Regional analysis of Occupational Diseases in the Automotive Industry in Czechia and the case study of Medical Geography RNDr. Jan Jarolímek, Ph.D., MBA prof. RNDr. Dagmar Dzúrová, CSc. 2015 – 2016, Grantová agentura Univerzity Karlovy Project summary: For fulfilling the national protection strategy, health promotion and illness prevention, it is necessary to know the distribution…
SOCIOECONOMIC INEQUALITIES IN MORTALITY
Socioeconomic inequalities in the spatial mortality distribution of the Czech urban population in the early 21st century
Principal investigator: mgr. L. Kážmér, doc. D. Dzúrová
Charles University Grant Foundation (GAUK) – 2013-2014
The project is aimed to identify socio-economic differentiation in mortality of the four largest Czech urban populations – Prague (1,257 mil.), Brno (371 thousand.), Ostrava (303 thousand.) and Plzen (169 thousand). It builds conceptually and methodically on results of the international INEQ-Cities project. For purpose of detailed analysis of mortality in its structural, temporal, and spatial context conditions within the inter-censual period of 2001-2011, new advanced statistical methods based on principles of generalized mixed modelling are applied, taking into account both the spatial structure and interdependence (autocorrelation) of spatial units (in literature often termed as Bayesian disease mapping methods).
Kážmér, L. 2014. Spatial-temporal Differentiation of the Mortality Structure by Causes of Death, Capital City of Prague, 2001-2011
Mgr. Ladislav Kážmér
Department of Social Geography and Regional Development
Faculty of Science
Charles University in Prague
128 43 Prague 2
Reviewers: RNDr. Boris Burcin, PhD., RNDr. Pavlína Netrdová, PhD.
The study was supported by the Charles University Grant Foundation Project No. 860213 Socioeconomic inequalities in the spatial mortality distribution of the Czech urban population in the early 21st century.
The full material of the study can be downloaded here [KÁŽMÉR, L. (2014)].
The study provides a detailed analytical view on the distribution of mortality of the population living in the Capital City of Prague within the period of 2001-2011. Mortality conditions were examined in relation to their structural, temporal and spatial characteristics. Both total and cause-specific aspects were applied in the analysis.
As the systematic monitoring of population mortality (and morbidity) conditions represent an important concern in both epidemiologic inquiry and public health policy agenda, research outcomes of this study can possibly serve as a starting point in addressing health inequality in structural as well as spatial context.
Recently, however, researchers tend to change the emphasis of their focus from national and regional level to intra-regional and local scale. Such a focus is, however, still underrepresented in the Czech Republic.
Thus, the aim of the paper is to give an in-depth analysis of structural and intra-urban spatial-temporal differences in mortality conditions of the population of Prague, evaluating the target population at a local scale. New advanced statistical methods were used in the study, based on principles of generalized linear mixed modelling and spatial data analysis.
Data were obtained from the routine death statistics of the Czech Statistical Office. New advanced statistical methods based on principles of generalized linear mixed modelling and spatial data analysis were applied in the analysis These so-called Bayesian mapping techniques were applied in order to take into account both the inner variance heterogeneity and the spatial structure of the analysed spatial units.
Stratified analysis were conducted within two separate (5-year) time periods: 2001-2005 vs. 2007-2010.
The age-standardized mortality ratios (SMR) were constructed separately for both genders and time periods using indirect standardization method. Current municipal districts of the Capital City of Prague (N=57) and their administrative boundaries were taken as spatial statistical units. High variability of the exposed population is characteristic to such intra-urban spatial areas. This results in unstable risk estimates (so-called inner variance heterogeneity problem). Because of that, the hierarchical Poisson-Gamma model (Clayton, Kaldor 1987; Lawson 2013) was applied in the next step in order to obtain smoothed mortality ratios (sSMR) considered as true unbiased risk estimates of the phenomena. Subsequently, these smoothed mortality ratios were applied within the spatial analysis of the phenomena (for full definition of both indirect standardization method and Poisson-Gamma model, see [Text] and [Appendix]).
In order to analyse temporal changes in the mortality of the total city population and compare it with the national population development, the annual age-standardized mortality rates (SDR) were constructed separately by gender (European Standard Population, 2013).
