How Many Americans Are Alive Because of Twentieth-Century Improvements in Mortality?
Citation: White, Preston, Samuel H. (1996) How Many Americans Are Alive Because of Twentieth-Century Improvements in Mortality?.
Internet Archive Scholar (search for fulltext): How Many Americans Are Alive Because of Twentieth-Century Improvements in Mortality?
Tagged: uw-madison (RSS), wisconsin (RSS), sociology (RSS), demography (RSS), prelim (RSS), qual (RSS), WisconsinDemographyPrelimAugust2009 (RSS), health progress (RSS)
Notes: Introduction: Because of rapid mortality decline during the twentieth century, life expectancy in the U.S. increased from 47.3 years in 1900 to 75.7 years in 1994. In this article, White and Preston "estimate the number of Americans currently alive who literally owe their lives to health progress. This calculation involves comparing the actual population to a hypothetical population in which mortality improvements did not occur" (p. 416). Methods: "To estimate what the population of 2000 would look like without the twentieth-century mortality improvements, the population of 1900 is projected forward using the U.S. life table of 1900. We use the standard component method of projection in five-year intervals of time and age. . . The rates of fertility and volume of immigration are assumed to be unaffected by the level of mortality. . . "(p. 416). The number of immigrants are estimated for each period and projected forward using U.S. fertility and mortality rates. In addition to the two main time series, one that uses actual demographic circumstances and one that keeps mortality fixed at 1900 levels, we apportion the difference between these two by generation. For example, people alive in 1905 because of health advances between 1900 and 1905 can be identified by age and sex. This hypothetical group is then projected forward to 2000 under actual conditions of fertility and mortality. Their offspring are people who would not have been born if mortality had not improved in the first five years of the century. . . The size and traits of each of these hypothetical populations in 2000 enable us to disentangle by 'generation' the persons living by virtue of twentieth-century mortality declines. (p. 417). Data are taken from a variety of sources, including the Social Security Administration, Census Bureau, and published studies which attempt to correct flawed national statistics on deaths, population and fertility from 1900 forward.
Results: "If mortality had remained at 1900 levels throughout the century, holding everything else constant, the population in the year 2000 would be almost exactly half its actual size: 139 million people instead of 276 million. Half of Americans living today can attribute their being alive to mortality improvements in the twentieth century. . . Mortality reductions during the first half of the century had a much larger impact on population size in 2000 than did reductions during the second half. If mortality rates beyond 1950 were fixed at 1950 levels, the US population in 2000 would still be 94 percent of its actual size, compared to 50 percent if mortality were fixed in 1900. Mortality declines in the first half of the century were concentrated in childhood and young adulthood. Hence their impact cumulated as those whose deaths were prevented bore children" (p. 420-422). Half of the "hypothetically dead" in 2000 are direct deaths (i.e., people who would have died without mortality improvements), but half are indirect deaths-people who never would have been born. Direct deaths are concentrated at ages above 75, whereas indirect deaths are concentrated below age 30. The structure of the population was not greatly affected by the mortality decline, except at young and particularly at old ages. As a result, the actual mean age of the population in 2000 is only .8 years greater than the projected mean age. Mortality reductions in the age interval 0-14 were responsible for 67 percent of the difference in size between the actual population and the hypothetical population. This demonstrates that saving lives at younger ages affects population growth via subsequent fertility. Finally, the sex ratio [male to female] for the entire population would have been 1.01 under 1900 mortality conditions, not the actual value of .98. "This small change masks much larger changes in certain age intervals, which counteract one another for the total population. The actual sex ratio of .38 at ages 85 and up would have been .69 in the absence of twentieth-century mortality decline. . . Improvements in infant and childhood mortality have worked in the opposite direction, however. In 1900, 15 percent of males but only 12 percent of females died before their first birthday. Fewer, than 1 percent of each sex now dies in infancy. . ." (pp. 427-428). This indicates that mortality improvements have disproportionately benefitted males at younger ages and females at older ages.
Conclusions: "Half of the population would not be alive in the absence of mortality improvements, including seven-eighths of women over age 84. The greater survival of people at very young and very old ages increased the dependency ratio by 22 percent. And women are a majority of the population because of mortality declines" (p. 427).
Theoretical and Practical Relevance
A dramatic demonstration of the effect of health progress on population!