Mothers at risk of postpartum depression in Sri Lanka: A population-based study using a validated screening tool.

BACKGROUND: Postpartum depression is an important public health concern. The prevalence of postpartum depression is estimated to be 18% worldwide. The purpose of this study was to estimate the prevalence of mothers at risk of postpartum depression in Sri Lanka and to investigate its associated risk factors.
METHODS: This was a cross-sectional study conducted among 975 mothers in Galle district, Sri Lanka. The prevalence of mothers at risk of postpartum depression was assessed using the Edinburgh Postpartum Depression scale (EPDS) which has been validated for screening for mothers at risk of postpartum depression in Sri Lanka with a cut-off score 9 or more. Prevalence was estimated using a cut-off 9 or more, 10 or more, 11 or more and 12 or more to assess the difference in prevalence using unvalidated cut-offs for screening. Data from routine records on pregnancy, delivery and postnatal care was collected to investigate possible predictors of EPDS score 9 or more (risk of postpartum depression). Univariate and multivariable logistic regressions were performed to identify risk factors for EPDS score 9 or more (risk of postpartum depression).
RESULTS: The prevalence of mothers with EPDS score 9 or more was found to be 9.4% (95%CI: 7.8-11.4); EPDS score 10 or more was 5.6% (95%CI: 4.4-7.3). EPDS score 9 or more (risk of postpartum depression) was associated with the following risk factors: Former history of mental illness (aOR 32.9, 95%CI: 7.9-136.2), high maternal age 30-39 (aOR 2.2, 95%CI: 1.3-3.8), BMI 25.0-29.9 (aOR 2.6, 95%CI: 1.5-4.5), hypertension (aOR 3.6, 95%CI: 1.2-10.9) and newborn death (aOR 28.9, 95%CI: 4.5-185.1). One in five women reported thoughts of self-harm.
CONCLUSION: Around one in ten mothers in Sri Lanka experience symptoms of postpartum depression, highest risk among mothers who reported former history of mental illness and newborn death. The prevalence estimates were lower with a higher cut-off for screening and this highlights the importance of using the validated cut-off for screening in future studies on postpartum depression in Sri Lanka. Mothers at increased risk should be identified in antenatal care and are important targets of referral.

Introduction

Depression after giving birth is considered a serious public health problem worldwide, and the global prevalence is estimated to be 18% [1]. Studies suggest that postpartum depression (PPD) not only affects the wellbeing of postpartum women, but also the mother-infant interaction [2]. There is a compelling body of evidence that PPD may influence the social, cognitive and emotional development of the child and infant growth [2].

PPD and its associated risk factors are well studied in high-income countries but data from low- and middle-income countries is scarce [3]. However, research suggests there is a higher prevalence of PPD in low- and middle-income countries compared to in high-income countries [3], which aligns well with low socioeconomic status as a known risk factor [2]. It is widely recognized that PPD is not attributable to one single cause, but rather influenced by complex interactions between environmental, psychosocial and biological risk factors [2]. Several risk factors associated with PPD have been investigated in the past; mothers with a former history of mental illness seem to be at higher risk of PPD [4], while maternal age as a risk factor have shown conflicting results [5].

Screening for and managing PPD is included in The Global Strategy for women’s, children’s and adolescent’s health (2016–2030) by the World Health Organization [6]. Several tools to screen for PPD have been used, and this is reflected in the diversity in reported prevalence in different countries [5]. The Edinburgh Postnatal Depression Scale (EPDS) is a 10-item questionnaire and the most widely used tool globally to screen for mothers at risk of PPD [2]. Each answer is scored 0–3 and a higher sum of scores indicates more symptoms of depression.

Originally, the EPDS was developed for screening for PPD in English-speaking women and a cut-off 10 or more for “possible depression” and a cut-off 13 or more for “probable depression” was suggested [7]. Since then, the EPDS has been translated to and validated in many languages using different cut-offs for screening in different countries. It is important to emphasize that most studies on PPD do not report on the real prevalence of PPD, but rather the prevalence of mothers at risk of PPD screened with the EPDS. In 2006, Matthey et al. addressed the increasing use of unvalidated EPDS cut-off scores in studies on PPD [8]. Furthermore, he stated that a difference of just one point in cut-off score can have a substantial impact on estimates of mothers at risk of PPD [8]. Optimal psychometric properties are important for assessing prevalence rates and the overall burden of PPD in countries [8].

A potentially alarming symptom of PPD are thoughts of self-harm, which can be identified by the EPDS item 10 “the thought of harming myself has occurred to me”. Thoughts of self-harm in pregnancy is not uncommon [9]; studies suggest that attention should be given to mothers with thoughts of self-harm as it can indicate more severe depressive symptoms and therefore increased risk of negative outcomes for mother and child [9]. Furthermore, postpartum mothers who reported thoughts of self-harm had increased risk of somatic and psychiatric disease seven years after delivery, which could not be fully attributed to depression [10].

The reported prevalence of PPD in South-Asia has ranged from 5–49% [1]. Sri Lanka is a lower middle-income country and has been considered a role model for maternal health promotion and for reducing maternal deaths. In Sri Lanka the Sinhala version of the EPDS was introduced and included as a part of postnatal care in 2012 [11] and it has been validated for screening for mothers at risk of PPD with a cut-off 9 or more [12]. In 2011, a cross-sectional study representing eight provinces in Sri Lanka estimated that 27.1% of the mothers had an EPDS score 10 or more [13]. A recent study from 2017 in Galle, Sri Lanka, found that 7.8% of the mothers had an EPDS score 10 or more [14]. To this date there are no published studies available on the prevalence of mothers at risk of PPD in Sri Lanka using the validated cut-off for screening EPDS score 9 or more. Using a higher cut-off may lead to an underestimation of mothers at risk of PPD in Sri Lanka. Therefore, in our study we aimed to: 1) estimate the prevalence of mothers at risk of PPD at postnatal clinics in Galle using an EPDS cut-off 9 or more 2) evaluate the difference on prevalence estimates of mothers at risk of PPD using several cut-offs 3) investigate risk factors associated with EPDS score 9 or more (risk of PPD) among mothers in Sri Lanka.

