Sci. Aging Knowl. Environ., 15 May 2002
Vol. 2002, Issue 19, p. nw62
[DOI: 10.1126/sageke.2002.19.nw62]


Forecasting Long Life: How low can human mortality go? (Demography)

Kendall Morgan;2002/19/nw62

Key Words: demography • life expectancy • longevity • survivorship • mortality rate

Abstract: Just as cars driven on highways outlast those taken off-road, people with modern resources to pave over life's bumps keep on rolling. Scientists have wondered, however, whether human longevity can climb forever. Previous work claimed that the limits were fast approaching, but new findings challenge that notion. The results predict that the number of centenarians will continue to swell and urge a reevaluation of social needs.

The idea that boundaries to long life lurk just around the corner is elderly. In 1928, demographer Louis Dublin, who became Chief Actuary of the Metropolitan Life Insurance Co., took the first stab at predicting maximum life expectancy, putting it at 65 years, based on evidence from the United States. He didn't realize that New Zealand women had already topped that figure. In 1990, S. Jay Olshansky, a demographer at the University of Illinois, Chicago, and colleagues predicted that life expectancy for men and women would stall at 85. Last year, when women in Japan reached that bar, he raised it to 88. Although life expectancy has repeatedly shattered records, the notion that longevity is hitting a wall persists.

Oeppen and Vaupel suspected that limits were made to be broken, so they brought historical evidence to bear on the question. If the bounds are looming, then the country with the longest lived people would be closest to the cutoff, they reasoned. Therefore, record life expectancy--held by different countries at different times--should be leveling off. Instead, they found, the lifetimes of the top-ranking country's citizens have increased at a constant rate for the last 160 years, with no hint of slowing. On average, people born in a given year live 3 months longer than those born the previous year. Record life expectancy could climb to 100 years within the next 6 decades, according to the researchers' extrapolations.

"It's a striking result," says demographer Shripad Tuljapurkar of Stanford University in California. The findings show that medical and social improvements can be effective consistently over the long term, he says. Olshansky agrees that the researchers make a good argument based on the numbers, "but it's a biological phenomenon, not a mathematical one," he says, and the future can't continue to resemble the past. Before 1950, life expectancy rose primarily due to a decline in infant mortality. Future jumps will occur by tacking on additional years late in life, a more difficult challenge, he explains.

An increase in life expectancy historically doesn't necessarily imply that the pattern will hold, concedes demographer Ronald Lee of the University of California, Berkeley, but it does "send a loud message." If the trend Oeppen and Vaupel point out continues, the Social Security Administration's projection for life expectancy in 2075 would rise by 13 years. "Following that path would require a major restructuring in retirement age or a tax increase for working people," he says. So it might be wise to put a little extra in the bank. Even if the path to immortality reaches a dead end, the new work suggests that it won't hit that roadblock soon.

--Kendall Morgan

J. Oeppen and J. W. Vaupel, Broken limits to life expectancy. Science 296, 1029-1031 (2002). [Abstract] [Full Text]

Citation: K. Morgan, Forecasting Long Life: How low can human mortality go? (Demography). Science's SAGE KE (15 May 2002),;2002/19/nw62

Science of Aging Knowledge Environment. ISSN 1539-6150