Development and evaluation of an early risk prediction model for gestational diabetes mellitus
Objective To analyze the risk factors of gestational diabetes mellitus(GDM),with which to develop and preliminarily evaluate a risk prediction model,so as to provide guidance for the prevention and treatment of GDM.Methods Five maternal and child health service centers in Hebei Province were randomly selected,from which a total of 1 738 pregnant women with complete information were enrolled and randomly divided into model development group and model validation group.Univariate and multivariate logistic regression models were used to select the independent risk factors of GDM and to fit the prediction model with the best performance.Homer-Lemeshow goodness-of-fit test and receiver operating characteristic(ROC)curve were used to evaluate the performance of the model.Results In the model development group comprised of 1 390 pregnant women,age over 35 years old(OR=1.82,95％CI:1.27-4.13),pre-pregnancy body mass index(BMI)being over 24.0 kg/m2(OR=2.26,95％CI:1.19-3.98),elevated fasting blood glucose during early pregnancy(OR=2.47,95％CI:1.62-3.70),history of diabetes among first-degree relatives(OR=2.39,95％CI:1.09-4.42),and anemia during early pregnancy(OR=1.22,95％CI:1.14-1.36)were independent risk factors for GDM(P＜0.05).The goodness-of-fit test of the selected logistic model incorporating the above risk factors showed P=0.796,the area under the ROC curve was 0.82.With a cut-off value of 0.451,the sensitivity of the model in the model validation group was 71.93％,and the specificity was 86.25％.Conclusions The prediction model based on the five identified risk factors demonstrated good fitting and high predictive performance,which may facilitate identifying high-risk populations for early management to reduce the incidence of GDM.
Gestational diabetes mellitusRisk prediction modelModel evaluationCohort study
NETL
NSTL
万方数据