Methods

Study design and study setting

A cross-sectional study was conducted among mothers who had delivered from 1st January 2019 to 31st December 2019 in Bope-Poddala, Medical Offices of Health (MOH) Division, Galle district, Sri Lanka. With a population of over 71 000, Bope-Poddala is a semi urban MOH division in Galle district. All three main population groups of Sri Lanka live there: Sinhalese, Tamils and Muslims. Here preventive care services are provided through 20 divisions of the MOH. Antenatal and postnatal care is provided to the mothers through five maternal and child health field clinics. In Sri Lanka the antenatal care attendance is very high, and nearly 100% of deliveries occur in health facilities. Maternal mental health services are provided during regular visits in the clinics by a medical officer of mental health.

Participants

In Bope-Poddala, around one thousand births take place each calendar year. The sample size for the postpartum mothers to be included in the study was estimated using OpenEpi.com calculator [15]. Assuming that the prevalence of mothers at risk of PPD in Galle would be around 18% [1] and aiming at estimating the proportion with a precision of +/- 2.5%, the required sample size for our study was estimated to be n = 907. Therefore, all the postpartum mothers who delivered from 1st January 2019 to 31st December 2019 were included in the study. Mothers with incomplete EPDS records were excluded from the study.

Study tool

In Sri Lanka, all mothers are routinely screened once at the clinics for depressive symptoms four to six weeks following delivery using the Sinhala or Tamil version of Edinburgh Postpartum Depression Scale (EPDS). An EPDS score 9 or more has been validated for screening for mothers at risk of PPD with the Sinhala version of the EPDS, with a sensitivity and specificity of 89.9% and 78.9% respectively [12]. The Tamil version of the EPDS is used for screening in the Tamil population and has been validated for screening in Tamil Nadu, India using the same value, EPDS score 9 or more, with a sensitivity of 94.1% and a specificity of 90.2% [16].

In current practice, the EPDS is filled on paper forms by the mothers at their postnatal visit at the clinic and is then collected and kept by the midwives. If the mother cannot read Tamil or Sinhala the EPDS is completed by the midwife through an interview with the mother. Routinely, mothers with an EPDS score 9 or more are referred to a medical officer of health at the clinics who then refer the mother to their designated medical officer of mental health (MOMH). MOMHs manage some of the cases, and if necessary, refer the mother to psychiatrist care at the nearest hospital. In severe cases the medical officer of health refers directly to a consultant psychiatrist. MOMHs and psychiatrists diagnose mothers with PPD using the ICD-10 classification for diagnosis, usually through clinical interview with the mother.

Data collection

All postpartum mothers have EPDS records and routine records on pregnancy, delivery and postnatal care in the B part of the pregnancy record which are kept in the clinics. For our study, determinants were collected as quantitative clinical data from these records into a data extraction sheet by a trained research assistant. The data extraction sheet included sociodemographic, obstetric and baby related variables. The collected data was translated to English and double-entered into the electronic database EpiData 3.1. After ensuring that data from various sources were linked to the same person, the names and identifiers were removed. In Sri Lanka, pregnancy outside marriage is very rare and thus there is no place for marital status in the pregnancy record.

Outcome measures and exposure variables

The main outcome measurement was rate of mothers with EPDS score 9 or more. Exposure variables were: maternal age (under 20, 20–29, 30–39 and 40 and more); educational level (completed secondary school ordinary level or below, completed above secondary school ordinary level); occupation (housewife or non-housewife); socioeconomic status (calculated according to the occupation of the mother and the husband, and categorized into SE 1–2 and SE 3–5); number of living children (0, 1, 2–3 and >4); former history of mental illness (yes/no); maternal Body Mass Index (BMI) (less than 18.5, 18.5–24.9, 25.0–29.9 and 30 and more); hypertension during pregnancy (yes/no); diabetes during pregnancy (yes/no); history of subfertility (yes/no); death of newborn defined as death of liveborn within 28 days (yes/no); gender of the baby (female or male); congenital abnormalities (yes/no); birthweight (less than 2500g, 2500g or more); establishment of breastfeeding (yes/no); other complications from delivery notes (yes/no).

Data analysis

The database was exported into IBM SPSS 26 version for statistical analysis. The prevalence of mothers at risk of PPD was calculated as percentages of mothers with an EPDS score 9 or more. One-way ANOVA was performed to compare the mean EPDS score for each independent variable. Possible predictors of EPDS score 9 or more (risk of PPD) were examined in univariate and multivariable models. In the final model we adjusted with maternal age, BMI, former history of mental illness, hypertension, history of subfertility and newborn death. Odds ratios (OR) and adjusted Odds ratios (aOR) with 95% confidence intervals (95%CI) were calculated. The prevalence of mothers reporting thoughts of self-harm were calculated as percentages of mothers reporting “never”, “hardly ever”, “sometimes” and “many times” on the EPDS item 10; mean EPDS score was calculated for these four categories.

Ethical issues

Ethical clearance for the present study was obtained by Ethics Review Committee, Faculty of Medicine, University of Ruhuna and by REK, Regional Committees for Medical and Health Research Ethics. Informed consent was not required as the EPDS was used as routine screening service and introduced into antenatal clinics several years before. The data archive had no names or identifiers.

