Risk factors and prediction model for postpartum psychiatric disorders: a retrospective cohort study of 1418 Chinese women from 2020 to 2022.

BACKGROUND

Postpartum psychiatric disorders (PPDs) have been deemed as a significant public health concern, affecting both maternal health and family dynamics. This study aimed to examine the current status of PPDs, identify the potential risk factors of PPDs, and further develop a clinical nomogram model for predicting PPDs in Chinese women.

METHOD

In this retrospective cohort study, 1418 postpartum women attending the routine postpartum examination at the 42nd day after delivery in Jiangsu Women and Children Health Hospital were recruited as participants from December 2020 to December 2022. The Symptom Checklist-90 (SCL-90) was utilized to assess the status of postpartum psychiatric disorders. A prediction model was constructed by multivariate logistic regression and presented as a nomogram. The performance of nomogram was measured by the receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA). The relationships between predictive factors of PPDs and SCL-90 were also evaluated using Pearson correlation analysis. The relationships between predictive factors of PPDs and SCL-90 were evaluated using Pearson correlation analysis.

RESULTS

With the SCL-90 cutoff value of 160, the incidence of postpartum psychiatric disorders was 9.17% among Chinese urban women. The univariate and multivariate logistic regression analyses indicated that age ≤ 25 years old (OR = 10.07, 95%CI = 1.83-55.33), prenatal mood disorder (OR = 4.12, 95%CI = 1.99-8.53), invasive prenatal diagnostic procedures (OR = 4.39, 95%CI = 1.16-16.56), poor relationship with husband (OR = 2.86, 95%CI = 1.58-5.16) and poor relationship with mother-in-law (OR = 5.10, 95%CI = 2.70-9.64) were significantly associated with PPDs. A nomogram prediction model for PPDs was further constructed based on these five independent risk factors, and the area under the receiver operating characteristic curve (AUC) of the nomogram model was 0.823 (95% CI = 0.781-0.865). The calibration curves showed remarkable accuracy of the nomogram and the DCA exhibited high clinical net benefit of the nomogram. Besides, we also explored the relationships between the five risk factors and different symptom dimensions of PPDs and found that the five risk factors were almost associated with increased levels of all symptom dimensions.

CONCLUSIONS

Five psychosocial risk factors for PPDs were identified in Chinese women and the nomogram prediction model constructed based on these five risk factors could predict the risk of PPDs intuitively and individually. Systematic screening these risk factors and further conducting psychosocial interventions earlier during the pregnancy period are crucial to prevent PPDs. For future research, we intend to incorporate additional risk factors, including blood biomarkers and facial expression indicators, to refine our risk model.