INTRODUCTION
We are all concerned about the care of the elderly, in part because of the growing awareness of the demographic realities and the fact that no society in the past has ever faced a similar situation. The rapid aging of the industrialized world calls for a degree of international cooperation that we have rarely seen.
I applaud the efforts of the World Health Organization and its Global Program on Health of the Elderly. The special program for research on aging, which is part of the Global Program, requires far more support than it has had in the past, and far more emphasis on the social, behavioral, and economic aspects of aging. Like the National Institutes of Health in the United States, the emphasis, in my judgement, is too biomedical. There is much to be learned about aging from many disciplines and people in the east and west. The WHO should not limit its vision in its research effort.
We also have much to learn from each other about services for an aging population. The rapidity with which Japan is aging poses even greater challenges than in the Americas, in Asia, and in Africa.
The United States, and other nations, have much to learn from Japan's 10-Year Strategy to Promote Health and Social Services for the Elderly (The Golden Plan) - because of the national commitment to a clearly defined policy and because of for the specifics of the plan, such as the emphasis on rehabilitation.
This conference is a critical step forward in creating a network of institutions working collaboratively to study, evaluate, and, hopefully, contribute to the solution of some of the most difficult social issues that we face.
Let me turn to my assigned topic: Elder care in the United States.
Elder care in the United States is often viewed in isolation from the care of the population under age 65 years because of the partial public financing of acute medical care for the elderly through the federal Medicare program and the partial financing of long-term care through the federal-state Medicaid program. I will say more about both of these programs a little later.
The fact is, however, that care of the elderly cannot be separated from care of individuals under the age of 65 years. In acute care the elderly use the same doctors and hospitals as do those under 65 years. In long-term care, the elderly represent 48.8 percent of those "noninstitutionalized" persons reporting a limitation requiring assistance. Slightly more that half of the disabled (51.2 percent) are under 65 years, including 12.4 percent under the age of 15 years. It is only in the area of nursing home care that the elderly represent the dominant group of users.
In order to describe the care of the elderly in the United States in an appropriate context, it is important at the outset to make a few observations about the U. S. health care system as a whole.
It is widely agreed that health care in the United States is in crisis. The crisis is threefold: cost, access, and quality.
In terms of cost, the United States is now sending ll.6 percent of its gross national product ($600 billion) on health care. In per capita terms, in 1988, the United States spent 38 percent more than Canada, 85 percent more than France, 124 percent more than Japan, and 170 percent more than the United Kingdom.
When expenditures for health care per capita are compared to gross domestic product per capita, two facts are clear: (1) as GDP rises, so does spending for health care; and (2) the United States spends far more than any other country. If the United States spent as much as Canada, it would spend $100 billion less per year than it now spends for health care.
Expenditures are not only high in the United States, but they have been rising more rapidly than in other industrialized countries for more than a decade. These cost increases are driven by the increasing intensity of care, technology, reimbursement policies, medical price inflation well above general inflation, the fragmented system of financing, and consumer demand. There is no set of policies governing public and private sector financing, reimbursement, or capital outlays.
To deal with this crisis in costs, a variety of different approaches have been attempted.
The emphasis at the federal level has been to regulate both hospital and physician payments in the Medicare program. At the state level, there has been an effort to contain increased expenditures for Medicaid through reducing the number of eligible - fewer than 50 percent of those with incomes below 100 percent of poverty are eligible for Medicaid. In some states, less than 15 percent of those with incomes below 100 percent of poverty are eligible for Medicaid .
In the private sector, the emphasis has been on cost-shifting from employers to employees; self-insurance on the part of large employers - to avoid state regulation - has become widespread; a variety of managed-care plans, including health maintenance organizations, has been adopted; and insurers have increasingly dropped community rating of health insurance premiums in favor of experience rating, with the result that the sick cannot get health insurance unless they work for a large employer.
The emphasis on competition and market forces in medical care in the United States contrasts sharply with the policies of universal entitlement, comprehensive benefits, and regulation of payments to providers that characterize most industrialized nations.
The second crisis relates to access to care. In the United States there are 35 million people (14 percent of the population) without health insurance at any one time. A study by the Census Bureau found that over 60 million people (25 percent of the population) were without insurance for varying periods of time during the 27-month study. People without health insurance often do not seek care; when they do, it is often late in the course of illness. Access is deteriorating for many individuals, including many who have health insurance that restricts payments or has high levels of consumer cost sharing.
Finally, the quality of medical care and long-term care is a serious problem. The United States has done more than most countries to begin to examine this issue. The findings of researchers suggest that for many major procedures, 25-30 percent of the care is unnecessary or inappropriate. In some cases, such as carotid endarterectomy, the problem is even more serious.
A recent report by the Special Committee on Aging, United States Senate summed up the problems:
Throughout the last two decades, the structure and delivery of health care have been plagued by perverse incentives, resulting in overutilization of services, inefficiency, and waste.
The quality of long-term care has not been examined systematically, but many investigations of nursing homes have revealed services of a low or questionable quality. Even less is known about the quality of home health services.
The shift in public policy in the United States during the past decade, coupled with the continued increase in physician supply, the erosion of physician autonomy, the changing role of the community hospital, and the loss of ability of hospitals to subsidize the care of the poor from third-party payers has resulted in what Professor Eli Ginzberg of Columbia has called the "destabilization" of health care. To deal with this problem and the general crisis in health care, federal leadership and a clear set of federal policies are needed.
It is within this context of crisis that I will discuss developments with respect to care of the elderly in the United States. I will first review briefly the demographic changes that are occurring and then I will consider issues related to the financing, organization, and delivery of acute and long-term care for the elderly.

