National Academies Press: OpenBook

The Aging Population in the Twenty-First Century: Statistics for Health Policy (1988)

Chapter: 3 Health Status and Quality of Life

« Previous: 2 Social, Economic, and Demographic Changes Among the Elderly
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 65
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 66
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 67
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 68
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 69
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 70
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 71
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 72
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 73
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 74
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 75
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 76
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 77
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 78
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 79
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 80
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 81
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 82
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 83
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 84
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 85
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 86
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 87
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 88
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 89
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 90
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 91
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 92
Suggested Citation:"3 Health Status and Quality of Life." National Research Council. 1988. The Aging Population in the Twenty-First Century: Statistics for Health Policy. Washington, DC: The National Academies Press. doi: 10.17226/737.
×
Page 93

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

3 Health Status and Quality of Life As individuals age, they are at risk for diseases and disabling con- ditions, use more medical care services, and incur medical expenses. While there is no doubt that age is a predictor of morbidity and mor- taTity, its predictive value is limited. The health status of the elderly is better than generally assumed, varies remarkably among individu- als, and is changing as successive cohorts progressively challenge the definition of old age. One major misconception in the health care field is that the el- derly are a homogeneous group of frail individuals progressing rapidly toward needs for long-term care. The elderly actually are a very het- erogeneous group. It has been noted that as individuals age, they become less like each other (Rowe, 1985~. From a physiological per- spective, differences between individuals characteristically increase with advancing age in those factors that change with age, such as blood glucose level and blood pressure. From a clinical perspective, specific subgroups of elderly individuals can be identified, including the 5 percent who at any one time are residing in long-term care facilities and the larger portion who have major functional declines. The marked effect of age itself on disability, morbidity, and mortality has led many workers in the field to divide the overall elderly popula- tion into at least two groups, a young-old population and an old-old population, which is characterized by frailty and marked increases in the need for acute and long-term care services (Besdine, 1982~. These clear age-related differences suggest the value of collecting dif- ferent types of data in different age subsets. The pane] has chosen 65

66 AGING POPULATION IN THE TWENTY-FIRST CENTURY to use three age subsets: young-old (age 65-74), old (age 75-84), and oldest-old (age 85 and older). The health care of elderly persons, perhaps more than any other age group, IS influenced by the social support system available to them (Brady, 1981~. The network of current and potential informal supports, such as family or friends, has an important role in modu- lating the clinical impact of underlying disease and is often the major determinant in decisions to institutionalize elderly people. For every impaired elderly person in a nursing home, there are approximately two equally impaired elderly people living in the community who of- ten can remain there by virtue of the critical role of informal support systems, which provide approximately 80 percent of their Tong-term care (Doty, 1986~. In choosing which data need to be collected and how they should be analyzed, it is important to recognize that the needs of the elderly differ from those of younger individuals, not only from a quantitative perspective (i.e., the elderly use more health care services), but also qualitatively. "Just as children are not merely young versions of adults, the elderly are not simply old adults. They require special approaches" (Rowe, 1985:827) and their health care needs reflect a complex interaction of the physiologic changes with age, their psychosocial concomitants, and the various pathologic processes that occur with advancing frequency in senescence. Although health status per se is not a policy issue, policy analysts need to be able to detect trends and to forecast changes among the elderly in their health status and utilization of services. Development of such trend data requires a stable program of periodic surveys of the health status of the elderly population. This chapter reviews and summarizes what is known about the elderly and the aging process in the later years with respect to health status, functional status, and quality of life and their determinants. The major sources of statistics relevant to these topics are reviewed and changes are recommended that will augment the available in- formation on the health status of the elderly, provide information to understand the health needs of the elderly, provide a dynamic description of aging as a process, and lead to improved measures of functional status of the active elderly that will help trace the transi- tions from their well state to states of disability as well as measure their need for assistance. The chapter concludes with a review of the relationship between health and functional status and the quality

HEALTH STATUS AND QUALITY OF LIFE 67 of life experienced by the elderly and discusses some of the compo- nents of quality of life with emphasis on the relationship between productive activity and quality of life. HEALTH STATUS Characteristics of Health Status and Data Successful Aging For too long, gerontological research has focused on the Tosses in function that occur with advancing age. We have lost sight of the fact that most elderly individuals function very well and report their health as good or excellent. Despite the physiologic Tosses and the psychosocial stresses often associated with advanced age, many elderly individuals have the vitality and resilience to function at a high level. Data collection should take into account the positive aspects of aging rather than purely functional decline, disease, and mortality. Similarly, there needs to be increasing recognition of the capacity of individuals to Improve over time; thus, data sets should be conceptualized with a view to following individuals as they increase in their functioning, rather than only to their entering a pathway of progressive loss of function leading to disability and ultimate mortality. Longevity Extension and Compression of Morbidity A major policy issue relates to the relationship between changes in the mortality experience of the elderly population and coincident changes in the underlying morbidity and disability experiences. In 1900, a man who reached the age of 65 could expect to live 11 more years, and a woman age 65 could expect to live 12 more years. By 1984, men turning 65 could expect to live 14.5 more years, and women turning 65 could expect to live 18.7 more years (National Center for Health Statistics, 1986c). This Longevity revolution" has even affected the very old, as the past two decades have brought dramatic reductions in mortality rates among those over 80, both men and women. The important issue, however, is the quality of the additional years of life for the elderly. Are the additional years ones of vigor and independence, or of frailty and dependence? Will future in- creases in longevity be associated with prolongation of dependency

68 AGING POPULATION IN THE TWENTY-FIRST CENTURY or will active life expectancy increase (~compression of morbidity") as health promotion and disease prevention strategies become in- creasingly effective (Rowe, 19853? "The initial claim that as mortality declines morbidity will also decline has recently been challenged by studies suggesting that the increased life span of the old-old is not accompanied by decreased morbidity and may actually result in more dramatic increases in the need for health care services" (Rowe, 1985:828~. While federal statistics have documented the changes in life epollutantsxpectancy during the twentieth century, there are no national data showing whether the period of vitality has also changed. Thus, it is not known whether either the actual number or the proportion of people's later years that are spent in a healthy state has increased, remained the same, or decreased during the period when the life span has increased. Chapter 4 discusses this issue in detail. Transitions Tom Morbidity to Disability and Mortality The World Health Organization has proposed a model (Man- ton, 1986a) that describes the linkages between mortality, disability' and morbidity (discussed in detail in Chapter 4~. While it is clear that the overall mortality experience of the elderly population has an underlying curve of morbidity experience in which individuals accumulate diseases and losses in specific capabilities, the specific interactions between the development of diseases and the subsequent development of disability have not been elucidated. It is particularly important to recognize that many different pathologic processes may result in, or contribute to, identical functional impairments. For any particular person, several coincident pathologic processes interact in a complex fashion to result in disability. This interaction is often strongly influenced by other factors, particularly in the psychosocial sphere. The marked variability in health status among the elderly and the uncertain nature of the link between the presence of pathologi- cal processes, their functional consequences, and eventual mortality underlie the importance of an approach to the elderly that permits analysis of the transitions that individuals make from one functional status or health state to another. It Is clearly inadequate to fo- cus only on mortality, hospitalization, or institutionalization as end points and increasingly important to detect changes in function when

