Background: Childhood malnutrition is a major public health challenge of global importance. It may result from either excessive or deficient nutrients. Despite investments and several efforts made by the South African government and civil society organizations to improve child health, the prevalence of childhood malnutrition remains high in South Africa. South Africa is still lagging in in achieving the sustainable development goals 1-3 (i.e., 1- no poverty, 2 – zero hunger and 3 –good health and wellbeing). This is because the indicators of childhood malnutrition are significantly higher with one in four children being stunted, 13% overweight, and 7.5% underweight. These figures highlight a troubling trend that is echoed in many other African nations, where malnutrition rates are similarly concerning. For instance, while countries like Nigeria and Ethiopia face severe challenges with stunting rates exceeding 30%, South Africa’s rates are comparatively lower but still indicative of a significant public health challenge. In contrast, developed nations such as the United States report much lower stunting rates—around 3.4%—and face different nutritional issues, such as rising obesity rates among children. The current malnutrition status is worrisome in South Africa given that these conditions have not changed much in nearly three decades. Among other factors recognised as the leading causes of poor nutrition outcomes is food insecurity in households -defined as the lack of regular access to safe, sufficient, and nutritious foods, disrupted eating patterns and reduced food intakes. Despite South Africa being a net exporter of food, it is characterised by high poverty, reduced opportunities for higher education, employment challenges, environmental hazards, substandard housing, and health disparities, still have challenges in access to affordable safe nutritious foods. Furthermore, due to the complexity of childhood malnutrition, an integrated multisectoral approach among families, communities, and government systems is critical to ensuring positive child health and nutritional outcomes. Addressing poor nutritional outcomes among under-5 children requires policy-relevant evidence. While the literature shows that childhood malnutrition is a multifaceted issue influenced by poverty and poor socio-economic outcomes, evidence is sparse on how structural and environmental factors operating at different levels influence childhood malnutrition. Therefore, an understanding of social contexts of childhood malnutrition is required to improve children’s health outcomes in South Africa. Hence, this study examined the social context of childhood malnutrition in South Africa with a focus on individual child, caregiver, and household-level characteristics. The study addressed five specific objectives: i) to determine the levels and patterns of childhood malnutrition in South Africa, (ii) to examine the individual child, caregiver, and household factors associated with childhood malnutrition in South Africa, (iii) to investigate the influence of food insecurity on childhood malnutrition, (iv) to explore the extent to which the socio-cultural and childcare practices of caregivers predispose under-5 children to malnutrition in selected low-income communities in South Africa, and (v) to investigate the role of a multi-sectorial approach in improving child nutritional outcomes in SA. This study was guided by the 2020 UNICEF conceptual Framework on Maternal and Child Nutrition as well as the Food and Nutrition Security Theory.
Methods: This study adopted an explanatory sequential mixed methods design (i.e., analysis of quantitative data followed by qualitative data collection and analysis). The research methodology was broken into the quantitative and qualitative study. The quantitative study entailed analysing the quantitative secondary data from the 2017 South Africa National Income Dynamics Study (NIDS Wave 5). The NIDS data was nationally representative. The sample was weighted using post-stratified weights. Data of 2 966 children and their mothers were analysed. These children were selected on the basis that they had complete anthropometric measurements (height and weight measurements) and were suitable and selected for the investigation of childhood malnutrition (stunting, overweight, and underweight). We also conducted qualitative in-depth interviews with Early Childhood Development (ECD) practitioners to gain a deeper understanding of their experiences in childcare and perceptions of feeding practices. They were key informants since under-5 children spent a lot of time at ECD centres. Data were analysed at the univariate level to obtain descriptive statistics, and at the bivariate level using the chi-square test of association. At the multivariate level, multi-level binary logistic regression was employed, and odds ratios were reported. The multilevel analysis involved two levels – the individual level (child and mother characteristics) and the household-level characteristics. Data were analysed using Stata software (version 17). The selection of the independent variables was guided by the literature review and conceptual framework of the study. The second part of the study was qualitative and was collected between June and August 2022. Twenty in-depth interviews, and five focus group discussions with mothers of under-5 children, and five in-depth interviews with early childhood development practitioners (ECD practitioners) were conducted. Interviews were conducted using semi-structured questionnaires in selected low-income communities in urban Gauteng (i.e., Thulani in Soweto), and in rural Limpopo (i.e., GaMasemola in Sekhukhune District). These communities were selected based on high poverty and unemployment rates, had substandard houses, insufficient infrastructure and environmental issues. The qualitative data provided deeper understanding about ethe quantitative findings and explored questions that were not available to the researcher in the NIDS dataset. The focus group discussions and key-in-depth interviews further provided a follow-up and an explanation of the quantitative findings. Thematic analysis was used to analyse qualitative data.
Key findings from objective 1: In terms of descriptive findings, found that 22.16% of children were stunted, 16.40% were overweight, and 5.04% were underweight. The distribution of children among female and male children in the study population was almost the same. About 40% of the children had a low birth weight (<3 kg), 80.59% relied on the child support grant, and 67.22% were cared for at home during the day. Different patterns of malnutrition were observed. The highest percentage of children ages 12-23 months were stunted (33.43%) and overweight (32.69%), while the highest proportion of children ages 0-11 months and 48-59 months were underweight. Among children with a low birth weight of 1-2.9 kg, the highest percentage of stunting (30.07%) (p = 0.001, χ² = 71.2) and underweight (7.05%) (p = 0.026, χ² = 16.9) was observed. There was a relationship between access to medical aid, access to the child support grant, and childhood stunting (p < 0.05), while being cared for at home during the day was associated with stunting (24.98%) and overweight (18.99%) (p = 0.002, χ² = 36.3). Caregivers’ religion was associated with overweight (p = 0.007, χ² = 25.6) among under-5 children, while caregiver’s ethnicity (p = 0.024, χ² = 18.4) was associated with underweight.
