Spatial Modeling of Mortality of the Working-Age Population in the Republic of Bashkortostan

Abstract

Introduction: The persisting high mortality rate among working-age adults affects both the production potential of the country and the prospects for achieving the goal of increasing healthy life expectancy, as set by the National Demography Project. Variations in mortality are attributed not only to structural differences in the population (sex, age, etc.) but also to significant differences in regional living and working conditions, thus necessitating a more careful study of spatial factors of its growth. Objective: To conduct a spatial analysis of mortality among the working-age population and to identify its determinants based on data for the Republic of Bashkortostan (RB). Materials and methods: The information basis of the study is a balanced panel of mortality rates and their socio-economic determinants for 74 areas (54 districts and 20 cities) and 19 time spans (2002–2020), formed for men and women separately. The Moran’s spatial autocorrelation coefficient and the spatial panel regression modeling were used for data analysis. The neighborhood matrix was used as a weight matrix of spatial connectivity of territories. Results: Our findings showed a pronounced spatial autocorrelation (p < 0.001) for mortality of the working-age population of both sexes. An increase in the gross municipal product per capita significantly (p < 0.001) reduced the overall mortality rate in both men and women of working age while an increase in the local crime rate, on the opposite, significantly increased it. Sex-specific differences were established in the impact of availability of medical resources on working-age mortality: for women this factor was found to be insignificant. Conclusions: The mortality of the working-age population has a non-local, but a spatially dependent nature.