The Study Of Influencing Factors Of Ten Top Birth Defects And Theirs Forecast Model In Hunan

Part One:The Study of Influencing Factors of Ten Top Birth Defects in HunanObjective:To explore the influencing factor of ten top birth defects in Hunan,and provide scientific evidence to comprehensive prevention of birth defects.Method:Adopting the 1:2 matched case-control study to explore the influencing factor of ten top birth defects in Hunan;all the objects using the unification questionnaire,filled out face-to-face through gynecologist's inquiry;the data was entryed by using Epidata3.2;the general information of mother and prenatal was get by descriptive analysis;condition logistic regression was adopted to conduct Single factor and multi-factor analysis of influencing factor of birth defects;the statistical analysis software is SPSS15.0.Results:The research got a total of 321 1:2 matched Study pairs,the analysis of single factor condition logistic regression shows that: high-income families;Participate in pre-marital medical examination, regular consumption of meat,fruits and vegetables,beans and milk during pregnancy are protective factors of birth defects;Family genetic history,multiple births,exposure to occupational risk factors before or during pregnancy,suffering from chronic diseases,early pregnancy flu, embryonic bleeding / abortion symptoms in early pregnancy,contact with pets before or during pregnancy,housing renovation before or during pregnancy,Place of residence exists environment pollution sources,dyed hair before or during pregnancy,fond of Kiamichi,pickled and smoked food during pregnancy,contact with agricultural chemicals,the pregnant woman has abnormal reproductive history,husband exposure to occupational risk factors,husband like drinking strong tea and regular smoking are suspicious risk factors of birth defects.The analysis of multi-factor condition logistic regression shows that:regular consumption of fish,meat,eggs,etc during pregnancy(OR=0.28,OR 95%CI:0.15-0.88),Family in high per capita income(OR=0.30,OR 95%CI:0.18-0.51),regular consumption of milk,beans during pregnancy,(OR=0.36, OR95%CI:0.22-0.88),Participate in pre-marital medical examination (OR=0.64,OR 95%CI:0.39-0.96)are the four protective factors of birth defects;Family genetic history(OR=2.59,OR 95%CI:1.05-6.45),twins or multiple births(OR=5.16,OR 95%CI:1.54-17.30),Place of residence exists environment pollution sources(OR =3.32,OR95%CI:1.65-7.82),suffering from chronic diseases before or during pregnancy(OR=3.02,OR 95%CI:1.37-6.65),maternal exposure to occupational risk factors before or during pregnancy(OR=2.26,OR 95%CI:1.24-17.30),has abnormal reproductive history(OR=2.16,OR95%CI:1.1-4.17),catch a cold during pregnancy(OR =1.71,OR95%CI:1.19-5.91)and husband like drinking strong tea(OR=1.70,OR95%CI:1.10-2.65)are the eight suspicious risk factors of birth defects.Conclusion:Participate in pre-marital medical examination,regular consumption of meat,beans and milk during pregnancy,have fewer child, avoid exposure to occupational risk factors,prevention disease during Pregnancy,away from environmental pollution,husband coordinate with maternal are the important measures to prevent birth defects. Part Two:The Research of Forecast model of Ten Top Birth Defects in HunanObjective:Introduce the principle of Decision Tree and its application in the forecasting of birth defects,in order to provide a new way in the study of birth defects.Methods:Adopting the 1:2 matched case-control study to explore the influencing factor of ten top birth defects in Hunan;all the objects using the unification questionnaire.Constructing forecast models by Decision Tree C5.0 algorithm and discriminate analysis.Clementine11.0 and SPSS 15.0 are used for Statistical analysis.Results:The coincident rate of Decision Tree categories with actual categories was 83.7%;the sensitivity was74.1%;the specificity was 88.6%.The coincident rate of discriminate categories with actual categories was 64.7%;the sensitivity was 54.0%;the specificity was 70.3%.Compared with discriminate analysis,Decision Tree methods had a better forecasting precision for its higher accuracy rate.Conclusions:Compared with the conventional statistic method,Decision Tree methods had better forecasting precision.Through the creation of pregnant women information databases,adopting Decision Tree method and choosing indicators of high quality with professional knowledge can play a very good forecasting effect to Birth Defects.