Prevalence, predictors, and functional impact of postpartum depression using the Patient Health Questionnaire-9 (PHQ-9) among 830 women

Abstract

Background

Postpartum depression (PPD) is a prevalent maternal mental health condition with significant consequences for mothers, infants, and families. The determinants and functional implications of PPD vary across populations, highlighting the need for population-specific data. This study aimed to determine the prevalence of PPD, identify associated risk factors, and assess functional impairment using the Patient Health Questionnaire-9 (PHQ-9).

Methods

This was a prospective cross-sectional observational study. Sociodemographic, obstetric, neonatal, and psychosocial characteristics were collected through structured face-to-face interviews during routine postpartum visits between days 7 and 21. Depressive symptoms were assessed using the validated Turkish version of the PHQ-9, with PPD defined as a score ≥ 10. Functional impairment was evaluated using item 10 of the PHQ-9. Initially, depression and non-depression groups were compared, followed by multivariable logistic regression to identify independent predictors.

Results

A total of 830 postpartum women were included, of whom 19.2% (159/830) met the diagnostic threshold for PPD. Multivariate logistic regression identified three independent predictors of PPD: functional impairment (strongest predictor, nearly eightfold increased odds), history of pre-gestational psychiatric disorder (over fourfold increased odds), and maternal unemployment (approximately twofold increased odds). No obstetric or neonatal variables—including mode of delivery, gestational age, or birth weight—remained significant after adjustment.

Conclusion

Postpartum depression affects nearly one in five women and is strongly associated with functional impairment, prior psychiatric illness, and maternal unemployment. Our findings highlight the need to incorporate routine mental health screening, functional assessment, and targeted follow-up into standard postpartum care to enable earlier identification and intervention, ultimately mitigating the burden of PPD on the mother, infant, family, and society.

Background

Although childbirth is widely considered a joyful life event, postpartum depression (PPD) remains a prevalent mental health concern among new mothers as one of the most common postpartum mood and anxiety disorders. PPD is broadly defined as a major depressive episode that occurs during the perinatal period [1]. The estimated prevalence of PPD ranges from 10% to 20%, with considerable cross-cultural variation [2]. Higher rates have consistently been reported in low-income populations compared with middle- and high-income groups [3]. Among high-risk populations—such as women who are poorly nourished, of low socioeconomic status, adolescent mothers, and single mothers—the prevalence may rise to as high as 26% [4]. Although PPD typically emerges within the first 1–3 months after delivery, it can develop at any point during the first year after childbirth. In some cases, symptoms may begin during pregnancy and subsequently intensify after delivery [5, 6].

Untreated PPD is associated with significant adverse outcomes across the maternal–infant–family system. Affected mothers may experience impaired bonding with their infants, difficulties in daily functioning, and an increased risk of persistent or recurrent depressive episodes [7–9]. Infants of mothers with untreated PPD are more likely to exhibit poor weight gain, sleep disturbances, delays in cognitive and emotional development, and insecure attachment patterns [10, 11]. At the family level, PPD has been linked to marital conflict, reduced partner support, elevated caregiving stress, and overall disruption of family dynamics [12]. Together, these well-documented consequences highlight the importance of evidence-based prevention strategies, early identification, and timely intervention to mitigate the burden of PPD.

According to the American College of Obstetricians and Gynecologists (ACOG), all women should be screened for depression and anxiety at least once during pregnancy and again during the postpartum period using a standardized, validated tool such as the Patient Health Questionnaire (PHQ-9) or Edinburgh Postnatal Depression Scale. Screening is recommended at the initial prenatal visit, at least once later in pregnancy, and during the comprehensive postpartum visit. ACOG also emphasizes that screening must be implemented within a system that ensures timely assessment, appropriate follow-up, and treatment when indicated. A positive screen, or the presence of severe symptoms, requires a more comprehensive diagnostic evaluation based on DSM-5 criteria, including assessment of mood disturbance, anhedonia, functional impairment, sleep disturbance, and suicidal ideation [13, 14]. Notably, functional impairment is an emerging and understudied dimension of postpartum mental health, warranting greater empirical attention.

As demonstrated, PPD is a multifaceted mental health condition that affects not only the mother but also infants, families, and the broader society in a cascading manner. Despite its clinical significance, the determinants, associated risk factors, and functional consequences of PPD vary across populations and require further investigation. As such, the present study seeks to determine the prevalence of PPD, identify its associated risk factors, and assess levels of functional impairment among postpartum women. By addressing these aims, the study contributes to a deeper understanding of PPD’s implications for maternal well-being and provides empirical guidance for enhancing prevention efforts and screening practices.