HEALTH STATUS AND QUALITY OF LIFE 69 they occur and to develop sensitive measures of the severity of dis- ease processes. It Is particularly important in studies of transitions in health status to maintain a perspective that permits detection of improvements over time, rather than just losses in functions. This approach may be more difficult in elderly patients than in their younger counterparts because there appears to be a very short recall period for health interactions in the elderly. In addition, self-reported data regarding some functional status measures, such as those referred to as instrumental activities of daily living, may not be adequate. Studies should be carried out to clarify whether the inaccuracy of such self-report data is a function of age or age- associated diseases, such as those that impair cognitive function. The Need for Acute and Chronic Care Over the past several years, the dramatic increase in Medicaid spending for nursing home care has led to a progressive interest in the long-term care needs of the elderly and to finding alternatives to high-cost institutionalization. The critical importance of acute care in the elderly appears to have been underemphasized: more than 40 percent of the health care expenditures of the elderly are for acute care (Waldo and Lazenby, 1984~. In addition, the common practice of identifying the type of care given by the site in which it is given is increasingly misleading. On one hand, many elderly individuals in acute care facilities receive chronic care. Some are admitted to an acute care hospital in order to gain access to a long-term care facility, and some are admitted to a hospital because of a worsening of a chronic condition. On the other hand, with the implementation of prospective payment and diagnosis-related groups under Medicare, there is a tendency for acute care hospitals to discharge patients to Tong-term care facilities earlier than they did previously. Many nursing homes are now faced with providing acute care, especially acute postoperative care, in facilities not designed for this care and by staffs who are often inadequately trained. In tracking patients' progress through the health care system, it is bemoaning increasingly important to know not only the locus in which care is provided, but also the specific nature of the care. Determinants of Disease In Old Age With advancing age, the general tendency is a shift from acute to chronic diseases. For example, hypertension and coronary heart

70 AGING POPULATION IN THE TWENTY-FIRST CENTURY disease increase with age, as does the risk of cancer, autoimmune dis- eases such as pernicious enema and Addison's disease, and diseases of the bones and joints such as arthritis and osteoporosis. The reasons for such increases are not fully understood. Increases in specific disease categories may reflect increased vulnerability with age, in turn caused by a wide range of age-related phenomena, from reduction of the immune functions to Tosses in physical agility. But age-related increases in pathology do not necessarily imply increased inherent vulnerability; they can also result from cumulative expo- sure to environmental pollutants or from reduced resources, both economic and interpersonal. Per capita income decreases with age, especially after work and retirement, and the likelihood of personal loss through bereavement increases, with possible concomitant re- duction in sources of health-promotive assistance and support. To the risk-enhancing aspects of old age must be added charac- teristics or capabilities of the elderly that may decrease risk. Each successive age group consists by definition of survivors, so there is an inevitable selection factor at work. Separate from the constitutional or genetic factors in such selection are the effects of learning. With increasing age people may be differentially successful in learning to recognize their own limits, avoid various environmental risk factors (e.g., quit smoking), or modify the severity of the risk (e.g., smoke less). Such learning then becomes a counterforce to the age-related factors that increase risk of disease and death. Current understanding of these complexities is fragmentary. For example, some demographic characteristics are known to be associ- ated with longevity (i.e., negatively related to standardized mortality ratios), but the causal pathways or the reasons that these relation- ships become weaker with age is not known. Such reductions in the older age groups have been observed for education, socioeco- nomic level of residence, race, and marital status. Some predictive relationships (e.g., the effect of high blood pressure on mortality and morbidity and the effect of supportive social ties on the risk of hospitalization), however, persist throughout the entire age range. And some relationships (e.g., the effect of marital dissatisfaction on coronary heart disease) seem to increase with age.

HEALTH STATUS AND QUALITY OF LIFE Special Characteristics of niness Behavior in the Elderly* Underreporting of Health Conditions 71 An important characteristic underlying functional impairment in the elderly is the failure of many persons to seek assistance despite the fact that both clear-cut illnesses and abnormalities are present. Studies in several countries with varying health care systems indicate that symptoms of serious and treatable diseases often go unreported (Rowe, 1983; Anderson, 1966~. Health problems reported by frail elderly persons are thus frequently only the tip of the iceberg of treatable conditions. As Besdine (1982) pointed out, nonreporting of symptoms of underlying disease in elderly persons is an especially dangerous phenomenon when coupled with the passive structure of American health care delivery, which lacks prevention-oriented or early detection efforts. He notes that aged persons, burdened by society's and their own views of functional loss with aging, cannot be relied on to initiate appropriate health care for themselves, especially early in the course of an illness. Multiple Pathology The coexistence of several diseases has a profoundly negative in- fluence on health and functional independence in the elderly, and the number of pathologic conditions in a person is strongly related to age. Elderly persons who live in the community have on the average three to four important disabilities (Anderson, 1966), and the hospital- ized elderly have evidence of six pathologic conditions (Wilson et al., 1962~. The entire array of diseases present in an individual patient must be considered as models are developed to describe pathways to dependency and as treatment plans are developed. Multiple pathol- ogy has obvious implications for health interview surveys, provider surveys, hospital discharge records, and the vital statistics on cause of death. Atypical or Altered Presentation of Disease A fundamental principle of geriatric medicine is that many dis- eases have signs and symptoms in the elderly that differ from those *This section closely parallels the section on "Illness Behavior in the Elderly" in Rowe (1985:830~.