1. HEALTH AND WELL BEING OF THE ELDERLY
Let me begin with the health and well being of the elderly. Our colleague at the University of California, San Francisco, Dorothy Rice, has summarized the demographic realities:
Americans are living longer today than ever before. Improvements in living conditions and lifestyles, as well as advances in science, medical technology, and pharmaceutical therapies have meant tremendous reductions in the number of deaths from fatal infections, dramatic gains in life expectancy and a rapid growth in the number of older Americans.
The number of elderly in the United States has increased at a rate more rapid than the rest of the population for most of this century. At the turn of the century, there were 3.1 million elderly in the United States, or 4.0 percent of the population. The number of elderly has now reached 31 million, more than 12 percent of the population. By the year 2030, the number is projected to be 64.3 million, or 20 percent of the population (Fig. 1).

Fig. 1 Number of persons 65 years & over and 85 years & over: United States, 1900-205O

During recent years, the most rapid rate of increase has been among those aged 85 years and older. There are now 18 million aged 65-75 years, l0.3 million aged 75-84 years, and 3.3 million aged 85 years and older. In the 50-year period from 198O to 2030, those aged 85 years and older are projected to be the fastest growing segment of the population - increasing from 4 percent to 14 percent of the elderly.
I will not dwell on the details of the declining mortality rates across the entire lifespan and the changing fertility rates which are contributing to the proportionate increase in the population aged 65 years and older in the United States. Similar developments are occurring in all of the countries represented at this conference.
Professor Carroll Estes has made four important observations about the demographic changes that are affecting all of our societies.
First, the aging society is a phenomenon so sweeping in its import and impact that it will exceed the capacity of any state or community to individually and adequately address the issues that population demographics generate.
Second, declining mortality and increased life expectancy do not automatically equal improved health; that is, longer life does not inevitably mean healthier life.
Third, in addressing the phenomenon of societal aging, it must be remembered that decades of research have consistently demonstrated a strong and persistent relationship between health and socioeconomic status, as measured by longevity, disability, and chronic illness.
Finally, health status in early life is likely to be predictive of health status in old age.
There are several factors related to the aging of the population - in addition to the numbers - that are important in terms of care of the elderly.

