# Spatial distribution and factors associated with co-occurrence of anemia and undernutrition among children aged 6–59 months in East Africa

## TL;DR

This study maps where anemia and undernutrition commonly occur together in East African children and identifies factors linked to their co-occurrence.

## Contribution

The study is the first to analyze the spatial distribution and predictors of co-occurring anemia and undernutrition in East Africa.

## Key findings

- The co-occurrence of anemia and undernutrition affects 26.08% of children aged 6–59 months in East Africa.
- Hotspots for co-occurrence were identified in regions like northeast Ethiopia, northern Uganda, and Burundi.
- Key predictors include young maternal age, maternal anemia, poor household wealth, and lack of health insurance.

## Abstract

Anemia and undernutrition are major public health concerns affecting child growth, development, and survival, particularly in low-resource settings like East Africa. These conditions often occur together and exacerbate the health and developmental challenges of young children. Despite their high prevalence, no research has explored their co-occurrence and the spatial variations across East African countries. Hence, this study aimed to analyze the spatial distribution and identify predictors of the co-existence of anemia and under-nutrition among children aged 6–59 months in East Africa, utilizing recent Demographic and Health Survey data. Understanding the burden and factors affecting their co-occurrence is crucial for mapping geographic hotspot areas and identifying location-specific factors, which guide policymakers in developing targeted interventions.

This study analyzed the latest Demographic and Health Survey data from East African countries, focusing on children aged 6–59 months. It included those with complete outcome and geospatial data while excluding those with missing or inaccurate survey coverage. The primary outcome was the co-occurrence of anemia and undernutrition, which was defined as the presence of anemia together with at least one form of undernutrition, including stunting, wasting, or underweight. The data was managed and analyzed using STATA version 17, ArcGIS version 10.7.1, SaTScan v10.1, and MGWR version 2.2 software. Hotspot analysis was performed to identify high or low prevalence, and ordinary kriging was utilized for interpolation. Furthermore, the Bernoulli-based model was used to identify the most likely clusters of co-occurrences of these conditions by SaTScan analysis. Finally, the geographical weighted regression with the multiscale geographical weighted regression extension analysis was fitted to identify the spatially varying determinants.

The overall prevalence of co-occurrence of anemia and undernutrition was 26.08% (95% CI: 25.70–26.46%). The spatial analysis revealed that the distribution was non-random (p-value<0.001), with significant hotspot areas identified in northeast Ethiopia, northern and northwest Uganda, most parts of Burundi, northeast Tanzania, northern Zambia, central to southern Malawi, eastern Mozambique, and southern Madagascar. The primary clusters were concentrated in Burundi and southern Rwanda (RR = 3.48, LLR = 82.53; p-value <0.001). Spatial regression analysis identified several key predictors with geographic variation, including young maternal age (15–24 years), maternal anemia, no maternal education, poor household wealth, lack of health insurance, maternal tobacco use, multiple births, recent child diarrhea and fever, lack of Vitamin A supplementation, and smaller perceived birth size.

The study highlights significant geographic disparities in the co-occurrence of anemia and undernutrition, with multiple maternal, child, and household factors contributing to the burden. These findings emphasize the need for region-specific public health interventions and future longitudinal research should explore additional confounders and refine regionally tailored approaches.