Workplace Stress, Overcommitment, and Maternal Health as Predictors of Postpartum Depression: A Case‐Control Study Among Employed Married Women in Bangladesh

ABSTRACT

Background

Postpartum depression (PPD) is a major mental health concern for employed mothers in low‐resource settings, where workplace stress, gendered expectations, and limited support are prevalent. This study aimed to examine the associations between workplace stress, overcommitment, family support, and selected maternal health factors with postpartum depression among employed married women in Dhaka, Bangladesh.

Methods

A hospital‐based matched case‐control study was conducted between March 2025 and May 2025 across five Surjer Hashi Clinics in Dhaka City. A total of 374 participants were recruited, consisting of 187 cases (EPDS score ≥ 12) and 187 matched controls (EPDS < 12). Cases and controls were individually matched on age and profession. Workplace stress was assessed using the Workplace Stress Scale (WSS), overcommitment was measured via the Effort–Reward Imbalance (ERI) model's OC subscale, and family support was evaluated using the Family APGAR scale. Conditional Logistic Regression (CLR) was employed to estimate matched adjusted odds ratios (mAOR) with 95% confidence intervals.

Results

Higher parity (≥ 3 children) (mAOR = 4.27, 95% CI: 2.11–8.38), cesarean delivery (mAOR = 2.29, 95% CI: 1.84–5.91), and anemia during pregnancy (mAOR = 4.11, 95% CI: 2.26–38.32) were significantly associated with PPD. Severe anxiety emerged as a strong predictor (mAOR = 14.61, 95% CI: 2.09–43.81). Workplace‐related factors demonstrated robust associations: women with severe workplace stress were over 18 times more likely to develop PPD (mAOR = 18.45, 95% CI: 5.33–57.96), while high overcommitment increased the odds nearly ninefold (mAOR = 8.16, 95% CI: 4.22–31.65). Family dysfunction was also a significant predictor, with women from severely dysfunctional families exhibiting 6.37 times higher odds of PPD compared to those from functional families.

Conclusion

Workplace stress, overcommitment, and dysfunctional family environments are strongly associated with PPD, highlighting the need for targeted workplace reforms, family support interventions, and maternal mental health policies in urban, low‐resource settings.

1. Introduction

Postpartum depression is a pressing maternal health issue, affecting an estimated 10%–20% mothers worldwide. It is a salient challenge for working mothers who need to navigate the complex balance between motherhood, career demands, and professional commitments. In South Asia, 17%–34% of employed married women experience postpartum depression, which is intensified and instilled by deep‐rooted patriarchal norms. The two major obstacles for employed married women are, firstly, a preference for a male child as traditionally sons are prioritized as male heirs and seen as economic providers and responsible for parents' care in old age; and secondly, a lack of support at the workplace, especially for new mothers, which often stems from rigid, inadequate, and gendered policies [1, 2, 3]. While many high‐income countries have implemented institutional supports like paid parental leave and flexible work arrangements to address and overcome this pressure, such systems remain unaddressed and limited in low‐resource settings such as Bangladesh. Beyond the limitations in policy implementation, gaps in labor protection laws, and gendered workplace settings, employed mothers encounter compounded pressure to remain in the job, which further negatively affects the ability to balance parenthood and work commitment [4].

Across South Asia, including Bangladesh, these universal workplace stresses are intersected by specific socio‐cultural factors, creating unique barriers for employed mothers. This intersection of professional and domestic factors dominantly contributes to the higher prevalence of PPD in the region, estimated to be between 17% and 34% among employed married women. In Bangladesh, previously, the majority of the research overlooked the intersectionality of workplace stress and postpartum depression, following the Western lens, even if the prevalence rate of PPD was increasing and conditions were rising rapidly. Studies have scarcely examined the complex relationship between workplace stress, physical maternal health (such as anemia and hypertension), and postpartum depression, despite the rapidly rising prevalence of PPD in its urban centers. The present study aimed to address this gap and examine the associations between workplace stress, overcommitment, perceived family support, and key maternal health factors with postpartum depression among employed mothers living in Dhaka, Bangladesh.

