Household structure, socioeconomic level, and child health in sub-Saharan Africa

The households studied are divided into two primary groups: elementary (parents and their biological children) and extended (either laterally, to include siblings, cousins, and other relatives of the household head, or three-generationally, to include parents or parents-in-law of the head). The study finds considerable variation across countries in children"s living arrangements. Elementary households predominate in 4 of the 11 countries, extended households in another 4, while in the remaining 3 countries children are almost equally divided between elementary and extended households. There are household structure differences in relative poverty, particularly in rural areas. In 9 countries, rural children from elementary households are socioeconomically more disadvantaged than those from extended households. In addition, children from elementary households tend to be worse off than those from laterally extended households in terms of mother"s education and are less likely to reside with adults who have secondary or higher levels of schooling. Full immunization coverage tends to be positively correlated with the socioeconomic level of the household and negatively correlated with elementary household structure. In some countries, however, household structure differences in full immunization coverage are explained by differences in socioeconomic level. Moreover, although the lowest levels of immunization coverage tend to be found in elementary households, in rural Rwanda and urban Madagascar, children from three generational households have significantly low levels of immunization coverage, even after controlling for other factors. In almost all countries, children are less likely to receive oral rehydration therapy (ORT) in elementary than in laterally extended households. Children from laterally extended households tend to be the most likely to be taken for medical treatment of diarrhea and to receive fluids. There is no consistent evidence that the relationship between household structure and diarrhea treatment practices is mediated by socioeconomic level. In Burkina Faso, Cameroon, Niger, and Zambia, household structure differences in ORT are largely explained by socioeconomic differences. However, in Senegal, household structure has a statistically significant effect on receipt of ORT, even after controlling for socioeconomic level and other factors. (Author abstract, modified)