72 AGING POPULATION IN THE TWENTY-FIRST CENTURY in their younger counterparts. These alterations can take two major forms. First, specific symptoms characteristic of a disease in middle age may be replaced by other symptoms in old age. For instance, in acute myocardial infarction, some studies have suggested that el- derly persons are less likely than younger adults to have chest pain. But acute myocardial infarction is not "silent" in older persons, who have instead a variety of other acute signs and symptoms. The sec- ond difference is that elderly persons may have nonspecific signs and symptoms, such as confusion, weakness, weight loss, or "failure to thrive" instead of specific symptoms indicating the organ or system affected. HEALTH DATA Major Sources of Health Statistics The United States is fortunate in having well-developed systems of data collection for demographic, social, and economic character- istics of the population and for its rates of morbidity and mortality, health expenditures, and utilization. These data collections are the major responsibility of two agencies: the Bureau of the Census in the Department of Commerce and the National Center for Health Statistics (NCHS) in the Department of Health and Human Services. Many other agencies collect data directly related to their missions. The two major agencies produce general-purpose data for pro- gram and policy decisions and devote a substantial portion of their staff to serve needs of researchers and policy makers and the general public. Historically, the programs of the National Center for Health Statistics have been the major source of information on the health status of the population. This report, in its discussion of policy is- sues related to health problems of the elderly, draws on only a small part of the data available on the health of the U.S. population. The surveys relevant to this chapter are the National Health and Nutri- tion Examination Survey and the National Health Interview Survey and its supplements. A related source of information is the series of national medical expenditure surveys conducted by the National Center for Health Services Research ant! Health Care Technology Assessment in 1977, 1980, and 1987, which are discussed in detail in Chapters 8 and 9 (on financing and utilization of health services). Recently a new source of health status information became available from the Bureau of the Census: in 1984 the Bureau fielded the first

HEALTH STY TUS AND Q BALI TY OF LIFE 73 wave of the Survey on Income and Program Participation (SIPP), which includes a topical module on health and disability. Although health status is not its major focus, SIPP does provide data that can be used to relate health status to income; SIPP is discussed in detail in Chapter 10. The National Health Inter~riew Surrey The National Health Interview Survey is a principal source of information on the health of the civilian, noninstitutionalized pop- ulation of the United States. The survey, conducted continuously since 1957, provides national data annually on the incidence of acute illness and accidental injuries, the prevalence of chronic conditions and impairments, the extent of disability, the utilization of health care services, and other health-related topics. The data are obtained from 40,000 household interviews that include about 110,00C) persons. Of these, 7,500 to 8,500 are ages 65-74, and 4,800 to 5,400 are age 75 and older. Because of budget constraints the sample size was reduced by 50 percent in 1986. To provide data on special topic areas in addition to the basic NHIS data, supplements to the NHIS have been conducted annually for the past 20 years. The THIS Supplement on Aging One of the NHIS supplements is the Supplement on Aging. Con- ducted in 1984, it is a major source of information on long-term care services used by the elderly particularly when used in conjunction with data collected by the I,ongitudinal Study on Aging, described below. Its purpose was to collect health and community service uti- lization information on the portion of the 1984 NHIS sample that was age 55 and over, which could be related back to the information regularly collected in the NHIS on these respondents. The SOA was administered to half the persons in the NHIS sample who were age 55-64 at the time of the interview and to all of those then age 65 and over. The Longitudinal Study on Aging The SOA is also the base for the I`ongitudinal Study of Aging, a prospective study of the same panel interviewed for the SOA. All respondents will be followed for at least six years by linkage with

74 AGING POPULATION IN THE TWENTY-FIRST CENTURY death records through the National Death Index. Respondents age 65 and over will be followed for at least six years by linkage with Medicare records. Those age 70 and over will be reinterviewed by telephone. The LSOA, which is based on a sample selected in 1984, provides extensive data on the health and medical history of a single cohort of the elderly. Rapid changes in health, medical care needs, living arrange- ments, and available support are characteristic of the older popu- lation. The environment in which these personal changes occur is also in a constant state of flux. In order to understand the changing health needs of the older population, a sample should be selected from a new cohort and should be followed for at least five years- a plan that would permit measurement of some of the transitions occurring in the elderly population. Recommendation 3.1: The pane! recommends that the Na- tional Health Interview Survey's supplement on aging for the noninstitutionalized population ages 55 and over be re- peated every 5 years. The sample should be followed for at least 5 years, and preferably 10 years, and interviewed for characteristics relevant to the older population as wed as for changes in these characteristics. The follow-up could be carried out by a combination of telephone calls and personal interviews with varying frequencies appropriate to the age of the sample member. The Integrated System of Health Interview Surveys The NCHS plans to replace the NHIS by 1989 with an Integrated System of Health Interview Surveys (ISHIS), which, although struc- turaBy different, will retain many of the attributes of the NHIS. It should still be possible to implement the panel's recommendations with the modified structure of the NHIS. The modification provides for core questions, the ability to link with other data bases, and surveys of special topics. The sample design consists of four panels, each a national probability sample of households. Core questions will be addressed to allfour panels, although core items may be ro- tated from year to year. Emphasis is placed on the use of screening questions in the core to identify special subgroups of the population, e.g., veterans, who might then be followed in a special topic survey. The modifications also recognize the importance of being able to link

HEALTH STATUS AND QUALITY OF LIFE 75 NHIS data with other data bases. The National Death Index will be checked routinely for all surveys to ascertain year and cause of death. Discussions are under way with the Health Care Financing Admin- istration concerning the possibility of a follow-on survey of ISHIS participants to obtain their Medicare history. Provision will be made to explore special topics of current policy relevance, on request, with one or more of the panels. These special topic surveys will be funded externally. The National Health and Nutrition Nomination Surveys The National Health and Nutrition Examination Surveys were established to collect those kinds of health data optimally obtained by direct physical examinations and physiological and biochemical measurements. The surveys measure the health and nutritional sta- tus of the U.S. population and permit estimation of the prevalence of certain diseases and the distributions of a broad variety of health- related measurements. NHANES I was conducted from 1971 to 1975; NHANES II was conducted from 1976 to 1980; current plans are un- der way for NHANES Ill to be fielded in 1988. NHANES ~ and NHANES IT did not include persons age 75 and over because it was anticipated that the logistics of bringing the frail elderly to the examination site would present serious problems, and any attempt to do so would result in a poor response rate. The growth of the older population has made it extremely important that this group be included in NHANES ITI. Ways must be found to make it possible to examine elderly people in their homes both to improve the response rate and to avoid the bias inherent in sampling only those able to come to the examination site. Recommendation 3.2: The pane] recommends that the Na- tional Institute on Aging and the National Center for Health Statistics support methodological work to increase the par- ticipation of older people in examination and interview sur- veys. Recommendation 3.3: The panel recommends that the sam- ple for NHANES IT! should not have an age cutoff. The panel further recommends that the sample of persons age 65 and over be adequate to provide suitably precise estimates of the prevalence of specific medical conditions among those over age 65.

