Spatiotemporal Modelling of Endemic-epidemic Cholera in Nigeria

Abstract

This study used a multivariate negative binomial model to capture the Spatiotemporal endemic-epidemic of infectious disease and explore the spatial and temporal patterns of cholera outbreaks in Nigeria. The model for the epidemic part measured spatial weights for the disease spread across the geographical neighboring regions and the endemic part accounted for temporal variation of disease incidence. Weekly count data on cholera from the Nigeria Department of Disease Control and Monitoring Epidemiology (NCDC SED) between January 1st and November 19th, 2018 was used to illustrate the model. In fitting the model, the study has shown that the model with seasonality and autoregressive components provided an adequate fit for the cholera count data and also perform better than the model without seasonality and autoregression for modelling the Spatiotemporal dependency structure of cholera disease