2. CHRONIC ILLNESS
The incidence and prevalence of chronic illness increases with age and becomes a major cause of disability requiring medical care. Today, the common association between old age and physical decline in health is attributed primarily to chronic arthritis, heart disease, hearing and vision impairments, and hypertension. Alzheimer's disease, depression, cancer, and alcoholism are also major problems among the elderly.
Many elderly suffer from multiple chronic conditions and disabilities. While only three percent of persons under age 17 years report three or more chronic conditions causing limited activity, this is reported by 16 percent of those aged 75 years or older. As the number of chronic conditions increases - and it increases with age - the number of days of restricted activity and the number of bed-disability days increases.
Forty percent of the population over the age of 65 years suffer limitation of activity. The percentage of persons extremely limited by chronic conditions is:
* 6.2 percent among 45-64 year olds
* 14.4 percent for 65-74 year olds, and
* 33 percent for those 85 and older.
In the aggregate, older persons, particularly women and minorities with low incomes and lower educational levels, have a higher incidence of chronic disease and disability. Rural elderly also report a greater number of days per year of restricted activity. In view of the lower income and educational levels of the rural elderly, this finding is not surprising.
In addition to the burden of illness and disability, income and social factors are important in determining whether an elderly person will be confined to a nursing home. It is estimated that for every nursing home resident, three people of equal functional impairment live in the community. Many functionally impaired elderly can be cared for at home, largely because of services provided by family members, usually a spouse or adult offspring. Widows and widowers are five times more likely to be institutionalized than married persons, and those elderly who never married, are divorced or separated may have up to ten times the rate of institutionalization of married individuals.
Social support networks between elderly persons, relatives, and friends have been found to have a positive effect on patients' mental functioning and serve as a buffer between decline and risk of institutionalization. These findings emphasize the need to develop an adequate base of social support for the elderly through family, friends, and organized community services.

3. THE HEALTH CARE SYSTEM
Governmental policies and programs play a major role in the planning, regulation, and financing of health care. The major function of governmental policies related to health care is to support services in the private sector, particularly those provided by physicians, hospitals, and nursing homes. Government financing is needed to compensate for failures of the marketplace to meet the needs of the elderly, the disabled, and the poor, including the working poor. The system of health and long-term care services in the United States is composed of a large private sector that includes thousands of non-profit institutions; public subsidy or direct financing of the system; multiple federal, state, and local government programs and agencies; and over 238 million consumers threading their way through the maze.
Health workers totaled 8.7 million people in 1988, including approximately 1.63 million registered nurses, 585,000 physicians, 156,000 pharmacists, 140,000 dentists, and more than six million workers in 150 other health service and support categories. More than eight million of these people are engaged directly in providing health care services. Health care now employs 37 of every 1000 Americans and it is expected to be one of the fastest growing sectors of the economy in the 1990s.
There are approximately 6000 hospitals in the country, including nonprofit community hospitals, public hospitals (federal, state, and local), and proprietary hospitals. Community hospital beds per 1000 population declined from 4.5 per 1000 population in 1980 to 4.0 per 1000 in 1988. States vary markedly with the high in the District of Columbia (7.6 per 1000) and the low in Alaska 2.4 per 1000) and Hawaii (2.5 per 1000).
Hospitals discharged 33.6 million patients in 1987, with the bulk from non-profit community hospitals (22.9), followed by state and local public hospitals (5.5), proprietary (3.2), and federal (1.9). The average length of stay was 7.4 days. Thus, the hospital provided more than 248 million days of inpatient care. They also provided more than 300 million out-patient visits.
Over 18,000 nursing homes provide care for more than 1.6 million patients, most of them elderly. Nursing home use by the elderly in 1985 is fairly typical of the pattern - increasing use with increasing age.
There are a variety of different arrangements for ambulatory medical care: solo practitioners and small partnerships, including primary care physicians, specialists, or sub-specialists in office-based private practice; large and small multi-specialty and single-specialty group practices, including health maintenance organizations; community health centers, mental health centers, and public health clinics, usually subsidized by government funds; hospital out-patient clinics; emergency rooms; adult day care centers; and a host of other arrangements. Any of these may serve as a point of entry into the health care system.
The use of health care services increases with age. In 1985, non-institutionalized elderly in the United States saw physicians on an average of 8.3 times per year, in contrast to 6.1 times for those aged 45-64 years. About 84 percent of the elderly had contact with a physician in 1985. Nine out of ten elderly had a regular source of care and eight out of ten saw a single doctor for their care.
Elderly people are more likely to be hospitalized or admitted to a nursing home than the non-elderly. When hospitalized, the elderly remain in the hospital longer than the non-elderly, but their hospital stays are remarkably short when compared to other countries. The average length of stay for persons 65-74 years was about 8.2 days in 1987 and it was 9.1 days for those aged 85 years and older.
It is not surprising that the elderly who suffer from chronic and disabling conditions are more likely to need medical and long-term care than those without disability. An elderly person with chronic activity limitation had 8.7 visits to a physician in 1985, in contrast to 4.3 visits for persons with no activity limitation. They had 41 .2 days of inpatient hospital care per 100 persons with activity limitation, in contrast to 14.8 for those with no limitation of activity. The 46 percent of the elderly who were activity limited accounted for 63 percent of physician contacts, 71 percent of hospitalizations, and 82 percent of all the days that the elderly spent in bed because of medical problems.