76 AGING POPULATION IN THE TWENTY-FIRST CENTURY The precision required in estimates depends on the planned analysis and therefore cannot be specified by the panel. When the preci- sion desired in the estimates has been specified, computation of the required sample size is straightforward.) Although NHANES ~ and IT included questions on cognitive impairment, a more comprehensive set of measures for cognitive functioning and mental health should be included in NHANES ITI. ., . Recommendation 3.4: The panel recommends that the Na- tional Institute of Mental Health continue its development of diagnostic instruments for specific psychiatric disorders of major public health importance (e.g., schizophrenia, ma- jor depressions, and anxiety disorders as well as for cognitive impairments) that can be utilized in national health surveys, such as the NHIS and NHANES. The instrument should be added to the other medical assessment procedures utilized in these surveys. The NHANES ~ Epidemiologic FoBow-up Survey The NHANES I Epidemiologic Follow-up Survey was first fielded in 1982-1984. The survey covered 14,407 persons who were ages 25- 74 at the time they were examined in NHANES I, about 10 years iThe effective estimation of distributions of characteristics for the elderly and important subgroups of elderly (such as the 65-74, 75-84, and 85 and corer eve ran~es~l requires sufficient sample size. Such sample size is necessary O O ~ ~ . _ in order for estimates to have sufficient precision to provide a meaningful description of these distributions. For attributes that apply to 20 to 80 percent of a subpopulation, a reasonable upper bound for the length of 95 percent confidence intervals from sample estimation is 10 percent. The minimum sample size required for this purpose is approximately n = 400. When the likelihood of an attribute is 10 to 20 percent (or 80 to 90 percent), narrower confidence intervals are needed 50 as to clearly indicate the extent to which estimates are bounded away from 0 percent (or 100 percent). As a result, a reasonable upper bound for the length of 95 percent confidence intervals in this setting is 6 percent, for which the minimum sample size needs to be n = 700. Even larger sample sizes are required for characteristics with likelihoods less than 10 percent (or greater than 90 percent). Two other considerations also are relevant to sample size. One is whether the survey design involves clustering or other components that tend to lead to greater sampling variability than simple random sampling. The other is the extent to which the sample needs to provide a basis with suitable statistical power for comparisons among subgroups. Each of these considerations can lead to a doubling or larger increase in the previously discussed sample sizes.

HEALTH STATUS AND QUALITY OF LIFE 77 before the follow-up visits; thus it included some persons in their eighties, but it did not reach the entire elderly population. The sur- vey identified chronic disease risk factors associated with morbidity and mortality, ascertained changes in risk factors, morbidity, func- tional limitation, and institutionalization between NHANES ~ and the follow-up recontacts, and provided information to map the nat- ural history of chronic diseases and functional impairments in the aging population. Continued follow-up of the elderly in the cohort was carried out in 1986 and of the entire cohort in 1987. It would be desirable to repeat the study following the NHANES IT! survey, particularly if it is expanded to reach the total elderly population. A longitudinal study for at least 10 years could produce useful informa- tion on changes in clinical assessments and laboratory measurements. Attrition in a follow-up survey may be very rapid among the very old. With an oversample of the older ages in NHANES ITI, a 2-year cycle for resurvey might be feasible and productive of measures of change. Recommendation 3.5: The pane! recommends that a follow- up study to NHANES Ill be carried out that is expanded to reach the older age groups and provides for repeated ob- servations, preferably at 2-year intervals but at a maximum interval of 5 years. Data on Cause of Death To ascertain which nonrespondents have died in a longitudinal survey of the health status of the elderly, it is essential to match non- respondents against the National Death Index (NDI), unless there are other definitive ways of identifying deaths. Social Security num- ber is a key item of information in such a survey, since it is needed for matching with the NDI. Once a death record has been found in the NDI, information on the cause of death can be obtained from the appropriate state health department. Obviously, the NDI can also be used to obtain information on date and cause of death for sample members in any survey, whether cross-sectional or longitudinal. Consistent follow-up with the NDI for sample members age 55 and over in surveys that collect data on health status could lead to an extremely rich repository of information for relating health status at one or more points in life to age at and cause of death. NCHS has provided for such follow-up with the NDI

78 AGING POPULATION IN THE TWENTY-FIRST CENTURY in the Tong-range plans for the Integrated System of Health Interview Surveys. Recommendation 3.6: The panel recommends that sample populations age 55 or over in all health-related surveys of individuals be followed by matching with the National Death Index to identify deaths and subsequently obtaining cause of death information from state health departments. All health-related surveys should include the respondent's Social Security number. The Importance of Longitudinal Studies Longitudinal data sets are increasingly important for two rea- sons. The first is to provide dynarn~c descriptions of aging as a process rather than static descriptions of elderly people at a particular his- torical moment. Aging is a continuous process, beginning at birth and ending at death. To understand that process, to identify and trace the pathways that individuals take through the successive years requires longitudinal data. Inferences drawn by comparing people in different age groups rather than following individuals through the ag- ing process are often tentative and subject to significant restrictions in interpretation. In the later years of life, longitudinal data track the movement of individuals through the stages of morbidity, disability, and mortal- ity. When coupled with measurement of specific endpoints, including dysfunction, disability, overt disease states, development of cognitive dysfunction, and cognitive competence, such data sets might pro- vide insight into the transition points in the development of frailty. Longitudinal data sets can also take into account the latency that characterizes the development of chronic diseases, the often long pe- riod between the recognition of a disease and its functional impact. Longitudinal data sets provide the capacity to establish patterns of change over time in individuals, which is increasingly important with the recognition of the tremendous variability among the elderly. A second reason for requiring longitudinal data on the aging pro- cess relates to causal inference. Static data can show, for example, the characteristics of men and women who are aging successfully, but they cannot show us which, if any, of those characteristics are causes of their success. Tentative inferences can be made from lon- gitudinal data (which pin down temporal order) and then tested by

