Do children’s feeding practices affect growth outcomes in sub-Saharan Africa? A multi-country study

  • Bedilu Alamirie EJIGU
  • , Khadijat ADELEYE*
  • , Linus BAATIEMA
  • , Edward Kwabena AMEYAW
  • , Sanni YAYA
  • *Corresponding author for this work

Research output: Journal PublicationsJournal Article (refereed)peer-review

Abstract

Introduction: Sub-Saharan Africa faces unique nutritional challenges, exacerbated by conditions such as poverty, limited access to healthcare, and diverse dietary traditions. Despite the scholarship on children's feeding and their nutritional conditions in sub-Saharan Africa, scanty evidence exists on the association between children's feeding practices and growth outcomes. Consequently, this study seeks to provide a nuanced understanding of children's feeding practices and their association with growth outcomes in sub-Saharan Africa.

Methods: Demographic and Health Survey data from 30 countries were used for this study, with a sample aged 0-59 (n = 174,281). The descriptive analysis involved the computation of child and maternal factors, presented in frequencies and percentages. We further analysed the weighted percentages of stunting, wasting, and underweight and their co-existence across included countries and children's socio-demographic characteristics. To investigate the association between child feeding practices and other demographic factors with child weight and growth outcome, we employed a multilevel logistic regression. Given the large amount of missing data on the covariate exclusive breastfeeding, it was excluded during the model fitting process.

Results: We noted that 6.94% (95%CI:6.79,7.09) were wasted, 32.17(95%CI:31.90,32.45) were stunted, 16.72%(95%CI:16.50,16.94) were underweight children, and 2.32% (95%CI:2.24,2.40) were wasted, stunted and underweight. Age-appropriate feeding (conceived as not wasted, stunted or underweight based on WHO's definition) increased co-occurrence risk (AOR = 1.68), while exclusive breastfeeding reduced it (AOR = 0.40). Children with low birth weight were more likely to be wasted [AOR = 1.65, 95% CI = 1.57-1.74], stunted [AOR = 1.49, 95% CI = 1.44-1.54] and underweight [AOR = 1.91, 95% CI = 1.84-1.98]. Females were observed to have lower odds of all three indicators as manifested in their co-existence (AOR = 0.59, 95% CI = 0.55-0.63). Children 12-23 months had the highest odds of experiencing the co-existence of stunting, wasting and underweight (AOR = 5.86, 95% CI = 4.94-6.95) relative to those below 6 months.

Conclusion: Feeding practices significantly predict growth outcomes, urging policies for age-appropriate diets, dietary diversity, and breastfeeding support. The findings underscore the urgent need to address child malnutrition, particularly stunting, wasting, and underweight, to improve child health and well-being. These findings should inform policymakers, healthcare professionals, and stakeholders in developing effective strategies to improve child nutrition and overall well-being in the region. Context-specific policies and interventions, prioritizing maternal nutrition, access to quality prenatal care, and appropriate infant feeding practices, are crucial in mitigating the impact of malnutrition on children in the region. Further research and collaboration are needed to develop targeted strategies and sustainable solutions to tackle child malnutrition effectively in sub-Saharan Africa.
Original languageEnglish
JournalJournal of Health, Population and Nutrition
DOIs
Publication statusE-pub ahead of print - 21 Feb 2026

Bibliographical note

We are grateful to the MEASURE DHS for data access.

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

  1. SDG 2 - Zero Hunger
    SDG 2 Zero Hunger

Keywords

  • child malnutrition
  • Growth outcomes
  • Feeding practices
  • Sub-Saharan Africa
  • Low birth weight

Fingerprint

Dive into the research topics of 'Do children’s feeding practices affect growth outcomes in sub-Saharan Africa? A multi-country study'. Together they form a unique fingerprint.

Cite this