<![CDATA[Abstract Maternal mortality is one of the key indicators in assessing public health status and the quality of maternal health services. East Java Province still exhibits considerable variation in maternal mortality rates across districts and cities, necessitating an analytical approach that accounts for regional characteristic differences. This study aims to analyze the factors influencing maternal mortality rates in East Java Province using the Geographically Weighted Regression (GWR) method. Secondary data from 2024 covering 38 districts and cities in East Java were used. The predictor variables analyzed include K1 coverage, K4 coverage, postpartum maternal health services, deliveries assisted by health professionals, vitamin A supplementation, Fe3 tablet distribution, as well as the ratio of hospitals and community health centers. The results indicate that the GWR model with an adaptive Tricube kernel and an optimum bandwidth of 35 is the best-fitting model, with an AICc value of 261.175 and R² of 0.8646, outperforming the OLS model which yielded an R² of 0.6612. Local parameter estimation results reveal that the influence of predictor variables differs across districts and cities, indicating that the factors affecting maternal mortality are spatially varying and heterogeneous across regions. Keywords: Maternal Mortality, Geographically Weighted Regression, Spatial Analysis, East Java.]]>
Copyrights © 2026
