Mortality in a Selected Population Subgroup: The Case of Brazilian Air Force Military Personnel

Vanessa di Lego, Vienna Institute of Demography

A great deal of research has used military data to examine survival under extreme conditions, such as war and famine, as well as the mechanisms responsible for mortality differentials over the life cycle (Buzzell and Preston 2007;Costa2010;Horiuchi1983). However, in countries that have not been involved in internal conflicts or foreign wars, military data may help to understand mortality differentials and the extent of inequality, particularly in places where substantial disparities in health persist. In Brazil, a free war country since the WWII, military officers have to go through a strict health selection process during enlistment. Also, they live in better conditions than the average citizen, enjoying stable economic status and having access to high-quality diet, healthcare, education, and social support (Castro2000; MoD2012). At the same time, the tradition of the military in collecting vital data is particularly valuable in Brazil, where accurate data to measure mortality patterns among population subgroups are usually lacking (Turra2016). The aim of this paper is to primarily update the estimates we have performed in a previous work by including not only a wider timeframe, but also other career types, and discuss the implications of those measures with cross country comparison among countries from the Human Mortality Database, using data on military personnel who enrolled in the institution from 1943 until 2000. It is composed of 37,026 living persons and 7,391 deceased individuals. The median follow-up time is 38.39 years, with a maximum follow-up of 72.91 years. We employ a smoothing spline Poisson regression model as a function of age for the analysis of mortality data and computing smoothed mortality rates with confidence intervals. This model is a robust non-parametric approach that uses penalized likelihood estimation method and provides approximate Bayesian posterior confidence intervals (Gu2014; Gu1993).

Presented in Session 1214: Environment, Development, and Space