Methods

This was a prospective cross-sectional observational study conducted between July 2025 and November 2025 in the Obstetrics and Gynecology outpatient clinic of our tertiary care center, with all variables assessed at a single, predefined postpartum time point (postpartum days 7–21). The postpartum assessment window of days 7–21 was selected to evaluate depressive symptoms beyond the immediate postpartum adjustment period (‘baby blues’) while enabling early identification during routine postpartum follow-up, despite the broader diagnostic window of postpartum depression extending up to one year after delivery.

An a priori power analysis was performed using GPower software (version 3.1.9.7). Based on a previously reported odds ratio of approximately 6.4 for the association between prenatal anxiety and peripartum depression, assuming a two-sided alpha level of 0.05, 80% power, an outcome probability of 4% in the non-anxious group, and an R² value of 0.20 for other covariates, the minimum sample size required for a single-predictor logistic regression model was estimated as 63 participants. However, as the primary analysis involved a multivariable logistic regression model, sample size determination was further guided by the events-per-variable principle. Assuming a peripartum depression prevalence of approximately 19% and a minimum of 10 outcome events per predictor variable, a total sample size of approximately 525 participants was required to ensure model stability and reliable effect estimates. However, as covariates that could affect the logistic regression analysis might arise during the study, the minimum number of patients included was calculated as 750.

This study was conducted in accordance with the Declaration of Helsinki (revised 2013). The study protocol was approved by the appropriate institutional ethics committee (approval date: June 15, 2025; decision number: 2025/190).

Inclusion criteria were as follows

(1) maternal age ≥ 18 years; (2) singleton pregnancy; (3) presentation for postpartum follow-up between days 7 and 21; (4) delivery at our hospital; (5) live birth; (6) absence of fetal structural or chromosomal abnormalities; (7) ability to read and understand the questionnaire; (8) voluntary written informed consent; (9) stable maternal medical condition without severe uncontrolled systemic disease; and (10) inclusion of both vaginal and cesarean deliveries to allow assessment of mode of delivery as a potential risk factor.

Exclusion criteria included

(1) women with a clinical diagnosis of major depression during pregnancy who required psychiatric treatment; (2) multiple pregnancy; (3) fetal or neonatal death; (4) incomplete questionnaire or missing data; (5) severe postpartum complications (e.g., postpartum hemorrhage requiring intensive care, sepsis, or need for surgical re-intervention); (6) maternal medical comorbidities that could influence mood, such as uncontrolled thyroid disorders, epilepsy, or chronic systemic diseases requiring continuous treatment; (7) Severe psychiatric disorders, such as bipolar disorder or schizophrenia, and treatment for these conditions, including the use of antipsychotic medications; (8) alcohol or substance abuse; and (9) language or literacy limitations interfering with reliable questionnaire completion.

Data collection was conducted during routine postpartum visits between days 7 and 21. After obtaining written informed consent, participants completed a structured questionnaire through face-to-face interviews conducted in a quiet and private room within the clinic. Sociodemographic variables (maternal age, education level, employment status, income level, marital status, smoking status), obstetric characteristics (gravidity, parity, pregnancy loss history, mode of delivery, gestational age at birth, gestational diabetes or hypertensive disorders), and neonatal outcomes (birth weight, sex, NICU admission) were obtained.

Pre-gestational psychological disorder was defined as any self-reported or medically documented history of depressive or anxiety disorders prior to the index pregnancy. These women were included in the analysis, as such conditions represent common mood or anxiety disturbances rather than severe psychiatric illness. In contrast, women with severe psychiatric disorders—such as bipolar disorder or schizophrenia—and those receiving treatment for these conditions, including antipsychotic medications, were excluded in accordance with the study’s exclusion criteria.

Depressive symptoms were assessed using the validated Turkish version of the PHQ-9 [15]. Postpartum depression was defined as a PHQ-9 score ≥ 10, consistent with widely used cut-off values in large validation studies and perinatal depression screening research [16, 17].

Functional impairment related to depressive symptoms was assessed using item 10 of the PHQ-9, which evaluates the extent to which symptoms interfere with daily functioning. All questionnaires were reviewed on-site for completeness, and incomplete forms were excluded according to the predefined criteria.

