumer preferences expressed through the market would simply determine the extent to which plans offered more than this threshold. It is at the level of determining the threshold standard of care for all plans, not the level of determining an individual plan's standard, that bioethics' discussion of deliberative democracy and fair decisionmaking procedures seems most relevant. (28) With this standard, the ethical duties of physicians and managers can be clarified. Both physicians and managers are ethically obliged to participate in societal deliberations about the content of the threshold SOC, and then to make sure that it is delivered at the level of the individual organization and individual patient.

Financial incentives must also be brought into alignment with these redefined ethical obligations of managers and physicians. If the ethical expectations of MCO managers are to be consistent with their organizations' economic survival, an essential element is a mechanism for risk adjustment. Risk adjustment redistributes premium revenue so that MCOs receive greater compensation when they enroll members with predictably greater costs. Without it, plans have strong financial incentives to compete by controlling the composition of their risk pools (favoring the lower, and cheaper, risks) rather than by providing high quality at low cost. Plans may hesitate to make the effort to provide excellent care to chronically ill patients if they receive only the usual compensation, notwithstanding the fact that these patients are more expensive than average to care for. Relatively small differences in the standard of care can have a significant influence on the mix of members. (29)

For physicians, the goal should be to eliminate financial incentives that are directly linked to clinical decisions that allocate resources to individual patients. The most straightforward way to do this would be to pay physicians salaries. In our economy, most highly skilled people who work for large organizations are paid on a salaried basis, not by piecework; their performance is judged in relation to general performance standards, and they are fired, promoted, and given merit raises and bonuses in accord with the quality of their work. In the past, it was argued that salaried employment in an organization would reduce the ability of a physician to be an independent agent for the patient, but now that changes in technology and financing have made linkages to organizations inescapable, this argument has lost its force.

Physicians are not suited by temperament or training to the careful analysis of complex payment structures. It seems to me, speaking both as an economist and as a potential patient, that it would be better for them to focus, as a profession, on working with others in society to establish a morally acceptable, cost-conscious standard of care, and then individually, on applying that standard appropriately to individual patients. They would be accountable for the skill with which they did this, not for reducing resource use per se.

Such changes would clarify the ethical responsibilities of MCO managers and physicians with respect to rationing. They would also be in the long-term self-interest of the members of MCOs. When market choice is the sole driver of the composition of the standard of care, consumers must be constantly vigilant to ensure that they know what they have bought and will actually get it if they need it; otherwise, they are in for nasty surprises. More important, a minimum standard of care set by consumer decisions about the standard of care they want for next year, given their personal risk profiles, does not protect people against changes in their own risk profiles over time, nor ensure that other people they care about will always be able to obtain basic health care.

Making the changes would not be easy. In particular, I am not sure that it would even be possible to design a mechanism that can satisfactorily define a morally acceptable threshold standard of care over time and hold competing for-profit MCOs accountable for delivering it. No one has done it yet. What I am sure of is that market competition does not perform these tasks automatically, even under the stylized conditions of perfectly functioning markets, let alone in the real world. However difficult the challenge may be, the market is not an easy way out of it. (30)

Acknowledgment

This paper was begun while I was a Fellow in the Institute for Ethics at the American Medical Association. I am grateful for the financial support of the AMA, but the opinions are mine and do not necessarily reflect the position of either the AMA or any of its divisions or affiliates.

References

(1.) See M.A. Baily, "Policies for the 1990s: Rationing Health Care," in Competitive Approaches to Health Care Reform, ed. R. Arnould, R. Rich and W. White (Washington, D.C.: Urban Institute Press, 1993), 313-40 and the references therein for further discussion of rationing definitions and the reasons for choosing this one. See also P. Ubel, Pricing Life: Why It's Time for Health Care Rationing (Cambridge, Mass.: MIT Press, 2000), chapter 2, and M. Hall, Making Medical Spending Decisions: The Law, Ethics and Economics of Rationing Mechanisms (New York: Oxford University Press, 1997), chapter 1.

