The house that Medicare built: remodeling for the 21st century.

Introduction

The current American Medicare system can be viewed as a house designed and built to fit on a very strangely shaped lot. The metaphorical building site that created the constraints within which the Medicare system was designed was the strange American fee-for-service (FFS), private practice, health care system of the early 1960s. Medicare's basic design flaws, shaped by those unique constraints, led to the inevitable problems that have emerged as the Medicare system followed the bulk of the American health care system into its destructive nosedive. I'll take this opportunity to develop some thoughts on the future of the relationship between Medicare and managed care, rather than cataloguing, in exquisite detail, where that relationship has been and what the current state of the relationship is. However, in order to think about the future of this relationship, it is necessary to frame it within the context of the larger American health care system.

In order to provide this frame, I'll reflect on the nature of the health care system and review some of the significant changes in the system, comparing the system at the time Social Security was first implemented with the time of Medicare's 20th anniversary. Then, as I propose a remodeling of the house that Medicare built, I'll speculate on the future of the health care system, particularly as it applies to the elderly in America. This reflection is of special value in contemplating the future of the relationship between Medicare and managed care, because that relationship will continue to be shaped by the role of the population-based clinical practice models in the overall health care system.

Let's look first at the structure of the health care system in 1935, the year of the passage of the Social Security Act. The national Social Security system had been designed for implementation without a health care component. The Nation was in the depths of a major depression. Unemployment was extremely high. Americans lived in a world without much in the way of social safety nets. Physicians practiced almost entirely in solo practice settings and the payment mechanism was entirely out-of-pocket. Many people had no access to care. When World War II began in 1941, a very large proportion of the military inductees examined were seeing a doctor for the first time in their lives. When medical care was given, it was given in the doctor's office, or perhaps in the patient's home. The role of technology was extremely limited and the function of medical care was just that, caring for sick patients. Because of the lack of technology and the focus on caring for sick people, the system and its practitioners were evaluated on the basis of how nice it was (or how nice they were).

But 50 years later, by 1986, things had changed dramatically. This was the middle of the Reagan years in America. Medicare had been in place for 20 years, but two major attempts at overall national health care reform had been defeated. There had been a dramatic expansion of medical school capacity, and medical care expenditures were eating up more than 12 percent of the U.S. gross domestic product. As a Nation, we were just beginning to doubt that we were living in a golden age, where everything was possible for all people, as we had believed at the implementation of Medicare in 1966. Medical care took place in small-to-medium-sized practices, with most doctors practicing in some form of organized group. Payment for medical care was covered by Medicare and Medicaid, by private insurance, or to some extent was paid out of pocket. We were witnessing the point of maximum coverage by private health insurance—more than 90 percent of the population were covered for hospital expenses, and about 80 percent were covered for physician expenses—and we were just starting to see the downward curve that has continued ever since.

The dominant site of care was the hospital. The role of the government in health care was as the payer of last resort. The form of physician payment in 1985 was mixed, with the major proportion still coming from FFS, but the fastest growing segment included capitation or salary arrangements. The future of managed care systems was beginning to look very bright, as a variety of new models were coming into the market, and a plethora of investor-owned corporations was bringing enormous amounts of capital into the field.

The role of technology was growing, but technology was mostly in the hospital and was mostly in the form of technologies of failure, such as coronary artery bypass graft surgery or diseased-organ transplantation. The function of the medical care system was curing disease or at least interfering with the disease process. The system (and its providers) was measured by how technical it was. For example, much of the dispute in malpractice cases was (is) around whether all appropriate/possible diagnostic tests were ordered in any situation. And finally, by 1985, the nature of the physician's obligation to patients was becoming increasingly unclear. Insurance had so sufficiently intervened between the doctor and the patient that the doctor's strong sense of obligation to the patient, at least with regard to economic issues, seemed to have become eroded.