The primary outcome of the study was the prevalence of PPD, with secondary outcomes including associated risk factors and levels of functional impairment.

Statistical analyses were performed using IBM SPSS Statistics version 26.0 (IBM Corp., Armonk, NY, USA). The distribution of continuous variables was assessed using visual inspection (histograms and Q–Q plots) and the Shapiro–Wilk test. As most continuous variables did not follow a normal distribution, they were presented as median (interquartile range, IQR) and compared using the Mann–Whitney U test. Categorical variables were expressed as number (percentage) and compared using the chi-square test when expected cell counts were adequate; Fisher’s exact test was used when any expected cell count was < 5. All statistical tests were conducted under the assumption of independent observations.

A multivariate logistic regression model was constructed to identify independent determinants of postpartum depression. Variables were entered into the multivariate model a priori based on clinical relevance and evidence from previous studies, rather than on statistical significance in univariate analyses. Results were reported as odds ratios (ORs) with 95% confidence intervals (CIs). A two-sided p-value < 0.05 was considered statistically significant.

Results

A total of 830 postpartum women were included in the final analysis, of whom 159 (19.2%) met the diagnostic threshold for PPD (PHQ-9 ≥ 10), while 671 (80.8%) were classified as the non-depression group. Sociodemographic characteristics were generally comparable between groups; maternal age did not differ significantly, with a median age of 28 years in both groups. Gravidity, parity, and the number of living children were all significantly higher among women with PPD. Employment status also differed significantly between groups, with unemployment more common in the depression group. In contrast, BMI, planned pregnancy, duration of marriage, and spouse’s employment status showed no significant associations with depression status (Table 1).

Table 1.

Comparison of demographic characteristics between depression (PHQ-9 ≥ 10) and non-depression (PHQ-9 < 10) groups

Variables	Non-depression (n = 671)	Depression (n = 159)	P-value
Maternal age (years), median (IQR)	28 (8)	28 (8)	0.80a
Age group ≤ 20 / 20–40 / ≥40, n (%)	46 (6.90%) / 608 (90.60%) / 15 (2.20%)	7 (4.40%) / 148 (93.10%) / 4 (2.50%)	0.77b
Gravidity, median (IQR)	2 (2)	2 (2)	0.01 a
Parity, median (IQR)	2 (1)	2 (1)	0.01 a
Pregnancy losses, median (IQR)	0 (0)	0 (0)	0.81a
Living children, median (IQR)	2 (1)	2 (1)	0.01 a
BMI (kg/m²), median (IQR)	28.50 (6.40)	28.30 (6)	0.23a
Planned pregnancy (No / Yes)	47 (7%) / 623 (92.80%)	11 (6.90%) / 148 (93.10%)	0.88b
Working status (No / Yes)	515 (76.80%) / 156 (23.20%)	138 (86.80%) / 21 (13.20%)	0.005 b
Spouse working status (No / Yes)	33 (4.90%) / 638 (95.10%)	10 (6.30%) / 149 (93.70%)	0.48b
Duration of marriage (years), median (IQR)	4 (5)	4 (5)	0.49a

PHQ-9 Patient Health Questionnaire-9, IQR Interquartile range, BMI Body mass index. Continuous variables are presented as median (interquartile range) and compared using the Mann–Whitney U test^a due to non-normal distribution. Categorical variables are presented as number (percentage) and compared using the chi-square test^b. A p-value < 0.05 was considered statistically significant

Obstetric, delivery, and neonatal characteristics were evaluated for their potential contribution to postpartum depression. Gestational age at delivery, mode of delivery, birth weight, Apgar scores, and neonatal sex did not differ significantly between groups, suggesting that these perinatal parameters did not influence depression risk. Although neonatal intensive care unit admission occurred more frequently among women with PPD, this difference did not reach statistical significance. In contrast, breastfeeding status differed significantly between the groups: women who were unable to breastfeed had higher rates of depressive symptoms (Table 2).

Table 2.

