The relationship between provision of community-based and home services on institutionalization is complex and depends in part on factors such as the level of cognitive impairment of the older person. Home health services are often used as a substitute for transient nursing home admissions but not for permanent nursing home admissions. While there is popular advocacy for respite services, research to date offers little support for their effectiveness in reducing costs or enhancing caregiver wellbeing. Such studies caution that critical factors affecting respite care (e.g., financing, eligibility, and staffing) are poorly understood.

The NIA proposes a funding level of about $1.5 million for research on this topic in FY 1992 as part of its initiative on long-term care.

PHYSICAL FRAILTY

Mr. Stokes: In your statement prepared for today, you indicated that costs of physical frailty range from $54 billion to $80 billion annually. Can you share with us some of the activities NIA is supporting in this area?

Dr. Williams: As noted in our report on physical frailty requested by the Committee, NIA is conducting a major research program to reduce physical frailty. The program includes two sets of clinical trials: FICSIT, Frailty and Injuries: Cooperative Studies of Intervention Techniques, and STOP/IT, Sites Testing Osteoporosis Prevention/Intervention Techniques. FICSIT has begun to test techniques to increase strength and prevent disabling falls. STOP/IT, scheduled to start in September 1991, will focus on reduction of osteoporosis in the hip, the most critical risk factor for hip fracture. Also in FY 1991, NIA will initiate research on special problems of frailty among minorities. In addition, NIA supports numerous basic and clinical studies to learn more about the causes of frailty as basis for developing better means of prevention and treatment.

The Claude D. Pepper Older Americans Independence Centers are a vital component of NIA's research program to reduce frailty. Testing the efficacy of interventions to reduce dependence is a required component of these Centers. Other key aspects of the Centers include career development of new investigators to test further interventions to increase independence, studies of factors affecting older persons' willingness and ability to participate in new interventions, studies of cost effectiveness of new treatments, and information dissemination to translate research results of the Centers into health care practice.

NIA has

There is intense research interest in these Centers. received 41 letters of intent to submit proposals. Potential interventions for testing include techniques to increase muscle strength, mobility, balance and/or endurance in frail older persons; prevention of disabling falls; prevention of institutionalization by improved identification and treatment of health problems in frail older persons; reducing disability from cardiovascular disease through better drug treatment; preventing

dehydration, malnutrition, and their debilitating consequences (common problems among frail older persons); and better treatments for urinary incontinence.

Recent findings are especially promising. Increasingly, we are learning that people are never too old to prevent or reduce frailty. Two NIA-supported studies recently published illustrate this: one showed that strength training improved the ability of frail 90-year old nursing home residents to walk; the other proved that a simple behavioral technique could dramatically reduce urinary incontinence among older women. The recent demonstration of the potential of growth hormone to reverse debilitating ageassociated changes has opened up the prospect of a fundamentally new approach to reducing and preventing frailty.

Mr. Stokes: How much is NIA allocating for research in this area? What are some of the most promising research opportunities in this area, and what additional resources would be required to support them?

Dr. Williams: In FY 1991, NIA's expected allocation for research on physical frailty is $41.9 million. Under the FY 1992 President's Budget, NIA will allocate approximately $43.3 million for research on physical frailty. Among the most promising new research opportunities are explorations of the role of trophic factors such as growth hormone in preventing frailty, reduction of disability from cardiovascular disease and osteoarthritis, and prevention of automobile driving disabilities. In addition, an expansion of the number of Claude Pepper Older Americans Independence Centers to the ten authorized by Congress would permit development and testing of numerous useful interventions. These lines of research could permit inauguration in the mid1990's of a large-scale demonstration of the impact of frailtyreducing interventions within the American health care system.

To initiate these new lines of research in FY 1992, an increase of $34.3 million over the President's Budget would be required, divided as follows:

Claude Pepper Older Americans Independence Centers

(increase of $11 million over President's Budget)

The annual cost for fully funding each Center is $1.5 million. Resources available to NIA in FY 1991 permit allocation of $3.9 million to begin this program. This will allow partial funding of up to four Centers. The President's FY 1992 budget does not include funding for new Pepper Centers. As noted, Congress authorized establishment of at least ten Older Americans Independence Centers, and at least ten institutions have the ability to establish high quality Centers. An increase of $11.0 million in FY 1992 would therefore be required to reach a total FY 1992 support of $15 million, the amount needed to fully fund ten Pepper Centers.

Other Research on Physical Frailty

(increase of $23.3 million over President's Budget)

As previously stated, to support studies of all promising treatments in NIA's STOP/IT clinical trials of osteoporosis treatments, which will begin in FY 1991, would require additional resources. The above figure would support six additional STOP/IT trials plus related research on osteoporosis, studies of use of growth hormone and other trophic factors against debilitating conditions, reduction of disability from cardiovascular disease and osteoarthritis, and reduction of automobile driving

disabilities.

Mr. Stokes: The justification notes that frailty poses special problems in different minority groups (page 183). What are some of the differences between minority groups?

Dr. Williams: Genetic, environmental, and cultural factors contribute to differences among minority groups both in disorders responsible for frailty and their means of coping with it. For example, data from a study in Los Angeles County found that Blacks, but not Hispanics, had elevated rates of strokes compared to Whites. These differences may not necessarily apply to Black and Hispanic populations in other areas. Blacks apparently have higher rates of hypertension, a major risk factor for stroke, than most or all Hispanic groups. Hispanics appear to have particularly severe disabling long-term consequences of diabetes compared to most other minorities and the national average. addition, the proportion of Hispanic elderly having difficulty with activities of daily living is two to three times the national average. Blacks apparently suffer from mobility problems due to arthritis at least as frequently as the national average; incidence of hip fractures among Blacks is considerably less than the national average.