Medicare and Managed Care: A Troubled Relationship

As I suggested before, Medicare was designed for the dominant American medical care system. Those design considerations included, of course, creating the reimbursement method. As Medicare was about to become a reality, the existing prepaid group practice plans in the United States became increasingly aware that they were in for a significant discontinuity in dealing with the aged members of the populations for which they were responsible. Most of the existing plans, such as Kaiser Permanente, Group Health Cooperative of Puget Sound, The Health Insurance Plan of Greater New York, Group Health, Inc., in Minneapolis, and the Group Health Association in Washington, DC, had members who were going to be eligible for Medicare coverage. Those aged members had been in the plans through their work, and they continued membership in the plans after retirement. Their premiums were paid by their former employer, or the members paid them themselves directly to plan. It was clear that the plans were neither interested in nor prepared to collect reimbursement according to the standard Medicare FFS-based procedures that were emerging.

The relationships between the Johnson administration and the plans had been excellent, because many physicians and executives of the plans supported the concept of Medicare. But as the plans negotiated for a different payment arrangement, they became increasingly frustrated. They were interested in developing capitation arrangements for Medicare beneficiaries; the administration was interested in avoiding any further complications. The administration was also particularly concerned about not offending organized medicine, which was ironic, considering the extraordinary opposition to Medicare on the part of organized medicine. The frustration level of the plans reached a kind of symbolic peak at one of the meetings, when a plan executive reportedly complained to Wilbur Cohen, the then-Acting Secretary of Health, Education, and Welfare, “Wilbur, it's enough to make you lose faith in socialized medicine.”

But eventually a patchwork method was developed, called the group practice prepayment plan (GPPP) approach, which allowed the plans to be paid in a way that approximated capitation methodology. This payment method was really a prospective cost-for-service (or almost-cost-for-service) method, with post facto reconciliation. The first modest remodeling of the house had been achieved, allowing managed care plans to enter.

The plans immediately began lobbying for some kind of true risk-based, capitation payment for managed care systems. We need to remember that the managed care systems of the time were non-profit group- and staff-model health maintenance organizations (HMOs) as well as a few county-based plans for physician services, such as the Marshfield Clinic in Wisconsin and the Physicians' Association of Clackamas County in Oregon. The 1972 Medicare amendments included the so-called “1876 amendments,” which allowed for experimentation with risk-based payments to the GPPP plans. But no plan, except the Group Health Cooperative of Puget Sound, ever took advantage of that approach to Medicare reimbursement.

In the half-decade following the 1972 amendments, the HMO Act was passed and implemented, and the idea that the organizations now called HMOs could provide care at a reasonable price became current on the national scene. Policymakers began inquiring as to why HMOs accepted Medicare beneficiaries only at the time they aged into Medicare. And they were told that HMOs worked on a prospective capitation payment basis and that the current model for Medicare reimbursement simply didn't cut it. Eventually, HCFA was moved to set up a risk-reimbursement demonstration project, and five demonstration sites were selected during 1978 to begin serving beneficiaries in 1980, under risk-based capitation.

There were several things to be tested in the demonstration. First of all, there was a widely held view that Medicare beneficiaries would not leave the security of an established relationship with a private physician to join a managed care program. Second, there was some doubt on the part of Medicare administration people and many policymakers in Washington about whether it was appropriate to contract for Medicare services in a prospective payment method, even if people would leave their private physician. The FFS ideology ran strongly in Washington. Also, the plans were skeptical that a payment methodology could be worked out that would adequately compensate them for the care of the elderly and still provide the savings that could be used to provide financial incentives for the beneficiaries to voluntarily join the plan. And finally, some plan physicians were reluctant to significantly increase the relative proportion of elderly in the plan because of the effect those changes would have on their medical practices.

Several plans actually did begin offering coverage to Medicare beneficiaries under the first round of the demonstration and under a larger second round that followed shortly. From most perspectives, the demonstrations were successful, or at least they were successful in enrolling new Medicare members into managed care from the FFS system. The belief that beneficiaries would not leave the FFS world for managed care proved not to be the case (Greenlick et al., 1983). Also, it was possible for the plans and HCFA to get some experience with a prospective payment, risk-based reimbursement method, based on the average adjusted per capita cost (AAPCC) methodology, although critics on both sides of the fence still harbor concerns about that specific methodology.