Comparison of Obstetric, Delivery, and neonatal outcomes between depression (PHQ-9 ≥ 10) and non-depression (PHQ-9 < 10) groups

Variables	Non-depression (n = 671)	Depression (n = 159)	P-value
Gestational age at delivery (weeks), median (IQR)	39 (1)	39 (1)	0.09a
Birth weight (g), median (IQR)	3210 (540)	3200 (570)	0.21a
PHQ-9 score, median (IQR)	5 (3)	12 (3)	< 0.001 a
Smoking, n (%)	91 (13.60%)	30 (18.90%)	0.08b
Mode of delivery (NVD / Cesarean), n (%)	431 (64.20%) / 240 (35.80%)	103 (64.80%) / 56 (35.20%)	0.89b
Delivery type (NVD / Elective C/S / Emergency C/S), n (%)	431 (64.23%) / 135 (20.12%) / 105 (15.65%)	102 (64.20%) / 33 (20.80%) / 24 (15.10%)	0.96b
Apgar score at 1 min, median (IQR)	9 (0)	9 (0)	0.44a
Apgar score at 5 min, median (IQR)	10 (0)	10 (0)	0.37a
Baby sex (Girl / Boy), n (%)	346 (51.60%) / 325 (48.40%)	83 (52.20%) / 76 (47.80%)	0.88b
Breastfeeding (No / Yes), n (%)	71 (10.60%) / 600 (89.40%)	27 (17%) / 132 (83%)	0.02 b
NICU admission, n (%)	92 (13.70%)	31 (19.50%)	0.16b

PHQ-9 Patient Health Questionnaire-9, NVD Normal vaginal delivery, NICU Neonatal intensive care unit. Continuous variables are presented as median (interquartile range) and were compared using the Mann–Whitney U test^a due to non-normal distribution. Categorical variables are presented as number (percentage) and were compared using the chi-square test^b. A p-value < 0.05 was considered statistically significant

Maternal comorbidities and pregnancy complications—including diabetes, anemia, thyroid disorders, hypertensive disorders of pregnancy, cholestasis, PPROM, and IUGR—did not differ significantly between groups. In contrast, preterm labor was significantly more common among women with PPD. Additionally, a history of pre-gestational psychiatric illness was markedly more frequent in the depression group (Table 3).

Table 3.

Comparison of maternal comorbidities, fetal complications, and medication use between depression (PHQ-9 ≥ 10) and non-depression (PHQ-9 < 10) groups

Variables	Non-depression group (n = 671)	Depression group (n = 159)	P-value
Diabetes (No / Yes), n (%)	599 (89.30%) / 72 (10.70%)	139 (87.40%) / 20 (12.60%)	0.50
Anemia (No / Yes), n (%)	535 (79.70%) / 136 (20.30%)	117 (73.60%) / 42 (26.40%)	0.09
Hypothyroidism (No / Yes), n (%)	607 (90.50%) / 64 (9.50%)	147 (92.50%) / 12 (7.50%)	0.43
Hyperthyroidism (No / Yes), n (%)	667(99.40%) / 4 (0.60%)	158 (99.40%) / 1 (0.60%)	0.96
Hypertensive disorders of pregnancy (No / Yes), n (%)	627 (93.40%) / 44 (6.60%)	146 (91.80%) / 13 (8.20%)	0.46
IUGR (No / Yes), n (%)	632 (94.20%) / 39 (5.80%)	149 (93.70%) / 10 (6.30%)	0.81
Cholestasis (No / Yes), n (%)	663 (98.80%) / 8 (1.20%)	159 (100%) / 0 (0.00%)	0.16
PPROM (No / Yes), n (%)	577 (86%) / 94 (14%)	140 (88.10%) / 19 (11.90%)	0.49
Preterm labor (No / Yes), n (%)	641 (95.50%) / 30 (4.50%)	144 (90.60%) / 15 (9.40%)	0.01
Pre-gestational psychiatric disorder (No / Yes), n (%)	666 (99.30%) / 5 (0.70%)	152 (95.60%) / 7 (4.40%)	0.001
Medication use during pregnancy (No / Yes), n (%)	527 (78.50%) / 144 (21.50%)	125 (78.60%) / 34 (21.40%)	0.98

IUGR Intrauterine growth restriction, PPROM Prelabor preterm rupture of membranes. Categorical variables are presented as number (percentage) and were compared using the chi-square test or Fisher’s exact test, as appropriate. A p-value < 0.05 was considered statistically significant

Functional impairment demonstrated the most striking difference between groups. Women with PPD reported significantly higher levels of functional impairment, whereas the majority of women without depression reported no limitations. Educational status, however, did not differ significantly between groups (Tables 4 and 5).

Table 4.

