Statistical analysis of regional heterogeneity of adverse pregnancy outcome in Ethiopia

Abstract

Adverse pregnancy outcome is defined as a pregnancy outcomes other than normal live birth. Adverse pregnancy outcome is still a major public health and socioeconomic status problems in developing countries including Ethiopia. GEE model and multilevel logistic regression models is an extension of GLM model. Multilevel logistic regression models to consider the unexplained variation within groups and between groups as random variability. Objectives: The main objectives of the study were determinants of adverse pregnancy outcomes using Generalized estimating equation model and multilevel logistic regression model as well as identify important determinant factors for the an adverse pregnancy outcomes in Ethiopia. Methods: The data for the study were taken from the 2016 EDHS and women in the age group of 15-49 years. Logistic regression model, GEE model and multilevel logistic regression models is used to explore the major risk factors and regional variations in adverse pregnancy outcome in Ethiopia, by using the STATA 14 software package. Results: Out of the 9694 reproductive age of women, 8412 (86.8%) not an experiencing adverse pregnancy outcome while 1282 (13.2%) of these women experiencing adverse pregnancy outcome of the time of the survey. All the fitted models Multiple logistic regression model, multilevel logistic regression model and generalized estimating equation model gave the same result that, Age of mother, place of residence, antenatal care visits, Parity, Education of mother, Marital status, Occupation of mother and Anemia level were found to be statistically significant at 5% level of significance a factors for an adverse pregnancy outcome among regions. The standard errors from the multilevel logistic regression models are smaller than GEE model. It was found that the random intercept with fixed slope model is an appropriate model for adverse pregnancy outcome. Conclusion: The random intercept with fixed slope multilevel logistic regression model provided the best fit for the data under consideration. There is a variation of adverse pregnancy outcome from region to region and within regions. This study found that not having Antenatal care, residing in rural area, working occupational status, being anemic, increased educational level, never marrieds marital status, being in age group of 15-24 or >35 years are associated with increased risk of adverse pregnancy outcome among reproductive age group women in Ethiopia. Mother’s, 35 years or older, should be able to avoid complications that come with age, should visit antenatal care during pregnancy.