2. Methods

2.1. Study Design and Setting

This study utilized a hospital‐based matched case‐control design to analyze the relationship between workplace stress, perceived family support, and postpartum depression among formally employed women. The study was conducted across five Surjer Hashi Clinics in Dhaka City, strategically located to ensure broad geographic coverage. These clinics specialize in providing affordable maternal healthcare, including antenatal care (ANC), delivery services, and post‐natal care (PNC). Women from diverse backgrounds (wealth quartile, geographic areas, education, etc.) visit these clinics for high‐quality, cost‐effective maternal healthcare services. Data for this study were collected from working mothers attending the clinics for PNC services. In this matched case‐control study, cases and controls were individually matched based on age and profession to reduce potential confounding. Maternal age is known to influence postpartum mental health through its relationship with reproductive history, psychological vulnerability, and coping capacity. Profession was also considered an important matching factor because occupational roles may shape exposure to workplace stress, job demands, work schedules, and work‐related expectations, which are central to the study's research focus. Matching on these variables helped ensure comparability between cases and controls in terms of basic demographic and occupational characteristics, allowing a more accurate assessment of the associations between workplace stress, overcommitment, family support, maternal health factors, and postpartum depression.

2.2. Participant Recruitment

The required sample size was estimated using Epitools, an open‐access online tool, based on a 95% confidence level, 80% statistical power, and an anticipated odds ratio of at least 3.0 for the primary predictors, workplace pressure and perceived family support, and a 1:1 case‐to‐control ratio. The proportion of controls having high workplace stress (25%) and low family support (18.5) was estimated based on previous research conducted in Spain [5]. Finally, to meet all the assumptions, the estimated sample size required was 187 cases and 187 controls. A convenient sampling approach was employed to recruit participants, ensuring informed consent was obtained before data collection.

2.3. Inclusion and Exclusion Criteria

The study included mothers who had delivered within the past 6 months and were actively employed in the formal sector, which refers to legally registered organizations that provide structured employment with regular wages/salaries and, in most cases, benefits such as fixed working hours, leave policies, or social security (e.g., government offices, private companies, NGOs, and schools/universities). Exclusion criteria were carefully designed to minimize confounding factors that could independently elevate the risk of postpartum depression; therefore, women were excluded if they had experienced severe illness during the observation period, had a history of illicit drug use or substance abuse, any deaths of close relatives, deaths of children, and undergoing any chronic diseases, after birth complications, or had recently undergone a life‐threatening event. These factors were excluded as they could disproportionately influence the risk of postpartum depression and obscure the relationship between workplace stress and family support.

2.4. Selection of Cases

Participants were classified into two groups based on their Edinburgh Postnatal Depression Scale (EPDS) scores. Participants were categorized into two groups based on their scores on the Edinburgh Postnatal Depression Scale (EPDS), a widely validated tool for assessing postpartum depression. The Bengali‐translated version of the EPDS, adapted by Gausia et al., was used alongside the original English version [6]. Women scoring 12 or higher on the EPDS were categorized as cases (i.e., experiencing postpartum depression), while those scoring below this threshold were considered controls (i.e., without postpartum depression). The EPDS consists of 10 items, each rated on a 0–3 scale, assessing symptoms such as anhedonia, guilt, anxiety, panic attacks, fatigue, sleep disturbances, sadness, tearfulness, and suicidal thoughts. The total score ranges from 0 to 30, with higher scores indicating a greater risk of postpartum depression [7]. While a threshold score of 12 was used for classification, prior research emphasizes caution even at slightly lower scores, particularly in cases where suicidal ideation is reported [8]. To ensure the reliability of the EPDS in this study, construct validity was assessed. The Average Variance Extracted (AVE) was 0.589, and Cronbach's alpha was (0.913), indicating strong internal consistency and a good fit for the study.

2.5. Selection of Controls

Following the identification of cases, controls were carefully selected to ensure comparability. Each control was individually matched to a case based on age and profession. Age matching was performed either exactly or within a ±12‐month range. For professional matching, six broad categories of formal sector occupations were established: (1) government and administration (including civil service and judiciary), (2) corporate and industrial sectors (covering banking, finance, telecommunications, garments, and pharmaceuticals), (3) academia and research (comprising university faculty, researchers, and policy analysts), (4) healthcare (including doctors, nurses, and therapists), and (5) social services (encompassing NGO workers, social workers, and project coordinators), and (6) skilled and technical professions (engineers, architects, auditors, accountants, and lawyers). When selecting controls, efforts were made to match them as precisely as possible to cases within these professional categories. Eligible controls were formal sector working mothers who visited the selected clinics for postnatal care (PNC) services within 6 months of delivering their last child. This 6‐month postpartum window was chosen based on evidence suggesting that the Edinburgh Postnatal Depression Scale (EPDS) is most effective in detecting postpartum depression within this timeframe [9].

