Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 52
2
Social, Economic, and Demographic
Changes Among the Elderly
The population of the United States is growing older, a phe-
nomenon widely noted and described, with significant implications
for the nation's health, social, and economic institutions. It is neces-
sary to understand the past demographic and socioeconomic trends
to better estimate the future size and characteristics of the older
population as well as to forecast their demand for services and the
extent to which those demands can be met. Analysis of the demo-
graphic and socioeconomic trends of the elderly population will also
help identify data needed to make informed policy decisions related
to the health of the future elderly population.
THE CHANGING DEMOGRAPHIC STRUCTURE OF THE
POPULATION
The distribution of the population in the United States has
shifted rapidly in both the number and proportion of the population
age 65 and over. This subgroup has grown faster than the rest of the
population in recent decades, will continue to grow at a more rapid
rate for the remainder of the twentieth century, and is expected to
continue to increase well into the next century. Between 1950 and
1980 the population age 65 and older more than doubled, from 12.3
million in 1950 to 25.5 minion in 1980 (Taeuber, 1983~. During this
30-year period, the percentage increase in the number of elderly was
74 percent larger than for the population under age 65 108 percent
compared with 62 percent. For the oldest-old, age 85 and over, the
52
OCR for page 53
SOCIAL, ECONOMIC, AND DEMO GRAPHIC CHANGES
rise was the largest, a 281 percent
2.2 million in 1980.
.
Population Forecasts
53
Increase from 577,000 in 1950 to
The size of the elderly population today and in the near future
is relatively simple to estimate: it depends on the births in the early
years of this century and the deaths in the birth cohort in subsequent
years. (A small portion of the total elderly population is accounted
for by net migration, which is not as accurately counted as births and
deaths.) The elderly population in the next century depends on the
births beginning in the 1930s and the estimated deaths in each year's
birth cohort. These estimates are subject to increasing uncertainty
as we move further into the future.
Birth rates were relatively high in the early part of this century,
low in 1920-1940, high in the postwar years 1946-1964, lower again
in 1965-1973, and slightly higher in more recent years. Throughout,
there have been important variations by age of mother, birth order,
and race. Death rates, meanwhile, have declined or remained level
throughout the twentieth century, although at rates that varied by
age, race, and sex. Declines in mortality rates have been consistently
greater for women than for men and, since 1968, almost as large
for the oldest-old as for young-old (ages 65-74) females. Current
indications are that the declines in mortality rates are continuing
(National Center for Health Statistics, 1986a). The future population
has been estimated by the Bureau of the Census on the basis of a
completed cohort fertility of 1.9 births per woman and a continued
decline in mortality rates. The most likely forecasts are identified
as the Bureau's "middle series," which are the basis for the analysis
in this report. Should there be great advances in medical care or
unpredictable epidemics, the estimated size of the elderly population
might be considerably different.
Table 2.1 shows the actual and projected growth of the older
population. The middle series estimates a steady rise in the elderly
(age 65 and over), from 25.5 million in 1980 (11.3 percent of the total
population) to a projected 64.3 million (21.1 percent) in 2030 more
than doubling over the 50-year period. The number of oldest-old
will continue to grow rapidly in the next 50 years, from 2.2 million
in 1980, to 8.8 million in 2030, and to 16.1 million in 2050. The
progression of the postwar baby-boom cohort, those born from 1946
to 1964 (Siege! and Davidson, 1984) may be seen in the peak for
OCR for page 54
54
AGING POPULATIONINTNE TWENTY-FIRST CENTURY
TABLE 2.1 Actual and Projected Growth of the Older Population,
1980-2050 (numbers in millions)
65-74 75-84 85 years 65 years
Total Years years and over and over
Year Population No. ~No. % No. % No. %
1980 226.5 15.6 6.9 7.7 3.4 2.2 1.0 25.5 11.3
1990 250.0 18.0 7.2 10.3 4.1 3.5 1.4 31.8 12.7
2000 268.0 17.7 6.6 12.2 4.6 5.1 1.9 35.0 13.1
2010 283.1 20.3 7.2 12.2 4.3 6.8 2.4 39.3 13.9
2020 296.3 29.8 10.2 14.3 4.8 7.3 2.5 51.4 17.3
2030 304.3 34.4 11.3 21.1 6.9 8.8 2.9 64.3 21.1
2040 308.0 29.2 9.5 24.5 8.0 12.9 4.2 66.6 21.6
2050 308.9 30.0 9.7 21.0 6.8 16.1 5.2 67.1 21.7
Source: Taeuber (19833.
the 65-74 age group in 2030, for the 75-84 age group in 2040, and
those age 85 and over in 2050. The oldest-old population group was
1 percent of the total population and 9 percent of the elderly in 1980;
by 2050, this group is projected to increase to 5 percent of the total
population and 24 percent of the elderly.
