Abstract
The early and late introduction of complementary food, both prevalent in Ethiopia, are associated with morbidities, growth faltering and developmental risks in children. The interhousehold network around the primary caregiver's intrahousehold network is critical in influencing the age of introducing complementary foods. This study examined the influence of household composition and structures on complementary food introduction. This is a secondary data analysis of four Ethiopian Demographic and Health Surveys conducted between 2000 and 2016. The household structure and composition variables were calculated from household members' kinship status and attribute, respectively. The introduction of solid, semisolid or soft foods was dichotomised as whether the children within 6 to 8 months have been given complementary foods. Multivariable logistic regression with adjustment for the primary caregiver and household characteristics was run to examine the associations between household structure and composition variables and the introduction of complementary foods. The marginal effects (ME) were calculated to facilitate the practical interpretation of the study findings. Large households (>3 nonredundant contacts) with extended family or unrelated people (high effective size, ME = 6.01%, 95% confidence interval [CI]: -8.53, -3.49) lowered the proportion of children starting food within the recommended 6-8 months. Households with close kins (high constraint) (ME = 7.22%, 95% CI: -13.65, 28.09) and greater age diversity (ME = 0.65%, 95% CI: 0.15, 1.15) increased the proportion of children receiving complementary food at an appropriate age. This study revealed that interhousehold structure and composition influence the age of introduction of complementary foods. These factors, therefore, need to be considered in designing interventions to improve age at the introduction of complementary foods.