Back

Geospatial Patterns of Cariogenic Feeding Practices Among Children Aged 6-23 Months in Ghana: A Spatial and Multilevel Analysis of the 2022 Ghana Demographic and Health Survey

Iddrisu, O. A.-F.; Isma-il, A.; Abubakar, H. S.; Siddiq, A. I.

2026-08-10 dentistry and oral medicine
10.64898/2026.08.06.26359897 medRxiv
Show abstract

Background: Early childhood caries remains one of the most prevalent and among the most neglected chronic diseases of childhood in low- and middle-income countries, and cariogenic feeding practices introduced in infancy are a principal, modifiable driver. National, spatially explicit estimates of such practices are lacking for Ghana, constraining geographically targeted interventions. Methods: We analyzed 3,007 children aged 6-23 months from the 2022 Ghana Demographic and Health Survey, linked to 600 georeferenced clusters. A composite cariogenic feeding index was built from bottle feeding, sugar-sweetened beverage consumption, and other cariogenic foods; a child was classified as exposed if any one component was present. Complex survey design was handled by Taylor series linearization. Determinants were examined using survey-weighted and two-level random-intercept logistic regression, and spatial clustering was assessed using global and local Morans I and Getis-Ord Gi* from a five-nearest-neighbour weights matrix. Results: Weighted national prevalence was 52.97% (95% confidence interval 50.19-55.75), ranging from 27.95% in the Savannah Region to 73.55% in the Western Region and patterned by residence (63.07% urban versus 43.90% rural) and wealth (32.87% in the poorest versus 74.45% in the richest quintile). Wealth and maternal education were the strongest multilevel correlates: the richest households showed roughly four times the odds of exposure relative to the poorest (adjusted odds ratio 4.00), and higher maternal education showed roughly two and a half times the odds relative to no education (adjusted odds ratio 2.51). The intraclass correlation coefficient was 0.219. Global Morans I was 0.231 (z = 9.73, p < 0.001), confirming significant positive spatial autocorrelation, with high-high clusters along the southern coastal belt (Western, Central, Greater Accra, Volta) and low-low clusters in the middle and northern belts (Bono, Ahafo, Upper West, Northern, North East). Conclusions: Cariogenic feeding among Ghanaian infants is common, socioeconomically patterned, and spatially clustered rather than random. Oral health promotion through routine growth monitoring and immunization platforms should be prioritized in the high-high clusters, while closing socioeconomic gradients would yield collateral benefit nationally. Trial registration: Not applicable. Keywords: cariogenic feeding, early childhood caries, infant and young child feeding, spatial analysis, Morans I, Getis-Ord Gi*, multilevel modeling, Ghana. Demographic and Health Survey; oral health epidemiology

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.