4. FINANCING OF HEALTH AND LONGTERM CARE SERVICES FOR THE ELDERLY
Current health care financing and social service programs for the elderly involve all levels of government, as well as the private sector. At the federal level, Medicare provides health insurance coverage of hospital and physician services for most individuals aged 65 and over, for disabled persons under age 65 who meet certain criteria, and for those suffering from end-stage renal disease. There are 30 million elderly and 3 million disabled eligible beneficiaries on Medicare. Medicare has two separate parts - a social security program of hospital insurance and a voluntary program of insurance for physicians services, funded by general revenues (75 percent) and premiums paid by beneficiaries (25 percent). In addition, there are copayments and deductibles that increase the elderly's out-of-pocket costs for medical care.
Of the total Medicare expenditures for 1989 ($94 billion), 55 percent was spent on hospital services and 27 percent on physician services. Medicare expenditures were negligible in covering nursing homes (less than 1 percent) and less than 3 percent in covering home health care. Although it still constitutes a small share of Medicare expenditures, home health care has been one of the most rapidly growing services in the Medicare program. Expenditures for home health services totaled $2.5 billion in 1989. Those over age 85 years are more than four times as likely as those 65-69 years to use home health services.
Medicare does not cover long-term care, out-of-institution drugs, dental care, eyeglasses, hearing aids, and other important health services for the elderly.
Medicaid is a federal-state matching entitlement program providing medical assistance for low-income persons who are aged, blind, disabled, or members of families with dependent children and certain other pregnant women and children.
An elderly person becomes eligible for Medicaid if he/she meets very stringent income eligibility requirements. Medicaid is important for the elderly for two reasons: (1) states are now required to "buy into" Medicare for the elderly whose incomes are below the poverty line; and (2) Medicaid is a major source of payment for nursing home care for the elderly. In 1986, $13 billion of Medicaid's $44 billion in expenditures were for nursing home care. Medicaid pays for nursing home care for the elderly who have met the government's "spend down" requirements. Basically, an individual must become impoverished before Medicaid will cover the costs of nursing home care.
Personal health care expenditures by the aged contrast sharply with those aged 19-64 years and those under 19 years, particularly because of the source of funds and higher proportion of expenditures for institutional care (Fig. 2).

Fig. 2 Personal health care expenditures by age and source of funds, 1987

The distribution of per capita health care expenditures for the aged by source of funds from 1965-1984 illustrates the changes since the enactment of Medicare (Fig. 3).

Fig. 3 Distribution of per capita health expenditures for the aged by source of funds, 1965-1984

It is often surprising to those not familiar with health care in the United States that the elderly must spend a significant portion of their income for out-of-pocket costs. The trends since 1966 show that out-of-pocket costs are rising (Fig. 4).

Fig. 4 Elderly out-of-pocket health care costs as a percent of elderly mean income

Out-of-pocket costs as a percent of per capita income illustrate the problem even more clearly (Fig. 5).

Fig. 5 Elderly out-of-pocket costs as a percentage of per capita income 1966-1987