The accelerated growth within the elderly population of those
age 85 and over has shifted attention to this subgroup and its unique
set of needs. The oldest-old are at risk for chronic illness, tend to be
functionally dependent, and have greater needs for medical, social,
and support services.
Forecasts by Sex
At birth, every cohort has a small excess of males but, owing to
the higher death rates for the male population and the more rapid
improvement in mortality for women there is a large excess of women
. ~ , _
~ ~ T ~ A^~ ~ ~ ~ ~ ~ ·11- 1 ~ 1 _ I or
at older ages. in 19bU there were 1U.z million elcterly men tage u
and over) and 15.2 elderly women, a ratio of 68 men to 100 women.
The Census Bureau population projections show that the sex ratio
of the population age 65 and over will continue to fall in the next few
decades, but more slowly than in the past, reaching 64 males per 100
females in the year 2000 (Siege] and Davidson, 1984~. Subsequently,
the trend will change, so that by the year 2020 the sex ratio or fine
elderly population will be 69 men per 100 females.
The sex ratio declines rapidly with increasing age: in 1980 there
OCR for page 55
SOCIAL, ECONOMIC, AND DEMOGRAPHIC CHANGES
55
were 80 males per 100 females for those age 65-69 and 44 mates per
100 females for those age 85 and older. For the latter group, the ratio
of men to women is projected to fall between 1980 and 2020, from 44
men to 36 per 100 women. Since the vast majority of the oldest-old
are female, many of the health, social, and economic problems of this
group are those of women.
Forecasts by Race
In 1980, 12 percent of the white population was age 65 and
older-a much larger proportion than the 8 percent of the black pop-
ulation (Siegel and Davidson, 1984~. The Census Bureau attributes
the difference to higher fertility of the black population and secon-
darily to higher mortality at ages below 65. The Census Bureau
projects that the black population of the future will continue to be a
younger population than the white, although improvements in mor-
tality rates for elderly blacks are expected. By 2020, 19 percent of
the total white population compared with 12 percent of the black
population is projected to be age 65 and over (U.S. Department of
Health and Human Services, 1985b).
Geographic Distribution of the Elderly Population
Older persons tend to move far less often than younger persons,
remaining in the state, county, or local area where they settled during
their adult years. Between 1975 and 1979, their rate of interstate
migration was 3.6 percent (Bureau of the Census, 19843. Between
1970 and 1980, the largest numerical increases in elderly people were
in the states of Florida, California, and Texas. Growth of more
than 50 percent in the number of elderly in that decade occurred in
Arizona, Florida, Nevada, New Mexico, Alaska, and Hawaii. In 1980,
almost half the elderly were living in eight states: California, Florida,
New York, Pennsylvania, Texas, Illinois, Ohio, and Michigan, but
these states also accounted for 47.9 percent of the total population.
In Florida, 17.3 percent of its 1980 population was elderly, the highest
proportion of any state, although Florida had only 4.3 percent of the
total population (Taeuber, 1983~.
Short-term population projections to the year 2000 by the Bu-
reau of the Census show significant differences in rates of change in
the population of the four regions of the United States. The West
and South will be the fastest-growing regions from 1980 to 2000,
OCR for page 56
56
AGING POPULATION IN THE TWENTY-FIRST CENTURY
increasing 45 percent and 31 percent, respectively. The North Cen-
tral region is projected to lose population during the same period.
The elderly population in all regions, however, is projected to rise,
ranging from a 12 percent increase in the Northeast to a 60 percent
increase in the South and West (Taeuber, 1983~.
These geographic data imply differential use of medical care
services by region. For example, in the Northeast and North Central
regions, the number of nursing home beds will need to increase by 44
percent. In the South and West, the number of nursing home beds
will have to more than double to meet the needs of the projected
elderly population (Rice, 1985~.