HEALTH STATUS AND QUALITY OF LIFE 79 experimental efforts to decrease morbidity and mortality and increase productive and satisfying life. The increasing geriatric population has stimulated interest in research in all aspects of care of the elderly. The field is chang- ing rapidly as these efforts yield new information regarding disease prevention, early detection of treatable causes of disability, and suc- cessfuT management of these disorders. Recognition of the special characteristics of the elderly wid guide collection of data, permitting further advances in understanding the interaction of normal aging and disease, the factors underlying the variability among the elderly, and the important health transitions they experience in the later years of life. To understand the illness histories of the elderly, recording the health status and lifestyle of these individuals earlier in life becomes important for many reasons. Among them are the association of psy- chological factors in middle age with morbidity later in life, the effects of risk-enhancing behavior in middle age on subsequent morbidity, and the need to identify the transition points from one functional status or health state to another. It is also important to track people who have certain diagnoses, such as high blood pressure, from an early point to see the functional consequences at later points in time; or to identify 55-year-olds who have a diagnosis of diabetes to see what functional losses accompany that early diagnosis and what the distribution of impairment is in that subpopulation at some later point in time. For these reasons longitudinal studies should start before age 65. Although there is no uniform age at which it is ideal to start tracking transitions in health that affect the elderly, for prac- tical reasons as well as the fact that changes appear to concentrate in the decade prior to age 65, age 55 is a desirable starting point for many longitudinal studies. Recommendation 3.7: The panel recommends that all sur veys designed to study the health status of the elderly, par ticularly longitudinal surveys, start at age 55 and have the capacity to provide information for five-year age groups. Even though there have been cohort studies ~ the past, studies with new cohorts are needed because of possible cohort differences. A cohort age 55 in 1966 may differ in important ways from a cohort of that age in 1986. The period of life between age 55 and 70 is a critical transition period that warrants current study. In summary, longitudinal studies can help address one of the major data gaps in

80 AGING POPULATION IN THE TWENTY-FIRST CENTURY the health status area-the elucidation of the illness histories of the elderly-by providing data to serve the following purposes: (1) clarify the relation between the presence of disease states (e.g., diagnosis of cancer, hypertension, diabetes) and the subsequent progressive emergence of symptoms of illness, functional impairment, and eventual mortality. (2) determine what kinds of individuals are more susceptible to a disease process; (3) focus on the factors influencing the transition from one health state to another; (4) determine the relation between improved longevity and changes in morbidity (i.e., compression of morbidity); recognize that health status can improve or deteriorate over time and measure the extent of changes in both directions in tracing an illness history; (6) identify the multiple potential pathways that individuals might take through the various stages of illness. FUNCTIONAL STATUS Health survey data show a pattern in which vigorous old age predominates. Most older Americans live in the community and are cognitively intact and fully independent in their activities of daily living (Rowe, 1985~. Dependency and institutionalization are the exception. However, major functional impairment is clearly age- related among the elderly; approximately 5 percent of individuals ages 65-74 require assistance in basic activities of daily living, while 35 percent require such assistance by age 85 (National Center for Health Statistics, 1983a). "Even if one maintains functional independence into old age, the risk of becoming frail for a long period is still high" (Rowe, 1985:828~. One study (Katz et al., 1983) found that for independent persons between the ages of 65 and 69, "active life expectancy"- that portion of the remaining years characterized by independence- represents about 60 percent of total life expectancy; by age 85, that portion fails to 40 percent (see Figure 3.1; see also Figure 4.13. Characteristics of Ebuctional Status "Health is a state of complete physical, mental, and social well- being ~.nd not merely the absence of disease or infirmity" (WorId

HEALTH STATUS AND QUALITY OF LIFE 20 18 16 14 12 co 10 8 6 4 2 o 81 Active Life Expectancies 1974- 1976 16.5 (6 5) 1 10.0 ~ \34.1 - - i 1.6 (6.0) - - - 1 8., - - (4. ~)\~.9 '_ ~ _ (4~2) 6.S 4.7 Life Expectancies . 1974 - - - - 65-69 70-74 75-79 AGE 80-84 85 + FIGURE 3.1 Life expectancies and active life expectancies in Massachusetts. Health Organization, 1948~. Although well-known, this definition has not been applied to produce measures of well-being. The famil- iar measures of health are generally measures of illness, and, for the elderly, measures of the need for assistance. Measures of the func- tional status of the active elderly are needed to trace the transitions from their well state to states of disability along the continuum of health. Most of the well-known measures describe functional status along one of four dimensions: physical function, cognitive function, mental health function, and social function. Some gerontologists add a fifth dimension, economic status, to these areas of measurement.

82 Physical Function AGING POPULATION IN THE TWENTY-FIRST CENTURY Physical function scales are of two types: self-reported and per- formance tests. The self-reported tests vary in the limitations they measure among the active elderly as contrasted with the elderly needing care. For example, one scale (the original or adaptation of the Katz Activities of Daily Living ADL Index) typically iden- tifies approximately 10 percent of the noninstitutionalized elderly population with limitations (Katz et al., 1983~. A second scale (an adaptation of the Rosow-Breslau Functional Health ScaTe) typically identifies approximately 40 percent of the well elderly who report limitations in one of three areas: the ability to walk a half mile, to climb stairs, to do heavy housework (Rosow and Breslau, 19663. The third common scale, based on items developed by Nagi (1976) for the Social Security Disability Surveys, typically identifies approximately 70 percent of the elderly who are living in the community as report- ing difficulty in performing one or more activities such as stooping, crouching, kneeling, fingering small objects, extending arms above shoulder level, pushing or pulling large objects, and lifting weights above 10 pounds. Performance tests of physical function include a 15-minute man- ual dexterity performance test described by Williams et al. (1982), a hand grasp performance test, a pulmonary function test, and blood pressure measurements (both standing and seated). Actual perfor- mance of some of the tasks on the ADL scale has been part of the examinations in the Framingham study. Cognitive Function In tests of cognitive function, performance tests are used more frequently than self-reported measures. Among the performance tests frequently used in community surveys of older people are a variety of tests (e.g., the Mini-Mental Scale Examination in the Epidemiologic Catchment Area Project) that are intended to measure orientation to time and place. In addition, community surveys often use subscaTes from intelligence tests and other sources to gauge short- term and Tong-term memory and certain other cognitive capacities. The self-reported tests of cognitive function include items asking the respondent's judgment of his or her memory capacity at present compared with specified earlier intervals.

HEALTH STATUS AND QUALITY OF LIFE Mental Health E unction 83 In the measurement of mental function, self-reported assessments are used almost exclusively in this country, where epidemiologic stud- ies are generally performed by lay interviewers. In contrast, clini- cians have been involved in studies in Europe, and have therefore made psychiatric diagnoses as traditionally done in a clinician's of- fice. Our understanding of some aspects of mental functioning would be enhanced through the combination of screening procedures and standardized diagnostic interviews. Standardized measures of morale and depression that can be used in community surveys are: the Law- ton Morale Scale, the Center for Epidem~ological Studies-Depression Scale (CES-D) (Radioff, 1977) designed by the National Institute of Mental Health (NIMH), and the Beck Depression Scale, among others. In addition, NIMH has sponsored the development and field testing of the Diagnostic Interview Schedule (DIS), which approxi- mates the DSMIT! (Diagnostic and Statistical Manual, ad edition) categorization of clinical diagnoses and can be administered by lay interviewers. The DIS was developed for use in the Epidemiological Catchment Area Projects. Social Function For measuring social function, numerous self-reported items are used in community surveys. The Alameda County Study (Berkman and Bresiow, 1983) and the National Institute on Aging's Estab- lished Populations for Epidemiologic Studies of the Elderly project (Cornoni-Huntley et al., 1986) are two examples. ABeessing Functional Status There are numerous functional tests in addition to those dis- cussed above. Several recent studies have listed and described the existing tests (Kane and Kane, 1981; Mangen and Peterson, 1982; Fillenbaum, 1984~. In her study published by the World Health Or- gar~ization (WHO) in 1984, Fillenbaum argues for a multidimensional approach to functional status (p.5~: There have been few attempts to make a comprehensive asness- ment of the well-being of representative groups of elderly people as a basis for policy decisions concerning the provision of appro- priate services. Rather than considering the elderly person as an integral human being, the tendency of care givers and research