Because the federal government plays a dominant role in the payment of hospital and physician services for the elderly, its reimbursement policies have been important in driving costs upward. A critical policy objective of Medicare, when enacted in 1965, was to assure access to "mainstream" medical care for the elderly. That objective has largely been achieved, but the dollar cost has been far higher than originally anticipated.
To guarantee provider acceptance, Congress required that payment to hospitals would be made on the basis of their costs, determined after the care was provided. Physicians were paid on the basis of their "usual, customary, and reasonable" (UCR) charges. These policies held for almost twenty years, in spite of steadily rising Medicare expenditures that exceeded increases in the consumer price index and the gross national product by a wide margin.
One immediate conse1quence of the implementation of Medicare, which had a significant impact on costs, was the dramatic increase in the use of short-stay hospital services by the elderly. There was, initially, an increase in both the rate of admission and the length of stay. Short-stay hospital admissions increased by more than 57 percent between 1965 and 1986. Since 1985, admissions declined slightly, only to rise again in 1987 and 1988. In 1986, l0.7 million elderly patients were discharged from short-stay hospitalizations, comprising over 31 percent of all short-stay hospital stays.
Surgical rates also increased dramatically after the enactment of Medicare. In 1965 there were 6,554 operations for every 100,000 persons aged 65 and older; in 1975 there were 15,483 operations for every 100,000 persons aged 65 and older-an increase of over 100 percent.
Although the number of hospital admissions and surgical procedures per 100,000 elderly rose dramatically, the use of physicians' services outside the hospital by the elderly changed relatively little. There was an increase in the use of physicians' services by the poor elderly and a decrease by the non-poor, with the overall average remaining close to 6.5 visits per year for 1965 through 1978.
The increased utilization of hospital services by the elderly and the gradually increasing number of elderly were factors affecting the rapid increase in Medicare expenditures. These factors were relatively minor, however, when compared to the impact of general inflation and the additional price increases by hospitals and physicians, and the increased complexity of care provided.
Because of the rapid rise in Medicare expenditures, Congress enacted a major reform in hospital payments in 1983. Instead of retroactive payments based on costs. Congress required prospective payments based on average costs per case determined on the basis of diagnosis groups. After the enactment of the prospective payment system, hospital admissions, length of stay, and inpatient expenditures increases slowed and hospital outpatient visits increased rapidly. Much of this increase was related to continued technological advances and increases in imaging, clinical laboratory, and other diagnostic procedures performed by the hospitals.
After enactment of the hospital payment reform, Congress turned its attention to physician payment. After a series of stop-gap measures in the mid-1980s, Congress enacted a comprehensive set of reforms in 1989, based on the recommendations of the Physician Payment Review Commission. The changes included a Medicare fee schedule, based primarily on resource costs, with a relative value scale reflecting these costs; a limit on balance billing (charges above the Medicare fee schedule) by physicians; and finally, Congress established volume performance standards to limit the rate of increase of Medicare expenditures. Increased payments to physicians in future years will depend on the extent to which the volume performance standards are not exceeded. If volume increases above the standard, the fee increases will be reduced.

