Lives saved in a population

Vladimir Canudas-Romo, Johns Hopkins University

The number of lives saved and the ages when this occurs is the focus of this paper. For presentation, I have divided the manuscript into two complementary parts. First, I study the change over time in life expectancy in term of lives saved, as alternative to the usual interpretation of average number of years gained. This alternative interpretation of changes in life expectancy can be used to rank periods of heaps or of few lives saved in a country. The measure can also be used to compare across countries. Secondly, we look at the mortality shift in age at death. To assess this change I look at a fixed mortality level at a given time and find the matching age at which this level of mortality is found in a more recent year. As an example, a person aged 50 in 1950 in Sweden has the same mortality level than a person aged 80 in 2000. However, when looking at life expectancy level for a 50 year old in 1950 this is found at age 56 in 2000. The reason for this paradox is the great disparity in life extension existent from age to age. Similarly, lives saved can be just a handful, but their life duration will vary from age to age, and we combine the two ideas in this study.