Modeling the Maternal Mortality Rate in Indonesia Using Geographically Weighted Poisson Regression Approach

Indonesia is one of the countries with a high Maternal Mortality Rate (MMR). In its implementation, reducing the MMR in Indonesia is one of the targets of the Indonesian government to achieve the Sustainable Development Goals (SDGs) targets. This target was stated in the third point of the SDGs targets, which is to ensure a healthy life and a better life for all the world's population at all ages. The purpose of this study is to model and analyze the MMR using the Geographically Weighted Poisson Regression (GWPR) method on maternal mortality data based on public health centers facilities in Indonesia in 2020. The health service facilities referred to are the ratio of public health services, coverage of K4 services (contact at least 4 times during pregnancy), implementation of the Delivery Planning and Complications Prevention Program (P4K), Tetanus Facility Services (TD2+), delivery coverage in health care facilities, complete postpartum visit coverage, and active contraception coverage. The AIC and deviance values of the GWPR model are smaller than the Poisson regression model, indicating that the GWPR model with the fixed gaussian kernel weighting function is better than the Poisson regression model in MMR modeling in Indonesia. The percentage of complete postpartum visit coverage is the most spatially significant independent variable.