Bayesian cluster detection of dengue incidences in Tamil Nadu

Abstract

In epidemiology, the local properties of the relative risk surface are important rather than modelling the relative risk by itself. These surface properties may consist of boundaries between areas of risk, peaks of risk, local heterogeneities in risk etc. These features do not directly relate to a value at a specific location. The main concern of the Relative risk estimation is the ‘global’ smoothing of risk and estimating the true underlying risk level. But the cluster detection is mainly focused on the local features of the risk surface where some elevations or depressions of risk are found. In this article, analysis for disease clustering has been carried out and the high and low levels of disease clusters in dengue incidences are identified in the Coimbatore district, and also for overall Tamil Nadu.