Marital Status and Living AITangements
Among the most important social characteristics affecting the
welfare of the elderly are those that pertain to their marital status
and living arrangements. Elderly men are most likely to be married;
elderly women are most likely to be widowed. In 1981, 79 percent of
elderly men and 39 percent of elderly women were married. For el-
derly women, the proportion of widows increases rapidly and remains
at a high level: for the 65-74 age group, 40 percent are widowed; for
the group age 75 and older, 68 percent are widowed.
Marital status has a direct bearing on the living arrangements
of the elderly. Among elderly men, 82 percent live in a family setting
and more than 74 percent are married and living with their wives. A
very different situation exists for elderly women; 55 percent live in a
family setting and only 36 percent are married and living with their
husbands. In short, women age 65 and older are more likely to be
widowed than married and living alone rather than with husbands.
The number of elderly women living alone has doubled in the last
15 years, and projections by the Census Bureau show a substantial
increase up to 1995 in the proportion of households with an elderly
female living alone or with nonrelatives (Siegel and Davidson, 1984~.
This trend has important implications for housing needs and the
demand for institutional care. With the decline in the proportion of
the elderly living with relatives likely to continue, there will probably
be a greater need for the provision of social support and health
services by the community or other public sources.
Education
The level of educational attainment of the elderly population
OCR for page 57
SOCIAL, ECONOMIC, AND DEMO GRAPHIC CHANGES
57
is currently less than that of the younger population. This educa-
tional gap by age group has narrowed since 1950 and is expected
to nearly close in the next decade, due to increased compulsory
secondary school requirements, as well as educational opportunities
made available by the G.~. bill after World War IT. A lower propor-
tion of foreign-born in the future elderly population due to changes
in immigration will also serve to increase the educational attainment
of the elderly population. The greater education of the future elderly
population implies a change in demand for services: combined with
rising income, they may seek and demand more and better health
care and other programs for their needs (Rosenwaike, 1985~.
Income
.
The income of the elderly has improved over time. According to
the Congressional Budget Office, "After accounting for inflation, the
average cash income of families with elderly members increased by
nearly 18 percent during the 15-year period from 1969 to 1984, the
latest year for which detailed data are available- while the average
income of unrelated elderly individuals rose by 34 percents (Gordon,
1986:2~. The income of younger families also rose in this period, but
not as much as for the elderly. Average elderly family income was 68
percent of average nonelderly family income in 1969 and 78 percent
in 1984. For unrelated individuals, the elderly-to-nonelderly income
ratio was 50 percent in 1969 and 60 percent in 1984.
The poverty rate among the elderly also declined in 1969-1984,
from 25 percent to 12 percent, but in 1984 an additional 9 percent
of the elderly had incomes of not more than 25 percent above the
poverty level. In 1984 incomes were below the poverty level for 9
percent of elderly men, 15 percent of elderly women, and 36 percent
of elderly black women.
Social Security benefits are the largest single source of money
income for the elderly (nearly 40 percent), followed by earnings,
property income, and private and public pensions. The most sig-
nificant change In source of income for the elderly population since
the 1960s has been a decline in the importance of earnings and an
increased reliance on retirement income from Social Security, public
and private pensions, and assets. Noncash benefits are estimated
to be 10 percent of the income of the elderly, the most important
ones being Medicare, Medicaid, food stamps, and publicly owned
or subsidized housing. Although asset ownership (including savings
OCR for page 58
58
AGING POPULATION IN TlIE TWENTY-FIRST CENTURY
and home ownership) is fairly common at the time of retirement, the
value of assets owned by the elderly is low.
Current expenditures by the elderly are highest for shelter, fol-
Towed by food, transportation, and health care, which, surprisingly,
uses less of the budget than transportation. These expenditures must
be considered along with available economic resources in planning
and developing public policies for the elderly.
Labor Force Participation
Sharp declines have occurred in the last few decades in the
labor force participation of men age 65 and older: from a third in
1960, to a quarter in 1970, to 17.6 percent in 1981 (U.S. Congress,
Senate, 1986a). This trend is associated with an increase in voluntary
early retirement and a drop in self-employment. With the growth
in retirement programs, more older workers have been financially
able to retire earlier. Projections by the Bureau of Labor Statistics
(BES) show a continued decline in labor force participation of elderly
men at least up to 2000 (Fullerton, 1980~. The proportion of older
women in the labor force has increased moderately since 1960, due
to economic necessity, more education, changes in social roles, and
increased divorce rates that result in more women heading their
own households. BES projections show a moderate decline in the
labor force participation of women age 65 and older and a continued
increase for women age 55-64 up to 2000 (Siege! and Davidson, 1984~.