3. Data Collection Technique and Quality Assurance

Data collection was conducted through face‐to‐face interviews using a structured questionnaire administered by trained interviewers, which lasted from 15 to 20 min. Data were collected from March 2025 to May 2025. To ensure data accuracy and sensitivity, interviews took place under the supervision of on‐duty physicians within the postnatal care departments of the selected clinics. The data collection team consisted of five fourth‐year female medical students from Dhaka Medical College, supervised by first author. The involvement of female medical students was intentional, due to their familiarity with maternal health concerns, clinical environments, and the cultural perspective of the respondents. A comprehensive 1‐day training session was conducted for the data collection team, focusing on the study's objectives, ethical considerations, and the standardized protocol for administering face‐to‐face interviews. This training was led by a specialist from the psychiatry department of Bangabandhu Sheikh Mujibur Rahman Medical University (BSMMU), Bangladesh. To validate the feasibility and clarity of the study instruments, a pilot study was conducted involving 20 respondents (10 cases and 10 controls) from the same clinics. Based on the pilot study findings, two variables, history of depression and family history of depression, were removed due to a high level of recall bias among respondents. Additionally, a new variable, hemoglobin levels during pregnancy, was introduced, as a significant number of participants spontaneously reported experiencing complications related to anemia during pregnancy.

4. Instruments Used in This Study

This study employed three instruments to measure workplace stress and family support. Workplace stress was assessed using the Workplace Stress Scale (WSS), a widely recognized tool developed by the American Institute of Stress (AIS) in 1978. This scale has been extensively used across different workplace settings and formal occupational positions worldwide, including assessments across genders [10]. The WSS consists of eight items, each scored on a five‐point Likert scale (ranging from 1 to 5), with total scores ranging from 8 to 40. Based on established AIS scoring guidelines, total scores were categorized into five levels: low stress (8–15), mild stress (16–20), moderate stress (21–25), high stress (26–30), and very high stress (31–40). However, for analytical parsimony and to ensure adequate distribution across categories, this study adopted a three‐level classification following Pristiawati et al. [11], categorizing scores as low stress (8–20), moderate stress (21–25), and high stress (26–40).

To further assess workplace stress, this study also incorporated the Over‐Commitment (OC) subscale from the Effort–Reward Imbalance (ERI) model [12], which measures excessive work‐related commitment. The OC subscale consists of six items, each rated on a four‐point Likert scale, with a final score range of 6– 24, which have been validated before in European societies [13]. Based on the total score, respondents were categorized into three groups: loosely committed (6–11), moderately committed (12–17), and highly committed (18–24). Both scales demonstrated strong reliability in this study. The Average Variance Extracted (AVE) for WSS was 0.66, and for OC, it was 0.69, and Cronbach's alpha reliability score for WSS was 0.821, and for OC, it was 0.918, confirming strong internal consistency.

Perceived family support was assessed using the Family APGAR Questionnaire, which consists of five items rated on a three‐point Likert scale, with a total score ranging from 0 to 10 (8). Based on the final score, family support was categorized into three groups: functional family (7–10), mild dysfunction (4–6), and severe dysfunction (0–3). The Family APGAR scale has been validated in various settings, such as European, Indian, and African [14], and demonstrated strong reliability in this study, with an AVE of 0.524 and a Cronbach's alpha score of 0.907. Additional information for scales and associated statistics can be found in the (Suppōrting Information S1: Tables S1 and S2).

5. Covariates

In addition to the primary explanatory variables, several covariates were included based on previous research that indicated their significant association with postpartum depression (PND) [7, 15, 16, 17]. Moreover, variables with p < 0.20 in bivariate analysis were entered into the multivariable model. Additionally, potential confounders were retained if their inclusion changed the primary exposure effect estimate by ≥ 10% (change‐in‐estimate criterion). Since this study follows a matched case‐control design, age and profession were used as matching variables. Other demographic characteristics, pregnancy‐related complications, and maternal healthcare utilization factors were also considered, as detailed in Table 1. Additionally, anxiety was included as an independent variable because substantial longitudinal evidence suggests that antenatal and perinatal anxiety are strong predictors of postpartum depression rather than merely comorbid conditions [18]. Studies have shown that anxiety symptoms during pregnancy increase vulnerability to subsequent depressive episodes in the postpartum period through shared cognitive, neurobiological, and stress‐response pathways [19, 20].

Table 1.

Description of independent variables.

