产后抑郁风险预测模型的构建与验证

Construction and validation of a risk prediction model for postpartum depression

摘要

目的 探讨构建和验证应用于女性分娩后早期初筛产后抑郁症状的风险预测模型。方法 选取2022年10月1日至2023年10月1日于川北医学院附属医院住院分娩的438例产妇作为研究对象.在产后1周内使用爱丁堡产后抑郁量表(EPDS)评估抑郁症状,选取9.5分作为阈值将研究对象分为产后抑郁组(n=114)与非产后抑郁组(n=324).多因素Logistic回归法筛选产后抑郁症状的危险因素,进一步构建可视化列线图模型,通过受试者工作特征(ROC)曲线、Hosmer-Lemeshow(H-L)拟合优度检验对模型的区分度、校准度进行评价.利用Bootstrap法对模型进行内部验证,最后使用决策曲线分析法(DCA)判断模型的临床适用性。结果 多因素Logistic回归分析显示年龄≥30岁、产前BMI越大、学历水平较高为产后抑郁的独立保护因素(OR值介于0.239~0.903之间,P<0.05);有抑郁家族史、孕晚期有抑郁症状、孕前一年期间存在持续两周情绪低落是产后抑郁的独立危险因素(OR值介于4.981~13.215之间,P<0.05).该模型ROC曲线下面积(AUC)为0.831(95%CI:0.785~0.877),灵敏度75.4%,特异度77.5%;H-L拟合优度检验显示χ²=2.773,P>0.05;该模型的DCA曲线在有效范围内,临床最大净收益阈值概率为10%~100%。结论 本研究构建的产后抑郁预测模型的预测价值及临床适用性良好,为早期且高效筛查产后抑郁症状提供了科学依据.

Abstract

Objective To explore the construction and validation of a risk prediction model applied to the early primary screening of women for postpartum depression in the early postnatal period. Methods A total of 438 women who delivered as inpatients in The Affiliated Hospital of North Sichuan Medical College from 1 October 2022 to 1 October 2023 were selected as the study subjects. Depressive symptoms were assessed using the Edinburgh postnatal depression scale (EPDS) within 1 week postpartum, with a threshold score of 9.5 used to divide the study subjects into the postpartum depression group (n=114) and the non-postpartum depression group (n=324). Multivariate Logistic regression was used to identify the risk factors for postpartum depressive symptoms, and a visual nomogram model was further constructed. The model's discrimination and calibration were evaluated using the receiver operating characteristic (ROC) curves and the Hosmer-Lemeshow (H-L) goodness-of-fit test. The model was internally validated using the Bootstrap method, and its clinical applicability was assessed using decision curve analysis (DCA). Results Multivariate logistic regression analysis showed that age ≥30 years, higher prenatal BMI, and higher education level were independent protective factors against postpartum depression (OR ranged from 0.239 to 0.903, P<0.05); a family history of depression, depressive symptoms in late pregnancy, and experiencing persistent low mood for at least two weeks in the year prior to pregnancy were independent risk factors for postpartum depression (OR ranged from 4.981 to 13.215, P<0.05). The area under the ROC curve (AUC) of the model was 0.831 (95%CI: 0.785~0.877), with a sensitivity of 75.4% and specificity of 77.5%; the H-L goodness-of-fit test showed χ²=2.773, P>0.05; and the decision curve analysis (DCA) demonstrated clinical applicability within the threshold probability range of 10% to 100%, with the maximum net benefit. Conclusion The prediction model of postpartum depression constructed in this study has good predictive value and clinical applicability, which provides a scientific basis for early and efficient screening of postpartum depressive symptoms.

关键词

孕产妇 / 产后抑郁 / 预测模型 / 列线图

Key words

pregnant and postpartum women / postpartum depression / prediction model / nomogram

参考文献

[1] Meaney M J. Perinatal maternal depressive symptoms as an issue for population health[J]. Am J Psychiatry, 2018, 175(11): 1084-1093.

[2] 冀荔, 李月琴, 王烈宏, 等. 产后母亲抑郁症状变化对儿童行为功能的影响[J]. 中国妇幼健康研究, 2023, 34(8): 19-25.

[3] Hahn-Holbrook J, Cornwell-Hinrichs T, Anaya I. Economic and health predictors of national postpartum depression prevalence: a systematic review, meta-analysis, and meta-regression of 291 studies from 56 countries[J]. Front Psychiatry, 2018, 8: 248.

[4] Shi X, Ying Y, Yu Z, et al. Risk factors for postpartum depression in Chinese women: a cross-sectional study at 6 weeks postpartum[J]. J Psychosom Res, 2021, 140: 110295.

[5] Keliyo E T, Jibril M K, Wodajo G T. Prevalence of antenatal depression and associated factors among pregnant women attending antenatal care at health institutions of Faafan Zone, Somali Region, Eastern Ethiopia[J]. Depress Res Treat, 2021, 2021(1): 2523789.