84 AGING POPUL,ATIONIN THE TWENTY-FIRST CENTURY workers alike has been to measure single dunensions of well- being, such as mental function, social support, economic status, physical morbidity, or capacity for self-care. However, elderly people are subject to multiple disadvantages, and their physical, mental, social, and economic well-being are closely interrelated- more so than at younger ages so that combined assessment of the various dunensions of well-be~ng is necessary. There is a general consensus that five basic dimensions should be included in any overall assessment of elderly individuals within a population, namely activities of daily living, mental health, physical health, and social and economic functioning. While it necessary and anport ant to have information on specific areas, functioning in one area has an impact on functioning in others. Assessment not only should be multidimensional, but also should be in terms of functional status. Fillenbaum focuses on three multidimensional functional as- sessments: the comprehensive assessment and referral evaluation (CARE), the Philadelphia Geriatric Center multilevel assessment in- strument (MAI), and the Older American Resources and Services multidimensional functional assessment questionnaire- selected be- cause they meet standards of validity and reliability. In addition, Fillenbaum also describes the Rand Health Insurance Study (RAND HIS) questionnaire. While it was designed for use at ages below 61, the original questionnaire was actually used in the Rand study with persons up to age 85. With minor alterations this questionnaire could be made fully relevant to an older population. Fillenbaum also notes the existence of the International CIassifi- cation of Impairments, Disabilities, and Handicaps (ICIDH), which can be used in the classification of handicaps on a Appoint scale, with O indicating the absence of a handicap. The classification was published by WHO in 1980, but it has not been used in the United States. It should be reviewed for possible use of those parts of the classification that could be adapted to descriptions of the well elderly. The U.S. Social Security Administration's Longitudinal Retire- ment History Survey (SSA [RHS) is cited as an example of the use of a multidimensional functional questionnaire over a Tong period 6 years. To illustrate the utility of such survey results, Fillenbaum uses data from the SSA [RHS to develop a transition matrix showing the estimated transition probabilities from each 1969 health state to each 1971 state for the [RHS sample (Fillenbaum, 1984:62~. Each of the five dimensions is treated as a bivariate (impaired or not impaired) yielding a matrix with 32 rows and 32 columns.

HEALTH STATUS AND QUALITY OF LIFE 85 There is a need to reexamine existing tests and questionnaires for use in surveys of the general population of the elderly and to develop new instruments if needed. One criterion for deterrn~ning the suitability of an instrument is its capacity to provide information for measuring the complete spectrum of health from well-being to illness and dependency and the transition rates from one state to another over time. The second criterion for selection should be the feasibility of applying the questionnaire in an interview survey. The third criterion should be whether the instrument is sufficiently brief for use in national information systems. Recommendation 3.8: The pane] recommends (a) that the National Center for Health Services Research and Health Care Technology Assessment, the National Institute of Men- tal Health, and the National Institute on Aging support research to develop, enhance, and evaluate instruments for measuring functional status (physical, mental health, cog- nitive, and social) based on the behavior of active older people; (b) that the National Center for Health Statistics continue and enhance its effort to determine how well these instruments address the health status of the elderly and how the instruments might be adapted (reduced) for application in national information systems; and (c) that the National Institute on Aging and the National Center for Health Statis- tics support methodological work to improve the validity and reliability of measures of cognitive and physical impairments of the elderly. The emphasis in the provision of health care to the elderly should be on maintaining functional capability and increasing active life expectancy. Evaluation of the elderly patient must focus on what the patient can do, relative to what the patient should be able or wishes to do, and on identification of recent functional deficits that may be reversible. Although a complete and precise diagnosis is essential, the functional impact of each diagnosis should be evaluated. Functional measures should include not only activities of daily living (bathing, continence, dressing, eating, mobility, and walking) and instrumental activities of daily living (e.g., doing housework/laundry, preparing meals, shopping, getting around outside, going places not within walking distance, managing money, taking medicines, using the telephone) but also more subtle ejects of disease on cognitive functions. Specific diagnoses often have little relation to functional

86 AGING POPULATION IN THE TWENTY-FIRST CENTURY status and the length of the diagnosis list provides little insight into the specific needs and capabilities of a particular patient. Too often a long diagnosis list biases physicians to think that the patient is multiply impaired and therefore frail, although this may not be the case at all. Diagnoses themselves are often a weak criterion for assessing the health care needs of the elderly (Besdine, 1983~. QUALITY OF LIFE The General Concept Several of the topics discussed earlier in this chapter are related to the quality of life, which for older persons includes the extension of longevity, the compression of morbidity, health and functional status, and psychosocial factors as modifiers of disease. This sec- tion discusses some of the components of quality of life, including measures of productive activity of the elderly, and shows their rela- tionship to those topics. Age is not a strong predictor of subjective well-being. By and large, older people are no less satisfied with their lives than those who are young. Indeed, when one looks at satisfaction with the various life domains income and standard of living, work, family and marriage, friendships, housing and community, leisure, and health itself only satisfaction with health declines with age. In other words, people tend to come to terms with their lives, and most people do not experience the terms as unduly harsh (Herzog et al., 19823. For persons of all ages, including the elderly, the quality of life experienced by an individual is related to his or her level of satis- faction with life, sense of well-being, and feeling of self-worth and self-esteem (Campbell et al., 1976; Andrews and Withey, 19763. In contrast to objective social indicators for the population or particular subgroups, which focus on socially significant observable events or characteristics (such as marital status and employment), quality-of- life measures are concerned with the subjective, psychological expe- riences that surround or accompany these events or characteristics, as perceived and reported by the individual (such as the level of pride and satisfaction derived from employment). Correlations be- tween objective social indicators and perceived quality of life for an individual tend to be weak (Vinokur et al., 1983:32~. Level of satisfaction has been measured globally, referring to overall satisfaction with life; with respect to domains of life ex- perience, such as health and mental health, marriage, family life,

