<![CDATA[Maternal and child health represents a critical aspect of national development, reflecting both the well-being of the population and the success of regional health equity programs. In Indonesia, disparities in maternal and child health outcomes remain evident across provinces due to socio-economic inequalities and unequal access to health services. This study aims to model the causal relationships between key health determinants and classify provinces based on maternal and child health degree using Structural Equation Modeling with Partial Least Squares (SEM-PLS) combined with Prediction-Oriented Segmentation (POS). The study uses secondary data from 2023 sourced from the Central Statistics Agency (BPS) and the Ministry of Health, covering 34 provinces at the provincial level. Nineteen indicators are grouped into four latent variables: Health Services, Clean and Healthy Living Behavior (PHBS), Environment, and Health Degree. SEM-PLS was applied to identify direct and indirect relationships among these variables, while POS was used to identify homogeneous segments of provinces. The results show that PHBS positively affects Environment (path coefficient = 0.896; p < 0.001), while Health Services negatively affect Health Degree (path coefficient = –0.668; p < 0.01), indicating the presence of indirect pathways influencing health outcomes. The segmentation analysis identified three segments: Segment 1 includes provinces with moderate outcomes but weak child health services; Segment 2 includes provinces with relatively better maternal outcomes but sanitation gaps; Segment 3 consists of provinces with the most critical health conditions, including high stunting and malnutrition rates. These findings demonstrate that PHBS is a dominant influencing factor, while improved service access alone does not always translate to better outcomes. The SEM-POS approach effectively identifies segment-specific health disparities, supporting more targeted policy interventions to improve maternal and child health in Indonesia.]]>
Copyrights © 2026