However, because of the apparent success of the demonstrations, risk-based, prospective payment methods were approved in the Tax Equity and Fiscal Responsibility Act (TEFRA) of 1982, and the use of managed care plans to provide service to Medicare beneficiaries was institutionalized. The Medicare house now had been remodeled sufficiently by TEFRA to set the stage for the growth of managed care programs and for the slow but increasing enrollment of Medicare beneficiaries into capitated managed care programs. The progress was not easy. The early days of the program were marred by managed care scandals in Florida and other places. Some of the apparently successful plans, such as the pioneering Marshfield Clinic, gave up their risk contracts because of financing inadequacies. And some plan physicians continued to resist increasing the proportion of elderly in their population.

But by 1987, about 1 million Medicare beneficiaries were enrolled under Medicare risk contracts. That number had grown to 2.8 million by 1995, and the rate of increase is still growing (U.S. General Accounting Office, 1996). But the important point about this growth is that it has been paralleled by the growth in managed care in the country generally. As the market share of managed care plans has grown, especially with the entry of profit-oriented managed care companies into the marketplace, plan managers have found significant potential in the Medicare market. This has been especially true in areas with high AAPCCs, such as Southern California and Florida. Consequently, in some of these areas, the market share has grown significantly, exceeding 20 percent of the Medicare market in California and Arizona and exceeding 10 percent in several other States, including Florida (U.S. General Accounting Office, 1996).

The relationship between Medicare and managed care continues to be a difficult one. For one thing, there has been a great deal of pressure on the payment system for risk contracts. Critics of managed care contend that the risk contractors use cream-skimming and other inappropriate methods to gain unfair advantage and accrue obscene profits. Many managed care programs continue to argue that the reimbursement methodology in place does not adequately pay for the legitimate costs of delivering services to Medicare beneficiaries. Physicians complain that the new managed care programs interfere inappropriately with the doctor-patient relationship to the detriment of the patient's best interest, and perhaps against the doctor's best interest as well. Beneficiaries complain that some managed care systems withhold needed services. There is probably some truth to each of the assertions, at least at the margin. But basically the system is currently working to the benefit of the enrolled beneficiaries.

I believe that there will be some significantly bad times before the system corrects itself and we create the kind of a health care system that we have long pretended to have. And I think that the care of the elderly will be in new and exceptional forms of population-based clinical practice models. Current managed care plans are primitive versions of the ultimate models that will emerge. And I think it is possible, perhaps even likely, that we will be able to develop humanistic forms of health care for the 21st century. But there are significant barriers to overcome on our path to the realistic “utopian” models that I believe are possible.

Humanistic Health Care in the 21st Century

Overview

We must accept the reality of the complex organization as the basic unit of the health care system. This gets us to the heart of the issue of the future of Medicare and the 21st century health care system and gives us a chance to look at the blueprints that could guide the remodeling. It's not too early to characterize the system over the next 20 years without much wild speculation (Table l). Most of the elements in this analysis are the inevitable product of forces already well under way. These features of the health care system do not depend on national health care reform being enacted. There are sufficient State initiatives and market developments to change the face of health care. Medical care costs have moved past a trillion dollars a year in the United States. Large numbers of people are uninsured. Distress about the system is at dangerous levels.

Table 1

Change in the Structure of the Health Care System in One Lifetime

Element	Year
	1935	1985	2005
Nature of Practice	Solo Practice	Small-to-Medium Organized Forms	Large Organizations
Payment	Out-of-Pocket	Private Insurance and Out-of-Pocket	Socially Organized Payment Mechanisms
Dominant Site of Care	Physician's Office	Hospital	Diffuse Networks
Role of Government	None	Payer of Last Resort	Primary Organizer of Financing
Form of Physician Payment	Fee-for-Service	Mixed	Capitation and Salary
Role of Technology	Minimal	Moderate (Mostly Hospital)	Extremely High
Function of Medical Care	Care	Curing Disease	Disease Prevention, Maintenance of Function
Measured by	How Nice?	How Technical?	How Cost-Effective?
Physician Obligation to Patient	1:1	Ambiguous	1:n

SOURCE: Greenlick, M.R.: Educating Physicians for the 21st Century. Academic Medicine 70(4): 179-185, March 1995. Reprinted with permission of the author.

