[BAYESIAN MODELLING AND INFERENCE OF MIXTURES OF DISTRIBUTIONS OF BLOOD TOTAL ANTIRADICAL ACTIVITY]

Spatial-temporal monitoring of the health status of the population plays an important role in public health. For identifying and considering the effects of heterogeneity at the subpopulation level, Bayesian methods for analyzing mixtures of the probability distributions are currently being intensively developed. This article presents the results of studies of the distribution spectrum of the blood total antiradical activity (TAA) levels of the Sachkhere district (Georgia) villages' (Chorvila, Sareki, Sairkhe) population. The research results indicate a non-uniform distribution of blood TAA levels in the populations of the villages of the Sachkhere district. The average blood TAA value in the village Chorvila was statistically significantly lower than the value of blood TAA in the villages of Sareki and Sairkhe. In the village Chorvila, the distribution of blood TAA indices can be described by the Gauss distribution; in Sareki and Sairkhe, a bimodal type of distribution of these values was revealed (the reliability of the difference between the mean values of the distribution components was be lower than in Sareki), which indicates the existence of at least two different subpopulations in this region, related to the impact a certain (unidentified) factor inducing the mobilization of blood TAA. The obtained results allow us to consider the TAA of blood as the most important marker of the oxidative homeostasis of the body in the population.