Introduction

Postpartum depression (PPD) is a common, non-psychotic mood or mental disorder which typically manifests in mothers within one year of delivery (first year postpartum) [1, 2]. Globally, the prevalence of PPD among mothers ranges from 0.5% to 60.8% [2]. In comparison to women of developed countries, women of developing countries showed higher rates of PPD [3]. A systematic review of 28 developed countries reported that the prevalence PPD symptom (PPDS) was 6–13% among women in high income nations [4]. An independent systematic review on low and middle income countries (LMIC) found the prevalence of postpartum common mental disorder was approximately 20% [5]. Asian countries reported between 3.5–63.3% prevalence rates of depression in postpartum women [6]. In India the prevalence of depression varied from 11% to 16% within fourteen weeks of delivery [7]. Several studies conducted in rural Bangladesh found the prevalence of PPD ranged from 18% to 35% among rural women [8–10].

PPD has negative impacts on children’s physical health. Existing literature of LMICs suggests that PPD leads to poor infant feeding practices, consequently leading to malnutrition and reduced physical growth [11, 12]. The effects of PPD are not limited to physical symptoms but can also impact themental health of the affected mothers’child. PPD leads to the lower levels of interaction and bonding between the mother and child, which leads to inadequate social, emotional and cognitive development of the child [13, 14]. Research also suggested that repeated level of depression leads to high level of chronic stress at the later age [13]. Further, PPD has a harmful impact on the family and social life [15].

There are various factors causing PPD and the subsequent consequences of PPD varies according to women ages, education levels, races, and ethnicities [16]. Globally, preterm or low birth-weight infants, unemployment, socio-economic deprivation, poor social or emotional support, housing problems, first-born child, childcare stress, infant temperament, high parity, obstetric complications, sleep disturbances, low self-esteem, negative attitude towards pregnancy, antenatal depression or anxiety, previous history of depression, poor marital relationship, history of domestic abuse, major adverse life events in preceding one year, neuroticism, perfectionism and level of daily hassles are strong predictors of PPD among women [1, 17–22].

The risk factors identified in the rural areas of Bangladesh were low economic status, nutritional status, physical violence, domestic quarrels with husband and in laws, stress, past mental and depressive symptoms during pregnancy, perinatal death, poor relationship between husband and in laws, morbidity during pregnancy, and current health condition [8, 9, 23, 24]. Comparing rural area, PPD was less explored in urban slum area. We found only one qualitative study explored the cultural attitude towards PPD in urban slum. Of the 36 women interviewed, 17 women were suffering from PPD [25]. In this study women reported that financial crises, adverse life events, intimate partner violence, health problems, and lack of practical social support were causes of daily emotional distress and sadness at postpartum period [25].

The population of urban Dhaka was estimated to be approximately 17 million in 2015 [26] and nearly one third of the population Dhaka city inhabit slums [27]. Within the next 14 years the urban population will increase by 50% due to rapid urbanization [28]. The status of PPD in women in urban slums has received less attention and the prevalence and associated risk factors for PPD in slum areas were unknown. The objective of this proposed study was to assess the burden and risk factors of PPD among women living in urban slums of Dhaka, Bangladesh.

Materials and methods

Results

Among the postpartum mothers 25.2% were adolescents and 31.7% were illiterate or with just enough skills to sign their names. On the other hand, 17.0% of mothers attained a secondary level of education or received a higher education. Approximately half, (48.7%) were first time mothers and 20.0% had three or more children. The mean number of household members was 4.5 and most of the families (83.8%) resided in one room. Only 7.7% of mothers were working after delivering their child, but before their most recent pregnancy around 49.7% of mothers had been engaged in income generating activities. Among the respondents employed before pregnancy, 54.6% were garments or industry workers, 27.3% were part-time house maid; whereas, 10.7% worked in private and other non-government organization and 7.0% involved some sort of home based work (Tailoring, handicraft, etc). Among the respondents we interviewed, 22.6% experienced at least one child death, or miscarriage, or stillbirth during her entire fertility period and 67.6% of respondents planned their most recent pregnancy. About 69.1% of mothers faced intimate partner violence before their last pregnancy, and 47.7% faced it during the pregnancy period as well (Table 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Respondent’s characteristics.

