Postpartum Depression in a Tertiary Maternity Service in Dhaka, Bangladesh: Prevalence, Risk Factors, Comorbid Anxiety, and Care Cascade — A Hospital-Based Observational Study (2024–2025)

Abstract

Background: Postpartum depression (PPD) affects approximately 17% of women globally but reaches substantially higher rates in Bangladesh, where estimates range from 22% to 60% depending on the setting and screening threshold. Despite recommendations for routine screening, the care cascade from positive screening to sustained treatment in Bangladeshi tertiary hospitals remains poorly characterised.

Methods: This hospital-based observational study was conducted at a tertiary maternity service affiliated with Bangladesh Medical University, Dhaka, during 2024–2025. A total of 1,200 postpartum women were screened using the validated Bangla Edinburgh Postnatal Depression Scale (EPDS; primary), the Patient Health Questionnaire-9 (PHQ-9), and the Generalised Anxiety Disorder scale (GAD-7) at 4–12 weeks postpartum. The study comprised (A) a cross-sectional prevalence and risk-factor component at first postpartum contact and (B) a prospective follow-up cohort for symptom trajectories and care cascade evaluation. Binary PPD was defined as EPDS ≥10 (primary) and ≥13 (sensitivity analysis). Multivariable logistic regression identified independent risk factors. The care cascade was quantified as proportions progressing through sequential steps from positive screen to treatment engagement. Reporting followed STROBE guidelines.

Results: Among 1,200 women (mean age 26.1 ± 4.6 years), the prevalence of PPD was 38.9% (95% CI: 36.2–41.7) at EPDS ≥10 and 16.8% (95% CI: 14.8–19.1) at EPDS ≥13. PHQ-9 ≥10 prevalence was 22.8% (95% CI: 20.5–25.3); postpartum anxiety (GAD-7 ≥10) was 9.2% (95% CI: 7.7–10.9). In multivariable analysis, the strongest independent predictors of PPD were prior depression/anxiety (aOR 11.61, 95% CI: 6.92–19.47), comorbid anxiety (aOR 9.50, 95% CI: 4.89–18.47), intimate partner violence (aOR 3.64, 95% CI: 2.69–4.93), emergency caesarean section (aOR 3.33, 95% CI: 2.26–4.90), unplanned pregnancy (aOR 3.19, 95% CI: 2.37–4.29), sleep problems (aOR 2.48, 95% CI: 1.85–3.33), and NICU admission (aOR 2.02, 95% CI: 1.36–3.02). Among screen-positive women, 69.6% received counselling, 50.3% were offered referral, but only 20.1% attended at least one mental health appointment. Cohen's kappa for EPDS ≥10 versus PHQ-9 ≥10 agreement was 0.498 (moderate).

Conclusions: PPD is highly prevalent in this urban tertiary-hospital population with substantial care-cascade attrition. Integration of routine EPDS screening into postnatal and immunisation clinic contacts, combined with on-site counselling and structured referral pathways, is urgently needed.