HEALTH STATUS AND QUALITY OF LIFE 87 friendships, housing, job or work, nonwork and leisure activities, and financial situation (personal assets, financial security, earnings); and on such dimensions as personal safety and the security of one's pos- sessions (the latter for institutionalized persons in particular) (see, e.g., Campbell et al., 1976; Gastil, 1978; George and Bearon, 1980; Morgan and Smith, 19693. This view of quality of life presupposes that an individual has aspirations and expectations that serve as the criterion or standard against which his or her current situation is per- ceived and assessed. The level of congruence between expectations and perceived or actual circumstances is viewed as generating a sub- jective evaluation of satisfaction or dissatisfaction, which is for that individual a measure or indicator of quality of life, either glob ally or for the particular domain in question (Franklin et al., 1984~. Personal well-being, also a contributor to one's perceived quality of life, is a broader concept than satisfaction with life or its domains. According to its proponents (including Bharadwaj and Wilkening, 1977, 1980), this notion involves self-development, the realization of one's own potential as an individual and as a social being (deriving from MasIow, 1954), and the capacity to adapt to change and is equivalent to a sense of personal efficacy (Vinokur et al., 1983:34~. Feelings of self-worth and self-esteem are also related to a per- son's assessment of the quality of his or her life (see George and Bearon, 1980~. The belief that one's achievements and activities are valued and needed by others and the perception that one is capa- ble of interacting and negotiating successfully with the environment enhance feelings of self-worth and self-esteem. Some Dimensions of Quality of Life Health Status While health status is only one among many dimensions that affect a person's assessment of the quality of his or her life, it is a very important one. George and Bearon (1980), for example, regard general health (and functional) status as one of four dimensions that define quality of life for older persons, along with socioeconorn~c sta- tus, general life satisfaction, and self-esteem and related measures. As defined here, health status consists of physical, psychological, and social components. The focus of survey questionnaires is typically on physical symptoms and the functioning of organs and systems, the ability to perform life roles, and the ability to move about. Measures

88 AGING POPULATION IN THE TWENTY-FIRST aENTURY of psychological health focus on psychological symptoms and disor- ders. Social health is most often defined in terms of an individual's ability to function as a member of the community, including the world of work, and involves the nature of family relationships, friendships, and community participation (Vinokur et al., 1983~. Yet here, as for the other domains of life, subjective estimates of one's physical health and well-being do not necessarily correlate with objective measures based on laboratory findings and physical examinations (Vinokur et al., 1983~. That is, the relationship between the presence of actual disease and disability and perceived well-being is neither obvious nor uniform across medical conditions. Similarly, neither increased life expectancy nor the use of medical interventions that objectively improve medical conditions invariably results in the perception that one's quality of life has been improved (Vinokur et al., 1983~. Satisfaction with Life The elderly tend to be as globally satisfied with their lives, on the whole, as younger persons, though their level of satisfaction with their health status tends to decline as they advance in years. They also tend to report their health as good or excellent and to see themselves as functioning well. Lengthening of the life span or longevity per se will not result in enhanced satisfaction with life for the elderly unless personal vigor and the ability to function in- dependently (without institutionaTization) are maintained. As noted above, it is not known yet whether increased life expectancy will be accompanied by an overall increase in active life expectancy for the elderly, or rather by a prolongation of the period of dependency and functional impairment, which does tend to increase with age. In this connection, it is worth reiterating that aging does not in- evitably result in decline of health status or function: older persons are very heterogeneous with respect to health and functional status, and improvement in both successful aging can and does occur. Functional Status Functional status including the ability to perform sel£care ac- tivities and to carry out activities of daily living such as cooking, shopping, and cleaning is more critical to quality of life than mor- bidity or diagnosis per se. Physical, cognitive, mental health, and

lIEALTH STATUS AND QUALITY OF LIFE 89 social functioning, individually and together, as they tend to rein- force each other, contribute to a sense of well-being and satisfaction with life. Availability of Social Support System The availability of a social support system, including actual and potential informal supports such as family and friends, is crucial to the quality of life for the elderly, whether residing in the community or in an institution. Not only are positive, close, and stable family relationships correlated with satisfaction with life (see, for example, Najman and Levine, 1981), but also the availability of social supports can prevent or delay institutionalization of hail elderly people. Productive Activity Increasing longevity should be accompanied not only by the maintenance or enhancement of physical, mental health, emotional, and social function, but also by productive activity that is recognized by oneself and others as socially useful. The development of statistics on the full range of productive activities, especially among older men and women, is important for both scientific reasons and national policy. Both require the broadening of national statistical series so that they represent more accurately the major activity patterns of men and women throughout the life course. The lengthening of life, if it is to be meaningful to individuals and enriching rather than burdensome to the society as a whole, must be made as productive as possible. Measurement of the extent to which this goal is achieved, requires accurate statistics on the current productivity of older people and statistical information on the factors that determine that productivity. National data of this kind are not now available. Instead, all paid employment is assumed to be productive and all unpaid con- tributions to society (except for work on a family farm or business) are ignored. These omissions are especially problematic for policy purposes because they are age-biased. They not only underestimate productive activity in the nation as a whole; they differentially un- derestimate the productive contributions of different age and gender groups. The underestimation is greatest for those past the usual age of retirement. When people leave paid employment, they drop from

go AGING POPULATION IN THE TWENTY-FIRST CENTURY the conventional statistics of productive behavior for example, the Current Population Survey no longer counts them as "in the labor force" and consequently they are also omitted from statistics derived from the CPS, for example, the gross national product (GNP). Peo- ple, including the elderly, who are homemakers, who care for ill or disabled family members, who do volunteer work in schools or hos- pitals are classified as nonworkers as recipients and consumers who do not produce. Policy debates over compulsory retirement, pension entitlements, Social Security funding, and the like are therefore con- ducted with only a partial base of knowledge, since they do not take cognizance of the important productive work of the elderly. The development of more comprehensive measures of productive behavior must begin with a conceptual task: the definition of activ- ities that are to be defined as productive. We propose a definition that is primarily economic: an activity is productive if it generates valued goods or services, and the key measure of its productivity is the market value (actual or attributed) of those goods and services minus the nonIabor costs involved in their production. Several properties of productive behavior follow from this defi- nition: (1) With the exception of paid employment, for which the mar- ket value is established by the payment, no other activity is currently counted as productive; market value must be esti- mated or attributed for these activities under the proposed definition. (2) Paid employment includes work done in both the regu- lar and irregular (reported and unreported) economies. It thus includes untaxed and typically unreported payment received for work, as well as pay or profit from regular employment. (3) Unpaid activities that generate goods or services, and for which market values can be determined, are likely to in- clude housework, child care, care of ill or disabled family members or friends, and some forms of volunteer work in organizations. For example, the activities of a volunteer who types letters or does bookkeeping for a civic organi- zation fall within our definition; mere attendance at the meetings of the organization would not. The proposed definition permits the inclusion of activities in which the producer is also the consumer of the product, (4)

