论文部分内容阅读
目的:探索在控制相关因素后,利用孕期膳食能量摄入和体力活动来预测巨大儿的方法。方法:收集866例孕妇在孕中期和孕晚期的膳食摄入和体力活动资料,待其分娩后记录新生儿出生体重。在控制相关混杂因素的条件下,拟合孕期膳食能量摄入与体力活动对巨大儿发生率的多元Logistic回归模型,建立孕期预测巨大儿发生的模型,使用ROC曲线来判断模型对巨大儿的预测能力。应用灵敏度、特异度和约登指数来探索巨大儿发生的预测函数值的分界点。结果:根据预测模型,拟合ROC曲线,得出曲线下面积AUC为0.846,表明模型有较好的预测能力。将预测函数值0.153为分界点判断是否发生巨大儿,此时的灵敏度为0.83,特异度为0.77。结论:利用孕期膳食能量摄入和体力活动探索孕期巨大儿预测模型可以经济、有效地提示胎儿生长过速的现象。建模方法可为日常围产保健工作提供借鉴。
OBJECTIVE: To explore ways to predict macrosomia by using dietary energy intake during pregnancy and physical activity after controlling for related factors. Methods: 866 pregnant women in the second trimester and third trimester of dietary intake and physical activity data, to be born after birth record newborn birth weight. Multivariate Logistic regression model fitted to the incidence of macrosomia in pregnant women during dietary intake and physical activity during pregnancy was established under the control of confounding factors to establish a model for predicting macrosomia in pregnancy and ROC curve was used to judge the prediction of macrosomia ability. Sensitivity, specificity, and the Youden index were used to explore the cut-off point for predictive function values occurring in macrosomia. Results: According to the prediction model, the ROC curve was fitted and the area under the curve AUC was 0.846, indicating that the model has a good predictive ability. The predicted value of 0.153 as the cut-off point to determine whether there is a huge child, the sensitivity of 0.83 at this time, specificity of 0.77. CONCLUSION: The use of dietary energy intake and physical activity during pregnancy to explore the prediction model of prenatal macrosomia can prompt the fetus to grow rapidly and economically. Modeling methods can provide reference for daily perinatal health care work.