Part-time employment is now an increasingly unport ant source
of employment for the elderly: half of those age 65 and over who
work do so on a part-time basis, up from a third in 1960. Age at
retirement and labor force participation of the elderly have a direct
erect on retirement programs and economic dependency. The age of
eligibility for Social Security and other pension benefits will affect the
age of retirement for many elderly, which in turn affects their level
of income and economic dependency. The projected decline in labor
force participation rates of older persons will lead to a continued rise
in the ratio of older nonworkers to the working population and an
associated increased dependency. The Social Security Act of 1983
advanced the age of retirement from 65 to 67 for payment of full
benefits. The change is to be phased in from age 65 in 1983, to 66 in
2006, to 67 in 2027. It is uncertain what effect the law will have on
actual age at retirement.
OCR for page 59
SOCIAL, ECONOMIC, AND DEMO GRAPHIC CHANGES
59
Dependency Ratio
The social support systems now in place reflect the current bal-
ance between the size of the working population and the retired.
The trends for people to live longer and for families to have fewer
children are changing the shape of the elderly dependency ratio-the
population age 65 and over divided by the population ages 18-64,
the working population. This ratio has risen steadily, from 11 per
100 in 1940 to 19 per 100 in 1980, and it is expected to reach 22 in
2010. The expected leveling off or slower increase in the next several
decacles will be followed by a sharp increase between 2010 and 2030,
when the baby-boom cohorts will reach old age; the ratio is expected
to be 29 per 100 by 2020 and 37 per 100 by 2030 (Siege} and David-
son, 19843. At the same time, projected low fertility rates will result
in fewer young persons and, thus, a declining young dependency ra-
tio, defined as the population under age 18 divided by the working
population, ages 18-64.
The total dependency ratio, the sum of the young and elderly ra-
tios, is a crude index of the total burden on the working population of
its support of both old and young dependents. The total dependency
ratio has declined since 1960, but it is expected to increase in the
next century, and the increase in the elderly dependency ratio will be
greater than the decline in the young dependency ratio. The elderly
are primarily supported by publicly funded programs while, except
for public education, mostly private (i.e., family) funds support the
young. Since the elderly will be the most rapidly growing age group
and more costly, the change in the dependency ratios will be a major
policy issue for both Social Security and the hospital insurance pro-
grams under Medicare that are financed by payroll taxes (Rice and
Feldman, 1983~.
In addition to the unknown effects of advancing the age of re-
tirement to 67 for payment of full benefits, fully effective in 2027,
other social and legislative changes in the next 50 years may change
the relationships between the working and the retired populations,
significantly changing the elderly dependency ratio.
Morbidity Patterns
There is considerable conjecture and controversy regarding fu-
ture morbidity patterns. One theory holds that the improvements
in lifestyle will delay the onset of disability and will result in a re-
duction in the prevalence of morbidity from chronic disease and a
OCR for page 60
60
AGING POPULATION IN THE TWENTY-FIRST CENTURY
compression of morbidity at older ages. This theory foresees a con-
tinuing decline in premature death and the emergence of a pattern
of natural death at the end of a natural life span (Eries, 1980~. An-
other theory argues that chronic disease prevalence and disability
will increase as life expectancy increases, leading to a pandemic of
mental disorders and chronic diseases (Kramer, 1980~. This theory
projects that the extension of life ~ bringing an extension of disease
and disability. The increases in the prevalence of chronic conditions
due to medical technology have been called the failures of success
(Gruenberg, 1977~.
A recent review of the evidence (Schneider and Brody, 1983) con-
cludes that the number of very old people is increasing rapidly, the
average period of diminished vigor will probably rise, chronic disease
will probably occupy a larger proportion of people's life spans, and
the needs for medical care in later life are likely to increase substan-
tially. Models linking morbidity and mortality can be developed to
predict how healthy or ill cohorts of the older population will be in
the future (Manton, 19823.
