Abstract:
Background:Anemia is one of the most widespread public health problems, especially in
developing countries. It is usually caused by iron deficiency, which is the most common nutrient
deficiency in the world. In Ethiopia, anemia is the severe public health problem affecting 62.7%
of pregnant mothers and 37.5% of women were not anemic. This study assesses the level of
anemia level of pregnant women in Ethiopia. The exploration and identifications of possible
factors related to anemia level of pregnant women across different regions formulate key
objectives and explain the regional-level variations.
Methods:The data source was a 2016 Ethiopian Demographic and Health survey (EDHS) which
involving 1053 pregnant women respondents in 11 regions. Single level ordered logistic
regression, multilevel ordered logistic, and Bayesian multilevel ordered logistic regression
analyses have been applied.
Results: Based on all models the result reveals that pregnant women living in urban area, higher
level of education, lower level of pregnancy duration, having improved toilet and having health
facility access are more likely to have lower prevalence of anemic level. According to empty
multilevel result, there is a significance evidence of heterogeneity prevalence of anemia level
across regions for pregnant women. On a regional- level, mean while, the prevalence of anemic
level is strongly associated with the health access facility. Health access facility and regional
variance have highly negative correlation. It makes a difference on anemic prevalence in
different region. Moreover, the result of the Bayesian multilevel random intercept model is best
fitted as compared to Bayesian multilevel random slope; and has smallest standared error as
compared to any classical model included in this study. Thus, it can be concluded that Bayesian
multilevel random intercept model is best fitted.
Conclusions: Based on the result, there exist significant differences of anemic distribution
among pregnant women in different regions of Ethiopia. As there is variation the government
should be balance health facility access, educational access, expand urbanization and improve
wealth distribution in all regions. Furthermore researchers should be applied Bayesian analysis
for further study in different fields.