	Variables	Groups
1.	Age	Continuous
2.	Profession	Government and administration workers, corporate and industrial workers, academia and research, healthcare, social services, skilled and technical professions
3.	Monthly income (BDT)	< 12,000 BDT, 12,000–30,000 BDT, > 30,000 BDT
4.	Education	No education, primary or secondary, higher
5.	Birth order	1st, 2nd, 3rd, or higher
6.	Pregnancy intention	Planned (assisted), unplanned
7.	Mode of delivery	Normal, cesarean
8.	Premature birth	No, yes
9.	Attending minimum 4 ANC visits for last birth	No, yes
10.	Pregnancy complication: gestational diabetes	No, yes
11.	Pregnancy complication: hypertension	No, yes
12.	Pregnancy complication: thyroid	No, yes
13.	Postpartum hemorrhage (PPH)	No, yes
14.	During pregnancy hemoglobin levels	Normal, non‐normal (anemia)
15.	Maternal pyrexia	No, yes
16.	Anxiety	0–4: Minimal anxiety; 5–9: mild anxiety; 10–14: moderate anxiety; 15–21: severe anxiety
17.	Effort‐reward imbalance scale (ERI) (over‐commitment part only)	Higher score means more commitment
18.	Workplace stress	Low stress (0–15), mild stress (16–20), moderate stress (21–25), severe (26–30), dangerous (31–40)
19.	Family APGAR questionnaire	Highly functional family (score 7–10), moderately dysfunctional family (score 4–6), severely dysfunctional family (score 0–3)

6. Statistical Analysis

The collected questionnaires were reviewed for completeness before being entered into SPSS (Version 26) for initial data cleaning. All further statistical analyses were conducted using STATA (Version 16). Each case‐control pair was assigned a matched ID number to ensure consistency during data entry and analysis. Descriptive statistics were used to summarize the study variables, expressed as percentages and frequencies. To assess the association between the explanatory variables and postpartum depression, Conditional Logistic Regression (CLR) was used. Multivariable models were constructed using a theory‐informed approach. Variables identified a priori as potential confounders were entered into the model, and their inclusion was further evaluated using a change‐in‐estimate criterion (≥ 10% change in the main exposure odds ratio). The matched crude odds ratios (mCOR) were estimated in bivariate analyses, while the matched adjusted odds ratios (mAOR) were derived from multivariable models to account for potential confounders. Age and profession were used as matching variables; therefore, conditional logistic regression automatically accounted for these factors, and they were not included as covariates in the multivariable model to avoid redundancy. Other demographic and maternal health variables were included to adjust for potential residual confounding. A p value < 0.05 was considered statistically significant, and the odds ratio (OR) was reported with a 95% confidence interval (95% CI). Additional information regarding the sensitivity and VIF analyses can be found in the Supporting Information S1: Tables S3, S4.

Table 4.

Maternal health complexities associated with postpartum depression.

Gestational diabetes	mCOR (95% CI)	p value	mAOR (95% CI)	p value
No	2.00 (1.05–3.80)	0.034	2.47 (0.31–19.71)	0.394
Yes	1		1
Hypertension
No	1		1
Yes	3.55 (2.21–5.69)	< 0.001	1.61 (1.16–4.62)	< 0.001
Thyroid
No	1		1
Yes	1.55 (1.12–2.76)	< 0.001	1.36 (1.03–3.12)	0.005
Postpartum hemorrhage (PPH)
No	1		1
Yes	1.53 (1.34–3.89)	0.017	0.37 (0.11–1.18)	0.092
During pregnancy hemoglobin levels (self‐reported)
Normal	1		1
Non‐normal (anemia)	6.82 (3.62–12.84)	< 0.001	4.11 (2.26–38. 32)	0.001
Maternal pyrexia
No	1		1
Yes	1.85 (1.16–2.96)	0.010	0.66 (0.20–2.15)	0.488
Anxiety (GAD score)
Minimal or mild anxiety	1		1
Moderate anxiety	3.49 (1.84–6.63)	< 0.001	2.24 (1.68–7.38)	< 0.001
Severe anxiety	16.17 (7.07–36.97)	< 0.001	14.61 (2.09–43.81)	< 0.001

7. Ethical Approval

In addition, the Ethical Clearance Committee of Research and Innovation Centre (RIC), Khulna University, Bangladesh, has approved this whole study protocol. The reference number is (KUECC‐2025‐11‐155). All the methods are in accordance with relevant guidelines and regulations.

