The Oldest Old

views updated


As many nations of the world experience aging populations, attention is increasingly turning to changes within the older population itself, especially changes in the age structure of this older population. As a general statement, over time the elderly population is likely to become older; such a phenomenon is important because there are likely to be important differences among age groups within the older population, such as health status and economic well-being.

The aging of the older population occurs primarily because of reductions in death rates, which in many countries have become concentrated in the older ages. In the United States, for example, life expectancy at birth increased by more than 50 percent during the twentieth century; not surprisingly, most of the increase was due to reduced mortality among children and the elderly. However, almost all of the declines in mortality among children occurred prior to 1950, while declining mortality for the elderly occurred much more recently, during the final decades of the twentieth century. Thus, while population aging (the number or share of the population that has reached some specified birthday, usually, the sixtieth or sixty-fifth) is mostly due to reductions in fertility, the aging of the older population and the rising number and share of the "oldest old" (conventionally those at least eighty-five years of age) is mostly due to reductions in old-age mortality.

To illustrate the dynamics of the oldest old population, consider census data from the United States, which first became available for this age group beginning in 1910. That census enumerated 167,500 very old Americans who accounted for about two of every thousand of the nation's 92 million persons and only 4 percent of the 3.9 million Americans aged sixty-five or older. By mid-century, there were 578,000 Americans who had reached their eighty-fifth birthday. Their share of total population, while still very small, was about twice what it had been in 1910. Current estimates of the very old population of the United States suggest nearly 4.3 million individuals or about fifteen of every thousand Americans. In the past half century, the number of oldest old Americans has risen more than seven fold while the national population has not even doubled. Forecasts by the U.S. Bureau of the Census for the next fifty years suggest 19.4 million very old Americans by 2050, nearly 5 percent of the projected national total (see Figure 1).

Gerontologists are particularly interested in these demographic trends because the oldest old have historically differed from the more general older population in many important respects. While it is not possible to state with any certainty that this pattern of differences will continue in the future, there are several interrelated issues that must be taken into account when contemplating the social and economic consequences of the demographic shifts summarized above.

Gender issues

The nexus of this issue is suggested by the sex ratio, or the number of men per one hundred women in a population. The oldest old are a largely female populationat present there are forty-three very old American men for every one hundred very old women, a level much lower than that found for the younger (persons sixty-five to eighty-four years of age) elderlyseventysix per one hundredor among the entire populationninety-six per hundred. Such a pattern, which would be characteristic of almost any national population, simply reflects the lifelong advantage women have with respect to mortality. Due to this mortality difference and the customary pattern of men marrying women a few years younger than themselves, the very old population is overwhelmingly composed of widows.

In countries such as the United States, where old-age pensions tend to be earnings based, and where women have only recently spent much of their adult life in the paid work force, the oldest old tend to fare relatively poorly from a financial perspective. Since a widow's pension income may diminish or disappear altogether, it is not surprising that nearly one in four very old women is impoverished, compared with less than one in six very old men, less than one in ten younger elderly, and about one in seven of all Americans.

Health status

The health status of the very old is also a matter of considerable interest, especially given that a significant share of their health-care expenses is paid with public funds. The proportion of the population with one or more limitations in the activities of daily living (ADL) tends to rise sharply with age. While nearly nine in ten persons aged sixty-five to sixty-nine reported no problems with ADLs, this was true for only about 40 percent of the oldest old. A closely related issue is the extent to which the oldest old population resides in a nursing home or similar institution, a living arrangement strongly conditioned by the health and marital status of the individual (persons with a surviving spouse are far less likely to be nursing home residents than are the widowed). Fewer than one tenth of Americans aged eighty to eighty-four resided in a nursing home when the 1990 census was taken, but the incidence of institutionalization was sharply higher among oldest Americans; for those age eighty-five and older, nearly one in four were in a long-term care environment. This issue is of great import to policy makers at the federal and state levels, since many nursing home costs are paid through the jointly funded Medicaid program.

The aging of the oldest old

Like the U.S. Census Bureau forecasts for a rapid expansion of the oldest old population, in both relative and absolute terms, forecasts by the United Nations Population Division and statistical agencies of various nations show a similar trend in practically all nations. In all likelihood, the aging process so characteristic of both overall and elderly populations will characterize the oldest old as well: currently two-thirds of those aged eighty-five and over are aged eighty-five to eighty-nine and less than 2 percent are at least one hundred. The population forecast for 2050 suggests that only half of that year's oldest population will be in the comparatively young range of eighty-five to eighty-nine, while more than 5 percent will be centenarians.

While these future numbers are uncertain, they should nevertheless give one considerable pause for reflection. Practically all of the characteristics that have been summarized here are correlated with age: widowhood, poverty, and ADL problems are not only more common among the oldest old than among younger individuals, but they are more prevalent among those aged ninety to ninety-four than among those aged eighty-five to eighty-nine, among those aged ninety-five to ninety-nine than among those aged ninety to ninety-four and so on. From a simple cross-sectional perspective, then, the increase in the numbers of the oldest old and the aging of this group would seem to augur poorly for society.

Outlook for the future

However, the outlook is not necessarily gloomy. Like any age group in a population, the oldest old at a specified point in time comprises individuals with differing lifetime experiences. Over time, the lifetime experiences of those currently aged eighty-five and over will be changing. Thus, the characteristics of the oldest old cannot be viewed as fixed, either absolutely or relative to the norm of some other age group. This process, known as cohort succession suggests that the oldest old of the future will differ from that of the present in some material respects. While health status deterioration and widowhood seem inevitable consequences of advancing age, the same need not be true for poverty.

Oldest old women of the future will generally have much more lifetime labor market experience than do their counterparts today and, accordingly, greater access to pension income in their own right. The availability of such pension income greatly reduces the prospect of poverty. Furthermore, the amount of education an individual has attained will condition her occupation and earnings history. In the aggregate, a person with more education will have had access to better paying jobs with greater benefits throughout her working life. As the oldest old of 2000 (persons born prior to the World War I) are gradually replaced over the next several decades first by the parents of the baby boom generation and then by the baby boomers themselves, the average level of education completed will rise from less than completion of secondary school to nearly one year of college.

In summary, it seems inevitable that those aged eighty-five and over will grow in numbers and proportions for the foreseeable future. At present, the economic well-being of this group is generally inferior to that of younger persons. This situation need not persist, at least to the same degree, over time because the process of cohort succession guarantees that the oldest old of future decades will differ from those of the present in several respects, including more favorable financial prospects.

William J. Serow

See also Centenarians; Cohort Change; Longevity; entries on Reproduction, Selection, Social Aspects; Population Aging.


Grundy, E. "Demography and Gerontology: Mortality Trends among the Oldest Old." Ageing and Society 17 (1997): 713725.

Perls, T. "The Oldest Old." Scientific American 272 (January 1995): 7075.

Serow, W. J., and Sly, D. F. "Trends in the Characteristics of the Oldest-Old: 1940 to 2020." Journal of Aging Studies 2 (1998): 145156.

Suzman, R. M.; Willis, D. P.; and Manton, K.G. The Oldest Old. New York: Oxford University Press, 1992.


See Federal agencies and aging; Internet resources; Political behavior; Professional organizations; Volunteer activities and programs

About this article

The Oldest Old

Updated About content Print Article