Comparison of functional impairment level (assessed by PHQ-9 item 10) between depression (PHQ-9 ≥ 10) and non-depression (PHQ-9 < 10) groups

Functional impairment level	Non-depression (n = 671)	Depression (n = 159)	P-value
None, n (%)	628 (93.59%)	93 (58.50%)
Quite difficult, n (%)	40 (5.96%)	58 (36.50%)
Very difficult, n (%)	3 (0.45%)	5 (3.10%)
Extremely difficult, n (%)	0 (0.00%)	3 (1.90%)	< 0.001

PHQ-9 Patient Health Questionnaire-9. Functional impairment was evaluated using item 10 of the PHQ-9, which assesses the degree of difficulty in daily functioning related to depressive symptoms (work, home responsibilities, and interpersonal relationships). Categorical variables are presented as number (percentage) and were compared using the chi-square test or Fisher’s exact test, as appropriate. A p-value < 0.05 was considered statistically significant

Table 5.

Comparison of educational status between depression (PHQ-9 ≥ 10) and non-depression (PHQ-9 < 10) groups

Educational status	Non-depression (n = 671)	Depression (n = 159)	P-value
Illiterate, n (%)	22 (3.30%)	3 (1.90%)
Primary school, n (%)	83 (12.40%)	18 (11.30%)
Middle school, n (%)	114 (17.00%)	35 (22.00%)
High school, n (%)	215 (32.00%)	52 (32.70%)
Associate degree, n (%)	74 (11.00%)	17 (10.70%)
Bachelor’s degree, n (%)	150 (22.40%)	30 (18.90%)
Master’s degree, n (%)	13 (1.90%)	4 (2.50%)
P-value (overall)			0.71

PHQ-9 Patient Health Questionnaire-9. Educational status was treated as a categorical variable. Data are presented as number (percentage). Group comparisons were performed using the chi-square test. A p-value < 0.05 was considered statistically significant

Multivariate logistic regression analysis identified three independent predictors of PPD: level of functional impairment, pre-gestational psychiatric disorder, and employment status. Functional impairment was the strongest independent predictor, increasing the likelihood of PPD nearly eightfold. A history of psychiatric illness increased the odds more than fourfold, and being unemployed nearly doubled the odds compared with being employed. No other maternal, obstetric, or neonatal variables remained significant after adjustment (Table 6).

Table 6.

Multivariate logistic regression analysis evaluating the independent determinants of postpartum depression (PHQ-9 ≥ 10)

Variable	OR	95% CI	P-value
Pre-gestational psychological disorder	4.53	1.18–17.42	0.028
Mode of delivery (cesarean)	0.93	0.62–1.40	0.721
Parity	1.09	0.93–1.29	0.260
Preterm labor	1.49	0.69–3.23	0.306
Functional impairment score	8.06	5.20–12.50	< 0.001
Breastfeeding (Yes)	0.66	0.37–1.16	0.150
Working status (No)	1.99	1.22–3.26	0.006

OR odds ratio, CI confidence interval, PHQ-9 Patient Health Questionnaire-9. “Breastfeeding (Yes)” indicates current breastfeeding; “Working status (No)” indicates unemployment. Variables were entered into the multivariate logistic regression model a priori based on clinical relevance. A p-value < 0.05 was considered statistically significant

Discussion

In our study, the prevalence of postpartum depression was found to be 19.2%, a rate that is highly consistent with the prevalence range reported in the international literature [14, 18]. This consistency with international literature reinforces the reliability and representativeness of our findings. The study adds to the growing body of evidence emphasizing that PPD remains a significant public health concern, affecting nearly one in five women during the early postpartum period. By identifying key predictors—including functional impairment, pre-gestational psychiatric history, and employment status—our results contribute valuable population-specific insights into the determinants and functional consequences of postpartum depressive symptoms.

Although several variables showed statistically significant differences between the depression and non-depression groups in the initial comparative analyses, these factors did not remain clinically or statistically significant in the logistic regression model. This indicates that certain characteristics—such as higher gravidity, parity, number of living children, preterm birth, or not breastfeeding—may differ between groups at a descriptive level, yet they do not exert an independent effect on postpartum depression when adjusted for stronger predictors such as functional impairment, pre-gestational psychiatric history, and employment status. These findings suggest that the observed group-level differences are likely attributable to confounding influences or indirect pathways rather than representing true independent determinants of postpartum depressive symptoms.

