Is sex ratio at birth changing in Italy? Spatial and temporal trends’ analysis

Francesca Rinesi, Università di Roma "La Sapienza"
Antonella Pinnelli, Università di Roma "La Sapienza"
Sabrina Prati, Instituto Nazionale di Statistica (ISTAT)
Francesco Lagona, Max Planck Institute for Demographic Research

Introduction and aim of the study: In societies where no prenatal sex selection have been observed, male to female ratio at birth (secondary sex ratio) is assumed to be relatively stable and it averages between 105 and 106 boys per 100 girls. Notwithstanding, some recent studies showed that in the last decades there has been a slightly but significant change in this ratio both considering Countries as whole and regions or small areas. A declining trend has been observed in Denmark, the Netherlands, Canada and US, whereas a research carried out in 29 Countries reveals mixed results. The sex of newborns is dictated by biological and genetic factors. At the same time there are exogenous variables that can affect them such as maternal and paternal age, parity, ovulation induction, socioeconomic status, living standards, quality of medical care and environmental factors. The aim of this study is to analyse temporal trend (from the end of Sixties to 2008) not only for Italy as a whole but also for sub national areas by considering explicitly the impact of the main demographic variables connected with reproductive behaviour and controlling for other confounding factors. Data and Method Data from Population Registers are used: they gives information on the main characteristics of births (sex, date and place of birth, nationality) and parents (date and place of birth, nationality, marital status) at municipal level. Environmental factors are considered to be crucial in understanding the recent trend in offspring sex ratio. However no accurate data on air pollution are available at sub national level: therefore regional and provincial data on mortality for specific causes of death will be used as proxy variables. Finally, 1-level mixed Poisson regressions will be estimated to disentangle the effects of demographic and health variables on sex ratio at birth.