Cataclysmic forces are changing the nature of health care organization and delivery in most communities in the United States. There is extraordinary vertical and horizontal consolidation taking place in the United States today. All but the very largest communities will end up with care being delivered by two or three major health care systems. During the 21st century, most citizens in the Western industrialized world will receive their care within some kind of a complex medical care organization, such as a group practice prepayment plan or some other form of managed care organization.

The payment mechanism for this care will most certainly be socially organized in some way. The one constant that ran through the recent health care reform debate, nationally and locally, was that it is not tolerable to have 35-50 million people uninsured. That problem will be solved during the next decade or so, and as a result of that solution, most people receiving health care services will have their expenses paid by employment-based insurance, by some form of government-subsidized insurance, or in some other way. The role of government will be as the primary organizer of financing for an increasingly large segment of our population. That is not to say that governments will be the primary payers for care, but that governments will be involved in developing complex new payment structures such as high-risk pools (as in Oregon) and State-organized purchasing alliances (as in Washington State) and other as-yet-undeveloped methods. And Medicare will be alive and well, albeit changed in many ways.

Care will be delivered across diffuse networks. Nothing I am saying should lead you to believe I think all care will be delivered in integrated, totally organized, prepaid staff- or group-practice models, such as Kaiser Permanente. A variety of new health care organization forms within which to deliver and receive care are being invented. But they will be organized. Most physicians will be paid according to capitation or salary schemes. FFS payment mechanisms, as we knew them in 1986, are virtually dead already, even if they are still writhing around like a snake with its head chopped off. This payment revolution is already having significant consequences.

In the FFS world of the past 50 years, the strategy of hospitals and specialist physicians has been to find ways to increase revenues. The vehicle for this has been such things as high-powered surgical specialty centers, e.g., heart surgery programs. In this case, the more the surgery, the greater the revenue, and the greater the income for surgeons. However, in a capitated world, more surgery means more expense, not more revenue. What once were enormously profitable revenue centers are becoming enormously expensive cost centers. This shift from revenue center to cost center for things like heart surgery units is driving more changes in the health care system than either ideology or health policy analysis. As vertically integrated systems become linked to capitation payment models, these systems begin looking for ways to decrease per capita costs. This leads to proposals for such things as gatekeepers controlling referrals, second-opinion approaches, and the development of strict guidelines for service. The community of specialists in most cities recognizes the danger as it feels uncontrollable forces closing in.

The role of technology in the 21st century health care system will be extremely high, but the technologies will not be the technologies of failure. The dominant technology for dealing with polio advanced from the iron lung in 1935 to polio vaccination in the 1960s. Similarly, microbiology, genetics, and other of our basic sciences will produce technologies that will revolutionize the next century's health care system. In the next 20 years, it will be possible to look ahead with confidence to the elimination of many forms of heart disease and cancer. These extraordinary technologies will change the nature of the health care system (as they will change the nature of our society). The focus of the health care system will, by necessity, be on preventing disease and on maintaining function. And this sea change in the focus of the medical care system is particularly apposite for care of the elderly.

The success of the system will be measured by how cost-effective it is and how well it works to maintain the mental, social, and physical functioning of its participants. And finally, the obligation of the physician will be not only to individual patients but also to the populations from which patients come, the 1-to-n obligations.

Definition

A humanistic health care system links each individual to his or her health care system, one person at a time, on the basis of that individual's needs, desires, aspirations, risks, disease condition, and health and functional status. Let's think about how these 21st century health care systems could look. The development of managed care during the 20th century featured innovations in the organization and financing of care, but major innovations in the delivery system were generally not undertaken. Standardization was the basic principle for saving money within prepaid staff and group practices, and participants in the systems were more or less forced into predetermined and standardized forms of care.

During the 21st century, that approach is not going to be necessary or appropriate. Managed care systems of various kinds have the potential for conceptual breakthroughs that will allow them to become not only efficient and effective but humanistic as well. As I said before, becoming humanistic depends on these systems' ability to turn what was a 20th century disadvantage into a 21st century advantage. During the first decade of the next century, we will test the hypothesis that organized care systems can turn their size and complexity into an advantage by using the resources of the system to create individualized links between participants and their medical care systems. This kind of a model could create a new “virtual” health care system for each participant.

