Health and social determinants of health. Methodological test for detecting spatial disparities in health in the State of Bavaria, Germany

Abstract

The health disparities observed between different population groups, and different spatial-temporal structures at different scales, have been increasingly associated with social determinants of health. The main objective of this work aims at analysing the association and spatial correlation between Bavarian’s public health status and specific socio-demographic and socio-economic determinants. A spatial autocorrelation analysis was conducted to detect spatial clusters of health. It was also executed a multiple linear regression analysis to identify the most appropriate predictive variables of health outcomes. Further, in order to explain spatial correlation between health outcomes and social determinants of health, a geographically weighted regression model was performed. Statistically relevant clusters were found in districts of the northern region of Bavaria (high-high, high-low) and in the metropolitan region of Munich (low-low, low-high). The multiple linear regression model indicated that the number of persons with professional qualification (p=0.015e-7), the unemployment rate (p=0.0334332e-3), the youth ratio (p=0.0072) and the old-age ratio (p=0.012e-7) do have a significant influence on the health status of a population. The results of the geographically weighted regression show a circular pattern with its highest values in the north-eastern part (local R²=0.8077) and its lowest values in the south-western part (local R²=0.5865) of Bavaria. It can be stated that the aforementioned variables do slightly better predict poor public health than good public health. The greatest spatial disparities regarding health remain between the northern districts and the districts in the metropolitan region of Munich. Variables from the educative, economic and demographic dimension are able to explain these differences.