Results

Included in the study were 975 postpartum mothers. All mothers had complete EPDS records and no mothers were excluded. Among the respondents the mean age was 29 years, ranging from 16 to 45 years. The majority of the mothers (70.5%) were housewives, and half of the mothers (48.1%) had completed education above ordinary level examination secondary school. Former history of mental illness was identified in 12 women (1.2%).

Table 1 shows the EPDS scores among the 975 women studied; 9.4% (n = 92) had an EPDS score 9 or more; and 5.6% (n = 55) had a EPDS score 10 or more. An EPDS score 12 or more was seen in 2.1% (n = 20) of the participants.

Table 1

Frequency distribution of the EPDS scores with sensitivity analysis among mothers in Galle, 2019.

EPDS	Frequency (n)	Percent (%)	95%CI
Total	975	100	-
EPDS score 9 or more	92	9.4	7.8–11.4
EPDS score 10 or more	55	5.6	4.4–7.3
EPDS score 11 or more	29	3.0	2.1–4.2
EPDS score 12 or more	20	2.1	1.3–3.1

Association of postpartum depression with risk factors

Table 2 shows mean EPDS score by exposure variables and corresponding logistic regression analysis of these determinants of EPDS score 9 or more (risk of PPD). Compared to mothers aged 20–29, those aged 30–39 had twice the risk of EPDS score 9 or more (aOR 2.2, 95%CI: 1.3–3.8). Mothers who had experienced newborn death had a higher risk of EPDS score 9 or more (aOR 28.9, 95%CI: 4.5–185.1), and mothers who reported a former history of mental illness had a highly increased risk of EPDS score 9 or more (aOR 32.9, 95%CI: 7.9–136.2). Mothers with a BMI <18.5 (aOR 3.0, 95%CI: 1.4–6.3) and BMI 25.0–29.9 (aOR 2.6, 95%CI: 1.5–4.5) had a higher risk of EPDS score 9 or more than those with normal BMI. Those who had hypertension were 3.6 times more likely (aOR 3.6, 95%CI: 1.2–10.9) to develop EPDS score 9 or more as compared to mothers without hypertension.

Table 2

Determinants of EPDS score 9 or more (risk of PPD) among mothers in Galle, 2019.

Variables		Total	EPDS 9 or more	Mean EPDS (95% CI)	OR (95% CI)	aOR (95% CI)
Maternal age
	≤19	52	2	4.6 (4.0–5.2)	0.6 (0.1–2.6)	0.8 (0.2–3.7)
	20–29	459	29	3.8 (3.6–4.1)	1	1
	30–39	447	59	4.6 (4.3–4.9)	2.3 (1.4–3.6)	2.2 (1.3–3.8)
	≥40	17	2	4.2 (2.6–5.9)	2.0 (0.4–9.0)	1.0 (0.1–8.6)
Education
	≤Ordinary level	505	46	4.4 (4.1–4.6)	1	1
	>Ordinary level	469	46	4.1 (3.8–4.4)	1.1 (0.7–1.7)	1.1 (0.6–1.7)
	Missing	1
Occupation
	Housewife	687	68	4.3 (4.1–4.6)	1	1
	Non-housewife	288	24	4.0 (3.7–4.3)	0.8 (0.5–1.3)	0.7 (0.4–1.2)
Socioeconomic class
	1–2	194	13	3.6 (3.3–4.0)	0.6 (0.3–1.2)	0.6 (0.3–1.2)
	3–5	781	79	4.4 (4.2–4.6)	1	1
Number of living children
	0	0	0	--	-	-
	1	428	34	4.1 (3.8–4.4)	1	1
	2–3	489	52	4.3 (4.0–4.5)	1.4 (0.9–2.2)	0.8 (0.4–1.3)
	>4	57	6	4.5 (3.9–5.2)	1.4 (0.5–3.4)	0.5 (0.2–1.4)
	Missing	1
History of mental illness
	Yes	12	9	11.0 (8.1–13.9)	31.8 (8.4–119.8)	32.9 (7.9–136.2)
	No	963	83	4.1 (4.0–4.3)	1	1
BMI
	≤18.4	119	13	4.4 (3.8–5.0)	2.0 (1.0–4.0)	3.0 (1.4–6.3)
	18.5–24.9	462	27	4.0 (3.7–4.2)	1	1
	25.0–29.9	290	43	4.6 (4.2–4.9)	2.8 (1.7–4.7)	2.6 (1.5–4.5)
	≥30	57	3	4.1 (3.4–4.9)	0.9 (0.3–3.1)	0.8 (0.2–2.8)
	Missing	47
Hypertension
	Yes	17	6	6.3 (4.4–8.1)	5.5 (2.0–15.3)	3.6 (1.2–10.9)
	No	958	86	4.2 (4.0–4.4)	1	1
Diabetes
	Yes	54	8	5.0 (4.0–5.9)	1.7 (0.8–3.8)	0.7 (0.2–1.9)
	No	921	84	4.2 (4.0–4.4)	1	1
History of subfertility
	Yes	39	7	5.3 (4.0–6.5)	2.2 (0.9–5.1)	1.0 (0.4–2.8)
	No	936	85	4.2 (4.0–4.4)	1	1
Newborn death
	Yes	6	4	9.0 (4.6–13.4)	20.0 (3.6–110.9)	28.9 (4.5–185.1)
	No	969	88	4.2 (4.0–4.4)	1	1
Baby gender
	Male	487	41	4.2 (3.9–4.4)	0.8 (0.5–1.2)	0.7 (0.4–1.2)
	Female	487	51	4.3 (4.0–4.6)	1	1
	Missing	1
Congenital abnormalities
	Yes	8	2	6.0 (2.1–9.9)	3.2 (0.6–16.3)	1.3 (0.1–11.8)
	No	966	90	4.2 (4.0–4.4)	1	1
	Missing	1
Birthweight
	<2500	89	6	4.3 (3.7–5.0)	0.7 (0.3–1.6)	0.5 (0.2–1.3)
	≥2500	886	86	4.2 (4.0–4.4)	1	1
Breast feeding
	Yes	429	35	4.2 (4.0–4.5)	0.8 (0.5–1.2)	0.6 (0.4–1.1)
	No	546	57	4.2 (4.0–4.5)	1	1
Delivery complication
	Yes	40	4	4.6 (3.4–5.8)	1.1 (0.4–3.1)	0.5 (0.2–2.0)
	No	934	88	4.2 (4.0–4.4)	1	1
	Missing	1
Baby birth complication
	Yes	23	2	5.0 (3.4–6.5)	0.9 (0.2–3.9)	0.1 (0.0–1.4)
	No	936	90	4.2 (4.0–4.4)	1	1
	Missing	16
Malpresentation
	Yes	45	1	4.1 (3.3–5.0)	0.2 (0.0–1.5)	0.1 (0.0–1.3)
	No	929	91	4.2 (4.0–4.4)	1	1
	Missing	1

