%0 Journal Article %J Archives of Public Health %D 2015 %T Mapping the cause-specific premature mortality reveals large between-districts disparity in Belgium, 2003-2009 %A Françoise Renard %A Deboosere,P. %A Jean Tafforeau %K a %K age %K ALL %K association %K at %K Belgium %K cause specific %K cause-specific %K Class %K Clustering %K conditions %K conference %K Correlation %K Correlations %K data %K Design %K Discussion %K disparities %K distribution %K district %K Ecological %K ECONOMIC %K European %K Expectancy %K Flanders %K Flemish %K Geographical %K Geographical-distribution %K health %K Health indicator %K HEALTH POLICIES %K HEALTH POLICY %K Heterogeneity %K Impact %K Income %K Income inequality %K Indicator %K inequalities %K inequality %K intervention %K interventions %K IS %K LEVEL %K Life %K Life expectancy %K Male %K males %K mapping %K method %K methods %K mortality %K Mortality rate %K Mortality rates %K Mortality-rates %K observed %K ON %K Paper %K pattern %K PATTERNS %K period %K POISSON %K Poisson regression %K POLICIES %K POLICY %K POPULATION %K premature %K Premature mortality %K PROGRESSION %K public %K public health %K Public-health %K RATES %K Ratio %K Ratios %K region %K regional %K regression %K Research %K result %K results %K scale %K Socio economic %K Socio-economic %K Socioeconomic %K Statistics %K Still %K Test %K unemployment %X Background: Belgium has a well-known regional pattern in life expectancy with higher life expectancy in the northern Flemish region. The aim of this paper is to analyse in how far this pattern is reproduced in premature mortality (1-<75 yr) in males, a health indicator sensitive to both health policy interventions and socio-economic conditions. Methods: Data on mortality, population and unemployment are provided by Statistics Belgium. We used maps to explore the geographical patterns of premature mortality, and look at the ecological association with socio-economic characteristics at the district level: age-adjusted premature mortality rates in males were computed and mapped at district level for the periods 1993-1997 and 2005-2009. Rates are grouped into classes with a geometrical progression of 1.1 and are represented with a diverging scale of colours. The disparity between districts was measured with the decile ratio (P90/P10). The association with the geographical distribution of unemployment was examined both visually and tested with a Poisson regression.Results: The clear geographical pattern with lower premature mortality rates in the North (Flanders) is confirmed. The highest premature mortality rates are observed in the South-West districts. The pattern is persistent over the 2 periods, with more heterogeneity in the 2005-2009 period, the between-district decile ratios being 1.56 and 1.75 respectively in the 1 %B Archives of Public Health %V 73 %8 0/0/2015 %G eng %N 13 %1 38004 %R http://dx.doi.org/10.1186/s13690-015-0060-5