HEALTH STATUS AND QUALITY OF LIFE 91 provided that such activities generate goods or services for which a clear market value can be attributed. As an example of the implications of this approach, consider the painting of a house, an activity that might be undertaken by a professional painter for wages on a regular job or as a contract for profit. The same activity, however, might be undertaken for pay by a person working outside a regular job. It might also be done by a relative or friend without pay of any kind and, finally, it might be done by the homeowner himself or herself. The definitions and procedures of current government statistics would include the first of these house painting done as part of a regular job for pay or profit in its measures of employment, and ult~rnately, its contribution to GNP. The second case house paint- ing done as part of the irregular economy might or might not be included. The third variation, in which the work is done as a favor to a relative or friend, would not be included. And the fourth case, in which the same work is done by a person for himself or herself, would certainly not be included. Yet in all these cases, the work done is identical. Ultimately, such work should be included in our national statistics in such a way that it would be possible to distinguish be- tween paid and unpaid work, and between work done for others and that done for oneself. Recommendation 3.9: The pane! recommends that the De- partment of Labor, in conjunction with the Bureau of the Census and the Department of Health and Human Services, develop a concept of productive activity that includes both paid and unpaid work and that can be measured and re- ported in surveys such as the Current Population Survey, the National Health Interview Survey, and the Survey of In- come and Program Participation, as well as in the decennial census. Implementation of this recommendation would recognize the fact that longer life means, for many elderly people, increased productive years that have the potential to contribute to an individual's sense of self-esteem and self-worth and thereby enhance his or her quality of life. Once the concept of productive activity that includes both paid and unpaid work has been developed and made operational for purposes of collecting data via surveys, the quality of the productive life of the elderly should be monitored. In the absence of such statistics, no authoritative measures of

92 AGING POPULATION IN THE TWENTY-FIRST CENTURY these categories of unpaid work are available, but estimates have been made (Goldschmidt-Clermont, 1982; Peskin, 1983~. They differ, but all of them indicate that the goods and services generated by unpaid] work are substantial relative to conventionally estimated monetary GNP. No estunates of unpaid work set its total value as less than 20 percent of GNP, and some estimates range as high as 44 percent of GNP. Some development of measures has been done on all of the broad activity categories with which we are concerned: paid employment, unpaid voluntary work in organizational and informal settings, and work that produces goods for one's own consumption. All of these available measures involve self-report to some extent. For example, unpaid work to operate a family farm or business is measured by the Bureau of the Census in comparable terms number of hours worked, occupation, and industry to those for paid employment. Other forms of unpaid work, either in organizational settings or on less formal bases, have been measured less frequently, less success- fully, and less completely. The Current Population Survey provides estimates of the number of people who are engaged in housework and are not working for pay or profit. Mutual assistance of various kinds has been measured mainly in terms of frequency of contact and self-rated importance of the relations involved. The CPS incorpo- rates no est~rnates of the monetary value of such work, but reviews of alternative methods for making such estimates are available (see, e.g., Goldschrn~dt-Clermont, 1982; Peskin, 1983~. Measures of self-care have been developed, with scales of physi- cal self-maintenance and daily living (Lawson, 1977; Kane and Kane, 1981~. The fuB range of activities in which the individual consumes the product of his or her own work, however, has not been systemat- ically measured. Given the state of measurement development and the goal of assessing the fuB range of productive activities throughout the life course, the pane! proposes five sets of measures: (1) categories of productive activity engaged in, (2) amount of input (hours devoted) to such activities, (2) monetary value of the product or service result- ing from each of these activity categories on an individual basis, (4) self-evaluation of the productivity of the activity, and (5) outcomes for the individual engaging in the activity and for others. These suggestions are based on a set of propositions for which substan- tial evidence is already available: that the reality and experience of producing something of value or of providing a service and the

HEALTH STATUS AND QUALITY OF LIFE 93 of producing something of value or of providing a service and the recognition and acknowledgment of that fact by oneself and others are conducive to improving quality of life. Quality of Life for the ~titutional~zed Elderly For older persons who reside in institutions such as nursing homes rather than in the community, the factors that contribute to satisfaction with life, a sense of well-being, and feelings of self- worth are somewhat different from those discussed above. Already [united in their health status and ability to function independently and autonomously in the community two contributors to a high quality of life such persons perceive their quality of life as heavily dependent on the quality of their medical and nursing care. This is especially true for the very ill and disabled (Institute of Medicine, 1986~. A recent empirical study shows that nursing home residents also attach high importance to the qualifications, competence, and attitudes of staff particularly nurse's aides who provide much of the hands-on care on a day-to-day basis. Friendliness, cheerfulness, and the treatment of residents with respect and dignity are quali- ties highly valued in staff and are viewed as contributing heavily to residents' quality of life (Institute of Medicine, 1986~. In addition, such amenities as the quality of the food, the ambiance, the ability to make personal choices and participate in planning one's own care, and activities that assist residents in maintaining or regaining inde- pendent function (such as rehabilitation exercises, encouragement in ambulation or self-feeding) contribute to a high quality of life for the institutionalized elderly (Institute of Medicine, 1986~. Conclusion In order to document the quality of life experienced by older persons, both community residents and those residing in institutions, new measures, suitable for cross-sectional and longitudinal surveys, will have to be developed. Recommendation 3.10: The panel recommends that na- tional population-based surveys such as the National Health Interview Survey and the National Nursing Home Survey (which includes a population component) include measures of factors that influence quality of life both positively and negatively as people progress to advanced ages.

Next: 4 Health Transitions and the Compression of Morbidity »
The Aging Population in the Twenty-First Century: Statistics for Health Policy Get This Book
×
Buy Paperback | $60.00
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

It is not news that each of us grows old. What is relatively new, however, is that the average age of the American population is increasing. More and better information is required to assess, plan for, and meet the needs of a graying population. The Aging Population in the Twenty-First Century examines social, economic, and demographic changes among the aged, as well as many health-related topics: health promotion and disease prevention; quality of life; health care system financing and use; and the quality of care—especially long-term care. Recommendations for increasing and improving the data available—as well as for ensuring timely access to them—are also included.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!