Analysed by ANOVA with mean EPDS score, and univariate and multivariable logistic regressions with odds ratios. Regression analysis included these variables in the adjustment model: Maternal age, BMI, former history of mental illness, hypertension, history of subfertility and newborn death. Other variables were adjusted with the same model.

No significant difference was found between EPDS score 9 or more and obstetric factors. Baby gender and late breast-feeding initiation was not significantly associated with the development of EPDS score 9 or more. Risk of PPD defined as EPDS score 10 or more showed similar associations with the same variables (not included in the table).

Thoughts of self-harm

Table 3 shows responses to EPDS Item 10 about self-harm thoughts and the associated mean total EPDS score. Nearly 1 out of 5 (n = 188) study participants reported “the thought of harming myself has occurred to me”. Out of those 188 participants, 26.1% had an EPDS score 9 or more. We see that the more frequent such thoughts are reported, the more likely depressive symptoms occurred.

Table 3

EPDS question 10: “The thought of harming myself has occurred to me”.

Response	Frequency (n)	Percent (%)	Mean total EPDS (95%CI)
Total	975	100	-
Never	787	80.7	3.6 (3.4–3.8)
Hardly ever	174	17.8	6.6 (6.2–7.0)
Sometimes	14	1.4	8.3 (7.0–9.6)
Many times	0	0	-

Discussion

In this study in Sri Lanka, among 975 women screened with the EPDS four to six weeks following delivery, we found that 9.4% (95%CI: 7.8–11.4) had an EPDS score 9 or more, suggesting risk of PPD, which means they are referred to mental health services for further management and diagnosis of PPD. The factor that showed the strongest association with EPDS score 9 or more was former history of mental illness, which was not investigated separately in the former publication from 2017 [14]. Several maternal characteristics and factors were also associated with EPDS score 9 or more: high maternal age, overweight, underweight, hypertension and experiencing death of the newborn.

Various studies on PPD in Sri Lanka have shown a wide variation in prevalence, perhaps partly due to differences in methodology. In 2017, a similar study conducted in Galle, Sri Lanka reported that 7.8% mothers had an EPDS score 10 or more [14]; our study showed 5.6% (95%CI: 4.4–7.3) with this cut-off. Our data support the previous finding of low risk of PPD in Galle, Sri Lanka. However, we see that when measuring the prevalence rates of mothers at risk of PPD using a EPDS cut-off 9 or more compared to a cut-off 10 or more, a difference of 3.8% in prevalence estimates occurred. This supports the statement by Matthey et al. that a one point difference in EPDS cut-off score can have a substantial impact on PPD rate estimates and highlights the importance of using the validated cut-off for screening in Sri Lanka [8]. In human development report for Sri Lanka in 2012, Galle ranked between mid and top on health index and high on education index [17]; this may contribute to a low prevalence of PPD. Furthermore, the prevalence of mothers at risk of PPD is lower than findings in other countries in South-Asia. A validation study from Nepal showed a prevalence of PPD at 17.1%. It defined PPD as EPDS score 13 or more and had a sensitivity of 92% and a specificity of 95.6% [18]., A meta-analysis from India estimated a pooled prevalence of PPD at 22% [19]. Compared to the neighboring countries, Sri Lanka has made more progress on maternal health and child mortality indicators, as well as other indicators such as GDP per capita and education [20].

In our study, mothers aged 30–39 had twice the risk of EPDS score 9 or more compared to those aged 20–29. The same findings are reported in the paper from the same place in 2017 [14]. A higher percentage of EPDS score 9 or more in the age group 30–39 compared to other age groups was observed in both primiparous mothers and in multiparous mothers. Other sociodemographic risk factors such as educational level and being a housewife did not show significant associations with EPDS score 9 or more.

In terms of maternal BMI, underweight women (BMI<18.5) had a three times higher risk of EPDS score 9 or more compared to women of normal weight, which is consistent with findings in Nepal [21] and Sweden [22]. It is widely recognized that BMI is likely to be influenced by income [17] and after adjusting for only socioeconomic class, the association between underweight and EPDS score 9 or more was no longer significant. In Sri Lanka, overweight (BMI 25.0–29.9) is far more prevalent than underweight and was in our study associated with more than twice the risk of EPDS score 9 or more compared to mothers with normal bodyweight. One study suggests that overweight in Sri Lanka is associated with urban living and being in middle age [23].