https://doi.org/10.1371/journal.pone.0215735.t001

Potential risk factors of postpartum depression

The potential risk factors of postpartum depression were organized in following categories: socio-economic factors; pregnancy related factors and family support and intimate partner violence (Table 2).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Bivariate and multivariable association of postpartum depression with socio-economic factors, pregnancy related factors, and family support and intimate partner violence.

https://doi.org/10.1371/journal.pone.0215735.t002

Discussion

This study explored the burden of postpartum depression and the associated factors in slum areas. Our study results show that about 40 women out of 100 women were suffering from PPD and the associated risk factors were current job involvement, job loss due to pregnancy, history of miscarriage, still birth and child death, unintended pregnancy, cost of delivery managed from borrowing/selling asset/mortgage, depressive symptom during pregnancy period, perceived antenatal stress, poor marital relationship with husband, and intimate partner violence.

The burden of PPD in women who inhabited slums was higher in comparison that of postpartum mothers living in rural areas [8–10]. Existing mental disorder data of urban slum women also suggested the higher level of mental disease burden in slum area. In one recent study at an urban slum, Khan et al. found that 46% mothers with children under the age of five suffered from common mental disorders [36].

Half of our respondents had been involved in income generating activities before the pregnancy, while only eight percent of survey respondents were involved in income generating activities during the postpartum period. Depression was more common for the women who worked but quit job due to pregnancy. There was the scarcity of data about the postpartum mental health and unemployment in low income countries. In a study of Canadian postpartum mothers, depression was less common in women who were in maternity leave [37]. Similarly, an Australian study found the evidence that paid maternal leave was beneficial for post-partum health and wellbeing [38]. According to the Labor Force Survey (2016–17), 89.1% female employment in urban Dhaka was informal employment[39]. In Bangladesh Labour Law, there is a provision for paid maternity leave[40]. A strict implementation of clause 45 to 50 of the Bangladesh Labour Law might improve the postpartum mental health of slum working women. We found that more than half of our respondents worked in garments or industry sector, however, the status of paid maternity leave was not satisfactory in garments industry [41, 42]. Among the informal employment garments and other industries were easy to track. So, at the first stage, at least in the garments and industry sector, the strict enforcement of paid maternity leave needed to initiate to reduce the burden of PPD in slum.

Postpartum depression was also high among them others who were working after child delivery. This was because with the professional working stress, taking care of new child added additional stress to women. These multiple roles leaded to role overload that could have negative effects on psychological well-being of postpartum mother [37, 43]. Some government oversight of maternity schemes for postpartum women, especially for the women working in slum areas, might improve the postpartum mental wellbeing of slum women.

PPD was not associated with household economic status within slum areas. This result contradicts the PPD findings within the rural context of Bangladesh [9, 24]. The possible explanation could be the people of slum area were concerned with immediate economic survival [44] rather than long term economic stability (measured by the wealth index).

However, debt and sudden financial hardship of family may affect the postpartum mental health of women [45, 46]. Depression was higher for women, whose child delivery cost was managed by borrowing, selling assets or through mortgages. The findings of recent national maternal mortality survey revealed that private sector delivery has increased several fold and has the highest delivery cost [47]. To minimize the costs of delivery, necessary steps are needed to promote deliveries in public and NGOs facilities.

In rural area of Bangladesh, association of postpartum depression with unintended pregnancy was found in bivariate analysis, but the association disappeared in multivariate regression analysis [8, 9, 23]. However, in the urban slum context, the unintended pregnancy was an independent risk factor for PPD. The association of PPD with unintended pregnancy persisted even after controlling other risk factors in the multivariate model. An unmet need of contraceptives was found in both urban and rural areas [48]. A patient centred approach and providing comprehensive information and access to contraceptive options may helpful for controlling the unintended pregnancy [49].

Among the obstetric factors assessed in our study, a history of miscarriage or still birth or child death was associated with postpartum depression; a finding that was consistent with those of other studies [8, 50, 51]. The fear of a repeat occurrence of a miscarriage, stillbirth, or child death are thought to be a contributory factors of depression in the antenatal, postnatal period, or in both periods. The recommended number of antenatal care (ANC) and postnatal care (PNC) visits may reduce miscarriage, still birth and postnatal child death; and consequently reduce PPD rates in subsequent pregnancies. Healthcare providers should consider the history of earlier loss as a factor of increased vulnerability of depression during the next pregnancy and early postpartum period [52].

