Bayesian Multilevel-modelling of Acute Respiratory Infections among Under Five Children in Ethiopia

Abstract

Introduction: Acute respiratory tract infection (ARI) is a leading cause of morbidity and mortality in children under the age of 5 years throughout the world. In Ethiopia the prevalence of ARI among under five children was 7% according to 2016 EDHS report. Objective: This study was aimed to determine factors associated with the prevalence of acute respiratory infection (ARI) among under-five children in Ethiopia using Bayesian multilevel approach. Methods: This study has been conducted in Ethiopia based on data obtained from EDHS 2016 which mainly focused on under-five children (age from 0-59 months). The survey was collected from a total of 10,641 children aged less than five years out of which 9,918 children were considered in this study. The study was used Bayesian Multilevel Logistic Regression Model to investigate the major risk factors and regional variations in ARI among under five children in Ethiopia. To determine the posterior marginal, the MCMC methods with non-informative priors have been applied. Deviance information criterion was used to compare models. Results:The analysis showed that the overall ARI prevalence rate between 9,918 under five children was 8.4%. Based on DIC, Bayesian multilevel logistic regression of random coefficient model was found to be more appropriate than empty and intercept model. According to the output of the model childs age, household wealth index, women’s educational level, vitamin A supplement, history of diarrhea, maternal work, stunting and source of drinking water were found to be significant factors for prevalence of ARI among children under five years. Conclusions: This study revealed that there is a significant variation of incidence of ARI among under-five children between and within the regions of Ethiopia. The study recommends all regional states to make remedial measures on public health policy and improve the abilities of stakeholder living in their region toward those major factors associated with the prevalence of ARI among under five children.