We found that hypertension was associated with increased risk of EPDS score 9 or more. In our study, the mothers with hypertension were more often of advanced age, which reflects the prevalence of hypertension in the general population [24]. Also, hypertensive mothers had more gestational diabetes, also increasing with age. The possible relationship between preeclampsia and PPD has previously been described in studies from South Korea [25] and Netherlands [26]. In pre-eclampsia, the blood level of serotonin is known to increase [27] and previous studies have hypothesized that this may be linked to a decreased level of serotonin in the brain [26]. A possible role of alterations in serotonin function has been noted in the pathophysiology of general depression [28]. A further complication may be that hypertension in pregnancy is commonly treated with methyldopa, of which a known side effect is depressive symptoms [29]. However, hypertension and preeclampsia could not be differentiated from the pregnancy record in our data.

In our study mothers who had experienced death of a newborn were at increased risk of EPDS score 9 or more, although grief reactions may be difficult to distinguish from PPD symptoms. We also found that former history of mental illness predicts EPDS score 9 or more, which is consistent with the findings by O’Hara and Swain [4]. In our study, a diagnosis of mental illness before the pregnancy was included but not further defined. However, the number of mothers identified (n = 12) may not represent the total number of mothers who ever had a mental illness; perhaps those with a more recent diagnosis were more likely to report on this item but we do not have this information. In Sri Lanka, stigmatizing beliefs on mental health have been reported to affect help-seeking behavior [30]. As in many Asian cultures, emotional restraint and self-control are perceived as desirable qualities in women, and to seek counselling for mental health issues may be considered unacceptable by family members [30]. Our findings suggest that attendants reporting a past mental illness may be considered at risk of PPD, and may benefit from referral.

Thoughts of self-harm was reported by almost 20%, of whom the vast majority responded “hardly ever”. In a randomized controlled trial in the UK, it was found that the response “hardly ever” on EPDS item 10 about thoughts of self-harm in mothers was not concordant with suicidality as measured by a clinical interview [31]. A study published in 2019 which included 475 pregnant women in Anuradhapura, Sri Lanka reported rates on thoughts of self-harm at 5.9% including “hardly ever” [32], and in Goa, India, it was 14.3% [9, 33]. In Pittsburgh, USA, with a study population of 10 000 mothers, the observed rate on thoughts of self-harm four to six weeks postpartum was 3.2%, out of whom the majority had screen-positive findings on depression [34]. This is in contrast to our study, where only one out of four women with thoughts of self-harm also had an EPDS score 9 or more; however, we noticed that the more frequent the mothers’ self-harm thoughts were, the more signs of depression occurred, shown by increasing mean EPDS score. In Sri Lanka, mothers are usually not referred to a psychiatrist based on the response on EPDS item 10. However, research in Sri Lanka on self-harm behavior following delivery is limited. The high proportion of mother with thoughts of self-harm in our study highlights the need for more research in this field.

Our study had several strengths. The study population provides some confidence to the precision in our results, and missing data was minimal. The transfer of information in paper forms into digital data was validated by double data entry. We think the study gives a good representation of the situation for pregnant women in the area, as almost all pregnant women there attend ANC. Furthermore, our study is the first study since the EPDS was included as a part of postnatal care in 2012 to use the validated cut-off for screening when estimating mothers at risk of PPD in Sri Lanka.

The study also had some limitations. The cross-sectional study design can only analyze associations and not causation, but this is in line with our aims. Recall is generally associated with some uncertainty, especially about feelings in the past, but limiting to the past week may provide more credibility to the findings. Misinterpretation of the questions and social taboos may have led to response bias, but the local author did not see such obvious mistakes. The cut-off threshold for categorizing depression is a matter of discussion; we used a validated cut-off for screening for mothers at risk of PPD in Sri Lanka, and we also tested the association with other cut-offs as well (Table 1) and found similar associations (not shown in table). The tool we used was tested and validated in both Sinhala and Tamil language. Nonetheless, great caution must be taken when comparing the results of EPDS screening for PPD in different countries as there is no international standard for how to translate, nor validate, the EPDS in local languages [35].

Conclusion

The prevalence of mothers at risk of PPD in Galle, Sri Lanka (as estimated by an EPDS score 9 or more) was 9.4% (95%CI: 7.8–11.4). The prevalence rate was lower when using higher cut-offs for screening which highlights the importance of using the validated cut-off EPDS score 9 or more for future studies on PPD in Sri Lanka. The major predisposing factors for EPDS score 9 or more (risk of PPD) were former history of mental illness, high maternal age, overweight, hypertension and newborn death. Furthermore, the number of mothers reporting thoughts of self-harm was high. Mothers at increased risk of PPD are important targets of referral to specialist.

(SAV)

Click here for additional data file.

6 Sep 2021

Please address the comments mentioned in the additional comments section below:

​Please submit your revised manuscript by October 10, 2021. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Yeetey Akpe Kwesi Enuameh, MD, MSc, DrPH

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For more information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially sensitive information, data are owned by a third-party organization, etc.) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide

Additional Editor Comments (if provided):

The topic is of great relevance to maternal and child health... there are some issues that when addressed should enhance the outlook of the paper... Please address the comments of the reviewer...

1. The title of the current paper and that by Fan et al., 2020 are quite similar. Could you please come out clearly as to what differs between your paper and the earlier published one, more so when the study population seems to be the same.

