A comprehensive dataset for postpartum depression prediction among bangladeshi mothers: Sociodemographic, obstetric, and psychosocial factors with class imbalance insights
6 affiliations
- 1Department of Computer Science, American International University Bangladesh, 1229, Dhaka, Bangladesh
- 2Department of Electrical & Computer Engineering, North South University, 1229, Dhaka, Bangladesh
- 3Department of Computer Science & Engineering, East West University, 1212, Dhaka, Bangladesh
- 4Department of Computer Science & Engineering, Southeast University, 1208, Dhaka, Bangladesh
- 5Department of Computer Science & Engineering, University of Chittagong, 4331, Chittagong, Bangladesh
- 6Department of Computer Science and Software Engineering, Malardalen University, Vastmanland, Sweden
Abstract
Postpartum depression (PPD) is currently a serious mental disorder that occurs after childbirth. It affects the well-being of both the mother and the child. However, early screening is still limited in low- and middle-income countries like Bangladesh due to socio-economic challenges and social stigma. This paper presents a comprehensive and multidimensional dataset consisting of 766 participants to facilitate data-driven PPD research. The participating women were between the ages of 18–41 years. The data were collected through a cross-sectional study using face-to-face interviews and hospital records at LABAID Diagnostics, Pabna, Bangladesh, between March and June 2025. The dataset integrates 51 variables, including relevant factors such as sociodemographic factors, obstetric history, neonatal details, and psychosocial indicators. Maternal health was assessed using standardized screening tools: PHQ-2 and PHQ-9 for symptoms before and during pregnancy, and Edinburgh Postpartum Depression Scale (EPDS) for postpartum symptoms. Results indicated that the prevalence of depression during pregnancy among the participants was 40.47%. A key feature of this dataset is the clear reporting of class imbalance, which can be reused for statistical analysis, epidemiological studies, and validation of depression screening tools, making it a valuable resource for building robust machine learning (ML) and interpretable AI (XAI) models for early risk prediction. This study fills an important gap in publicly available, high-quality maternal health information in the South Asian context. It also provides the basis for evidence-based interventions and targeted support for at-risk mothers.
Related
- Sociodemographic, psychosocial, and contextual factors associated with HIV testing uptake among adults across five districts in Ghana: a cross-sectional study
- Maternal bonding difficulties and postpartum depression in pediatric clinical practice: a longitudinal study in a Japanese cohort
- Vitamin K2 ameliorates pregnancy‐ and lactation‐associated osteopenia and postpartum depression
- Analysis of the subject-object effects and chain mediation of psychosocial factors in couples during late pregnancy on depression: a cross-sectional study of pregnant women and their spouses
- Contemporary Approaches to Postpartum Hemorrhage with Focus on Uterine Compression Sutures—An Updated Comprehensive Systematic Review
BibTeX
@article{ayon2026comprehensive,
title = {{A comprehensive dataset for postpartum depression prediction among bangladeshi mothers: Sociodemographic, obstetric, and psychosocial factors with class imbalance insights}},
author = {Shahriar Siddique Ayon and Shahidul Morsalin Jahin and Md. Ebrahim Hossain and Md. Saef Ullah Miah and Kazi Mostaq Hridoy and B.M. Mredul Arafin and Anik Dey Shaon and Md Rakibul Islam and M.F. Mridha},
journal = {Data in Brief},
year = {2026},
month = aug,
doi = {10.1016/j.dib.2026.113210},
url = {https://doi.org/10.1016/j.dib.2026.113210},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - A comprehensive dataset for postpartum depression prediction among bangladeshi mothers: Sociodemographic, obstetric, and psychosocial factors with class imbalance insights
AU - Shahriar Siddique Ayon
AU - Shahidul Morsalin Jahin
AU - Md. Ebrahim Hossain
AU - Md. Saef Ullah Miah
AU - Kazi Mostaq Hridoy
AU - B.M. Mredul Arafin
AU - Anik Dey Shaon
AU - Md Rakibul Islam
AU - M.F. Mridha
PY - 2026
DA - 2026/08/30
JO - Data in Brief
DO - 10.1016/j.dib.2026.113210
UR - https://doi.org/10.1016/j.dib.2026.113210
AB - Postpartum depression (PPD) is currently a serious mental disorder that occurs after childbirth. It affects the well-being of both the mother and the child. However, early screening is still limited in low- and middle-income countries like Bangladesh due to socio-economic challenges and social stigma. This paper presents a comprehensive and multidimensional dataset consisting of 766 participants to facilitate data-driven PPD research. The participating women were between the ages of 18–41 years. The data were collected through a cross-sectional study using face-to-face interviews and hospital records at LABAID Diagnostics, Pabna, Bangladesh, between March and June 2025. The dataset integrates 51 variables, including relevant factors such as sociodemographic factors, obstetric history, neonatal details, and psychosocial indicators. Maternal health was assessed using standardized screening tools: PHQ-2 and PHQ-9 for symptoms before and during pregnancy, and Edinburgh Postpartum Depression Scale (EPDS) for postpartum symptoms. Results indicated that the prevalence of depression during pregnancy among the participants was 40.47%. A key feature of this dataset is the clear reporting of class imbalance, which can be reused for statistical analysis, epidemiological studies, and validation of depression screening tools, making it a valuable resource for building robust machine learning (ML) and interpretable AI (XAI) models for early risk prediction. This study fills an important gap in publicly available, high-quality maternal health information in the South Asian context. It also provides the basis for evidence-based interventions and targeted support for at-risk mothers.
SN - 2352-3409
ER -