Diagnosing some important factors that affect spontaneous abortions using the linear discriminant model

Abstract

Discriminant analysis is one of the statistical tools for multivariate descriptive analysis that has been widely used in medical studies. The research problem was the increase in spontaneous abortions by 10% to 50% of pregnancies with clear clinical miscarriages. This percentage is considered high, and its accompanying negative repercussions affect pregnancy. It was represented by introversion, an intense feeling of sadness and loss, irritability, self-rejection and guilt, preoccupation with thinking about getting pregnant again, anxiety and nightmares, eating disorders, thoughts or attempts at suicide...etc , The research aimed to use discriminant analysis to determine the most important factors and their relative importance influencing spontaneous abortion through the extent to which each factor contributes to distinguishing between the two groups. The research also aims to predict the possibility of spontaneous abortion for pregnant women based on the influencing factors, leading to early diagnosis and avoiding spontaneous abortion, A sample size of 100 pregnant was used, including 50 spontaneous abortions, 50 live births, and 14 independent variables. It appeared that the variables that had a significant effect on spontaneous abortion were (age x₁, first pregnancy x₈, diabetes in the pregnant woman x₁₃, blood type matching between spouses x₄, Educational level of the pregnant woman x ₃, previous miscarriages x ₉, high blood pressure in the pregnant woman x ₁₁, previous dead births x ₁₄). The contribution rate of variables in distinguishing and classifying between the two groups was (6.54%)