Our findings show substantial concordance with the comprehensive systematic review by Yaqoob et al. Both studies identified a history of pre-gestational psychiatric illness as one of the strongest and most consistent predictors of postpartum depression. Similarly, psychosocial vulnerability factors demonstrated parallel patterns: while maternal unemployment was associated with depression in our cohort, Yaqoob et al. emphasized the importance of low social support, increased domestic workload, and psychosocial stress. The relationship between breastfeeding and depressive symptoms also showed a comparable direction; although women with postpartum depression in our study were more likely not to breastfeed, breastfeeding status did not remain an independent determinant in the multivariate analysis. Nevertheless, the bidirectional interaction model described in the literature—where not breastfeeding may exacerbate depressive symptoms and existing depressive symptoms may hinder the initiation or continuation of breastfeeding—is consistent with our findings. Additionally, preterm birth was more common among women with PPD in both studies, but it did not remain an independent predictor in multivariate models, a result that aligns with the moderate level of evidence reported in the review [19].

In contrast, certain factors highlighted in the systematic review were not confirmed in our population. Notably, unintended pregnancy, premenstrual syndrome, and domestic or partner violence—frequently reported as strong predictors in the global literature—were either not associated with PPD in our cohort or were not evaluated in the present study. Similarly, medical comorbidities such as gestational diabetes, anemia, and thyroid disorders, which have been suggested as potential risk factors in earlier research, showed no meaningful association in our cohort. Moreover, unlike the modest association between cesarean delivery and PPD described in the review, mode of delivery was not related to depressive symptoms in our sample. This finding may be explained by the unique obstetric context in Turkey, where cesarean delivery is highly prevalent and has become a common and normalized mode of birth. A recent study using the Robson Ten-Group Classification System demonstrated that cesarean section rates in Turkey are high across nearly all obstetric risk groups—including low- and intermediate-risk women—and are largely driven by institutional practices and repeat cesarean policies rather than emergency obstetric indications [20]. In such a setting, cesarean delivery may not represent an unexpected or psychologically distressing event for women, which could attenuate its association with postpartum depressive symptoms in this population.

Finally, our study contributes two noteworthy findings that are less extensively addressed in previous reviews: functional impairment and maternal unemployment emerged as strong independent determinants of PPD. While functional impairment has traditionally been viewed as a consequence of depressive symptoms, our results suggest that it may also serve as an important concurrent predictor. Similarly, maternal unemployment, independent of general socioeconomic status, may be associated with reduced social engagement and greater vulnerability. In culturally conservative or socially restrictive settings like ours, employment often represents a key pathway to social interaction and community participation; thus, identifying maternal unemployment as a predictor of PPD can be considered an expected finding within this sociocultural context. These differences underscore the importance of considering population-specific dynamics when evaluating predictors of postpartum depression.

Postpartum depression is shaped by the interplay of biological, psychosocial, and behavioral mechanisms. The rapid decline in estrogen and progesterone levels immediately after delivery can influence serotonergic activity and lead to marked fluctuations in mood [2]. Additionally, common postpartum challenges such as sleep disturbances, fatigue, physical recovery, and the metabolic demands of breastfeeding may further heighten hormonal sensitivity to depressive symptoms. From a psychosocial perspective, adaptation to motherhood, increased caregiving responsibilities, and the level of social support are important determinants [19]. Functional impairment may contribute through behavioral pathways, as limitations in daily activities, reduced social engagement, and diminished self-efficacy can increase vulnerability to depressive symptoms. Some obstetric or demographic variables losing significance in the multivariate model is likely due to the inclusion of stronger biopsychosocial predictors. This reflects the multidimensional nature of postpartum depression and the potential confounding effects among contributing variables. The absence of significant associations between PPD and demographic or obstetric characteristics suggests that psychosocial and functional factors play a more central role in postpartum vulnerability. This finding highlights the importance of universal screening rather than selective screening based on medical risk factors.

This study has several notable strengths that enhance the validity and clinical relevance of its findings. First, the large sample size of 830 postpartum women provides robust statistical power and increases the representativeness of the study population. Second, the use of the validated Turkish version of the PHQ-9 ensures reliable and standardized assessment of depressive symptoms, allowing for meaningful comparison with international research. Third, data were collected during a well-defined postpartum interval (days 7–21), minimizing temporal variability and capturing depressive symptoms during a clinically important period. The face-to-face administration of questionnaires in a private setting further strengthened data quality by reducing misunderstandings and ensuring completeness. Additionally, the study comprehensively evaluated a wide range of sociodemographic, obstetric, neonatal, and psychosocial variables, allowing for an in-depth exploration of potential risk factors. Importantly, the inclusion of a multivariate logistic regression model enabled the identification of truly independent predictors, eliminating the influence of confounding variables. Together, these design features contribute to the methodological rigor of the study and support the reliability and clinical applicability of its conclusions.

