Background and senescent component of adult mortality and its connection to mortality shifting

Klara Hulikova, Charles University in Prague

Because of the already very low level of mortality at lower ages, the mortality analysis started to be focused mainly on higher ages and the population group of the “oldest-old”. Within the analysis of the force of mortality the decomposition of the total mortality into two different components could be seen to be applied (among other methods) – decomposition to the senescent and background mortality. Both these components could be studied separately. In some papers the decomposition is based on the application of the logistic model, in the others on the Gompertz-Makeham formula. Usually this type of analysis was applied to the data from developed and mostly European countries. This paper follows the issue of senescent and background mortality and extends the analysis to wider range of countries data, also on the post-communist countries. Moreover, where possible, longer time interval is considered in comparison with previously published works. The shifting hypothesis, involved in this paper, is closely related to the previously mentioned decomposition of the total mortality because it describes the shifts of the mortality curve not only as a whole but also separately the shifts of the senescent mortality curve only. The first goal of this paper is to show the differences in the development of background and senescent mortality between post-communist and other developed countries involved in the analysis. Then the paper deals the topic of the shifting hypothesis of the senescent mortality. It is shown that for more historical data one important assumption couldn’t be taken as true and “age-specific-shifts” were defined and estimated. Analyses in this paper are based mostly on the usage of data from the Human Mortality Database. For the analysis almost all the countries with data available were used.