Methodological alternatives for measuring premature mortality.

Although crude and age-adjusted mortality statistics are frequently used to quantify public health problems, they are heavily influenced by the underlying disease processes of the elderly. Alternative measures have been developed to reflect the mortality experience of younger age groups (i.e., premature mortality). We evaluated four different methods for tabulating premature mortality, one method weighted by the remaining life expectancy at death and three methods with constant end points using age spans from birth to 65 years, birth to 75 years, and 1 to 65 years. These alternatives provide dramatically different descriptions of premature mortality in the United States in 1984. In general, the constant end-point methods emphasize the different pattern of mortality among younger persons, while premature mortality computed by the remaining life expectancy method more closely resembles the pattern of crude mortality. Although no single method is preferable for all purposes, the constant end-point method best differentiates the leading causes of premature death.