(2.) This phenomenon is called moral hazard in insurance terminology. A homely analogy is the situation that arises when people dining out together decide in advance to split the bill evenly. Each person has a financial incentive to order a more expensive meal, yet the group as a whole must pay the total cost. If the check is to be evenly divided, it would be wise to agree to choose from a limited menu, i.e. to submit to some (non-price) rationing.

(3.) For detailed discussion of the arguments for and the nature of this societal obligation to ensure universal access to an adequate level of health care, see, for example, President's Commission for the Study of Ethical Problems in Medicine, Securing Access to Health Care, Volume One: Report (Washington, D.C.: U.S. Government Printing Office, 1983) and A. Buchanan, "Health-Care Delivery and Resource Allocation," in Medical Ethics, ed. R. Veatch (Sudbury, Mass.: Jones and Bartlett Publishers, 1997), 321-61.

(4.) M.A. Baily, "Rationing Medical Care: Processes for Defining Adequacy," in The Price of Health, ed. G. Agich and C.D. Begley (Dordrecht, The Netherlands: D.Reidel Publ. Co., 1986), 165-84.

(5.) The best-known real world example of this type of rationing comes from the first dialysis program in Seattle in the early 1960s, in which a small group of people from different walks of life used social worth criteria to decide which patients should be admitted to the program and thereby saved from certain death. See R. Fox and J. Swazey, The Courage to Fail: A Social View of Organ Transplants and Dialysis, 2nd ed., revised (Chicago: The University of Chicago Press, 1978) for a classic discussion of this so-called "God Committee" and R. Rettig, "Historical Perspective," in Ethics and the Kidney, ed. N. Levinsky (Oxford, U.K.: Oxford University Press, 2001), for the history of the rationing of kidney dialysis and kidneys for transplant from 1960-99.

(6.) Buchanan, "Health-Care Delivery and Resource Allocation."

(7.) For discussion of the difficulties see, for example, N. Daniels, "Four Unsolved Rationing Problems: A Challenge," Hastings Center Report 24, no. 4 (1994): 27-29; F. Kamm, "To Whom?" Hastings Center Report 24, no. 4 (1994): 29-32; E. Rakowski, "The Aggregation Problem," Hastings Center Report 24, no. 4 (1994): 33-36; J. Broome, "Fairness versus Doing the Most Good," Hastings Center Report 24, no. 4 (1994): 36-39; M.A. Baily, "The Democracy Problem," Hastings Center Report 24, no. 4 (1994): 39-42.

(8.) For example, M. Strosberg et al., eds. Rationing America's Medical Care: The Oregon Plan and Beyond, (Washington, D.C.: Brookings Institution, 1992); N. Daniels, "Is the Oregon Rationing Plan Fair?" JAMA 265, no. 17 (1991): 2232-35; L. Fleck, "Just Caring: Oregon, Health Care Rationing, and Informed Democratic Deliberation," Journal of Medicine & Philosophy 19, no. 4 (1994): 367-88; M. Garland, "Justice, Politics and Community: Expanding Access and Rationing Health Services in Oregon," Law, Medicine & Health Care 20, nos. 1 and 2 (1992): 67-81.

(9.) See L. Jacobs, T. Marmor and H. Oberlander, "The Oregon Health Plan and the Political Paradox of Rationing: What Advocates and Critics Have Claimed and What Oregon Did," Journal of Health Politics, Policy and Law 24, no. 1 (1999): 161-79.

(10.) P. Kongstvedt, Essentials of Managed Health Care, 2nd ed. (Gaithersburg, MD: Aspen Publishers, Inc., 1997). The book, by the way, has 557 pages.

(11.) N. Daniels and J. Sabin, "Limits to Health Care: Fair Procedures, Democratic Deliberation, and the Legitimacy Problem for Insurers," Philosophy & Public Affairs 26, no. 4 (1997): 303-50; N. Daniels and J. Sabin, "Closure, Fair Procedures, and Setting Limits within Managed Care Organizations," Journal of the American Geriatrics Society 46 (1998): 351-54.

(12.) E. Spencer et al., Organization Ethics in Health Care (New York, Oxford: Oxford University Press, 2000).