THE LIMITATIONS O1? AVAILABLE DATA
Needs for data in several areas related to the demographic and
socioeconomic trends described can be identified. The sources of
currently available demographic and socioeconomic data on the aging
population are the decennial census and sample surveys, including
the Current Population Survey (CPS), the Survey of Income and
Program Participation (SIPP), and the National Health Interview
Survey (NHIS). These surveys have sample sizes that are too small
to provide detailed age breaks and characteristics of subgroups of the
elderly population. Analytic studies of the oldest-old, for example,
are severely hampered because the size of the sample age 85 and
older in these surveys is not large enough to provide a socioeconomic
profile of this growing segment of the elderly. Future planning to
meet the needs of this age group, especially the need for institutional
care, will require more in-depth knowledge of the sex, race, marital
composition, and living arrangements of the oldest-old, along with a
more detailed income and wealth profile than is currently available
from survey statistics. A major requirement is to obtain a large-
enough sample of the elderly to provide detailed age and subgroup
characteristics.
Detailed data on income, wealth, and pension statistics for the
OCR for page 61
SOCIAL, ECONOMIC, AND DEMO GRAPHIC CHANGES
61
elderly are essential for policy analysis. Better income measures for
example, the use of income distributions rather than averages- are
needed to more accurately assess the economic status of the older
population. Noncash benefits also need to be quantified in determin-
ing the financial well-being of elderly persons. And since employment
status has a direct bearing on economic resources, various measures
of income should separate the older population into working versus
retired categories. In addition to income and wealth data, there is a
need for data on detailed pension programs and retirement income.
A related issue is the need to analyze retirement trends in terms
of a measure such as the average age at retirement. Better income
data and measures will provide a more accurate picture of the finan-
cial well-being and economic resources of the elderly and also aid in
planning public and private retirement, health, and social programs.
Methods of forecasting income at the time of retirement also need to
be developed with those planning efforts.
:Lim~ted data are now available on the pension prospects of
women. Due to women's increased labor force participation, the
situation of women as they turn 65 in future years is likely to be
quite different in terms of anticipated future income and pension
coverage and benefits. In order to forecast the pensions of women,
there is a need to monitor the pension benefits being accrued by the
more recent cohorts of women.
The increased labor force participation of women implies greater
financial prospects from pension benefits on one hand, and lesser
availability to provide a caregiver role for age parents on the other.
With the rapid growth of the oldest-old, there is a greater likelihood
for the young-old to have extremely aged parents in need of care.
Fewer women will be available to provide such support and greater
reliance wall shift to formal caregiver and support systems.
Forecasts of the older population are especially useful for Tong-
range planning, and several types of demographic and socioeconomic
data are needed for more accurate and useful projections of the
elderly. First, greater age detail at extreme ages (i.e., over 75 or
80 years) is needed, which would allow analysm of such factors as
institutionalization (Myers, 19853. Second, forecasts of the living
arrangements and future housing needs of the elderly are needed to
aid in planning efforts in the areas of publicly subsidized housing and
institutional facilities. Projections of both the noninstitutionalized
and institutionalized long-term care populations are needed.
Data on ethnicity and the foreign-born population of the elderly
OCR for page 62
62
AGING POPULATION IN THE TWENTY-FIRST CENTURY
have been largely ignored. Although the proportion of foreign-born
elderly persons is expected to decline ~ future years, the need to
study changes in the foreign-born composition of the older population
arises due in part to different needs of such subgroups.
Greater geographic detail in national projections is also needed,
along with forecasts of migratory flows of the elderly (Myers, 1985~.
Better data are needed on migration and mobility of the elderly and
the subsequent population redistribution, which is especially valu-
able in projection work. The need to better forecast migratory flows
for states and local areas coincides with the need to know about the
concentration or distribution of the elderly population. The following
questions become increasingly important for regional planning and
the provision of health and social services: Will there be a residential
turnover among the elderly, from central cities to suburban areas?
What is the likelihood of certain suburbs' becoming essentially el-
derly communities? Information related to such questions can help
regional planners decide where to put facilities such as nursing homes
and board and care homes. Knowledge about interstate migration as
well as residential mobility of the elderly is important for projection
purposes.
Short-term projections (to 2000) show a regional shift among the
elderly from the Northeast and North Central regions to the South
and West regions. In a detailed study of migration patterns of the
elderly based on decennial census data, Longino observed that even
while migration into the sunbelt states continued, outmigrants from
Florida to northern states were characterized by high proportions of
persons age 75 and over returning to their state of birth (Longing
et al., 19843. Study of migration streams by age can contribute to
more accurate projections, and thereby to better state and regional
planning.
