The relationship between household welfare and infant mortality in Turkey: evidence from TDHS-2008

Baris Ucar, Hacettepe University

This study tries to understand whether there is a relationship between infant mortality and household welfare in Turkey. Data from TDHS-2008 was used in this study. Wealth index was used to measure household welfare. Besides descriptive analyses logistic regression method was realized to understand the determinants of infant mortality for the 1998-2008 birth cohort. In the descriptive analyses it was seen that infants were more likely to die in the poor households. While infant mortality rate is 11,6 in the rich households, it is 17,3 in the middle wealth group households and as high as 35,8 in the poor households. Although wealth index has an effect on infant mortality when no other determinant is included in the model, the results of the logistic regression showed that wealth index is not a determinant of infant mortality when other factors are included in the model. This implies that when other factors are equalized wealth index does not have an effect on infant mortality. Besides smoking in the house, some of the maternal factors such as preceeding and succeding birth intervals, and age of mother at birth were found to be effective on infant mortality in the final model. Also a model was constructed for different wealth groups. Analyses on the determinants of infant mortality in poorer hoseholds shows that besides the determinants of infant mortality in the general model, for poorer households sex of the child, health insurance status and family type was found to be significant. The explanatory power of the model for poorer households was lower than that of the general model.