2. Lines 17 & 40: The prevalence estimates provided here were 1996 figures... could there not have been much more recent ones? Also add the estimates from South Asia - as you compare it in the conclusion...

3. Lines 56 - 58: I am not clear about this point... what has this to do with PPD? I read over the cited article, did not come across a relationship between poverty and quality of parenting... please clarify...

4. Lines 92 - 93: Why the assumption based on 1996 figures? More so, when there was a 2017 study in the same population?

5. Lines 94 - 95: Why all mothers????? We the readers are not privy to information on deliveries in the study community... Also, the results section (line 159) has 975 persons... you have not clearly stated how was that arrived at... were those all the mothers interviewed? any mothers not included? response rate?

6. Line 98: Will suggest this section be structured as e.g., data collection methods, tools, and process.

7. Lines 106 - 107: check this.... specificity????

8. Lines 109 - 115: Is the information presented here related to data collection?

9. Line 114: Is the ICD used for classification or diagnosis?

10. Line 126: Label this as a sub-section e.g., variables considered.

11. Lines 159 - 162: Please make reference to table 2.

12. Lines 193 - 195: Revise the title to be much more concise.

13. Line 201: Mention "EPDS Question 10" in the "Methods section" as part of the data collection tools.

14. Lines 214 - 216: Were these characteristics also in the 2017 study? If so, state that clearly... if not, then provide the appropriate citation.

15. Finally, please follow the reporting guidelines for Cross-sectional studies as provided in the link to guide your paper: https://www.equator-network.org/reporting-guidelines/strobe/. Also review the PLOS ONE guidelines on publications: https://journals.plos.org/plosone/s/criteria-for-publication

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The study addresses a key issue however there are some modifications required.

Background

The authors provide some useful information about postpartum depression in the background. Considering that a very similar study had already been conducted in the catchment area however, I think the authors could have provided more information on PPD in Sri Lanka and Galle. In addition, information on the validated tools used in measuring PPD in Sri Lanka as stated in the previous study are also relevant to this study. They could have been stated and appropriately referenced in this section as well.

For example, the fact that the EPDs have been translated into Sinhalese and validated in Sri Lanka is important information that should be made known in the background paper. Additionally, PPD in Sri Lanka is reported to have decreased since 2011 and the background of this paper should have captured this information.

Justification of study: Lines 71-74

The study was justified on the grounds that the similar study conducted in the same area in 2017 showed a PPD lower than that of another study conducted in Sri Lanka in 2004. However, a lot could have happened between 2004 to 2017 when the new study was conducted. Various reasons could have contributed to the difference and a proper review of the programs and interventions put in place between 2004 to 2017 could have provided more information as to why this difference was observed. Additionally, the 2017 paper indicated that there had been a general reduction in PPD from 2011, so the reduction should not really come as a surprise anyway. In my view therefore, there is not a strong justification for this study and it does not really offer anything new as it stands.

Recommendation: A comparative analysis or trend analysis between the previous studies from 2011 to 2017 will give a better justification for the study and provide a clearer picture of PPD in Sri Lanka. Therefore, if the data is available, I recommend that the authors provide some more information on PPD between 2011 to 2017. The authors can also make a stronger justification for the study using the timing of their administration of their EPDs if it differs significantly from other studies in Sri Lanka

Methods

The collection of data using the EPD form is not really clear. Was the PPD information collected as part of the routine data collection or the team specifically went out to the field and collected this data using midwives?

This distinction should be made clearer in the write up. It is clear that the team made use of routine records for the other variables but as things stand now one gets the impression that the PPD information was collected in the same way. I am not sure that is the case however if the previous 2017 study is anything to go by.

Additionally, if this is not the case then, I am also wondering why informed consent (verbal at least) was not required? Was this study simply a secondary analysis of routine data?

How often is screening done for PPD at the facility? Was the EPD form only employed because of this current study? If the screening form and PPD data is readily available, then the team could also do a trend analysis or provide some more information on PPD in the catchment area prior to the study.

Lines 104-109: The study made use of the Tamil and Sinhala versions of the EPDS. Additionally, the study also talks about illiterate participants. Who are the illiterate participants? Are illiterate participants in this case participants who could not read Tamil and Sinhala?

Results

The results are well presented. Some of the confidence intervals are really wide though and the authors should take a second look at their analyses and the assumptions made for their models. Example New born death aOR 28.9 (4.5-185.1). The authors did not present any findings from their chi square analysis.

Discussion

Line 209: According to the authors they dealt with women who had just given birth. However, in earlier chapters, they spoke about women who had given birth 4 to 6 weeks ago. The authors should be consistent as time is important in measurement of PPD. The study in 2017 for example had two time points (10 days and 4 weeks).

Recommendation: In the 2017 study, it was recommended that further studies on the effect of time since delivery on PPD should be looked at. Since the form is administered routinely, the authors can look at the different time periods that the forms are administered to make a stronger case about PPD in Sri Lanka.

Conclusion

The conclusion here is much better than that the one in the abstract. The authors could look at revising the one in the abstract

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

6 Oct 2021

EDITOR

Additional Editor Comments:

Comment 1: The title of the current paper and that by Fan et al., 2020 are quite similar. Could you please come out clearly as to what differs between your paper and the earlier published one, more so when the study population seems to be the same

Response: Thank you for pointing this out. In the revised version we have chosen to highlight that this study uses the validated cut-off for screening for PPD. The title is changed to “Risk of postpartum depression in Sri Lanka: A population-based study using a validated screening tool”.