8. Results

8.1. Background Participants' Characteristics

Table 2 presents the demographic characteristics of 187 cases (with postpartum depression) and 187 age‐ and profession‐matched controls. The mean age was similar between cases (28.1 years) and controls (27.7 years), and the distribution of professional categories was identical across groups. Cases were more likely to report higher monthly income, with 83.5% earning above 12,000 BDT compared to 46.3% of controls. Educational attainment was also higher among cases, with 61.7% having completed higher education versus 53.2% among controls. Significant differences were observed between cases and controls across several obstetric and clinical characteristics. Controls were more often first‐time mothers (51.1%), whereas cases more frequently had a third or later birth (35.6%). Unplanned pregnancies were markedly higher among cases (37.2%) than controls (18.1%). Although cesarean section was the predominant mode of delivery in both groups, it was more common among controls (69.1%) than cases (59.6%). Cases had significantly higher rates of premature birth (28.2% vs. 12.8%), gestational diabetes (18.1% vs. 10.6%), hypertension (55.3% vs. 25.5%), anemia during pregnancy (46.8% vs. 12.8%), and maternal pyrexia (32.4% vs. 20.2%). Antenatal care attendance (≥ 4 visits) was slightly lower among cases (66.5%) than controls (70.7%). Thyroid disorders showed minimal difference between groups, while postpartum hemorrhage was more prevalent among controls (26.6% vs. 16.5%). Cases exhibited substantially higher psychological and psychosocial distress than controls. Severe anxiety (GAD‐7) was far more common among cases (41.5%) than controls (6.4%), while most controls reported minimal or mild anxiety (68.6% vs. 29.8%). High overcommitment (ERI–OC) was reported by 70.2% of cases compared with 33.0% of controls. Severe workplace stress was also more prevalent among cases (33.0% vs. 8.5%). Family functioning differed markedly, with over half of cases (53.7%) reporting severely dysfunctional families, compared to 29.8% of controls, whereas highly functional families were more common among controls (23.4% vs. 9.6%).

Table 2.

Participant's background characteristics.

	Control (%)	Case (%)
Age (matched)
Mean (±)	27.7 (16.3%)	28.1 (13.5%)
Profession (matched)
Government and administration workers	23 (12.23%)	23 (12.23%)
Corporate and industrial workers	22 (11.70%)	22 (11.70%)
Academia and research	56 (29.78%)	56 (29.78)
Healthcare	47 (25.00%)	47 (25.00%)
Social services	19 (10.10%)	19 (10.10%)
Skilled and technical professions	21 (11.17%)	21 (11.17%)
Monthly income (BDT)
< 12,000	101 (53.7%)	31 (16.5%)
12,000–30,000	62 (33.0%)	108 (57.4%)
> 30,000	25 (13.3%)	49 (26.1%)
Education
No education	18 (9.6%)	19 (10.1%)
Primary or secondary	70 (37.2%)	53 (28.2%)
Higher	100 (53.2%)	116 (61.7%)
Birth order
First	96 (51.1%)	57 (30.3%)
Second	63 (33.5%)	64 (34.0%)
≥ 3	29 (15.4%)	67 (35.6%)
Pregnancy intention
Planned	154 (81.9%)	118 (62.8%)
Unplanned	34 (18.1%)	70 (37.2%)
Mode of delivery
Vaginal	58 (30.9%)	76 (40.4%)
C‐section	130 (69.1%)	112 (59.6%)
Premature birth
No	164 (87.2%)	135 (71.8%)
Yes	24 (12.8%)	53 (28.2%)
Attending minimum 4 ANC visits for last birth
No	55 (29.3%)	63 (33.5%)
Yes	133 (70.7%)	125 (66.5%)
Gestational diabetes
No	168 (89.4%)	154 (81.9%)
Yes	20 (10.6%)	34 (18.1%)
Hypertension
No	140 (74.5%)	84 (44.7%)
Yes	48 (25.5%)	104 (55.3%)
Thyroid
No	119 (63.3%)	113 (60.1%)
Yes	69 (36.7%)	75 (39.9%)
Postpartum hemorrhage (PPH)
No	138 (73.4%)	157 (83.5%)
Yes	50 (26.6%)	31 (16.5%)
During pregnancy hemoglobin levels (self‐reported)
Normal	164 (87.2%)	100 (53.2%)
Non‐normal (anemia)	24 (12.8%)	88 (46.8%)
Maternal pyrexia
No	150 (79.8%)	127 (67.6%)
Yes	38 (20.2%)	61 (32.4%)
Anxiety (GAD score)	Mean (8.38)	Mean (12.63)
Minimal or Mild anxiety	129 (68.6%)	56 (29.8%)
Moderate anxiety	47 (25.0%)	54 (28.7%)
Severe anxiety	12 (6.4%)	78 (41.5%)
Overcommitment	Mean (10.87)	Mean (14.05)
Lower over commitment	126 (67.0%)	56 (29.8%)
Higher over commitment	62 (33.0%)	132 (70.2%)
Workplace stress	Mean (15.64)	Mean (21.94)
Low stress	164 (87.2%)	83 (44.1%)
Moderate stress	8 (4.3%)	43 (22.9%)
Severe stress	16 (8.5%)	62 (33.0%)
Family functionality	Mean (4.66)	Mean (3.40)
Severely dysfunctional family	56 (29.8%)	101 (53.7%)
Moderately dysfunctional family	88 (46.8%)	69 (36.7%)
Highly functional family	44 (23.4%)	18 (9.6%)

