Machine Learning-Based Predictive Modeling of Postpartum Depression.

Dayeon Shin,Kyung Ju Lee,Temidayo Adeluwa,Junguk Hur

doi:10.3390/jcm9092899

Abstract

Postpartum depression is a serious health issue beyond the mental health problems that affect mothers after childbirth. There are no predictive tools available to screen postpartum depression that also allow early interventions. We aimed to develop predictive models for postpartum depression using machine learning (ML) approaches. We performed a retrospective cohort study using data from the Pregnancy Risk Assessment Monitoring System 2012–2013 with 28,755 records (3339 postpartum depression and 25,416 normal cases). The imbalance between the two groups was addressed by a balanced resampling using both random down-sampling and the synthetic minority over-sampling technique. Nine different ML algorithms, including random forest (RF), stochastic gradient boosting, support vector machines (SVM), recursive partitioning and regression trees, naïve Bayes, k-nearest neighbor (kNN), logistic regression, and neural network, were employed with 10-fold cross-validation to evaluate the models. The overall classification accuracies of the nine models ranged from 0.650 (kNN) to 0.791 (RF). The RF method achieved the highest area under the receiver-operating-characteristic curve (AUC) value of 0.884, followed by SVM, which achieved the second-best performance with an AUC value of 0.864. Predictive modeling developed using ML-approaches may thus be used as a prediction (screening) tool for postpartum depression in future studies.

Highlights

Postpartum depression is a mood disorder that affects up to 15% and 13% of mothers after childbirth in the United States and worldwide, respectively [1,2]
Previous research has revealed that risk factors for postpartum depression include a history of mental illness, such as past history of postpartum depression, other depression or psychiatric illnesses, and a family history of affective disorder [7]; low social support [8]; poor marital relationship [9]; pregnancy-related complications, including emergency cesarean sections [10]; unplanned/unwanted pregnancy [11]; stressful life events during pregnancy [12]; and preterm birth [13]
The status of postpartum depression significantly differed by maternal age, maternal race/ethnicities, education, small-for-gestational-age based on the 10th percentile, pre-pregnancy exercise for more than three days, depression before pregnancy, drinking three months before pregnancy, changing smoking in the last three months of pregnancy and postpartum period, and marital status

Summary

Introduction

Postpartum depression is a mood disorder that affects up to 15% and 13% of mothers after childbirth in the United States and worldwide, respectively [1,2]. Previous research has revealed that risk factors for postpartum depression include a history of mental illness, such as past history of postpartum depression, other depression or psychiatric illnesses, and a family history of affective disorder [7]; low social support [8]; poor marital relationship [9]; pregnancy-related complications, including emergency cesarean sections [10]; unplanned/unwanted pregnancy [11]; stressful life events during pregnancy [12]; and preterm birth [13] These independent risk factors for postpartum depression are known, little is known about the predictive modeling of postpartum depression that includes maternal and paternal risk factors. One of the objectives of the Healthy People 2020 initiative is to decrease the proportion of women delivering live births who experience postpartum depressive symptoms, so it is imperative to develop a screening tool for postpartum depression for prevention and intervention purposes.

Objectives

Results

Discussion

Conclusion