Comment 2: Lines 17 & 40: The prevalence estimates provided here were 1996 figures... could there not have been much more recent ones? Also add the estimates from South Asia - as you compare it in the conclusion

Response: Good point. We have changed the reference on prevalence estimates in line 16 and line 104 to a more recent one by Hahn-Holbrook (2018). We also added prevalence ranges of PPD in South Asia in the introduction in line 73. However, we removed the sentence “this is lower than other countries in South Asia” in the conclusion as we have put more focus to screening for PPD in Sri Lanka in the revised version.

Comment 3: Lines 56 - 58: I am not clear about this point... what has this to do with PPD? I read over the cited article, did not come across a relationship between poverty and quality of parenting... please clarify...

Response: We have deleted this sentence.

Comment 4: Lines 92 - 93: Why the assumption based on 1996 figures? More so, when there was a 2017 study in the same population?

Response: At the time of planning and estimating our sample size we did not have the 2017 figures. As we knew that around 1000 births take place each calendar year in Galle, we estimated that in our sample 1000 mothers would give a reasonable precision and power. We calculated samples size both using the 1996 figures and using lower hypothetical figures with not related references. Basically our considerations tested more but only referred to the published old number. As we have now changed the 1996 reference in the introduction to a more recent one, we also updated the sample size estimation reference to be 18% Hahn-Holbrook (2018).

Comment 5: Lines 94 - 95: Why all mothers????? We the readers are not privy to information on deliveries in the study community... Also, the results section (line 159) has 975 persons... you have not clearly stated how was that arrived at... were those all the mothers interviewed? any mothers not included? response rate?

Response: Thank you for pointing this out. Basically all women who attended clinic after delivery are asked to fill the EPDS questionnaire screening them for postpartum depression. It is part of the routine public services at Galle, and we do not know whether any of them refused to fill the form, as we have only the filled forms. Exclusion criteria was incomplete EPDS records. We adjusted the methods section according to you recommendation, and added a line in the first paragraph in results section in line 177 to make this clearer.

Comment 6: Line 98: Will suggest this section be structured as e.g., data collection methods, tools, and process.

Response: Thank you for this great suggestion. We have now made the following subsections under methods “study design and study setting”, “participants”, “study tool”, “data collection”, “outcome measures and exposure variables”, “data analysis” and “ethical issues”.

Comment 7: Lines 106 - 107: check this.... specificity????

Response: Thank you for notifying this mistake. It has now been corrected in line 117-118.

Comment 8: Lines 109 - 115: Is the information presented here related to data collection?

Response: No, the information presented was related to the current practice on screening for PPD in Sri Lanka. Thank you for notifying this. We have now divided the methods section into more subheadings to separate our data collection from the study tool according to your recommendation.

Comment 9: Line 114: Is the ICD used for classification or diagnosis?

Response: Yes, we have now changed the line 127 to be more precise.

Comment 10: Line 126: Label this as a sub-section e.g., variables considered.

Response: Thank you for this recommendation, we agree to add a subsection before this paragraph and have adjusted it accordingly.

Comment 11: Lines 159 - 162: Please make reference to table 2.

Response: Table 2 is now referred to in the text line 178.

Comment 12: Lines 193 - 195: Revise the title to be much more concise.

Response: We have shortened the title of table 2.

Comment 13: Line 201: Mention "EPDS Question 10" in the "Methods section" as part of the data collection tools.

Response: Thank you for this recommendation. We added this information under methods section in line 118.

Comment 14: Lines 214 - 216: Were these characteristics also in the 2017 study? If so, state that clearly... if not, then provide the appropriate citation.

Response: Thank you, this needed clarification. The factors in our study that showed the strongest association with PPD was former history of mental illness and death of the newborn, both not mentioned in the 2017 study. Hypertension was associated with PPD and not studied separately in the former study. High maternal age was associated with PPD in the former study, and this is mentioned in maternal age section in discussion.

Comment 15: Finally, please follow the reporting guidelines for Cross-sectional studies as provided in the link to guide your paper: https://www.equator-network.org/reporting-guidelines/strobe/. Also review the PLOS ONE guidelines on publications: https://journals.plos.org/plosone/s/criteria-for-publication

Response: Thank you for providing us with this information. We will adjust the paper according to the guidelines provided.

REVIEWER

Reviewer #1: The study addresses a key issue however there are some modifications required.

Background

Comment 16: The authors provide some useful information about postpartum depression in the background. Considering that a very similar study had already been conducted in the catchment area however, I think the authors could have provided more information on PPD in Sri Lanka and Galle. In addition, information on the validated tools used in measuring PPD in Sri Lanka as stated in the previous study are also relevant to this study. They could have been stated and appropriately referenced in this section as well.

For example, the fact that the EPDS have been translated into Sinhalese and validated in Sri Lanka is important information that should be made known in the background paper. Additionally, PPD in Sri Lanka is reported to have decreased since 2011 and the background of this paper should have captured this information.

Response: Thank you for these great suggestions. We agree that the findings from 2011 should have been included in the introduction and we have adjusted it accordingly. In the introduction we also added new information about use of cut-off scores when screening for PPD, as well as information the validated cut-off for screening in Sri Lanka and current practice.

Comment 17: Justification of study: Lines 71-74

The study was justified on the grounds that the similar study conducted in the same area in 2017 showed a PPD lower than that of another study conducted in Sri Lanka in 2004. However, a lot could have happened between 2004 to 2017 when the new study was conducted. Various reasons could have contributed to the difference and a proper review of the programs and interventions put in place between 2004 to 2017 could have provided more information as to why this difference was observed. Additionally, the 2017 paper indicated that there had been a general reduction in PPD from 2011, so the reduction should not really come as a surprise anyway. In my view therefore, there is not a strong justification for this study and it does not really offer anything new as it stands.

