| 引用本文： |
-
胡海萍,王志豪,张琳,等.基于产前社会心理因素的妊娠晚期围产期抑郁障碍的预测研究[J].同济大学学报（医学版）,2022,43(5):658-665. [点击复制]
- HU Haiping,WANG Zhihao,ZHANG Lin,et al.Establishment of a prediction model of perinatal depression disorder for late pregnant women based on psychosocial factors[J].同济大学学报（医学版）,2022,43(5):658-665. [点击复制]
|
|
| 摘要: |
| 目的建立预测妊娠晚期围产期抑郁障碍(perinatal depression, PND)的社会心理因素模型，为早期识别和及时采取相关干预措施提供参考。方法选取2020年2月—2020年6月期间在中国福利会国际和平妇幼保健院(以下简称国妇幼)、静安区妇幼保健所(以下简称区妇幼)就诊的孕妇进行问卷调查，采用线上自愿形式发放问卷，共计873名孕妇作为调查对象。采用一般情况调查表、病人健康问卷抑郁量表、广泛性焦虑障碍量表、社会支持评定量表、自编相关危险因素问卷对妊娠晚期社会心理因素调查，同时进行PND的评估诊断，对社会心理因素进行单因素分析和多因素逐步Logistic回归构建预测模型，运用ROC曲线评价该模型的预测价值。结果妊娠晚期PND的发生率为10.19%。妊娠晚期孕妇社会心理因素单因素分析显示，年龄、性格、有无经期情绪不良、有无抚养新生儿经验、孕妇是否性别歧视、公婆是否性别歧视、对收入的满意度、有无两系三代抑郁史、经前紧张史、家庭居住条件、孕期是否一次接触手机大于15min、SSRS总分差异均有统计学意义(P<0.05)。对上述指标进行多因素逐步Logistic回归分析，建立了如下预测模型: Logit(PND)=-0.537×性格-1.257×有无经期情绪不良+0.684×有无抚养新生儿经验-1.694×两系三代抑郁史-0.890×经前紧张史+0.950×家庭居住条件+2.625。该模型曲线下面积为0.793(P＜0.001；95%CI: 0.747~0.838)。PND=-2.368时，约登指数最大，灵敏度为0.854，特异度为0.602。结论有抚养新生儿的经验、家庭居住条件满意是妊娠晚期PND的保护因素，性格内向、有经期情绪不良、有两系三代抑郁史及有经前紧张史是PND的独立危险因素；本研究建立的基于社会心理因素的妊娠晚期PND预测模型可以有效预测妊娠晚期PND的发生。 |
| 关键词: 围产期抑郁障碍 妊娠晚期 社会心理因素 预测模型 |
| DOI：10.12289/j.issn.1008-0392.22011 |
| 投稿时间：2022-01-11 |
| 基金项目:上海市静安区科学技术委员会与静安区卫生健康委员会医学科研课题(2019GW03) |
|
| Establishment of a prediction model of perinatal depression disorder for late pregnant women based on psychosocial factors |
| HU Haiping,WANG Zhihao,ZHANG Lin,CHEN Yuming |
| (Department of Clinical Psychotherapy, Shanghai Jing’an Mental Health Center, Shanghai 200436, China;Department of Obstetrics, The International Peace Maternity and Child Health Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai 200030, China) |
| Abstract: |
| ObjectiveTo establish a prediction model of perinatal depression disorder(PND) for women of the third trimester pregnancy based on psychosocial factors. MethodsAn online questio-nnaire survey was conducted among 873 late pregnant women who were visiting International Maternal and Child Health Hospital and Jing’an District Maternal and Child Health Institute from January 2020 to June 2020. The patient health questionnaire-9(PHQ-9), Generalized Anxiety Disorder(GAD-7), Social Support Rating Scale(SSRS) and a self-made related risk factor questionnaire were used to investigate the social psychological factors in women with the third trimester of pregnancy. The PND was evaluated and diagnosed. The correlation of PND with psychosocial factors was analyzed by univariate and multivariate stepwise logistic regression, and a prediction model was established and evaluated by ROC curve. ResultsThe incidence of PND in the third trimester of pregnancy was 10.19%. Univariate analysis showed that age, personality, bad mood, experience of newborn-raising, gender discrimination by pregnant women or husbands’ parents, satisfaction with income, family history of depression, history of premenstrual tension, housing condition, continuing use mobile phone>15 min during pregnancy, and total score of SSRS were significantly correlated with PND(P<0.05). Based on the results of multivariate stepwise logistic regression analysis a prediction model was established: Logit(PND)=-0.537×Personality-1.257×emotional disturbance during menstruation+0.684×experience of raising newborns-1.694×depression history-0.890×premenstrual tension history+0.950×housing condition+2.625. The area under the curve(AUC) of this model was 0.793(P<0.001；95%CI: 0.747-0.838). When PND=-2.368 was set as cut-off value, the Yoden index was the highest, the sensitivity was 0.854, and the specificity was 0.602. ConclusionThe prediction model established in this study based on psychosocial factors can effectively predict the risk of PND in women with the third trimester of pregnancy. |
| Key words: perinatal depressive disorder third trimester of pregnancy social psychological factors forecasting model |
