Prediction of peripartum depression by oxytocin levels, Edinburgh Postnatal Depression Scale scores and lack of social support: a prospective study.

Purpose: This study aimed to determine predictive factors for peripartum depression (PPD).

Methods: We collected pre-natal sociodemographic data, blood markers and psychosocial questionnaires from 237 mothers. We detected the oxytocin (OXT) concentrations in saliva and administered the Edinburgh Postnatal Depression Scale (EPDS) within 1-2 days postpartum. The multivariable logistic regression analysis model was applied to analyze the factors related to PPD at 4 weeks postpartum. A nomogram model was constructed and evaluated.

Results: Multivariable logistic regression showed that OXT levels (OR = 0.984, 95% CI 0.971-0.997), EPDS scores (OR = 1.284, 95% CI 1.154-1.428), and lack of social support (OR = 3.562, 95% CI 1.444-8.784) were predictive factors for PPD (all p < 0.05). A nomogram model was constructed based on these factors. The model's C-index was 0.846 (95% CI 0.777-0.908). The calibration curve indicated good agreement between the predicted probabilities of PPD and the actual probabilities (Hosmer-Lemeshow test p = 0.928). The area under the Receiver Operating Characteristic (ROC) curve was 0.820 (95% CI 0.749-0.891). The results of the decision curve analysis showed that the model had good clinical utility.

Conclusions: The predictive model can be used to identify high-risk parturients of PPD early.