(13.) N. Daniels and J. Sabin, "Last Chance Therapies and Managed Care: Pluralism, Fair Procedures, and Legitimacy," Hastings Center Report 28, no. 2 (1998): 27-41.

(14.) Spencer et al, Organization Ethics.

(15.) See Baily, "Policies for the 1990s," and the references therein for a full discussion.

(16.) The classic economics article on the role of market imperfections in health care and health insurance markets is K. Arrow, "Uncertainty and the Welfare Economics of Medical Care," American Economic Review 53, no. 5 (1963): 941-73.

(17.) U.S. Congress, Congressional Budget Office, The Tax Treatment of Employment-Based Health Insurance (Washington, D.C.: CBO Publications Office, March, 1994).

(18.) For further discussion, see Baily, "Policies for the 1990s."

(19.) Economists are as committed to individual autonomy as bioethicists. The main difference between the two perspectives is that economists give greater attention to who is bearing the resource costs associated with an individual's autonomous choice.

(20.) Baily, "Policies for the 1990s," and references therein; also, A. Enthoven, "The History and Principles of Managed Competition," Health Affairs 12, Supplement (1993): 24-46.

(21.) It is interesting to note that what actually happened in Oregon resembled this picture of rationing care to Medicaid recipients more closely than it resembled the stylized picture of rationing in terms of ranked condition-treatment pairs. Most of the people covered by the reform were enrolled in existing managed care plans. Reimbursement rates were high enough and the people enrolled were similar enough in health status to the regular managed care population that the plans could provide their usual standard of care and ignore the condition-treatment rankings. For most of those covered by the reform, the resulting access to care was as good as or better than their previous situation. See Jacobs, et al., "The Oregon Health Plan."

(22.) For an excellent survey of the bioethics literature on the pure patient advocate view, with references and representative quotes, see Hall, Making Medical Spending Decisions, chapter 4.

(23.) See, for example, S. Wolf, "Health Care Reform and the Future of Physician Ethics," Hastings Center Report 24, no. 2 (1994): 28-41.

(24.) This quote is from a superb analysis and ultimate rejection of the physician/lawyer analogy, W. Sage, "Physicians as Advocates," Houston Law Review 35 (1999): 1529-1630, at 1621.

(25.) See, for example, P. Menzel, Strong Medicine: The Ethical Rationing of Health Care (New York: Oxford University Press, 1990); E. Morreim, Balancing Act: The New Medical Ethics of Medicine's New Economics (Dordrecht, The Netherlands: Kluwer Academic Publishers, 1991).

(26.) See, for example, Medical Professionalism Project, ABIM Foundation, "Medical Professionalism in the New Millennium: A Physician Charter," Annals of Internal Medicine 136, no. 3 (2002): 243-46.

(27.) The MCO need not be assigned responsibility for the entire standard. Some kinds of care might be guaranteed through separate social mechanisms. Of course, careful attention would have to be given to the boundaries between delivery systems to avoid the inefficiencies and perverse incentives that can arise.

28. See also A. Buchanan, "Managed Care: Rationing without Justice, But Not Unjustly," Journal of Health Politics, Policy and Law 23, no. 4 (1998): 617-34 on the importance of defining the morally required standard of care.

(29.) See Baily, "Policies for the 1990s," for further discussion of this point.

(30.) Given the widespread belief that it is impossible to define the morally required level of health care, it is ironic that we do not see the contradiction in maintaining institutional structures that only make sense on the assumption that we already have defined it. The malpractice system, civil litigation over insurance contracts, and the external review systems being proposed to allow patients to appeal allocation decisions implicitly assume a normative standard of care that can be applied in individual cases. For discussion, see, for example, M. Hall and G. Anderson, "Models of Rationing: Health Insurers' Assessment of Medical Necessity," University of Pennsylvania Law Review 140 (1992): 1637.

Mary Ann Baily, "Managed Care Organizations and the Rationing Problem," Hastings Center Report 33, no. 1 (2003): 34-42.

Mary Ann Baily is associate for ethics and health policy at The Hasting Center, and is an economist by training. She has written on a range of issues in health policy and biomedical reseach.