The population ages 45-64 today will become the "new aged"
population in the next 20 years, and longitudinal data are needed to
monitor this group. These cohorts require special attention, as they
represent cohorts very different from the current elderly population
in terms of educational, marital, income, and perhaps health char-
acteristics (Myers, 1985~. Changes ~ composition, behavior, and
needs of the future elderly can best be foreseen by analyzing the
differences between newly entering cohorts of the elderly and their
immediate predecessors (Serow and Sly, 1985~. Serow contends that
the group ages 55-64 in 1980 is critically different in composition and
life-course experiences, separating the elderly of yesterday from the
OCR for page 63
SOCIAL, ECONOMIC, AND DEMO GRAPHIC CHANGES
63
elderly of the future. This leads to several implications for the later
years of life for the new aged. The future elderly are also expected to
have relatively higher incomes and greater assets, along with better
health. On the basis of the changes in the labor force experiences of
women noted earlier, future generations of women can be expected
to enter their retirement years with greater financial resources, from
their own pension and Social Security entitlements, but with fewer
familial resources to provide necessary support. Although tomor-
row's elderly may have an improved financial position, the size of
the future population will increase the need for long-term home care
and may require a greater supporting role by society. As the elderly
population ages, the number of nursing home residents is projected
to increase by 60 percent by 2003.
Successive cohorts of the elderly are projected to have both in-
creased educational attainment and higher incomes. As educational
level has been shown to be associated with various measures of health
status as well as demands for better health care, it becomes increas-
ingly important to study the implications of a more highly educated
older population. Higher income levels of tomorrow's elderly imply
a greater ability to pay for better health care for acute conditions.
Data on changing conditions for future generations of the elderly also
need to incorporate the implications of housing and transportation
available to the elderly, as these factors affect people's quality of life,
which in turn affects measures of health and illness.
Related to data needs on the older population are several method-
ological issues. These areas include the development of appropriate
health status measures for major socioeconomic surveys, the linkage
of data bases, the need for longitudinal studies of socioeconomic and
health characteristics of the elderly, and studies of the elderly on an
age cohort basis.
Appropriate health measures need to be developed and incorpo-
rated into the major socioeconomic surveys. We need new approaches
to the development of measures of physical, cognitive, emotional, and
social functioning. Several measures are available and have been used
in national surveys. These need to be improved and broadened to
take into account the positive or successful aspects as well as the
negative aspects of aging and to reflect ordinary behaviors and activ-
ities of older persons that indicate their quality of life and affect their
relationships with those close to them (National Research Council,
1986).
Longitudinal studies of socioeconomic and health characteristics
OCR for page 64
64
AGING POPULATION IN THE TWENTY-FIRST CENTURY
of the elderly are especially relevant. Such data are essential for
assessment of transitions over the life cycle. Despite the physiologic
losses and psychosocial stresses often associated with advanced age,
many elderly individuals have the vitality and resilience to func-
tion effectively or to recover and function independently, once again,
following a disabling condition. Data are needed that measure the
extent to which older individuals remain in good health and the
changes that occur as they move from one state of health to another,
whether this marks an improvement or progressive loss of function
leading to disability, dependency and, ultimately, mortality. Measur-
ing this requires repeated observations on the same people over time,
i.e., longitudinal information on both the wed and the impaired in
the population (National Research Council, 1986~.
Data for the elderly on an age cohort basis are needed in addition
to those being collected on an age period basis. For example, the
onset of a particular disease or condition could vary by birth cohort.
Analysis of data on a cohort basis would thus reveal if a shift in the
age of onset has occurred for successive cohorts. Such data would
aid greatly in understanding tomorrow's elderly.
Data base linkage could be invaluable to research on the elderly.
Separate analyses have been conducted based on survey data, medi-
cal records, and administrative data. By linking these sources, more
detailed analyses will be possible along with the testing of new rela-
tionships. For example, linking Medicare files with social survey data
and medical records could provide information on Medicare use by
the availability of a social network or by the severity of the medical
condition. Such linkage could supply a more comprehensive health
profile of the elderly.
IMPLICATIONS 1?0R THE STUDY
The social, economic, and demographic changes in successive
cohorts of elderly people highlighted in this chapter have implications
for the topics discussed in each of the chapters in this report. Indeed,
they helped determine the topics selected for discussion.
Representative terms from entire chapter:
labor force
