Socioeconomic mortality differences in Italy – new insights from indirect orphanhood-based estimates

Marc Luy, Vienna Institute of Demography
Paola Di Giulio, Vienna Institute of Demography
Graziella Caselli, Università di Roma "La Sapienza"

The study of socioeconomic differences in mortality, especially mortality differences by occupation, is of great interest not only to evaluate potential risks related to the type of job performed but also for policy purposes. Nonetheless, in many countries the quantification of occupational mortality differences is still a problem of difficult solution. In this paper we propose the use of the so-called “orphanhood method” to study mortality differences by socioeconomic status in Italy. Specifically we analyse the mortality of the parents of the interviewed people. Data stem from the multipurpose survey “Family and social subjects” carried on by the Italian National Statistical Institute in the years 1998 and 2003. Both surveys are representative at the repartition level. The sample case numbers are over 54,000 respondents in 1998 and almost 50,000 in 2003. The data contain information about the parents of the respondents, such as their survival status, their highest education level reached and several characteristics of their job activity when the respondent was about 14 years old. The latter aspect provides an especially interesting information since it enables to link the mortality experiences of respondents’ parents with their job in a way that is independent of the current age of the parents. With the orphanhood method we estimate a series of period life tables by characteristics of jobs, with the corresponding reference periods. Thus, we obtain a time series of education- and occupations-specific life tables for women and men that enable both, an analysis of more recent socioeconomic mortality differences in Italy and their development over some decades. Moreover, the use of the orphanhood method provides estimates for differences in life expectancy. For most people interested in socioeconomic mortality differences this kind of information is more useful than the traditional calculation of relative risks.