Mapping districts in West Java by under-five pneumonia indicators: an agglomerative hierarchical clustering study (Open Data Jabar 2023)

Abstract

In West Java Province, Indonesia, pneumonia remains a major contributor to morbidity and mortality among children under five years of age. This study aims to employ an agglomerative hierarchical clustering approach to classify districts and cities based on indicators associated with under-five pneumonia. The analysis utilized secondary data from Open Data Jabar (2023), encompassing nine cities and eighteen districts. The variables considered included malnutrition prevalence, vitamin A supplementation coverage, incidence of low birth weight, DPT-HB-HIB 3 immunization coverage, the proportion of households practicing Clean and Healthy Living Behavior (PHBS), and the number of reported pneumonia cases. Euclidean distance was applied to compare five linkage methods, with the Single Linkage method selected for final clustering due to its highest cophenetic correlation coefficient (0.8681). The optimal clustering solution yielded four distinct profiles, ranging from regions with high PHBS coverage and immunization rates to those with low immunization coverage and high malnutrition prevalence. These findings provide an evidence-based framework for designing region-specific pneumonia prevention and control strategies targeting under-five populations in West Java.