Abbreviation: SD, standard deviation.

8.2. Socio‐Demographic and Household Factors Associated With Postpartum Depression

Table 3 presents the conditional logistic regression results, showing that birth order was significantly associated with PPD. Compared with first‐time mothers, women with a second child had higher odds of PPD (mAOR = 1.52, 95% CI: 1.12–2.82, p = 0.021), while those with a third or higher‐order child had more than fourfold increased odds (mAOR = 4.27, 95% CI: 2.11–8.38, p = 0.002). Cesarean delivery was also associated with increased odds of PPD compared to vaginal delivery (mAOR = 2.29, 95% CI: 1.84–5.91, p = 0.001). Monthly income, education level, pregnancy planning, premature birth, and antenatal care attendance were not significantly associated with PPD in the adjusted models (p > 0.05).

Table 3.

Socio‐demographic and household factors associated with postpartum depression.

	mCOR (95% CI)	p value	mAOR (95% CI)	p value
Monthly income
< 12,000	1.14 (1.05–3.29)	< 0.001	0.12 (0.07–0.31)	0.783
12,000–30,000	2.80 (1.44–4.47)	0.003	0.91 (0.47–1.76)	0.175
> 30,000	1		1
Education
No education	0.95 (0.45–2.00)	0.171	1.39 (0.39–3.69)	0.400
Primary or secondary	0.65 (0.41–1.02)	0.898	0.64 (0.33–1.09)	0.128
Higher	1		1
Birth order
First	1		1
Second	1.58 (1.25–2.63)	< 0.001	1.52 (1.12–2.82)	0.021
≥ 3	3.81 (2.13–6.81)	0.081	4.27 (2.11–8.38)	0.002
Pregnancy
Planned	1		1
Unplanned	2.90 (1.72–4.88)	< 0.001	2.13 (0.74–6.11)	0.159
Mode of delivery
Normal	1		1
C‐section	4.59 (1.36–12.96)	< 0.001	2.29 (1.84–5.91)	0.001
Premature birth
No	1		1
Yes	2.81 (1.59–4.98)	< 0.001	1.45 (0.44–7.12)	0.453
Attending minimum 4 ANC visits for last birth
No	0.83 (0.54–1.27)	0.383	0.68 (0.25–1.88)	0.462
Yes	1		1

8.3. Maternal Health Factors Associated With Postpartum Depression

Table 4 shows that women with hypertension had higher odds of PPD (mAOR = 1.61, 95% CI: 1.16–4.62, p = < 0.001), as did those with thyroid disease (mAOR = 1.36, 95% CI: 1.03–3.12, p = 0.005). Anemia during pregnancy showed a strong association, with affected women having markedly higher odds of PPD (mAOR = 4.11, 95% CI: 2.26–38. 32 p = 0.001). Anxiety severity demonstrated a dose–response relationship: moderate anxiety was associated with increased odds of PPD (mAOR = 2.24, 95% CI: 0.68–7.38, p = 0.001), while severe anxiety was associated with substantially higher odds (mAOR = 14.61, 95% CI: 2.09–43.81, p < 0.001). Gestational diabetes, postpartum hemorrhage, and maternal pyrexia were not significantly associated with PPD.

8.4. Associations of Workplace Stress, Overcommitment, and Family Support With PPD

Table 5, portrays that women with high overcommitment had markedly increased odds of PPD (mAOR = 8.16, 95% CI: 4.22–31.65, p < 0.001). Severe workplace stress showed the strongest association, with affected women over 18 times more likely to experience PPD (mAOR = 18.45, 95% CI: 5.33–57.96, p < 0.001), while moderate stress was also associated with higher odds (mAOR = 4.55, 95% CI: 1.44–8.45, p < 0.001). Family dysfunction demonstrated a graded relationship with PPD: women from moderately dysfunctional families had increased odds (mAOR = 2.12, 95% CI: 1.51–11.12, p = 0.006), and those from severely dysfunctional families had substantially higher odds compared to women from severely functional families (mAOR = 6.37, 95% CI: 1.11–34.20, p < 0.001).