Depressive symptom during pregnancy was mostly associated with postpartum depression, similar to the findings of other studies [4, 8, 53]. Perceived family, working, and any other mental stress during pregnancy was also associated with PPD like other findings [54, 55]. Perceived antenatal stress may contribute to development of depressive symptoms in the antenatal period [56]. These findings highlighted PPD is not only a matter of postpartum period and noted the importance of intervention in the prenatal period to prevent PPD [53, 54].

Intimate partner violence (IPV) is the one of the most important risk factors found in our study, which was consistent with most of the literatures of postpartum depression [9, 23, 57–62]. In a recent study of Bangladesh, Ziaie found that all forms of domestic violence were strongly associated with higher levels of emotional distress during the pregnancy period as well [63]. In our study, we also found that poor marital relationship (sharing personal feelings and practical support in household work) with husbands was a strong predictor of PPD, consistent with the findings of other studies [8, 22]. IPV and poor marital relationship might be associated with each other [64]. Therefore, these findings suggested that couples focused intervention from the prenatal period may reduce the risk of PPD.

Strengths and limitations

The main strength of the study was the study focused the rapid increasing segment of urban population of Bangladesh with a large sample size. To our knowledge it was the first study to assess the prevalence and risk factors of postpartum depression in urban slum of Bangladesh. In addition, the study included the wide range of risk factors found in the postpartum literatures of Bangladesh and similar context. We used GLM and GEE model to estimate Prevalence Ratio (PR) and to adjust the clustering effect at slum level. Statistically, estimating PR has some methodological advantages as it is a comparable measure of relative risk (RR) for cross-sectional data, and the estimation models of PR able to control the confounders and interactions more adequately than the Logistic regression models [65, 66]. Most importantly, the study used the repeatedly validated Bangla version of instruments and scale which was nationally and internationally recognized and widely used.

The study had some limitations worth to acknowledge. The design of the study was cross-sectional and therefore could not measure the incidence of PPD. Data about most of the risk factors were collected from mother’s recall, which may lead to under or over reporting of symptoms. However, data collectors were intensively trained on the interview techniques to reduce the recall bias. In Dhaka city, there were about 4,966 slums of different sizes [67]. In this study, we purposively selected 3 large slums, as slum level randomization was not possible due to time and resource constraint. This may lead some selection bias in terms of representativeness of slums of Dhaka city. However, in terms of generalisability of study results, this selection bias was less, as overall contextual factors of slums are representatively prevailing in the large slums of Dhaka city. In addition, the frequent movement of slum dwellers across the slums of [28, 68] keeps the urban slums typically heterogeneous.We used a likert scale to measure the support of mother in law and husband to the mother, which were not validated before our use. However, we believe that this likert scale was a good instrument to measure the ‘Family Support’. Some working mothers may not be reached due to strict data collection timeline and not having schedule of household revisit. We tried to overcome this constraint by extending the data collection time beyond the office hour and collecting data in weekend as well.

Conclusions

The higher prevalence of PPD suggested the importance of mental health support system for the low income women in slum area. Maternal mental health services should be integrated with existing maternal health services. The primary maternal health care staffs could be provided the basic PPD screening and its primary management training, so that they can refer the PPD cases for appropriate mental health services when needed. They are also needed to educate about the contextually relevant risk factors of PPD as part of the training component. Additionally, the existing maternal health services in slum area should be strengthened and pro-poor friendly.

Moreover, research is required to develop the low cost non-pharmacological management of PPD cases in the informal settlement of urban poor. In addition to that, research for developing the culturally appropriate preventive interventions to control the risk factors should be undertaken in the urban slum of Bangladesh.

Supporting information

Acknowledgments

The study was supported by the James P. Grant School of Public Health (JPGSPH). This paper is based on the data collected for Summative Learning Project (SLP) for author’s partial requirement of Master of Public Health (MPH) degree of JPGSPH at BRAC University. The author is grateful to Mahfuz Al Mamun, gender expert and Farzana Ramee, research psychologist of icddr,b for their training support. He is also thankful to Aniqa Hassan for her final language editing of author’s manuscript. Finally he wants to thank icddr,b for providing the scholarship opportunity for pursuing the MPH degree in JPGSPH.