Abstract:
|
In this work, we present a spatio-temporal model for use by epidemiologists when monitoring the spread of communicable diseases. Thematic maps had been quite used for representation and interpretation of the spatial patterns, but they did not provide information about future behavior of the phenomena. We here consider a hierarchical model including transmission, infection, and removal process usually associated with infectious diseases, defining a relationship between number of cases and number of exposed for each time. The starting point was the updating equations of the specific SEAIR model for measles. Heterogeneity and spatial dependency--using a conditional auto regressive prior distribution (CAR), allowing for over dispersion--were also included in the model. The application of interest is measles diffusion and notified, and localised cases of measles for administrative districts of São Paulo Municipality in 1997 were used. Models were evaluated using some graphical results, where it can be seen that the proposed model had a good fit to the observed data. It is shown that predictions are easily produced and can be used for decision-making and Public Health surveillance.
|