Table 5.

Workplace stress, overcommitment, and family support associations to postpartum depression.

Overcommitment	mCOR (95% CI)	p value	mAOR (95% CI)	p value
Lower overcommitment	1		1
Higher overcommitment	5.12 (3.04–8.61)	< 0.001	8.16 (4.22–31.65)	< 0.001
Workplace stress
Low stress	1		1
Moderate stress	9.26 (4.28–20.01)	< 0.00	4.55 (1.44–8.45)	< 0.001
Severe stress	16.54 (5.93–46.10)	< 0.001	18.45 (5.33–57.96)	< 0.001
Family functionality
Severely dysfunctional family	4.87 (2.38–9.94)	< 0.001	6.37 (1.11–34.20)	< 0.001
Moderately dysfunctional family	2.04 (1.02–4.10)	0.044	2.12 (1.51–11.12)	0.006
Highly functional family	1		1

9. Discussion

This study examined the multifactorial predictors of PPD among employed mothers in Bangladesh, emphasizing workplace stress, psychosocial vulnerabilities, and maternal health factors, which are less explored in non‐Western contexts [18, 20, 21]. Using a matched case‐control design, we found that severe workplace stress, high overcommitment, moderate to severe anxiety, dysfunctional family dynamics, and obstetric or maternal health factors such as anemia, hypertension, thyroid disorders, cesarean delivery, and higher parity were significantly associated with PPD. These findings underscore the intersection of occupational, social, and biological predictors in shaping maternal mental health in low‐ and middle‐income countries, where women often navigate the dual responsibilities of work and motherhood within rapidly changing socioeconomic landscapes.

Workplace stress showed close association with PPD, aligning with global research linking higher job‐related stress, low job satisfaction, and lack of organizational support to postpartum depressive and anxiety symptoms [4, 19]. The impact of workplace stress is gendered and systemic, disproportionately affecting women who experience work–family conflict, particularly when married, responsible for childcare, or engaged in inflexible work environments [1, 22, 23]. Role theory posits that the “multiple burden” of employment and unpaid household labor creates inter‐role conflict, generating emotional exhaustion and reduced coping capacity when expectations from both spheres collide [24, 25]. High work‐related overcommitment was associated with higher odds of postpartum depression, consistent with prior studies reporting that excessive job demands are linked to greater emotional exhaustion and depressive symptoms [26, 27]. Overcommitment is often a response to systemic pressures, where women must work beyond capacity to balance employment and family responsibilities [28]. In South Asia, gender‐based workplace discrimination and limited recognition exacerbate this imbalance, aligning with the effort–reward imbalance model, which highlights mental health risks when effort is high, but rewards are inadequate [12, 29].

Family dysfunction was also heavily linked with PDD. Women from moderately or severely dysfunctional families had higher odds of PPD compared to those from highly functional families, consistent with prior literature [30, 31, 32]. Family function encompasses communication, emotional bonding, conflict resolution, and decision‐making [33]. Declines in these areas increase feelings of isolation, familial tension, and poor coping capacity, contributing to maternal mental health vulnerability [34]. In South Asian households, limited support from mothers‐in‐law or generational conflicts in postpartum care can intensify stress and feelings of inadequacy, highlighting the importance of culturally sensitive family support [28, 35, 36, 37, 38].

Anxiety severity is significantly associated with higher odds of PPD, corroborating evidence that antenatal or perinatal anxiety heightens vulnerability to clinically significant postpartum depression [38, 39, 40]. Biological factors, including hypertension, thyroid disorders, and anemia, were independently associated with higher odds of postpartum depression (PPD). Hypertension has been linked to higher PPD prevalence and emotional strain due to chronic disease management or complications such as preeclampsia [41, 42]. Thyroid dysfunction and elevated TSH levels during delivery are associated with postpartum mood disturbances [43, 44, 45]. Anemia contributes to fatigue, irritability, and impaired concentration, further affecting maternal mood and caregiving capacity [46, 47, 48]. Cesarean delivery was also significantly associated with postpartum depression, particularly emergency procedures, possibly due to unexpected complications, perceived loss of control, and traumatic birth experiences, although previous studies report mixed findings [49]. Higher parity (≥ 2 children) was associated with higher odds of postpartum depression, potentially reflecting cumulative caregiving responsibilities, financial pressures, and reduced self‐care time; however, some research suggests that experienced mothers may develop better coping mechanisms [23, 50, 51].

