Socio-economic Inequalities in Neonatal Mortality in Kenya: a Decomposition Analysis

Abstract

Introduction: Neonatal mortality represents a major proportion of all the under-five deaths in Kenya. Socioeconomic inequalities have been found to be related to the distribution of health variables. This study aimed at measuring the socioeconomic inequalities in neonatal mortality in Kenya as well as to decompose the inequality into its various determinants. In addition, the study determined the inequality trends in the socioeconomic inequality over time. Methods: This study used the data collected during the Kenya Demographic Health Surveys (KDHS) of 2014 and 2008/09. Data on the household demographic, environmental and socioeconomic characteristics were obtained from the household questionnaire while child mortality data was derived from the woman’s questionnaire. Neonatal death was the binary dependent variable with various independent variables. Univariate and bivariate analyses were used to show frequencies and distribution of variables with respect to the dependent variable. Multiple logistic regression analysis was done to depict the association between neonatal mortality and various independent variables. Concentration curve was plotted to show the graph of the inequality in neonatal mortality. Concentration Index was used to measure the socioeconomic inequalities in neonatal mortality. Decomposition analysis of the concentration index was done to determine the extent to which various variables contribute to the inequalities in neonatal mortality. STATA version 14.2 and R 3.4.4 software were used to conduct the statistical analyses. Results: There were 1954 neonatal deaths compared to the 81637 neonates who survived beyond the neonatal stage. Neonatal mortality was significantly associated with sex of the child, twin status of the child, place of residence and mother’s education level. For both the 2008 and 2014 surveys, there was a pro-poor inequality in neonatal mortality evidenced by a negative concentration index. Decomposition results reveal that wealth status and education levels explain most of the inequality in neonatal mortality for both years. Conclusions: Most of the inequality in neonatal mortality occurs because of the disparities in education and income levels. Health insurance is also an important determinant of the inequality in neonatal mortality. Access to education should be promoted especially among the poor households as it will reduce the inequality in neonatal mortality. Economic empowerment programs targeting the poor will reduce the wealth disparities hence reducing the inequalities in neonatal mortality.