Radiation epidemiological classification of cerebrovascular morbidity and mortality among Chernobyl clean up workers with use of claster analysis and method of principal components

Structure and classification of cerebrovascular morbidity and mortality among Chernobyl clean up workers were examined for determination of models and objects for radiation epidemiological analysis. For investigation cluster analysis and method of principal components were applied in a space of variables used for estimating relationship between dose and tumor diseases development. Analysis was carried out with the use of statistical software “Statistika”. Sources of data were personal information about 127912 clean up workers (men) accumulated in the National Radiation Epidemiological Registry for the period from 1986 to 2012. The number of deaths from cerebrovascular diseases (CVD) was 2412, the number of diagnoses - 87917. As a result of the study the following was identified: 1) a set of variables, which were main contributors to dispersion of data on cerebrovascular morbidity and mortality among clean up workers, they are age, radiation dose, date of entering the 30-km dose, region, from which a worker came in the radiation zone. The set of variables can serve as the basis for mathematical models and calculation of radiation risk; 2) the diseases, which can be considered as confounded factors for calculating radiation risks of CVD, they are G81 for I61, I63, I64, I69, D33, G45 and blocks D60-D68, E10-E13, I10, I20 - for I60, I67; 3) groups of diseases, which can depend on radiation doses, these are I61, I63, I64, I67, I69.