Barriers

There are at least four major sets of barriers to the development of extraordinarily new health care delivery models, in addition to all of the financing and policy barriers that will be and have been discussed in the literature. I'll focus on these particular barriers in this article because they are especially relevant to the future of Medicare and managed care. The first special barrier relates to the need to develop the appropriate social control model for the health care system. The second barrier has to do with developing the areas of knowledge we will need to deliver care appropriately in a system as complex as the new systems must, perforce, become. Essentially, we will need a totally new level of publicly and privately financed health services research. Third, we need to develop new and more appropriate, risk-based payment mechanisms for paying for care in population-based clinical practice organizations. And finally, we need to find a way to refocus clinical care away from the disease-curing model that finished the 20th century to a new approach that focuses on the maintenance and improvement of function.

Barrier: There must be a form of control of the health care system that assures and enhances trust. Perhaps the most critical issue facing the health care system worldwide is the destruction of the social contract to assure quality health care among members of a society. The United States is leading the world in destroying that social contract, but other countries are closing in on our record of the erosion of trust in our health care institutions. I've come to believe that to understand the trust issue we need to explicate the dramatic shift in power in the health care system over the last 50 years. It is in that shift of power that we find the underlying cause for the erosion of trust.

Prior to the end of World War II, it is pretty clear that the power, in the United States, was in the hands of “the people,” whatever that means. Certainly physicians had control of the profession prior to World War II. But it didn't matter very much, because they didn't have very powerful tools with which to intervene in the disease process. At least in the United States, the financing system was quite primitive prior to the diffusion of health insurance, since we hadn't developed the Social Security-based health care financing that was emerging across the industrialized world. Trust in individual physicians, among those who used physicians, was very high in those days, even though most of the physicians were less technically effective than physicians are today.

But as we moved into the 1960s and 1970s, power in the health care system shifted to the providers—to physicians and hospitals—although as patients used the system, they didn't see much difference. Their interaction was with individual physicians, so when they faced power it was clearly in the hands of those individual physicians. This period, including the first decade of Medicare, was the golden age in a lot of ways. The new house seemed bright and shiny, there on its funny-shaped lot. Society was generally protected because social control was embedded in the role obligations inherent in the one-to-one physician-patient relationship. Trust was very high in the era, even though there is compelling evidence that this trust was systematically violated in the United States, at least in terms of intervening far too much and without the best interest of individual patients in mind. The culture of FFS medicine supported the systematic exploitation of society and of individual patients in the process. But all of this happened within the existing normative structure of the medical profession. The protection of individual patients was hard-wired in the physician role and in the socialization of physicians to those role obligations.

But the changes in the health care system that have taken place since the early 1980s have truly changed the balance of power in the United States. Power first shifted to the purchasers of care—large employers and the government. And because of the takeover of power by the purchasers, especially the corporate purchasers, the shift to profit-making, managed care companies was facilitated. Because price was the major objective of the corporate purchasers, investor-owned organizers of care were able to enter the market, buy services at the margin, and take over major shares in many markets. And because the primary interest of the corporate organizers of care was in manipulating the stock value of their corporations, rather than long-term market stability, they were able to prosper and grow. The secondary shift of power to corporate organizers of care has been accelerating very dramatically, and the balance of power has shifted in their direction. And this shift must lead us to address the issue of the erosion of trust in the health care system.

I believe that the resultant erosion of trust is the rational response of skeptical Americans. There is no earthly reason, given events of the past 50 years, why individuals should trust their employers, their government, or the corporate organizers of care. From a social control perspective, there is nothing in the culture of these institutions, except for government, that works to protect the individual. As a society, we need to really worry about this. There have been many critics in the United States of non-profit group- and staff-model prepaid group practices, such as Kaiser Permanente. But my experience with these eleemosynary organizations tells me that they have in them the potential for trust on the part of their constituencies. The basis of that trust is in the nature of their cultures (Greenlick, 1988). And I think, given the nature of our current health care system, organizational culture is the only possible existing basis for trust—if we only had organizations we could trust.

