About 25% of all children under 5 years in South Africa are severely malnourished. The objective of this study was to assess the comparative effect of person-related and household characteristics on nutritional status of children <5years old in South Africa before 1994 realization of democracy. A secondary analysis of 4 027 children under 5 years from the 1993 Living Standards and Measurement Survey (LSMS) conducted by the Southern Africa Labour and Development Research Unit (SALDRU) in South Africa was performed. Only children with plausible Z-scores (n = 4 027) were selected for subsequent analyses.
The influence of person-related and household related characteristics on the nutritional status of children were assessed, taking into account variables such as, gender of household head, de jure and de facto household head, relationship of child to household head, size of household, type of toilet facility and type of dwelling. Statistical Package for Social Scientists (SPSS/PC+
) was used for the analysis. Chronic malnutrition and underweight were significantly pronounced in children from households with de jure household heads (P<0.05).
However, chronic malnutrition was markedly prevalent among children from male-headed households. Thus, the presence of the de facto household head and the gender of the head of the household are important in determining nutrition outcomes. Furthermore, children of the head of the household had better nutritional status (P<0.05) than the grandchildren and other children in the household. Both person related and household related variables were significantly associated with malnutrition, but at varied levels. Household expenditure, particularly on the basic needs of life, is the most frequently used measure of socio-economic status in nutritional analysis studies. Expenditure is considered as precisely representing the household’s reserves that influence the health status of its members. In the absence of variables used for measuring household expenditure proxy variables are used.
The proxy variables (type of dwelling, household size, water source, and toilet location) for economic status of households seem to influence nutritional status more directly while the person related variables seem to indirectly influence nutritional status.