Modifiable risk factors for child stunting in sub-Saharan Africa

Demographic and Health Survey data from 2014–2021 was analysed from 25 sub-Saharan African countries, providing a study population of 145,900 children aged under five years. The aim of this cross-sectional study was to determine the modifiable risk factors associated with childhood stunting. Across the 25 countries, survey design and data were consistent, as they were based on standardised DHS data collection methods and survey tools. Data was obtained from eligible women: females aged 15–49 years residing permanently or visiting the household on the night before the survey.

Stunting (<-2SD) and severe stunting (<-3SD), defined using height-for-age z scores from WHO standards, were the primary outcomes. Modifiable risk factors examined included child, maternal, and household factors. As well as reporting relative risks for each risk factor exposure, adjusted for confounding, the study also presents Population Attributable Fractions (PAF) – the proportional reduction in stunting if exposure to risk factors were reduced – which are utilised in this summary. Relative risk, which may be more intuitive for some readers to interpret, can be found in the original paper.

In the two weeks preceding the survey, 15.1% of children had episodes of diarrhoea (≥3 episodes per day), roughly 84.6% did not consume dairy products in the past 24 hours, and 39.4% of mothers did not attain a formal level of education, with 66.6% unemployed. Over half of children (55%) lived in households without an improved toilet system and 63.9% were born at health facilities. Mean overall stunting prevalence was 30.6%, ranging from 55.8% (Burundi) to 17.1% (The Gambia). Overall severe stunting prevalence was 10.5%, ranging from 24.8% (Burundi) to 3.5% (The Gambia).

A lack of childhood dairy consumption (PAF 15.1%), mothers with only primary education (PAF 8.1%), unclean cooking fuel (PAF 9.5%), and low-income household (PAF 5.4%) were the largest risk factors for stunting. When combined, these four risk factors were associated with 40.7% (95% CI 34.0–47.2) of stunting cases.

”œThe combined PAFs showed that 51.6% (95% CI 40.5–60.9) of cases of severe childhood stunting in sub-Saharan Africa were associated with no formal education among mothers, children lacking dairy products, unclean cooking fuel, home birth, and low-income household.” – Ahmed et al., 2023

The authors note that this may be the first time that PAFs have been used to estimate childhood stunting in this setting, with this measure offering a clear picture on where the most efficient interventions can be targeted. We can clearly see that some risk factors have an ‘outsized’ impact on childhood stunting, so these should be prioritised in a tight funding environment. The use of nationally representative Demographic and Health Survey data increases the generalisability of the results to the region; however, more specific, locally available data will always provide a more nuanced picture. Nonetheless, this large-scale study offers a useful first step for policymakers and programmers to consider.