As I pointed out in an earlier article (Greenlick, 1992), I believe the culture of the non-profits has changed the role-set of participating physicians, thus creating a modern version of the physician-patient relationship that adds to the traditional obligations the 1-to-n set of obligations to the population from which patients derive. I think that the new definition of the physician role provides a rational mechanism for social control and, therefore, for trust.

This analysis leads me to believe that we need to take seriously the question of how we get the power in the American health care system out of the hands of corporate managers and back to “the people.” I think we need to consider such options as banning for-profit companies from the organization and financing of health care. And we need to create and debate other such solutions, including regulatory solutions. And to policymakers abroad I say that you need to resist with your last breath the siren songs of free-enterprise consultants who argue that the solution to health care financing and organizing problems is to privatize the system. Rather, the issue is how to put societal structures in place that provide for the social control mechanisms that can be the basis of both individual trust and efficiency. This will certainly require changes from where we are and where we are going.

Barrier: We must overcome a difficult set of knowledge deficits. This new health care system must be based on a revolution in formal health services research and development (R&D). The original concepts underlying the development of programs like Kaiser Permanente in the United States, the British National Health Service in the United Kingdom, and the sick fund systems of other Western countries were extremely powerful. But time has caught up with us. These are 50- to 100-year-old medical care systems. What once produced advantage now creates inertia. The solution is to address the situation directly and create a new and unique medical care system.

There are some obvious areas that represent targets of opportunity for this R&D effort. We must get health care systems out onto the leading edge in several technological domains. Therefore, there are a set of content areas to be covered as we struggle to change the way medical care is provided in these new health care systems. These are: technology development and assessment; care management science; and human interaction sciences.

Technology Development and Assessment

There are three technology areas in which we must undertake serious R&D activities. They are:

Information Technology: This must be a keystone of a national R&D strategy, because it underlies the ability of systems to link individuals uniquely to their medical care source.

Clinical Technology: We need to be developing new kinds of clinical technology, those designed to fit a model of care oriented toward prevention and maintenance of function. We have accepted the disease-intervention-oriented clinical sciences long enough. We particularly need to be scanning the horizon and implementing application-level clinical trial models.

Technology Intelligence and Assessment: We need the capability to scan the technology world, including the non-health care world, to pick the potential winners, to support their development, and to test the most promising applications.

Care Management Science

The ultimate health care systems will feature advanced care-management systems. The Nation's R&D activity will continue to have the care-management focus that has recently developed in the work of the Agency for Health Care Policy and Research. Learning how to link each member to his or her medical care system, designing the individual links based on specific health status, needs, and aspirations, must be within the capability of the 21st century health care program. We need to mount the R&D effort that will give us the tools.

Human Interaction Sciences

Developing a 21st century health care system will require 21st century organizational tools. The current archaic approach to management won't do it. This R&D effort should bring together management scientists, cultural psychologists, and other behavioral scientists to develop, test, and implement the organizational systems that are analogous to the clinical, clinical information, and care-management systems produced by the other elements of the R&D program.

And finally, we need to develop special research settings to allow for demonstrating, testing, and observing the system implications of the integration of the state of the art into a health care system. I propose the development of a worldwide network of what I call “Experimental, Prototype, Health Care of Tomorrow Sites” (EPHCOTS). We should create health care system subunits, across the United States and in other countries, to form EPHCOTS. Although these sites would deliver care to an identified population, they would also serve as an experimental laboratory for the health care world. The R&D leadership and the system leadership would be closely integrated.

The EPHCOTS would be prototypes in that they would, at any point in time, have in place the most advanced methodologies and technological components. They would be experimental in that each new innovation would be implemented in a carefully controlled experimental approach, to allow the proximate and distal effects to be carefully measured and assessed. I see them as actual medical care systems taking care of 250,000 to 1 million people. They would need to have integrated management and R&D program leadership. An integrated system of specific projects would take place within any one of the sites. Each would have an advanced technology intelligence unit, and together, they could become a part of a worldwide network of strategic alliances for the development of the 21st century health care system. The EPHCOTS would be a place where medical care organization experiments could take place, where the culture would support a meeting place for program innovators, and where medical care system leaders, in the public and private world, could observe the state of the art on-line and in color.

