The Study Of The Risk Factors And Forecasting Models Of Small For Gestational Age

Objectives:the definition of SGA, which is short for small for gestational age, defines a sort of neonates with lower birthweight, which is one of the main risk factors for infant morbidity and mortality. To be specific, they are more susceptible to neonatal respiratory distress Syndrome (NRDS), retinopathy of prematurity (ROP), necrotizing enterocolitis (NEC), so on and so forth. Moreover, when they come to school stage, they are still more liable to be bothered by cognitive impairment as well as learning disability. What’s more, when the SGAs are grown up, they are prone to suffer from short statures and metabolic syndrome. For all the reasons that have been mentioned above, it is of great importance to know the risk factors of SGA babies, so that early detection and treatment will be realistic in the future. To date, a variety of factors have been confirmed to have influence on foetal growth and birthweight, and can be grouped into several general categories, namely genetic factors, parental socio-demographic factors, maternal living conditions, maternal psychologic status and maternal health status as well. In addition, many researchers have already successfully utilized some of these factors and developed several models to predict the risk of SGA. However, all the present studies were based on samples from limited populations, and most of the predict models were built by using classic statistics method, which means more studies are still need.Patients and Methods:the cases with pregnancy outcomes of the National Pre-pregnancy Examination Program were obtained from the database, after simple data processing, Chi-squared test, Mann-Whitney U test and Logistic regression were used to analyze these data, then we built a series of forecasting models based on the SVM theory.Results:according to our study, the incidence of SGA is 6.0% when 3rd percentile of birthweight is used as the diagnostic cutoff point of SGA, and the number rises to 11.2% when we use 10th percentile of the birthweight as the diagnostic criteria. And the incidence of SGA varies from 5.6% to 12.0%, depending on different baby sex and geography area. And among all kinds of influence factors, parental constitutional factors, social-psychological status, living habits, previous history on reproductive systems, environmental risk factors, the health conditions during pregnancy as well as the results of routine physical examinations are associated with SGA independently. Furthermore, the models we built to predict the risk of SGA by utilizing SVM theory are of relatively high accuracy.Conclusions:the outcomes of our study show that parental constitutional factors, social-psychological status, living habits, previous history on reproductive systems, environmental risk factors, the health conditions during pregnancy as well as the results of routine physical examinations are related to the incidence of SGA, which consist with some of the previous studies, and our study have also proved that SVM theory can be used in establishing forecasting model of diseases.