Household environmental factors and childhood malnutrition: evidence from the SANHANES-1

Abstract

Background: Food insecurity impacts childhood nutritional status as well as physical and cognitive development, and increases lifetime risk for chronic disease. Malnutrition, as evidenced by stunting, underweight, overweight and obesity, remains pervasive in the paediatric population. Children’s dietary knowledge also has long-term implications on overall health and nutritional status via the development of lifetime feeding habits that endure into adulthood, predisposing to non-communicable diseases (NCDs) such as hypertension, heart disease, stroke, and diabetes. Previous South African studies have examined hunger at sub-national levels with a limited focus on children in the context of their immediate household environment. The true extent of childhood food insecurity as well as contributing factors in the environment have not yet been elucidated completely.

Aim: We sought to determine the national prevalence of childhood food insecurity (0- 19 years old) and identify factors associated with hunger within the household, with a particular focus on the household head. We also examined food insecurity in households without children. Furthermore, we explored the impact of food insecurity and dietary knowledge on childhood anthropometric indicators of malnutrition and ascertained the percentage of self-reported NCDs among adults (>19 years) living in households with children. Methods: Individual and, householdlevel data were extracted from the first wave of the South African National Health and Nutrition Examination Survey (SANHANES-1) to conduct secondary data analyses. Food insecurity was assessed using the Community Childhood Hunger Identification Project (CCHIP) index. Nutritional status for children aged 0-60 months was assessed using Weight for Height (WHZ) and Height for Age (HAZ) scores which referenced the 2006 WHO Child Growth Standards. For children aged >5-19 years, nutritional status was assessed using BMI for age (BAZ) and HAZ scores which referenced the 2007 WHO Growth Reference for children aged >5-19 years. Dietary knowledge of children aged 10-14 years was assessed using a general nutrition knowledge questionnaire. Multinomial logistic regression analyses were conducted on all households (with and without children) to determine the predictors of food insecurity with adjusted odds ratios (AOR) and 95% CI as measures of association. Two-way ANOVA tests with Tukey post hoc corrections were employed to assess the associations between anthropometric indicators and food security status among all children.

Results: Of 5 098 households with complete CCHIP scores, 68.64% had children (0-19 years). Of the households with children, barely 40.25% were food secure while 32.55% [95% CI (29.51 – 35.73)] were experiencing hunger and 26.37% [95% CI (23.96 – 28.82)] were at risk of hunger. Among all the households, significant associations for households experiencing hunger were: the presence of children: AOR [95% CI]: 1.68 [1.12 – 2.53]; being female-headed: AOR [95% CI]: 1.53 [1.21 – 1.94] and informally-located; AOR [95% CI]: 1.61 [1.07 – 2.43]. Having a non-African household head (Coloured: AOR [95% CI]: 0.29 [0.19 – 0.44] and White/Indian/Asian: 0.12 [0.04 – 0.33]) was protective against experiencing hunger. Having a household head with a secondary/tertiary educational attainment was also protective against experiencing hunger; AOR [95% CI]: 0.40 [0.28 – 0.56] and being at risk of hunger; AOR [95% CI]: 0.69 [0.52 – 0.92]. Receiving social grants or remittances more than doubled the risk of experiencing hunger; AOR [95% CI]: 2.15 [1.49 – 3.09]. Children and adolescents (>5-19 years) from households that were at risk of hunger and those that experienced hunger were thinner (had lower BAZ scores) compared to children from food secure households (p=0.013) and (p=0.019), respectively. Also, children >5 – 19 years old from food insecure households (both at risk of hunger and experiencing hunger) were shorter (had lower HAZ scores) than their food secure counterparts (p=0.004) and (p<0.001), respectively. Among younger children (0-60 months), there was no association between food security status with WHZ; however, experiencing hunger was significant for lower HAZ scores (p=0.038). Among children aged 10-14 years, dietary knowledge lacked associations with nutritional status. Lastly, hypertension (23.04%) was the most frequently selfreported NCD among adults from these households with children, followed by diabetes (7.59%)

Conclusion: This secondary analysis of the SANHANES-1 data points to the continued vulnerability of children and their maternal caregivers to the risks of food insecurity and poor nutritional status, with implications across the life course. This is in line with other South African research, including the finding that social grants are not protective against food insecurity. Teenagers have poor nutritional knowledge, which also impacts better food choices in late adolescence and into adulthood. The low self-reporting of NCDs in this adult sample is surprising but is perhaps indicative of people being unaware of their chronic health problems. These findings point to the need for renewed efforts to address the constitutional right to food and basic nutrition, and other social determinants of health, for children and the households they inhabit.