5. LONGTERM CARE
Although Medicare has been the most important source of payments for hospitals and physicians caring for the elderly since 1965, it is limited in scope of benefits and reimbursement policies required to meet the needs of the chronically ill and disabled elderly. The cost of care, including a full range of services addressing the health, personal, and social needs of the elderly, is borne by both the private (the elderly themselves and their families) and the public sectors, including Supplemental Security Income (SSD), Medicare, and Medicaid. The Social Services Block Grant and Older Americans Act programs, which support nonprofit voluntary agencies at the local level, are a vital part of the long-term-care picture in the community, but they garner considerably less public resources than the strictly medically defined long-term-care services.
State policy, because of the decentralization of policies relative to Medicaid and social services, has become a major factor determining the scope, structure, and availability of long-term care. The result has been a variety of approaches, in part dependent on the fiscal condition of the states. The recession of 1981-1982 and the subsequent economic recovery have affected states quite differently; as a result, the resources available for public programs at the state level varies markedly.
Three developments are of great importance in the organization of health and long-term care services if the needs of the chronically ill and disabled elderly are to be met effectively:
(1) the need to better link and integrate acute-care and long-term-care services in the community;
(2) the need to strengthen ambulatory care, community-based services (e.g., adult day care, congregate meals, senior centers), and in-home services to reduce the emphasis on inpatient care in hospitals and nursing homes; and
(3) the need to recognize the benefits (as well as the limits) and potential roles of family members and other sources of social support, including the full spectrum of non-profit community agencies serving the elderly.
There have been many positive professional and community efforts (largely demonstration projects) directed toward a more comprehensive and humane long-term-care policy. Koff (1982) envisions a long-term-care system in which institution-based and community-based services are integrated and appropriately utilized in a "continuum of care." In contrast to this ideal, there is generally no systematic link between the myriad health and social services that have emerged as alternatives to institutionalization; nor is there a systematic link between the acute and chronic care systems.
In addition to the changes in financing and organization that are needed, effective care of the elderly requires changes at the clinical level. Clinical care, whether provided by physicians, nurses, dentists or other health professionals, must take account not only of the biological and medical factors that contribute to morbidity among the elderly, but also the social, behavioral, and economic factors as well. Physicians can no longer isolate themselves in an office-based practice or in a hospital and expect to fully meet the needs of elderly patients with multiple social and emotional, as well as medical, needs. Linkages are also needed between levels of care - primary, secondary, and tertiary - as well as between acute and long-term care. One approach to the better integration of acute and long-term care could be through the expansion of health maintenance organizations (HMOs). HMOs could encompass the full spectrum of social and health (and long-term care) needs, including home care, ambulatory care (including adult day care, congregate meals), and nursing home care, on a prepaid capitation basis to control costs. Called social health maintenance organizations (SHMOs), these new types of prepaid plans not only place providers at risk and change incentives (as do HMOs), they also have the potential for redesigning both the delivery and financing of long-term care.
Another linkage model that has emerged focuses on those in greatest need of medical and social services and provides a comprehensive range of services, primarily in the home and in the community. An example of such a model of comprehensive care for the very frail, sick, and disabled elderly is On Lok in San Francisco. Here, medical, nursing and social services; physical, occupational, and recreational therapy; counseling; congregate meals, housing; transportation; respite care; and in-home services are provided by a single agency. For a patient population of over 300, all of whom were eligible for nursing home admission, it has been possible to meet the patients' needs in a humane, compassionate, and cost-effective manner. Physicians and all other health professionals involved are team players, adjusting their respective roles to the patients' needs.
The emerging hospital model of acute and long-term care involves a "vertical" integration of traditional hospital inpatient services, with ambulatory care, home care, and nursing home care. Whether this approach will further fragment community-based care and whether it is the most cost-effective use of community resources remain to be seen.
These changes in health and long-term care will not be possible without substantial changes in medical, nursing, pharmacy, and dental education. Today's entering medical and nursing students particularly, as well as many other health professionals, will spend an increasing part of their professional lives dealing with chronic illness and functional disability in their elderly patients. To do their jobs well, their education and training will need to place more emphasis on chronic diseases, aging, management of chronic disability, prevention and rehabilitation, and social and behavioral factors in health and disease.
Research is needed to develop an understanding of the limits and potential of an enlightened health and aging policy; it is also needed to disentangle the effects of disease from those of psychological or biological aging; and from the effects of social, economic, and political factors and forces. As has been discussed, the achievement of health is dependent upon a complex web of interactions among biological, behavioral, sociocultural, and environmental factors. It is possible and important to construct policies that recognize the determinants of health and that foster healthy aging.

6. SUMMARY
This has been a brief overview of elderly care in the United States. The problem of providing adequate and appropriate care for the elderly was viewed within the context of the crisis in health care-in the United States. It is a crisis of cost, access, and quality. The health care system has become "destabilized" in the past decade under the pressure of growing physician supply; a emphasis on market forces and regulation of physicians by private payers and government; pricing of insurance premiums on the basis of experience rating rather than community rating; unrestrained proliferation of technology; inflationary reimbursement policies; the changing role of community hospitals; and the loss of the hospitals' ability to subsidize care of the poor from other third-party payers.
The federal government, particularly the Congress, have taken the lead in the United States in the development of policies within the Medicare program to deal with the health care crisis as it affects the elderly, particularly designing new prospective payment policies for the hospitals and a fee schedule for physicians that will go into effect on January 1, 1992. Problems persist, however, because of the lack of integration of federal Medicare policies, state Medicaid policies, and private sector policies.
The United States has more to learn from Japan and other nations as it strives to develop sound policies and programs to meet the needs of its aging population.
The situation is critical from all viewpoints-payers, providers, and the public. To serve the needs of the nation's population, including the elderly, a long-range and multigenerational perspective that will achieve an equitable as well as an efficient use of the nation's resources is needed. An intergenerational agenda is needed on health policies. National policies must provide the framework for both public and private action in order to continue to improve the health of the nation at a price we can all afford.