Generalized linear mixed models for spatio-temporal data with an application to Leptospirosis in Thailand
Applied Mathematical Sciences
The research objectives were to estimate the morbidity rates of Leptospirosis in each month of all provinces in Thailand, to determine the trend of Leptospirosis over time, to investigate factors influencing on the morbidity rates, and to construct the disease maps of the Leptospirosis. The modified generalized linear mixed model (GLMM) to include spatial effects and temporal effects was applied for spatio-temporal data analysis. The estimated morbidity rates were used to construct the disease
... struct the disease maps. The dependent variables were the number of Leptospirosis patients in each month of each province and were assumed to have a Poisson distribution. The factors considered were the amount of rainfall, averaged temperatures, and regions. The results showed that the factors influencing on the morbidity rates were the amount of rainfall, average temperature, northern region, northeastern region, southern region, western region, and eastern region, where the central region was a reference region, and there was a linear trend. The Leptospirosis maps are easy for readers to identify which areas are at high risk. They are a useful tool for planning and controlling the Leptospirosis.