Barrier: We must develop a sufficiently powerful reimbursement method to adequately and appropriately reimburse for the care of Medicare beneficiaries. As I said before, several problems result from the inadequacy of the payment mechanisms under risk contracting, and to a great extent, in the marketplace for younger populations. First, misspecified payment models may result in windfall revenues for some health plans. Specifically, health plans that skim healthier members either through conscious strategies to avoid enrolling sick and frail persons or through the switching selectivity effect (which operates because sicker persons are tied more closely to their physicians) may be overcharging enrollees and purchasers (Billi et al., 1993). Second, health plans may fail because adverse selection is not reflected in their premiums. Efficient health plans for low-risk persons drive out efficient health plans for high-risk persons (Luft, Trauner, and Maerki, 1985). Third, payment models containing implicit cross-subsidization arising from differences in the relative cost structures among health plans may create financial incentives to alter patterns of access and treatment across subgroups of patients.

Efficient and equitable payments to health plans require that both payers and plans have better information on the nature of health plan outputs—namely, the distribution of medical risks of their enrolled populations relative to non-enrolled populations. Health plans need to know whether payments are adequate for the level of risk being carried, and payers need to know whether the rates quoted by health plans are affected by differences in risks among plans. Finally, we need to remove the effects of risk from the contributions paid by consumers so that their choice of health plans can be made on the basis of true efficiency, not risk-skimming.

It is clear to me that all payers should be required to deal with health plans on a risk-adjusted, community-rated basis, using a nationally standardized risk-rating approach. This kind of model would provide all payers with the same information regarding the prices they face for their beneficiaries and would reinforce a single standard of care for all citizens. Plan prices to consumers should reflect only efficiency and benefits. Consumers should be able to include in their choice model for the selection of their health plans an estimate of price relative to benefit value, in addition to other factors such as their co-workers' or neighbors' opinions of various plans or the new plan scorecards that provide satisfaction and quality data.

The array of competitive strategies would be greatly enhanced by a predictive measure that was cheap to use and not gameable and that explained enough of the variance to remove financial incentives for health plans to select patients. Removing the medical risk-selectivity component from variation in premiums would leave consumers facing relative prices reflecting internal plan efficiencies rather than plan enrollment and disenrollment policies (Hornbrook and Goodman, 1991, 1995, 1996; Hornbrook et al., 1996; Hornbrook, 1994; Gruenberg, Kaganova, and Hornbrook, 1996; and van de Ven and van Vliet, 1992). The research in this field is moving very quickly, but the solution in the real world requires more than a correct technical solution. It requires the political will to change basic concepts about how care is financed and reimbursed, and the ultimate repudiation of the FFS methodology (Hornbrook, 1994; Hornbrook and Goodman, 1991; Hornbrook et al., to be published).

Barrier: We must redefine the purpose of clinical care to include the prevention of disease and the maintenance and improvement of function as primary objectives of clinical care. Our health care system has defined too narrowly the scope of the basic clinical services for our aged and other chronically ill citizens. And this is particularly true within Medicare. We define the scope of services available in a health care system within a “medical model.” Even in the area of long-term care (LTC), we restrict coverage to services that are “illness-related” in the strictest sense. This is because the system is disease-oriented, focused more on curing than on maximizing function. Changing the model has a variety of important consequences, one of the most critical of which is the need to provide for the organization and financing of expanded services beyond what are Medicare-covered services.

The question of providing home and community-based long-term services is one of the most critical. The traditional Medicare definition of services, invented and refined during the 1960s as a part of the original construction of the new Medicare house, created a gulf between skilled home care and other services that are needed for the proper care of a patient in the home. This gulf was hard-wired into our care definition when these services were made available to participants of all ages of managed care systems and, to some extent, even in the FFS world. As we move to the 21st century, it is time to rethink the dimensions of our scope of care and to make the next marginal addition to this scope by including home and community-based LTC services.

