Predictive Model of Stunting in Children 6-59 Months of Age in Kirundo Health District, Burundi

Abstract

An analytical cross-sectional study was conducted among a randomly selected sample of 374 households
with at least one child aged 6 to 59 months in the Kirundo health district, Burundi. Sociodemographic, socioeconomic,
socio-sanitary factors, food insecurity, behavioral, and environmental data were collected using a structured
questionnaire. Children's weight was measured using a standard procedure (SECA scale), their height using a standard
UNICEF height rod, and their age was obtained from the birth certificate. Anthropometric data were analyzed using
Emergency Nutrition Assessment (ENA for Smart) software.
Modeling was performed using logistic regression to eliminate confounding factors, and all independent variables with a
significance level less than or equal to 20% in the bivariate analysis were included to explore factors associated with
stunting in children aged 6 to 59 months.
In this study, the prevalence of stunting is estimated at 61.5%. According to multivariate logistic regression, sex (AOR =
2.83; 95% CI:1.40-5.75), age (AOR= 10.40; 95% CI: 1.21-88.30), food insecurity (AOR = 10.47;95% CI: 3.58-30.61),
latrine type (AOR = 6.83; 95% CI: 3.12-14.94), diarrhea (AOR = 2.56; 95% CI: 1.19-5.48), water source (AOR = 3.17;
95% CI: 1.54-6.52), media exposure (AOR = 0.24, 95% CI: 0.11-0.51), nutritional knowledge (AOR = 0.11; 95% CI: 0.05-
0.25), birth spacing (AOR = 0.39, 95% CI: 0.16-0.93), complete vaccination (AOR = 0.06; 95% CI: 0.02-0.21), father's
occupation (AOR = 0.25; 95% CI: 0.09-0.72), and mother's education (AOR = 0.21; 95% CI: 0.07-0.64) were significantly
associated with stunting.
The predictive model showed an area under the curve (AUC) of 0.95, indicating excellent discrimination ability.
The high prevalence of stunting in this study highlights the importance of urgent action to end this problem.