Spatial Autoregressive Model of Tuberculosis Cases in Central Java Province 2019

Abstract

Tuberculosis is an infectious disease caused by infection with the bacterium Mycobacterium tuberculosis. Central Java is one of the three provinces with the highest tuberculosis cases in Indonesia. Some of the risk factors used in this research are the spatial lag of the number of tuberculosis cases representing the agent component, the morbidity rate representing the host component, population density, proper sanitation, and proper drinking water which represent environmental components. This study uses the Spatial Autoregressive (SAR) model. The SAR model is a regression model where the response variable has a spatial correlation. The estimation method usually used in SAR model is maximum likelihood. The value of Moran's I on the number of tuberculosis cases in Central Java is 0.499 and is significant, which means that there is a positive spatial autocorrelation. The model was chosen based on the LM test and AIC. The best model is the SAR model. The results of the analysis obtained show that the greater the number of tuberculosis cases is influenced by the number of tuberculosis cases in the surrounding area. Proper sanitation has a negative effect, on the contrary, the dense population has a positive effect on the number of tuberculosis cases in the province of Central Java.

Keywords

• maximum likelihood
• sar
• spatial
• tuberculosis