The legislation that created Medicare permitted posthospital coverage in extended care facilities and services by home health agencies only if the care was “primarily engaged in providing skilled nursing care or related services.” The definition of “related services” has been continually restricted, leaving out essential clinical services, specifically the home and community-based LTC services. The problem is that a patient's need for care is defined by a composite of physical, functional, emotional, social, and medical levels. Unfortunately, the skilled services directed to demonstrable acute needs are frequently not sufficient to care for the aggregate needs of patients. The scope of available services needs to be expanded because of the clear link between skilled services and community-based LTC services that are needed to foster stability for chronic care patients.

Community-based LTC services include those supportive personal care services that are needed to enhance or maintain normal body function, to address emotional comfort, and to assist the patient in independent living. These services are required in a variety of care situations, including convalescence from a specific acute episode, for a medical flare-up of a relatively stable chronic condition, during an end stage illness episode, for respite for informal caregivers, or to care for a very frail and declining patient.

There have been two major national Medicare demonstrations that have proven the potential of providing these kinds of services to an aged population: the PACE (On-Lok) demonstration and the national Social/HMO demonstration. I'm, of course, most familiar with the S/HMO demonstration, which is the most relevant for this discussion anyway, because S/HMOs cover total populations, rather than just the peak of the pyramid, the frail elderly. The S/HMO demonstration was conceived by scholars at the Heller School of Brandeis University. The HCFA-funded demonstration had four service sites, including the Medicare Plus II program in the Northwest Region of Kaiser Permanente. The National S/HMO Research Consortium's Data Center is housed at the Kaiser Permanente Center for Health Research in Portland. The integrated managed care program demonstration currently enrolls about 25,000 Medicare beneficiaries across the sites. Each of the beneficiaries receives the full range of medical care, including community-based LTC services.

Members are regularly assessed for needed services using annual screening questionnaires and data gathered through other methods. These data are very useful for assessing the need for linking skilled and community-based LTC services. Because the S/HMO includes the resources for providing community-based LTC services, it is possible to assess a concurrent need for skilled and LTC services. Although utilization of Medicare skilled care services and community-based LTC services is bound to overlap to some degree, S/HMO data show that this overlap is indeed substantial.

Among S/HMO patients receiving Medicare skilled care, 37 percent were found to concurrently qualify for and receive care from the S/HMO's LTC benefit during their first month of skilled care. On the other hand, new community LTC patients often need and qualify for Medicare skilled services. Also, 37 percent of the community LTC plans for S/HMO members made during their first month of community LTC eligibility contained concurrent authorizations for Medicare-covered skilled services. That is, more than one-third of newly identified LTC patients were eligible for, needed, and received Medicare-covered skilled services during that same month (Leutz, Greenlick, and Capitman, 1994).

But, equally important is the notion that the health care system of the future is going to be judged on the extent to which function is maintained. Clinicians today have not been too successful in incorporating that approach into their basic clinical perception. Services like community-based LTC are still perceived as useful for social objectives, but rather irrelevant to clinical objectives, because these are seen as totally separate domains. When these and similar services, which are critical to maintaining and improving function, are viewed as central to providing care to a population, as are, for example, clinical laboratory services, then we will have moved closer to developing the humanistic health care system of the future.

Challenge

The challenge of moving to humanistic health care is enormous, especially because we have a health care system with such a mature culture. But there is a pressing need for change. Citizens, group purchasers of care, and all segments of government are demanding new kinds of responsiveness to the problems of the system. New technologies are forcing changing clinical perspectives. New requirements for community responsibility are before us. And the systems of the 21st century will be judged on new criteria, especially the extent to which they can change their population's health and functional status and the extent to which they can achieve those outcomes in a cost-effective way.

We can be sure of one thing: The situation is going to get a great deal more problematic than it is now. We will see cataclysmic changes take place in the health care system of the next decade, with more negative than positive changes. But overall, I'm optimistic that the ultimate solution to the American health care dilemma will move us to a more effective, efficient, and humanistic approach to the organization and delivery of care to our populations. But to get to this ultimate objective, we need to totally redefine and reorient our health care perspectives. And we need to be prepared to rebuild a bright new model house, in the ruins of the house that we've virtually destroyed.