Background: Recent evidence suggests that the prevalence of postnatal depression (PND) is highest in low-income developing countries. This study aimed to estimate the prevalence of PND and its associated risk factors among Bangladeshi women.Method: The study was conducted in the Matlab subdistrict of rural Bangladesh. A cohort of 346 women was followed up from late pregnancy to post-partum. Sociodemographic and other related information on risk factors was collected on structured questionnaires by trained interviewers at 34–35 weeks of pregnancy at the woman's home. A validated local language (Bangla) version of the Edinburgh Postnatal Depression Scale (EPDS-B) was used to measure depression status at 34–35 weeks of pregnancy and at 6–8 weeks after delivery.Results: The prevalence of PND was 22% [95% confidence interval (CI) 17.7–26.7%] at 6–8 weeks post-partum. After adjustment in a multivariate logistic model, PND could be predicted by history of past mental illness [odds ratio (OR) 5.6, 95% CI 1.1–27.3], depression in current pregnancy (OR 6.0, 95% CI 3.0–12.0), perinatal death (OR 14.1, 95% CI 2.5–78.0), poor relationship with mother-in-law (OR 3.6, 95% CI 1.1–11.8) and either the husband or the wife leaving home after a domestic quarrel (OR 4.0, 95% CI 1.6–10.2).Conclusions: The high prevalence of PND in the study was similar to other countries in the South Asian region. The study findings highlight the need for programme managers and policy makers to allocate resources and develop strategies to address PND in Bangladesh.

In developing countries, postnatal depression (PND) is estimated to affect a high proportion of women following childbirth. There are no reliable estimates for the magnitude of the problem in Bangladesh, a country of 140 million people. The lack of a validated Bangla version of screening scale such as the Edinburgh Postnatal Depression Scale (EPDS) has hindered attempts to quantify and address the problem in the country. This study was carried out among Bangladeshi women to validate the Bangla version of the EPDS with the help of a multi‐disciplinary research consultative group. A research assistant administered the Bangla EPDS to a convenience sample of 100 mothers at 6–8 weeks postpartum attending an urban childhood immunization clinic. A psychiatrist examined all women using the Structured Clinical Interview (SCID) for DSM‐IV diagnosis of depression. Nine (9%) women were found to have depression. Internal consistency was tested using Cronbach's alpha coefficient (0.84). The optimal Receiver Operating Characteristic (ROC) cut‐off score of Bangla EPDS was 10 with a sensitivity of 89%, specificity of 87 %, positive predictive value of 40%, and negative predictive value of 99%. The Bangla EPDS is thus a valid and reliable screening scale for identifying PND in Bangladesh.