This study has several limitations. A potential limitation is that functional impairment was assessed using item 10 of the PHQ-9, which also contributes to the total PHQ-9 score; therefore, a degree of item overlap cannot be entirely excluded. First, its cross-sectional design prevents establishing causal relationships. Second, as a single-center study, the findings may not fully reflect populations with different cultural or socioeconomic backgrounds. Third, several important psychosocial variables—such as social support, intimate partner violence, premenstrual syndrome, and life stress—were not assessed, which may have limited the ability to capture all relevant risk factors. The postpartum assessment window (7–21 days) may also have missed cases of depression that emerge later in the postpartum period. Additionally, the low prevalence of some obstetric complications may have reduced the statistical power to detect meaningful associations. Finally, the influence of unmeasured confounders cannot be excluded.

Our findings offer several important clinical implications for the early identification and management of postpartum depression. Women who exhibit any of the three independent predictors—functional impairment, a history of pre-gestational psychiatric illness, or maternal unemployment—should be considered priority groups for targeted screening and closer follow-up during the postpartum period. Notably, while the employment status of the spouse did not influence depression risk, maternal unemployment emerged as a significant determinant; this suggests that the mother’s engagement in daily life, social interactions, and personal activities may play a protective role against depressive symptoms. Furthermore, the bidirectional relationship between breastfeeding and postpartum depression requires careful clinical attention, as difficulty initiating or maintaining breastfeeding may increase susceptibility to depressive symptoms, while existing depressive symptoms may impair breastfeeding ability. Understanding this reciprocal interaction is critical for implementing timely interventions such as breastfeeding counseling and psychosocial support.

One of the most feared complications of postpartum depression is suicide. Although completed suicide rates during the postpartum period are generally lower than those observed in the broader female population, PPD is associated with increased suicidal ideation and self-harm thoughts [21]. Therefore, the management of PPD should include both interventions aimed at preventing suicide and self-harm ideation and a comprehensive care approach that protects maternal functioning, mother–infant bonding, and overall family well-being. Taken together, these findings highlight the need to integrate routine mental health screening, functional assessment, and individualized support strategies into postpartum care to reduce the burden of untreated depression on mothers, infants, and families.

Conclusion

Postpartum depression remains a significant public health concern with important implications for mothers, infants, families, and society. In accordance with current clinical guidelines, universal screening is essential, and women who exhibit the independent predictors identified in our study—functional impairment, a history of pre-gestational psychiatric illness, and maternal unemployment—should receive closer monitoring throughout the postpartum period. Health-care professionals should maintain heightened awareness of early depressive symptoms and adopt a proactive, multidisciplinary, and family-centered approach to ensure timely intervention and prevent long-term adverse outcomes.

Authors’ contributions

Conceptualization: Mehmet Can Keven, Ebru Yücel, Özer Birge, Atakan TanaçanData acquisition: Mehmet Can Keven, Melike Savaş, Ece Akça Salık, Seda Yıldız, Banu Derim Yeğen, Halime Kılıç, Sevgi Güleç, Keziban Saylam Dönmez, Kıymet Çiçek, Ayşe Yılmaz Altay, Eylül Bakır, Alperen KaradağData analysis: Mehmet Can Keven, Ebru Yücel, Atakan TanaçanData interpretation: Mehmet Can Keven, Ebru Yücel, Özer Birge, Nadi Keskin, Ayşe Gülçin BaştemurDrafting: Mehmet Can Keven, Ebru YücelRevision: Ebru Yücel, Özer Birge, Nadi Keskin, Zafer Bütün, Ayşe Gülçin Baştemur.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Data availability

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

Declarations

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Ethical statement

This study was conducted in accordance with the principles of the Declaration of Helsinki (revised 2013). The study protocol was approved by the Ethics Committee of the Republic of Turkey Ministry of Health, Eskişehir City Hospital (approval date: June 15, 2025; decision number: ESH/BAEK 2025/190). Written informed consent was obtained from all participants prior to data collection during routine postpartum visits.

Consent to participate

Written informed consent was obtained from all individual participants included in the study.