Multilevel Modeling on the Anemia status of Women in Ethiopia

Abstract

Background: In developing country like Ethiopia as a whole and globally Anemia is the most common public problem caused by nutritional deficiency diseases among women at reproductive age. Hence, this study was determining the regional variation and associated factors of anemia status among women at reproductive age in Ethiopia using multilevel model. Method: A cross-sectional study was conducted among 14489 women enrolled in Ethiopia demographic and health survey of 2016 which nested under nine regions and 2 administrative city .The data was entered into Stata version-14 for cleaning and data analysis. Binary and multilevel logistic regression was carried out for variables to determine associated factors with anemia status of women and its regional variations at ascertained of 5% level. Result: The random intercept with fixed effect of the covariates multilevel model was considered as best fit of the data set based on Akaike information criteria when compared with other candidates models. In determining the potential factors associated with anemia status of women, the study indicated that women who use improved source of drinking water (OR=1.98, 95%CI=1.05, 3.72), being in middle wealth index (OR=0.25, 95%CI=0.10, 0.63), being in rich wealth index (OR=0.42, 95%CI=0.19, 0.94), having age at 1st birth in 20-24 years(OR=0.24, 95%CI=0.11, 0.53), having number of living children 1-2(OR=3.68, 95%CI=3.48, 4.98), having number of living children 3-4(OR=3.03, 95%CI=2.48, 4.05)and used government place of delivery(OR=0.96, 95%CI=0.22, 1.70) were significantly related to anemia status of women at 5 % level of significance. Conclusions: The findings of this study showed that wealth index, age of women at 1st birth, number of living children, source of drinking water and place of delivery were potential covariates associated to anemia status of women and there were regional variability. It is recommended that health workers should be give attention to these proximate determinants on anemia at regional level.

Keywords

• Multilevel model; nutritional Deficiency; women; Ethiopia demographic and health survey.