Mortality partitions and their relevance to research on senescence.

The reasons for classifying causes of death into aggregate groups are discussed and the impact of mortality partitions on analyses of mortality is described. Special emphasis is given to a mortality partition that distinguishes between intrinsic causes of death that arise primarily from the failure of biological processes that originate within an organism, and extrinsic causes of death that are primarily imposed on the organism by outside forces. Examples involving mortality data for mice, dogs, and humans are used to illustrate how this mortality partition infuses biological reasoning into mathematical models used to analyze and predict senescent-determined mortality, enhances the information content of the mortality schedules generated from these models, improves mortality comparisons between populations within species separated by time or geographic location, and provides a logical pathology endpoint for making interspecies comparisons of mortality. By bridging biology and the statistics of mortality, a mortality partition based on intrinsic and extrinsic causes of death provides both structure and direction for research on senescent-determined mortality.