Claim Missing Document
Check
Articles

Found 1 Documents
Search

Geographically Weighted Poisson Regression Modeling Using Adaptive Gaussian Kernel Weighting For Mapping Maternal Mortality Rates In East Java Ngoro, Inayati; Pramoedyo, Henny; Astuti, Ani Budi
Jambura Journal of Biomathematics (JJBM) Volume 6, Issue 4: December 2025
Publisher : Department of Mathematics, Universitas Negeri Gorontalo

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37905/jjbm.v6i4.30411

Abstract

Maternal Mortality Rate (MMR) is a key public health indicator that reflects spatial variation across districts in East Java.  This study aims to model the spatial distribution of MMR using Geographically Weighted Poisson Regression (GWPR) with an Adaptive Gaussian Kernel weighting function. Secondary data were obtained from the 2022 East Java Provincial Health Profile, covering 38 districts and municipalities. The results indicate that GWPR outperforms the classical Poisson regression. The intercept β=2.889 (exp=17.95) suggests an average of 18 maternal deaths in the absence of predictor effects. The coverage of the fourth antenatal care visit (K4) has a significant negative effect ( β=-0.027; exp = 0.973), indicating that a 1% increase in K4 coverage reduces MMR by approximately 2.7%. Conversely, obstetric complications managed by midwives show a significant positive effect (β= = 0.0173; exp = 1.017), meaning that a 1% increase in complications raises MMR by 1.7%. Other predictorsfirst antenatal care visit (K1), ironfolic acid (IFA) supplementation, and number of health workersare not statistically significant. This study underscores the importance of expanding K4 coverage and strengthening complication management as priority strategies to reduce maternal mortality.  Furthermore, GWPR-based mapping enables more targeted maternal health interventions tailored to local characteristics.