Abstract:
Background: Neonatal mortality in low-and middle-income countries has shown slow decline and marked disparities across regions and countries. In Ethiopia, despite making overall progress in child survival, deaths among newborn babies still remain high at 29 deaths per 1,000 live births. Evidence on trend and inequalities in neonatal mortality in Ethiopia is scarce and yet, such evidence is important to contribute to the effort towards reducing neonatal mortality. Objectives: This study aims to assess the trend and inequalities in neonatal mortality by the five equity stratifiers, namely region, place of residence, wealth status, maternal education and composite coverage index (CCI) using the four Ethiopia Demographic and Heath Survey (EDHS) rounds. Methods: Using the four nationally representative survey conducted in 2000, 2005, 2011 and 2016, direct estimate of neonatal mortality rate were computed from birth history information provided by mothers. The trend in neonatal mortality were measured using annual rate of reduction and 95% CI to indicate the period when significant difference was observed. Once the data from the four rounds pooled together, we used both absolute and relative inequality measures to measure inequality in neonatal mortality. Concentration index was computed for the socioeconomic inequality and for area-based inequality, difference and ratio were calculated. We further did decomposition analysis for the wealth-based inequality to understand individual percentage influence to the observed inequality in neonatal death of the commonly known factors of neonatal death. A p-value of < 0.05 and 95% CI was reported as a measure of magnitude of significance. Result: Overall trend in neonatal mortality rate was slowly declining with the annual rate of reduction of 1.98 percent per annum between 2000 and 2016, and also great variation in average decline by selected equity stratifiers. Well served regions in reproductive maternal and child health service had comparatively lower mortality rate than underserved regions. In all the survey years there is wealth and education-based inequality in neonatal mortality, however the levels were not significant except for the 2011. An estimate of concentration index indicates (-.002, .009, -.012, .012) for wealth and (-.011, -.001, -.009, -.003) for education with p-value of (0.68, 0.07, 0.02, 0.05) and (0.00, 0.67, 0.00, 0.53) respectively. The absolute and relative inequalities in terms of regions showed greatest inequalities between regions with the highest mortality(Amhara) and regions with lowest mortality(Addis Ababa) with a difference = 0.0141, 95% CI (0.0059, 0.0222) and Ratio = 1.55 with 95% CI (1.1584, 2.0767) were as for place of residence (rural Vs urban) it showed non-significant, Difference = -0.0089 (-0.0222, 0.0045) and Ratio = 0.8054 (0.5972, 1.0863) respectively. The major contributors for observed wealth-based inequalities were different levels of wealth status, underserved regions and rural residence. Conclusion: Trend in neonatal mortality have showed slow decline and it was not uniform among the selected equity stratifiers, progress in annual rate of reduction was also sluggish. Across the survey years, though mortality is concentrated among socioeconomically disadvantaged, the levels of wealth and education-based inequalities were significant only for the 2011 survey. Furthermore, estimates of absolute and relative inequalities for regions shows significant inequalities. It is recommended to maintain improvement in Reproductive, Maternal & Newborn Child Health (RMNCH) services and scaling up of interventions that directly act on leading determinants of neonatal mortality by focusing on the general population in addition to targeting the disadvantaged society, rural residents and underserved regions.