9.1. Strengths and Limitations of the Study

Strengths include a matched case‐control design, use of validated instruments (Workplace Stress Scale, Effort–Reward Imbalance model, Family APGAR), and adjusted analyses that enhance the reliability of observed associations. Findings are policy‐relevant, highlighting modifiable occupational and family‐level risk factors. This study has several limitations. Participants were recruited exclusively from Surjer Hashi Clinics in Dhaka, which may introduce selection bias, as clinic attendees may differ from women delivering at public hospitals, at home, or in rural areas in terms of socioeconomic status, healthcare‐seeking behavior, and employment. Therefore, findings may not be generalizable beyond an urban, clinic‐attending population. Additionally, several maternal health variables, including anemia and other pregnancy‐related conditions (e.g., hypertension, thyroid disorders), were self‐reported without clinical verification, introducing the potential for recall bias or misclassification, particularly if cases were more likely than controls to report health problems.

9.2. Policy Recommendations

Addressing postpartum depression (PPD) among formally employed women in Bangladesh requires coordinated policy attention within existing labor and maternal health frameworks. Strengthening enforcement and potential extension of current maternity protection policies could help reduce the combined pressures of employment and early motherhood. While Bangladesh's labor law currently provides paid maternity leave provisions, consideration may be given to gradually expanding both the duration and wage replacement level in alignment with international recommendations and national economic feasibility. Such reforms may be approached as phased or long‐term policy goals. Employers in medium‐ and large‐scale organizations should be required to implement perinatal mental health support programs, including workplace stress screening, monitoring of overcommitment, and access to counseling services. Within the healthcare system, strengthening routine screening for depression and anxiety during pregnancy and the postpartum period, using validated tools such as the EPDS or PHQ‐9, could enhance early identification and referral, particularly if incorporated into existing maternal health service delivery platforms. Finally, expanding access to affordable, state‐subsidized community daycare centers would reduce caregiving burden, particularly for multiparous women, and support maternal mental well‐being and workforce participation.

10. Conclusion

This study identifies key occupational, psychosocial, and maternal health factors that increase postpartum depression risk among employed married women in Bangladesh. Severe workplace stress, high overcommitment, moderate‐to‐severe anxiety, and dysfunctional family support were major psychosocial contributors, while hypertension, thyroid disorders, anemia, higher parity, and cesarean delivery elevated biological risk. These findings highlight the complex interplay of workplace, social, and health predictors on maternal mental health. Future research should include longitudinal studies and intervention trials targeting workplace stress reduction, flexible work policies, and structured family support to effectively mitigate postpartum depression in low‐ and middle‐income settings.

Author Contributions

M.M.T. contributed to the literature search, data analysis and interpretation, figures and tables, and wrote the manuscript draft and contributed to the revision. S.B.M. and M.A.R. contributed to the review of the study design, data interpretation, and writing of the manuscript draft. M.M.T., S.B.M.T.R and S.Z. contributed to the writing of the manuscript and the preparation of draft tables. M.M.T. and M.A.R. contributed to data analysis and writing the manuscript draft, and contributed to the revision. M.A.R and T.M. contributed to the design of the critical review of the design, analysis, and interpretation involved in the manuscript. M.M.T. had access to data in the study and took responsibility for the data's integrity and the accuracy of data analysis. All authors gave final approval of the version to be published.

Funding

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

GenAI Declarations

The authors did not use any AI tools for data analysis or writing, but in order to improve readability and solve grammatical errors, they used GrammarlyPro and the free version of CoPilot. But they ensure all edits were checked by in‐house authors to make sure all edits were coherent and academically readable.

Transparency Statement

The corresponding author, Md. Ashfikur Rahman, affirms that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.

Supporting information

Table S1: The Workplace Stress Scale.

Table S2: Effort‐Reward Imbalance Scale (ERI) [over‐commitment part only).

Table S3 (Sensitivity Analysis): Maternal Health Complexities Associated with Postpartum Depression (Model Excluding Anxiety).

Table S4: Variance Inflation Factors (VIF) for Predictors in Multivariable Logistic Regression Model.

Tohan M. M., Mahbub S. B., Zaman S., Roy T., McAuliffe T., and M. A.Rahman, “Workplace Stress, Overcommitment, and Maternal Health as Predictors of Postpartum Depression: A Case‐Control Study Among Employed Married Women in Bangladesh,” Health Science Reports 9 (2026): e72500, 10.1002/hsr2.72500.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.