[6] 王文静, 庄惠人, 黄贤莉, 等. 孕产妇围产期抑郁预防和管理的最佳证据总结[J]. 上海护理, 2023, 23(8): 13-19.

[7] 邵珠燕, 宗可敬, 范清美, 等. 产后抑郁症风险预测模型的系统评价[J]. 中国循证医学杂志, 2023, 23(7): 807-813.

[8] McCauley M, White S, Bar-Zeev S, et al. Physical morbidity and psychological and social comorbidities at five stages during pregnancy and after childbirth: a multi-country cross-sectional survey[J]. BMJ Open, 2022, 12(4): e050287.

[9] 符夏瑜, 陈雄, 张云志. 爱丁堡产后抑郁量表在孕晚期妇女产前抑郁筛查中的临界值分析[J]. 中国妇产科临床杂志, 2018, 19(5): 453-454.

[10] Biaggi A, Conroy S, Pawlby S, et al. Identifying the women at risk of antenatal anxiety and depression: a systematic review[J]. J Affect Disord, 2016, 191: 62-77.

[11] 何茜, 孙静, 高梦云. 初产妇产后抑郁现状调查及其影响因素分析[J]. 中国计划生育学杂志, 2022, 30(12): 2683-2687.

[12] 汪立. 初产妇产后抑郁风险列线图模型的建立[J]. 护理学杂志, 2020, 35(13): 30-33.

[13] 于莹, 刘凯波, 沈汝冈, 等. 北京市产后抑郁状态的影响因素分析[J]. 中国妇幼健康研究, 2022, 33(11): 76-82.

[14] 杜兴梅, 邓淦元, 杨中婷, 等. 不同年龄组女性围产期抑郁发生风险及影响因素分析[J]. 中国预防医学杂志, 2024, 25(3): 300-306.

[15] 王佳, 马力凤, 吴奎, 等. 产妇产后3天内抑郁发生现状调查及影响因素分析[J]. 中国计划生育学杂志, 2023, 31(3): 500-504.

[16] 孙梦云, 黄星, 杨业环, 等. 各孕期及产后焦虑状况自然转归情况及影响因素分析[J]. 中国妇幼健康研究, 2021, 32(8): 1112-1117.

[17] Silverman M E, Smith L, Lichtenstein P, et al. The association between body mass index and postpartum depression: a population-based study[J]. J Affect Disord, 2018, 240: 193-198.

[18] 薛香莉, 刘微娜, 漆正堂, 等. 脂肪细胞因子与运动的抗抑郁作用[J]. 体育科学, 2016, 36(11): 66-74.

[19] Ge T, Fan J, Yang W, et al. Leptin in depression: a potential therapeutic target[J]. Cell Death Dis, 2018, 9(11): 1096.

[20] 陈璐, 拓明花, 胡金萍, 等. 瘦素与产后抑郁症状的相关性研究[J]. 中国神经精神疾病杂志, 2019, 45(6): 340-345.

[21] Yu J, Zhang Z, Deng Y, et al. Risk factors for the development of postpartum depression in individuals who screened positive for antenatal depression[J]. BMC Psychiatry, 2023, 23(1): 557.

[22] Al Rawahi A, Al Kiyumi M H, Al Kimyani R, et al. The effect of antepartum depression on the outcomes of pregnancy and development of postpartum depression: a prospective cohort study of Omani women[J]. Sultan Qaboos Univ Med J, 2020, 20(2): e179-e186.

[23] Okunola T O, Awoleke J O, Olofinbiyi B, et al. Predictors of postpartum depression among an obstetric population in South-Western Nigeria[J]. J Reprod Infant Psychol, 2022, 40(4): 420-432.

[24] Jagini K K. Temporal binding in multisensory and motor-sensory contexts: toward a unified model[J]. Front Hum Neurosci, 2021, 15: 629437.

[25] Goffin K, Viera G. Emotions in time: the temporal unity of emotion phenomenology[J]. MIND LANG, 2024, 39(3): 348-363.

[26] Hoerl C, Lorimer S, McCormack T, et al. Temporal binding, causation, and agency: developing a new theoretical framework[J]. Cogn Sci, 2020, 44(5): e12843.

[27] Qi D, Chen K. Bioinformatics analysis of potential biomarkers and pathway identification for major depressive disorder[J]. Comput Math Methods Med, 2021, 2021: 3036741.

[28] 程夏龙, 张许来, 关建军, 等. 家族史阳性首发未用药抑郁症患者的神经认知功能危险标记物[J]. 神经损伤与功能重建, 2023, 18(7): 404-408.

基金资助

国家工业和信息化部、国家卫生健康委员会«5G+医疗健康应用试点项目»批准立项项目(JKZX2022G5G03)