Abstract:
Introduction: Even though remarkable declines in under-five mortality rates, nearly 5.3
million children still die annually worldwide before their fifth birthday. The 2016 Ethiopian
Demographic and Health Survey report revealed that 67 children per 1,000 live births died
before the fifth birthday. This study will aimed to evaluate spatial distribution and identifying
potential determinant factors of under-five child mortality in Ethiopia.
Methods: The data will retrieved from EDHS 2016, from which 10,641 total numbers of
under-five children will considered. Global Moran’s I and Moran scatter plot will used to
assess the spatial distribution of under-five child mortality rate. WAIC and DIC will be
applied to compare cox regression and Accelerated Failure Time survival models. The
Survival frailty models will be applied to identify the potential determinants of under-five
mortality using Bayesian approach.
Results: The test result of Moran’s I test statistics showed that the presence of significant
global positive spatial autocorrelation for the prevalence of Under-five child mortality.
Urban residence, Mother attained secondary and higher, age of mother at first birth,
improved source of drinking water; having toilet facility, multiple birth type, month of
breastfeeding, birth order of child, mother’s occupation, sex of child, mothers delivered in
health center were statistically significant factors to under-five mortality in Ethiopia.
Conclusion: Spatial distribution of under-five child mortality rates is clustered in Ethiopia.
The Weibull survival model under accelerated failure time with INLA has better fit the
dataset. Mothers aged greater than 35 at their first birth and birth order of seventh and
greater were the significant factors that increase the risk of mortality of children. This finding
suggests that enhancing opportunities to mother education, addressing regional disparities,
and encouraging mothers to deliver at health institutions will help to combat the burden of
under-five mortality
