Why do Death Rates Decline?

"Since 1960, mortality reductions have been associated with two newer factors: the frequent conquest of cardiovascular disease in the elderly and the prevention of death caused by low birth weight in infants."

During the twentieth century, mortality rates declined quite rapidly in the United States and in all developed countries. In 1900, the annual mortality rate was one in 42 Americans. In 1998, on an age-adjusted basis, the rate had dropped to one in 125 people. That's a cumulative decline of 67 percent. In Changes in the Age Distribution of Mortality Over the 20th Century (NBER Working Paper No. 8556), co-authors David Cutler and Ellen Meara explore how we achieved such gains in health: that is, which innovations or policies contributed most to these gains.

Except for a 10-year period between 1955 and 1965 when the mortality rate was essentially flat, mortality rates have declined at the relatively constant rate of approximately 1 to 2 percent per year since 1900. That mortality reduction used to be concentrated at younger ages, but then became increasingly concentrated among the aged. In the first four decades of the twentieth century, 80 percent of life expectancy improvements resulted from reduced mortality for those below age 45, the bulk of these for infants and children. In the next two decades, life expectancy improvements were split relatively evenly by age group. In the latter four decades of the century, about two-thirds of life expectancy improvements resulted from mortality reductions for those over age 45.

During the first half of the century, changes in the ability to avoid and withstand infectious diseases were the prime factors in reducing mortality. Infectious diseases were the leading cause of death in 1900, accounting for 32 percent of deaths. Pneumonia and influenza were the biggest killers. Therefore, improved nutrition and public health measures, particularly important for the young, were vastly more important in this period than medical interventions. Better nutrition allowed people to avoid contracting disease and to withstand disease once contracted; public health measures reduced the spread of disease. During this period, reduced infant mortality contributed 4.5 years to overall improvements in life expectancy; reduced child mortality contributed nearly as much, and reduced mortality among young adults added about 3.5 years.

Between 1940 and 1960, infectious diseases as a cause of death continued to decline. But more of this decline was attributable to medical factors, such as the use of penicillin, sulfa drugs (discovered in 1935), and other antibiotics. These help the elderly as well as the young, thereby reducing mortality across the age spectrum. By 1960, 70 percent of infants could be expected to survive to age 65.

Since 1960, mortality reductions have been associated with two newer factors: the frequent conquest of cardiovascular disease in the elderly and the prevention of death caused by low birth weight in infants. Traditional killers such as pneumonia in the young also have continued to decline, but mortality from these causes was already so low that further improvements did not add greatly to overall longevity.

Increasingly, mortality reductions are attributed to medical care, including high tech medical treatment, and not to social or environmental improvements. Smoking cessation and better diets also are factors: per capita consumption of cigarettes rose from essentially zero in 1900 to more than 4,000 per year per capita in 1960, or over two packs per smoker per day. Since then, per capita consumption has fallen by more than 50 percent. These trends affect death from heart disease and from smoking-sensitive cancers with a 10 to 20-year lag.

For several important causes of death, rising incomes and a variety of social programs have accompanied significant reductions in mortality. Higher incomes make possible the use of expensive medical technology and are correlated with less smoking than lower incomes. Medicare increases the access of the elderly to medical care while Medicaid does the same for the poor. A large increase in the incomes of the elderly stemming from pre-Social Security social programs and the phase-in of the Social Security system has coincided with suicide rates for that group dropping 56 percent since 1930.

Finally, the authors make the broad point that dramatic swings in the ratio of non-white to white infant deaths over the century suggest that a variety of factors influence health differentially for the different races.

-- David R. Francis

The Digest is not copyrighted and may be reproduced freely with appropriate attribution of source.

National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email:

Contact Us