Research On The Risk Factors And The Risk Prediction Model Of Gestational Diabetes Mellitus

Objective: Gestational diabetes mellitus(GDM)has been an important public health problem and a hot clinical research topic,which will cause short-term and long-term adverse effects on the health of mother and offspring.In the past decade,observational studies have identified a number of dietary and lifestyle factors associated with the risk of GDM,which demonstrated that the two consecutive phases before and during pregnancy may be closely related to the risk of GDM.In recent years,with the increasing incidence of GDM,the number of studies on the risk prediction model of GDM has been increasing year by year.However,the risk prediction models of GDM are based on age,nationality,pre-pregnancy BMI,still lack of the basic features in addition to the pregnant woman other than the risk factors associated with GDM(e.g.,dietary intake,sleep status,physical activity)into the risk prediction models.The purpose of this study is to establish a risk prediction model of GDM by exploring the association between the basic characteristics,physical activity,sleep factors,dietary intake before and during pregnancy,psychological status and other potential risk factors of pregnant women and the risk of GDM.Methods: A retrospective cohort study design was used to investigate the pregnancy and birth outcomes of pregnant women who came to our hospital for routine prenatal examination.The outpatient department of obstetrics of the affiliated hospital of China Medical University was used as the survey field,and pregnant women of 24~28 gestational weeks were selected as the object of study.Sleep quality was measured by Pittsburgh Sleep Quality Scale,mental state was measured by Edinburgh Postpartum Depression Scale and General Anxiety Disorder Scale,physical activity was measured by Chinese Maternal Physical Activity Questionnaire,and dietary intake before and during pregnancy was measured by semi-quantitative food frequency method.According to the guidelines for the diagnosis and treatment of gestational diabetes mellitus in China(2014)and the diagnostic criteria for GDM formulated by the American Diabetes Association,75 g oral glucose tolerance test was conducted for pregnant women between 24 and 28 weeks of gestation.Factor analysis was used to explore the dietary pattern of pregnant women.Logistic regression and generalized linear regression were used to analyze the risk factors of gestational diabetes mellitus.Predictive factors were screened according to AIC principle,and a predictive model of gestational diabetes mellitus was established.Results: The incidence of gestational diabetes mellitus was 22.5%.Age over 35 years(OR,6.21;95% CI,2.40-16.08),ethnic minorities(OR,3.93;95% CI,1.40-11.05),pre-pregnancy obesity(OR,2.62;95% CI,1.52-4.50),family history of diabetes mellitus(OR,1.69;1.01-2.86),history of spontaneous abortion(OR,3.15;95% CI,1.56-6.38),history of drug abortion(OR,4.02;95% CI,1.52-10.67),assisted reproductive technology(OR,4.28;2.40-7.66)and changes in dietary habits during pregnancy(OR,1.83;95% CI,1.26-2.64)are independent risk factors for gestational diabetes mellitus.One score of sleep quality(OR,0.37;95% CI,0.22-0.61)and total energy consumption of physical activity of pregnant women was 76.1~116.1 MET-h/week(OR,0.35;95% CI,0.20-0.62)are independent protected factors for GDM.Before pregnancy,pregnant women often adopt four dietary modes: fruit vegetable-protein mode,leaf vegetable-seafood mode,junk food mode and fruit mode.During pregnancy,they often adopt five dietary modes: mixed mode,fruit vegetable-fungus-seafood model,leaf-rhizome vegetable model,drinking model and fruit model.Pre-pregnancy fruit vegetable-protein model and junk food model were risk factors for GDM.Fruit vegetable-fungus-seafood pattern during pregnancy was the protective factor of GDM,while leaf-rhizome vegetable pattern and drinking pattern were the risk factors of GDM.The predictive factors of GDM risk prediction model for pregnant women were age,ethnicity,BMI before pregnancy,sleep quality,sleep efficiency,sleep disorder,physical activity,intake of junk food before pregnancy and intake of fruits,vegetables,fungi and seafood during pregnancy.The AUC of GDM risk prediction model for pregnant women in the validation cohort was 0.743(95%CI: 0.687,0.798)which had better distinguishing ability.The sensitivity and specificity of the model were 0.847 and 0.526,respectively.Emax = 0.113,Eavg = 0.038,and P Hosmer-Lemeshow = 0.549,which indicate that the goodness of fitting of model is good(P> 0.05).Conclusions: The risk prediction model of GDM established in this study has good performance.Diet before and during pregnancy,physical activity and sleep during pregnancy are important predictors and influencing factors of GDM.