In

For many minority older people, problems of physical frailty are intensified by poverty and poor access to rehabilitative and supportive services and medical care. For example, the poverty rate for elderly Hispanics is nearly double that for all elderly, with rural Hispanic women the most impoverished group of all. Studies of older Blacks show they are more likely to experience disability earlier than Whites with the greatest disparity in prevalence in disabilities occurring between Blacks and Whites, ages 55-65. NIA-supported researchers reported that AsianAmericans as a whole have the largest proportion of foreign-born persons. As younger Asians arrange for older relatives to immigrate, language barriers hamper access to care for older Asians seeking assistance in their own language or dialect. social, economic, and language difficulties also hamper the conduct in clinical and epidemiological studies of frailty in older minorities. Recent NIA initiatives to recruit and retain minority researchers is part of the Institute's effort to increase research emphasis on this underserved older population.

These

Better knowledge of genetic, environmental, and nutritional factors affecting the severity and progression of frailty in different minority groups would be useful in the design of intervention and prevention programs. To achieve this, NIA has solicited research proposals on the above issues, including:

Prevalence and incidence of conditions responsible for impaired physical functioning in specific older minority populations.

O Genetic, environmental, and nutritional factors affecting severity and progression of these conditions in specific minority populations.

The relationship of specific functional impairments to severity of specific degenerative conditions in minority populations.

O Efficacy in specific minority groups of interventions to prevent and/or reduce physical frailty.

Beginning in FY 1991, $1.2 million per year for three years has been set aside to support these projects, which will begin in September 1991.

HEALTH AND RETIREMENT SURVEY

Mr. Stokes: Early retirement is now costing the nation an estimated $15-20 billion in lost earnings, lost taxes, reduction in savings rates, and other costs every year. I understand the NIA is initiating a Health and Retirement Survey. What kinds of information will this survey provide and how will it relate to public policy? What is its current status?

Dr. Williams: The survey will include questions regarding health condition and disability status, work history, economic situation, pension policies, employer retirement incentives, family structure, and family responsibilities. Moreover, much

needed information focusing on the causes and consequences of retirement for women and minorities will be collected as part of this study. The resulting data will allow researchers to model the retirement decision-making process.

As

This study will provide policy-makers and researchers with important and timely information concerning the role of key factors affecting the causes and consequences of retirement. findings emerge, they will provide the necessary information base for federal and private sector retirement policies. They will assist in evaluating current policies aimed at allowing those who wish to continue working to have the maximum opportunity to do so. Retirement decisions can affect activities and financial resources for up to three decades beyond the date of retirement. The study will provide data on how people are making these decisions today in light of increased longevity and risk for long term disability.

In regard to the status of the survey, a cooperative agreement to support the Health and Retirement Survey was awarded in September, 1990, to the University of Michigan. The first year of the study includes an intensive planning phase, and the study is now five months into that phase. The survey will likely focus on individuals who are initially between the ages of 51 and 61. About 8,000 households will be included in the survey, yielding a sample of about 14,000 persons. Blacks will be oversampled, and plans are underway to oversample Hispanic persons also. The first

interviews for this study are planned for April 1992. Follow-up interviews are planned at a two-year interval.

Consideration is being given to the advisability of a special one-year follow-up that would include a physical exam that would provide direct physical measures including blood samples. Such measures might be needed in order to ensure the long-term epidemiologic viability of the study. A decision has also been made to collect, provided funds are available, information on the availability of employer provided health insurance including postretirement insurance. During the planning phase it has become clear that employer provided health insurance (or its absence) is an increasingly major factor in the retirement decision.

Mr. Stokes:

OLDEST OLD

It is my understanding that while the Census labels this group of those 85 and older as the fastest growing population age group in the country, the Bureau's estimates might be extremely conservative (page 188). Your justification notes that "the growth of this age group has the makings of a slowly but inexorably emerging national crisis." Why does the NIA classify the Census data as "extremely conservative" in this area? What are NIA estimates?

Dr. Williams: It is the Census population projections of the oldest old population and not their current estimates that might be considered extremely conservative. The projections developed by the Social Security Administration and the Bureau of the Census over the last thirty years have underestimated the size of the oldest old population, due in part to the unexpected decrease in mortality among persons age 65 and older. The Census Bureau projects in their middle estimates that there will be about 12 million people aged 85 and over by year 2040, while their high estimate (assuming lower mortality levels) forecasts about 18 million.

Those who will become the oldest old over the next fifty years are already over age 35 and so are not subject to the hazards of infancy, childhood and early adulthood. Population projections anticipating the size of the oldest old population are dependent upon assumptions about the changing force of mortality at advanced old age. Projections continue to vary widely depending on the mortality assumptions employed. Mortality rates at extreme ages declined very rapidly during the 1970s and early 1980s, leveled off during the mid-1980s, and have begun to decline again over the most recent period.

A number of researchers have developed forecasts that far exceed the Census Bureau's high range projections for the middle of the next century. One of these projections done by an NIA intramural scientist assumes a constant two percent rate of decline in mortality and projects about 23 million. Another set of projections carried out by an NIA grantee, using a modelling or simulation method, has forecast that healthier life styles, such as greatly reduced smoking, hypertension, cholesterol, etc., could plausibly result in some 40-45 million people over age 85 in the year 2040.