An ecological analysis of leukemia incidence around the highest 137Cs concentration in Poland.

Cancer has long been known to be a hazard of exposure to ionizing radiation. However, the assessment of health effects from exposure to radiation is a matter of considerable controversy. This paper presents results of a retrospective study of leukemia incidence (203-207, ICD-9) around the highest 137Cs pollution in Poland (as an effect of the Czarnobyl disaster and/or military bomb tests). The data relating to all the registered leukemias in males and females originated from the Regional Cancer Registry in Opole. The information on 137Cs concentration rates in Opole province was derived from the state monitoring provided by the Polish Geological Institute in Warsaw. The spatial analysis--based on the random-effects Poisson regression model--was carried out via the Markov Chain Monte Carlo (MCMC) technique (Gibbs sampling) using BUGS software. The model incorporated epidemiological data and an ecological covariate--isotope concentrations--and provided a framework for estimating the strength of a dose-response relationship. The differences in incidence levels were quantified by traditional standardized morbidity ratios (SMRs) and presented in thematic maps as well as in combined charts of distance-disease-dose relations. Additionally, to assess spatial disease clustering, a Tango test was adopted. The results of this ecological study suggest that the 137Cs concentrations did not have any negative influence on the exposed population.