Spatial mathematical analysis : an application to mapping of dengue hemorrhagic fever in Thailand

Abstract

The objectives of this research are to estimate the morbidity rate of Dengue Hemorrhagic fever in each province of Thailand and to construct the map of the Dengue Hemorrhagic fever. A Bayesian model is used for the data analysis. The data are mathematical spatial correlated in the form of the conditional autoregressive model (CAR). The secondary data were collected at a provincial level. They were the number of Dengue Hemorrhagic fever patients, rainfall and average temperature. The research found that the rainfall (RR = 1.0026) and the average temperature (RR = 1.2681) had influence on the morbidity of Dengue Hemorrhagic fever. The season during July to September had more influence on the morbidity of Dengue Hemorrhagic fever than during the other seasons. The top 5 highest morbidity rates, per 100,000 population, were Chiangmai in October (1,288.00), July (1,096.00), September (804.30), Surin in October (772.30), and Nakhon Si Thammarat in August (666.20). The disease map constructed for showing the distribution of the disease is easy to understand and useful for public health.