以Logistic回归模型构建病人健康问卷与广泛性焦虑量表联合预测因子对妊娠晚期围产期抑郁障碍预测的临床运用

Formulation of combined predictive indicators with patients health questionnaire depression scale-9 and generalized anxiety disorder 7-item scale using Logistic regression model in predicting perinatal depression disorder in the third trimester of pregnancy

目的:探索妊娠晚期病人健康问卷(PHQ-9)联合广泛性焦虑量表(GAD-7)评分对围产期抑郁障碍(PND)的预测效果,为早期识别提供依据。方法:选取2020年2月至2020年6月于中国福利会国际和平妇幼保健院、静安区妇幼保健所就诊孕妇873名,采用线上问卷调查形式,收集人口学资料、PHQ-9和GAD-7量表评分,并通过Logistic回归模型,构建基于PHQ-9与GAD-7评分的联合预测因子。结果:①围产期抑郁障碍组相较非围产期抑郁障碍组,在年龄、性格、孕期并发症、两系三代抑郁史因素中差异有统计学意义(P<0.05)。②PHQ-9与GAD-7联合预测因子预测妊娠晚期PND的AUC值高于PHQ-9和GAD-7量表单独预测(0.942 vs 0.929和0.912)。结论:使用Logistic回归模型构建的PHQ-9与GAD-7评分的联合预测因子可以提高妊娠晚期对PND预测的准确性。

Objective:To explore the predictive effect of the Patients Health Questionnaire Depression Scale-9(PHQ-9)and Generalized Anxiety Disorder 7-item Scale(GAD-7)on perinatal depressive disorder(PND),providing evidence for early identification.Method:Demographic data,PHQ-9,and GAD-7 of 873 pregnant women in the International Peace Maternity&Child Health Hospital of China welfare institute and Shanghai Jingan Mental Health Center from February 2020 to June 2020 were measured through online questionnaire.Logistic regression model was formulated to create the new combined predictor.Results:①Compared with non-PND group,there were significant differences in age,personality,pregnancy complications and three generations of depression history in the perinatal depressive disorder group(P<0.05).②The AUC value of PND in the third trimester predicted by PHQ-9 and GAD-7 combination was higher than that predicted by PHQ-9 or GAD-7 alone(0.942 vs 0.929 and 0.912).Conclusion:The combined prediction of PHQ-9 and GAD-7 constructed by Logistic regression model can improve the accuracy of predicting PND in the third trimester of pregnancy.