For the spatial clustering analysis and cluster detection, 2 global indicators and 1 local one were used. Regarding global indicators, general G statistics (Getis, Ord 1992) together with Moran´s I index (Moran 1950) were employed. The spatial cluster detection was evaluated by means of the Local Moran´s I index (Anselin 1995) (for detailed definition of spatial statistical indicators, see [Text] and [Appendix]).
The analysis was performed for the following population age-groups and leading causes of death:
– total population standardized mortality;
– economically active population (deaths between 15 and 65 yrs.);
– premature mortality (deaths before the age of 75 yrs.);
– senior population (deaths after the age 65 yrs.);
– malignant neoplasms (C00-C97);
– diseases of the circulatory system (I00-I99);
– diseases of the respiratory system (J00-J99);
– diseases of the digestive system (K00-K93);
– external causes of death (S00-T98);
– avoidable mortality (as the sum of treatable mortality, preventable mortality and mortality on ischaemic heart diseases) (for full definition, see [Table – Avoidable Mortality];
– treatable mortality;
– preventable mortality;
– mortality on ischaemic heart diseases (I20-I25).
– List of individual map sheets (see below)
– For interpretation and discussion of the particular map, see [Text].
Mortality conditions improved during the analysed period in the case of the capital city as well as in the country. However, the downward trend in the mortality rates has detectably slowed down in the Capital City of Prague in 2010. The analyses of the data from the next years will show whether this is only a temporary phenomenon or whether we should look for some important factor behind this stagnation.
Although the results indicate considerable spatial dynamics in the mortality by causes of death at the intra-urban spatial scale after the adjustment to both structural and temporal confounders, the spatial differences in the total standardized mortality seem to be relatively stable in both cross-sectional periods. Further investigation needs to be done in order to reveal the key factors underlying such intra-urban differences in the population of the Capital City of Prague (e.g. in relation to further (socio-) structural differences / confounders of the target population).
Anselin, L. (1995). Local Indicators of Spatial Association – LISA. Geographical Analysis, 27 (2), 93-115.
Burcin, B. (2008): Vývoj odvratitelné úmrtnosti v České republice v období 1990–2006 (Avoidable Mortality in the Czech Republic in 1990-2006). Demografie, 50 (1), 15–31.
Environmental Systems Research Institute (2013): ArcGIS Desktop 10.2. Redlands, California.
Eurostat (2013): Revision of the European Standard Population – Report of Eurostat’s Task Force. Eurostat Methodologies and Working Papers, European Commission, ISSN 1977-0375.
Getis, A., Ord, J.K. (1992): The Analysis of Spatial Association by Use of Distance Statistics. Geographical Analysis, 24 (3), 189-206.
Lawson, A.B. (2013): Bayesian Disease Mapping: Hierarchical Modeling in Spatial Epidemiology. Chapman & Hall/CRC Interdisciplinary Statistics, Second Edition, ISBN: 1466504811, 378 p.
Moran, P.A.P. (1950): Notes on Continuous Stochastic Phenomena. Biometrika, 37 (1), 17–23.
Newey, C., Nolte, E., McKee, M., Mossialos, E. (2004): Avoidable Mortality in the Enlarged European Union. Institut des Sciences de la Sante Statistics, Brussels.
R Development Core Team (2004): R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing,Vienna, Austria. URL http://www.R-project.org.
Rogerson, P.A., Han, D. (2002): The Effect of Migration on the Detection of Geographic Differences in Disease Risk. Social Science and Medicine, 55 (10), 1817-1828.
Spiegelhalter, D.J., Thomas, A., Best, N.G., Lunn, D. (2003): WinBUGS Version 1.4 User Manual. MRC Biostatistics Unit, Cambridge, United Kingdom. URL http://www.mrc-bsu.cam.ac.uk/bugs/.
Sturtz, S., Ligges, U., Gelman, A. (2005): R2WinBUGS: A Package for Running WinBUGS from R. Journal of Statistical Software, 12 (3), 1–16.
Tomášek, I., Tomášková, H. (2009): Použití bayesovských metod pro mapování onemocnění (Bayesian methods in diseases mapping). Demografie, 51, 190–201.
University College London: The INEQ-CITIES Project (Socio-Economic Inequalities in Mortality: Evidence and Policies in Cities of Europe). United Kingdom. URL https://www.ucl.ac.uk/ineqcities/