[Geographical analysis of mortality in a municipality of the Veneto region, where a landfill is located, and the surrounding area. Years 1995-2003].

OBJECTIVE: the study explores whether a potential source of environmental pollution (a dumping ground with different kinds of waste, in Spinea, an area adjacent to Venice, population 25,000) could have led to an excess of mortality from certain pathologies, and in particular some cancers which have been reported to be associated to the presence of dumping grounds. Besides traditional estimation techniques, Bayesian estimators (BMR) have been used, which--if based on appropriate statistical analysis techniques--allow to consider the spatial dependence of the data. The smoothed geographical distribution of mortality in the area surrounding the pollution source is then represented as a map and the presence of particular mortality patterns is verified. Compared to traditional techniques, this approach produces more reliable data in a relatively short time and leads to an analysis with a better information level. Communication to the decision makers and to the population should be based on these data and results. DESIGN: the data were derived from ISTAT mortality reports coded at a local health district level. The following analysis have been carried out: a. a traditional descriptive analysis, i.e. comparison of age. standardized rates in the Spinea municipality and the surrounding area with crude regional rates; b. an analysis of heterogeneity of BMR distribution (reference rates = age-specific rates in the population of the investigated area) in the area itself c. the application of execution of the Martuzzi-Hills test and d. the creation of a mortality distribution map (divided into BMR value classes) in the investigated area. SETTING: the examined area includes Spinea and the surrounding municipalities within the Veneto Region borders, considering Spinea in the centre and a 15 km radius. RESULTS: the total number of deaths in the examined area in the 9 years covered by the present analysis is 49,739 (13% of the regional total). The annual age-standardized rate was 89.91 deaths/10,000. The results of the analysis do not suggest any particular mortality patterns either in the area (compared with the Veneto Region) or within it. CONCLUSIONS: the study has not highlighted geographical mortality clusters of deaths from the causes which have been selected for the analysis.