Determinants of infant and child mortality in Kilimanjaro: logistic regression approach

This study attempts to identify socio-economic and demographic factors which influence infant and child mortality in Kilimanjaro. The extent to which socio-economic factors operate through proximate determinants to bring the influence is also examined. The variables covered in the study are: Maternal education, maternal occupation, household income, residence, source of water, type of toilet facility, duration of breast-feeding, maternal age, parity and birth interval. Data on mortality and related variables were collected in a retrospective survey conducted in Kilimanjaro in July 1993. Logistic regression analysis is utilized to identify variables which have a significant effect on infant and child mortality. Maternal demographic status (an index of age and parity), breast-feeding duration and birth interval were found to have statistically significant effect on infant and child mortality. Maternal education and occupation had little significant effect on infant and child mortality. The significance however, disappeared when the intervening and proximate determinants were controlled for. The absence of significant effects on child survival of residence, source of water and type of toilet facility was also noticed.