The spatial epidemiology of alcohol-related mortality among the Czech adult population

Abstract

Abstract Background In Czechia, alcohol-induced deaths account for a significant portion of preventable mortality. As inequalities in health are both socially and spatially determined, the paper aims at the detailed examination of socio-geographic inequalities of this phenomenon. Methods The 2011-2015 annual data on both ICD-10 cause-specific deaths (K70; F10; X45/64; Y15) and mid-year population were obtained from the official Czech registries - the data were cross-classified by gender, 5-year age-groups, and permanent residence (N = 6,302 small area spatial units). The selected socio-demographic indicators (education, unemployment, religious population) from the Czech 2011 Census were spatially merged to the mortality dataset. From the data on education and unemployment, composite deprivation index (DI) was derived. In the adult population aged 25+, the age-standardised mortality ratios (SMR) were computed for each of the spatial units, separately by genders. The SMRs were spatially modelled by the Besag-York-Mollié (BYM) autoregressive approach, applying a fully bayesian framework integrated within the INLA R-package. The study applied cross-sectional design and employed ecological regression conducted on observational data. Results Compared to the Czech average, the highest SMRs were located in the historical regions of Moravia [SMR=1.15; 95%CI: 1.11-1.19] and Silesia [SMR=1.59; 95%CI: 1.52-1.66]. The SMRs were significantly correlated with DI among males [Rel.Risk=1.15; 95%CI: 1.11-1.19], and with religiousness rate among females [Rel.Risk=0.83; 95%CI: 0.77-0.90]. Conclusions Significant socio-geographic inequalities were detected, particularly with respect to the Czech historical regions. Among males, higher mortality was associated with a structural deprivation. Among females, protective effect of religiousness rate was found to be significant. The results highlight an importance of both socially and spatially integrated efforts for public health promotion. Key messages The inequalities in health are both socially and spatially contextualised. The paper presents robust empirical evidence in favour of the proposition, as examined on alcohol-related mortality data. The health determinants may be gender sensitive. Males might be more responsive to a structural disadvantage. Among females, cultural factors related to a local community might be more relevant.