Investigating the Relationship between Maternal Socioeconomic Influences and Child Stunting in South Africa, using DHS Data

Abstract

Understanding disparities in children’s nutritional status is crucial to reducing health inequality, as stunting increases the risk of mortality and long-term noncommunicable diseases. Global stunting rates have been declining, while South Africa maintains high rates. Stunting occurs early in life, so it is critical to consider the maternal influences on child health. Aspects of maternal empowerment may be particularly impactful in South Africa, as the country is known to have high levels of gender inequality.

Previous research has established independent relationships between stunting and elements of women’s empowerment. These include a negative relationship between high rates of stunting and low maternal education levels, a positive relationship between intimate partner violence and stunting, and an inconclusive relationship between stunting and maternal employment. Previous research also finds a positive association between stunting and poverty. However, there is a lack of research on these maternal influences on stunting in conjunction with one another, and additionally considering the number of children’s antenatal visits and wealth. These factors are important to consider, as they impact social interactions and healthcare. This thesis evaluates the influences of maternal socioeconomic factors on childhood stunting using the 2016 South African Demographic and Health Survey (DHS). It employs several means to do so. First, it examines the known social determinants of health that affect women in South Africa, as well as policies surrounding gender and nutrition. Next, it uses individuals’ responses and children’s anthropometric measurements to quantify the relationship using logistic regressions. This thesis then looks at two locations that have successfully reduced their stunting rates through gender equality-based interventions. The statistical model demonstrates that, while holding these variables independent from one another, stunting has statistically significant relationships with maternal education level, employment outside of the household, head of household status, intimate partner violence, and the third and fifth wealth index in South Africa. It finds that women are 9.1% less likely to have a stunted child for each additional year of education, those that work outside their household are 6.0% less likely, and women that are head of their household are 7.2% less likely to have a stunted child. It also finds women facing intimate partner violence are 42% more likely to have a stunted child. This thesis concludes by examining the healthcare system and providing policy reflections. It offers modifications of current policies such as the Integrated Nutrition Programme and Domestic Violence Act and suggests that targeting domains of women’s empowerment may improve overall program efficacy and reduce stunting rates.