Statistical Analysis of Anemic Status among Pregnant Women in Ethiopia

Abstract

Anemia in pregnancy is related to different socio-demographic, dietary and economic factors, Mother’s age, educational status, economic position, and antenatal care were significantly associated with anemia during pregnancy. Objective: To fit an appropriate statistical model and identify potential factors of anemic status among pregnant women in Ethiopia Methods: A cross-sectional but cluster study carried out based on the secondary data of the Ethiopia Demographic Health Survey. Data of a total of 1277 pregnant women of reproductive age (15-49) were included in the analysis. Data were mainly analyzed using that SAS software offers for the analysis of binary responses with correlated data (GENMOD procedure), and both marginal and cluster specific data (NLMIXED procedure). It has been showed that each of them estimates parameters from among different statistical models and comment on the interpretation of parameters and the statistical properties of the methods involved. For the categorized response variable (non normal response), General linear model, over dispersion, Generalized Estimating Equation, Generalized Linear Mixed model, and Marginalized Multilevel model were used to identify the associated factors of anemic status among pregnant women. Result: some of covariates for the marginal model revealed that pregnant women those lived in urban had 0.862 (p = 0.0012) times lower risk than those who lived in rural or the probability that the pregnant women those who lived in urban had 13.8% times less likely to develop anemia than those who lived in rural, on the other hand the pregnant women whose education status was poor had 2.087(p-value=0.0001) times higher risk to develop anemia than those whose education status was higher .Similarly, the number of pregnant women who had HIV –positive had 1.39 (p = 0.0001) times higher risk than their counterparts and similar results were obtained in cluster specific and marginalized multilevel model. Recommendation: Government should design strategies and policies to enhance women education to make them independent in socio-economic and cultural decision, which directly and indirectly affect women health status due to anemia. It is recommended that the remaining factors that have not been included in this study could be included in future studies.