Count Regression Modeling of Infant Mortality in Ethiopia.

Abstract

Infant mortality is the death of a child at any time after birth and before reaching his or her first birth day. This death toll is measured by the infant mortality rate. Sub-Saharan Africa has the world’s highest rate of infant and child mortality. Mortality rates tend to be higher for infants and child mortality. The aim of this study was to examine the spatial distribution and factors that affect infant mortality based on 2016 Ethiopia Demographic and health survey dataset using spatial and multilevel Count Regression modeling models. The descriptive result showed that a total of 10547 mothers’ were covered from 11 regions. And also the descriptive Statistics states that the number of infant death calculated showed (2.18) that the variance (0.994) is greater than the mean (0.526) indicating over-dispersion. And the highest mean number of infant mortality is occurred in Somali (0.69) and the lowest is, in Addis Ababa (0.089). The spatial distribution result using ArcGis software reveal that 71% of the mothers have not face death, whereas 0.6% of them faced at least 5 infant death per mother in their life time before survey. According to the result of GeoDa results infant mortality per mother were clustered in national level. High risk of infant mortality was mostly occurred in the eastern part of all regions and low risk of infant mortality in central and western part of the regions. According to multilevel Poisson model using R out put the random intercept and slope model of multilevel ZINB model are the best fit the data, and the study also showed that there is a significant regional variations of infant mortality and also the result also revealed that infant deaths per mother differs among regions of the country in terms of residence, age of mother and household size. Accordingly random ZINB result, the variables residence, age of mother, household size, age of mother at first birth, breast feeding, weight of child, use of contraceptive, birth order, wealth index, father education level and birth index were found to have significant effect with infant mortality. This study also suggest that efforts are needed to extend educational programmers aimed at educating mothers on the benefits of contraceptive use, age of first birth, and spacing birth interval in order to reduce infant mortality.