PEMODELAN JUMLAH KEMATIAN BAYI AKIBAT TETANUS NEONATORUM DENGAN METODE GEOGRAPHICALLY WEIGHTED ZERO-INFLATED POISSON REGRESSION

Abstract

Tetanus Neonatorum (TN) is an infection in infants caused by the Clostridium tetani bacteria. In 2020, the Case Fatality Rate (CFR) due to TN in Indonesia increased to 50% compared to 2019, which was 11.76%. So it is necessary to study the number of infant deaths due to TN. This study discusses the modeling and factors that influence TN disease in Indonesia using the Geographically-Weighted Zero-Inflated Poisson Regression (GWZIPR) method. The GWZIPR model is divided into two based on the state: the ln model for the Poisson state and the logit model for the zero states. The data in this study are the number of infant deaths due to TN, the percentage of pregnant women carrying out Td2+ immunization, the percentage of pregnant women delivering at health facilities, and the percentage of puskesmas carrying out P4K in 34 provinces in Indonesia in 2020. The results of this study are that there is an excess zero of 58.82% and spatial heterogeneity occurs so that each region has a different model based on significant variables. The factors that influence the number of infant deaths due to TN are divided into four groups based on significant variables in the ln and logit models.