Key findings from objective 2: Female children had a lower likelihood (0.63 times) of being stunted compared to males. Children aged 12-23 months face a 60% higher risk of being overweight than those aged 0-11 months (AOR = 1.6). However, the risk of overweight declines steadily as age increases. Children aged 48-59 months are 83% less likely to be overweight compared to the youngest group of 0-11 months (AOR = 0.17). Children with a birthweight of 3 kg are 63% less likely to be underweight compared to those weighing 1-2 kg at birth (AOR = 0.37). Children attending crèches/day moms are 69% less likely to be underweight compared to those cared for at home (AOR = 0.31). Children cared for at home are 1.5 times more likely to be stunted (AOR=1.49) compared to children at a creche/day mom. Caregivers who were Nguni had a 26% lower likelihood of having stunted children. Caregivers of other religions had 2 times higher likelihood of having overweight children compared to Christian caregivers (AOR=1.21). Middle-income households were associated with having overweight children (AOR=1.35) compared to low-income households. Children from structurally sound households had a 54% of high risk of being overweight compared to children from dilapidated household structures. The study found that a significant portion of the variation in child malnutrition (stunting, overweight, and underweight) occurred within communities. This is evident from the intraclass correlation of stunting (ICC) values from 27.9% to 30.2% variation, 34.3% to 38.2% overweight variation and 19.6% to 33,9% underweight variation within communities. The increase in ICC after adding additional variables suggest that these factors explain more of the variation within communities.
Key findings from objective 3: The results showed that nearly 30% of the households were below the lower-bound food poverty line of R890 per person per month in South Africa, and just about half of the households did not always have enough available foods all the time. The qualitative findings show that the COVID-19 pandemic exacerbated the food insecurity during the COVID-19 lockdown, when many caregivers lost their income sources due to job losses. Food affordability and availability in the households became a major issue, forcing households to make hard decisions between deciding on foods with high nutrition that should be eaten against diverting financial resources and paying for other household expenses such as rent or electricity. Caregivers understood that they should be feeding their children nutritious foods but due to financial constraints, they were forced to give children the available but less nutritious foods in the households.
Key findings from objective 4: Qualitative findings further showed that caregivers had various socio-cultural and childcare practices which influenced children’s nutritional and health outcomes. Socio-cultural practices that influenced childhood malnutrition included dietary choices – these were not necessarily affected by cultural beliefs, but they were rather influenced by the lack of income. Traditional beliefs on food- such as foods like eggs and dairy products such as milk or yoghurts were not given to girls. This was from a belief that this food would make girls more fertile and grow much faster. Traditional healing practices influence the dietary restrictions, limiting access to some nutritious foods, which are based on superstitions and lead to stigma. With regards to the childcare practices, there was also a lack of clarity by caregivers on the duration of exclusive breastfeeding as well as the duration when the children should stop breastfeeding. Caregivers did not have adequate knowledge about when to resume weaning. Some caregivers highlighted that the last time they received nutrition knowledge was when their children were infants, and they had taken the children for vaccinations. Caregivers were not aware of how responsive caregiving such as child feeding frequency and portion sizes could improve children’s nutritional outcomes.
Key findings from objective 5: From the qualitative interviews with early childhood development (ECD) practitioners, findings indicated a growing disintegration of childcare systems, including the family, health, and social systems, where a lack of parental support in nutrition programmes, a lack of support in health services and other social services when making referrals. Furthermore, various systems of care were working in silos in childcare service provision, resulting in children facing multiple adversities.
Conclusions: The study demonstrated that individual-level child characteristics appear to exacerbate childhood malnutrition more than the mother and household-level characteristics. For example, the child level characteristics showed high significance, with age, sex, and child support grant, compared to the caregiver characteristics such as education, employment, and income. At the household level, variables such as household size and income did not show any significance. While this is the case, it does not necessarily mean that the mother and household-level characteristics were not important. This gap can be explained by the small sample, which can cause challenges of limited statistical power, making it harder to detect statistically significant differences. Furthermore, the qualitative assessment filled some gaps regarding these findings and gave an in-depth understanding on how the income disparities among caregivers and households result from high unemployment rates, highlighting the importance of socio-economic status and food security in child nutritional outcomes. From the ECD practitioners’ interviews, given the disintegration of childcare systems, the coordination and multisectoral collaboration of different sectors of care for children is urgently needed to improve children’s nutritional outcomes. Understanding the social context in which a child is brought up is important for the design of programmes and policies that will be effective in addressing this public health challenge. This understanding will enable efficient and effective service referral and service delivery to improve childhood nutrition in South Africa. This study highlights the need for a good coordination of food, family, health, and social systems to ensure a positive childhood nutritional outcome.
Keywords: malnutrition, stunting, overweight, underweight, social context, food insecurity, childcare, socio-cultural practises, low-income, South Africa