Modelling the Number of Cases of Dengue Hemorragic Fever with Mixed Geographically Negative Binomial Regression in West Java Province

Abstract

Dengue hemorrhagic fever (DHF) is an infectious disease caused by the dengue virus of the genus Flavivirus, which is transmitted by the bite of the Aedes Aegypti. Different regional demographics cause the number of DHF cases to differ in each region following by environmental conditions in the area. The model applied is GWNBR (Geographically Weigthed Negative Binomial Regression) due to count data outcome affected by geographical effect. In some instances not all in the GWNBR model have spatial effects, sometimes the estimate parameter are constant, so the GWNBR model can be developed using a mixed model to become MGWNBR. Determination of global and local parameters using the confidence interval. This study aims to analyze the factors that influence the number of dengue cases in West Java Province in 2015 using the MGWNBR approach. Based on the comparison of AIC values, the MGWNBR model has a smaller AIC value compared to the negative binomial regression model. The variables that significant globally are population density (X1) and health worker (X2)The variables that significant locally are number of health facilities (X3) PHBS (X4) and healthy homes (X5)