Background: Malnutrition is both a significant cause and a result of poverty and deprivation. In developing nations, child malnutrition is still the main public health issue. Severe malnutrition affects every system of the body and leads to medical instability. The assessment of the burden of severe malnutrition is important for ready-to-use therapeutic foods and preparing therapy for these conditions. Therefore, this study aimed to assess the prevalence and spatial distribution of severe malnutrition and the factors associated with it.

Methods: Data from the 2019 Mini-EDHS (Ethiopian Demographic and Health Surveys) with stratified sampling techniques were used. The data were weighted using sample weight to restore the data's representativeness and provide accurate statistical estimations. A total of 5,006 weighted samples of children under the age of five were used to analyze the study. A multilevel binary logistic regression model was built, and a cutoff P-value of 0.05 was used. The wag staff normalized concentration index and curve as well as spatial analysis were used.

Results: The prevalence of severe malnutrition practice among under five years children in Ethiopia was 14.89% (95%CI: 13.93%, 15.91%), and ranges from 4.58% in Addis Ababa to 25.81% in the Afar region. Women with secondary and above education status as compared to uneducated [AOR = 0.17; 95%CI;[0.06, 0.48], high community women's education as compared to low [AOR = 0.54; 95%CI; 0.36, 0.78], women from richest household as compared to poorest [AOR = 0.63; 95%CI; 0.26, 0.94] and living in Oromia region as compared to Tigray [AOR = 0.33: 95%CI; 0.15, 0.74] were preventive factors. Whereas children 24-59 months of age as compared to under six months [AOR = 1.62; 95%CI; 1.50, 1.75], and being multiple births as compared to single [AOR = 5.34; 95%CI; 1.36,2 1.01] have significant risk factors for severe malnutrition. There was a pro-poor distribution of severe malnutrition among under-five children in Ethiopia with a concentration index of -0.23 [95%CI: -0.27, -0.19]. Severe malnutrition has significant spatial variation over regions in the country where the entire Afar, Eastern Amhara, Southern, and eastern Tigray regions were severely affected (RR = 1.72, P-value < 0.01).

Conclusion and recommendations: The prevalence of severe malnutrition in Ethiopia is relatively high as compared to other studies and most of them were severe chronic malnutrition. Having an educated mother/caregiver, and living in a cluster with high community women's education were preventive factors for severe malnutrition in children. Whereas having an unmarried mother/caregiver, old age of the child, plurality of birth, and having double children in the family have a positive association with it. Moreover, it was disproportionately concentrated in poor households (pro-poor distribution). The spatial distribution of childhood severe malnutrition was not random. Regions like Tigray, Afar, Eastern parts of Amhara, and Somalia regions should be considered priority areas for nutritional interventions for reducing severe malnutrition. Equity-focused nutritional interventions could be needed to curb the wealth-related inequalities of childhood severe malnutrition.