Bayesian Spatio-temporal Analysis for D...

Bayesian spatio-temporal models are very common to analyze epidemiological data and that works with both small areas and large areas under study. We have used combinations of predictors to model the best combinations of the factors that are most responsible for the spread of dengue fever and the analysis is based on the Deviance Information Criterion (DIC). Our results show that rainfall, sanitation and the availability of hospitals are the main driving forces in the increase of the number of dengue cases in Pakistan. Sanitation and hospitals are significantly responsible for the continuation of disease especially in less developed areas, but also in a poorly maintained metropolitan city like Karachi, on smaller numbers, also literacy in the study area is significantly affecting. With higher temperatures the chances of dengue spread are less. Bayesian spatiotemporal models have yielded better analysis in the cases where the intercept of the model is kept fixed. Karachi has shown to be most prone to the continuation of the dengue larvae in the period under analysis. Erratic patterns of the rainfall are a reason for the increase in the cases of dengue in the northern Pakistan.