Spatial analysis of stunting determinants in 514 Indonesian districts/cities: Implications for intervention and setting of priority

Abstract

While the national prevalence of stunting in Indonesia has decreased, the level remains high in many districts/cities and there is significant variation. This ecological study employed aggregated data from the Basic Health Research Report and the District/City Poverty Data from 2018. We investigated the determinants of stunting prevalence at the district/city level, including autocorrelation applying the spatial autoregressive (SAR) model. The analyses revealed stunting prevalence above the national average in 282 districts/cities (54.9%), i.e. ≥30% in 297 districts/cities (57.8%) and ≥40% in 91 districts/cities (17.7%). Autocorrelation was found between Sumatra, Java, Sulawesi as well as Bali, East Nusa Tenggara and West Nusa Tenggara (Bali NTT NTB). The SAR modelling revealed the following variables with significant impact on the stunting prevalence in various parts of the country: closet defecation, hand washing, at least four antenatal care visits during pregnancy, poverty, immunisation and supplementary food for children under 5 years.

Spatial autocorrelation of stunting prevalence among children under five years in West Bandung Regency in 2022

“…Figure 1 shows the visualization of stunting Based on the results of the significance analysis in The findings of this research are contradicted to several spatial studies on stunting prevalence in Indonesia and several other countries that have been conducted previously, which stated that there is spatial autocorrelation related to the prevalence of stunting [13][14][15][16][17][18][19][20][21]. A study in Ethiopia examining the spatial trends of stunting in children under five using DHS data from 2011, 2016, and 2019 showed that all three had a p-value < 0,001 with respective Moran's Index values of 0.9185, 0.3968, and 0.3530 [16].…”

“…The spatial analysis results related to stunting in several previous studies have shown significant spatial autocorrelation based on the prevalence of stunting [13][14][15][16][17][18][19][20]. Based on the spatial analysis of stunting conducted in 514 districts/cities in Indonesia, spatial autocorrelation was found based on stunting in the regions of Sumatra, Java, Sulawesi, Bali, Nusa Tenggara Timur (NTT), and Nusa Tenggara Barat (NTB) [20].…”

Correlation between Basic Immunization Status and IHC Visits to Stunting Incidents on Toddlers

“…Next, the two data were combined and categorized for visits to integrated healthcare (IHC). ANC examinations are categorized as routine if ANC examinations are carried out at least four times during the mother's womb, namely once in the first trimester, once in the second trimester, and twice in the third trimester (Eryando et al, 2022). In the same article, the ANC examination category is said to be non-routine if the ANC examination is not carried out at least four times during the mother's womb, namely once in the first trimester, once in the second trimester, and twice in the third trimester.…”