Response: We thank the reviewer for the very interesting comment. We would like to mention that the study from 2017 was published after we had collected our data in 2019 and the results were not available at the time when we made the research protocol. From your input we now realize that we should have been clearer on this issue, and not use the paper from 2017 as an argument for conducting a similar study over again as this study was not available at the time we gathered data. Nevertheless, we believe that our study provide new information about PPD in Sri Lanka as our study is the first study to use the validated cut-off 9 or more for screening for PPD in Sri Lanka. Previous studies on PPD in Sri Lanka have used the cut-off 10 or more for reasons we are not aware of. However, the validated cut-off for screening is 9 or more and we believe that our study provides more credibility to the findings on prevalence on PPD in Sri Lanka. Furthermore, our study focuses on individual risk factors including diseases during pregnancy as well as former history of mental illness, of which the 2017 study did not provide information. For future studies on postpartum and antenatal depression in Sri Lanka it is important to use the validated cut-off for screening for postnatal depression.

Comment 18: Recommendation: A comparative analysis or trend analysis between the previous studies from 2011 to 2017 will give a better justification for the study and provide a clearer picture of PPD in Sri Lanka. Therefore, if the data is available, I recommend that the authors provide some more information on PPD between 2011 to 2017. The authors can also make a stronger justification for the study using the timing of their administration of their EPDs if it differs significantly from other studies in Sri Lanka

Response: Thank you for good suggestions. In our study, the mothers are screened once at clinics for PPD between 4-6 weeks following delivery, not at two different timepoints. The 2017 study screened once for postpartum depression at 10 days following delivery in Dankotuwa district and once in Galle district 4 weeks following delivery. As two different times for screening was not completed within the same population, we believe that there are several sociodemographic confounding factors that could have contributed to these results. Furthermore, when testing after 10 days the findings of PPD prevalence may be confounded by the common “postpartum blues” which is limited to 2 weeks following delivery. We think by using 4 weeks instead of the first 10 days postpartum we give a better picture of the real depressive postpartum conditions that appear.

Methods

Comment 19: The collection of data using the EPD form is not really clear. Was the PPD information collected as part of the routine data collection or the team specifically went out to the field and collected this data using midwives?

This distinction should be made clearer in the write up. It is clear that the team made use of routine records for the other variables but as things stand now one gets the impression that the PPD information was collected in the same way. I am not sure that is the case however if the previous 2017 study is anything to go by

Response: Thank you for pointing this out. We agree and have adjusted the text to be clearer.

Comment 20: Additionally, if this is not the case then, I am also wondering why informed consent (verbal at least) was not required? Was this study simply a secondary analysis of routine data?

Response: The EPDS and pregnancy records are routinely administered in all mothers in Sri Lanka, meant for screening and potentially referral for further assessment and management. Therefore, as you have stated, the study is a secondary analysis of routine data. However, we understand that we were not clear on this and have adjusted the manuscript accordingly.

Comment 21: How often is screening done for PPD at the facility?

Response: The screening is completed once in Galle, Sri Lanka, and it is administered between 4-6 following delivery. This study digitalized the routine copies of the forms to analyze the data that is stored in the records of the clinic clients.

Comment 22: Was the EPD form only employed because of this current study? If the screening form and PPD data is readily available, then the team could also do a trend analysis or provide some more information on PPD in the catchment area prior to the study.

Response: Thank you for good suggestions. The EPDS is a part of a routine screening service. However, we are not in the possession of PPD data from years before, and accessing this data is not possible. Therefore, a trend analysis is unfortunately not possible to conduct.

Comment 23: Lines 104-109: The study made use of the Tamil and Sinhala versions of the EPDS. Additionally, the study also talks about illiterate participants. Who are the illiterate participants? Are illiterate participants in this case participants who could not read Tamil and Sinhala?

Response: Thank you for pointing this out. Agree, we have accordingly revised line 121-122 to emphasize who the illiterate participants are.

Results

Comment 24: The results are well presented. Some of the confidence intervals are really wide though and the authors should take a second look at their analyses and the assumptions made for their models. Example New born death aOR 28.9 (4.5-185.1). The authors did not present any findings from their chi square analysis.

Response: Thank you. We agree with this comment. As you know when the cases are few the confidence intervals of the estimates are wide, which is what we see with the mothers who lost their child. We prefer to give estimates and statistical significance shown as 95% confidence limits instead of p-values; however, if the editor insists, we can easily provide p-values in addition to the 95% CI.

Discussion

Comment 25: Line 209: According to the authors they dealt with women who had just given birth. However, in earlier chapters, they spoke about women who had given birth 4 to 6 weeks ago. The authors should be consistent as time is important in measurement of PPD. The study in 2017 for example had two time points (10 days and 4 weeks). Recommendation: In the 2017 study, it was recommended that further studies on the effect of time since delivery on PPD should be looked at. Since the form is administered routinely, the authors can look at the different time periods that the forms are administered to make a stronger case about PPD in Sri Lanka.

Response: We agree and we have adjusted the text to be clearer in line 230. Regarding effect of time since delivery, please see the response to comment 18 which explains why a study on different times for screening is not possible to conduct with the data we have available.

Conclusion

Comment 26: The conclusion here is much better than that the one in the abstract. The authors could look at revising the one in the abstract

Response: Thank you. We have adjusted the conclusion in the abstract accordingly.

Submitted filename: Response to reviewers.docx

Click here for additional data file.

22 Feb 2022

Submitted filename: PONE-D-21-20712_R1.pdf

Click here for additional data file.

22 Mar 2022

General comment

We have now corrected grammatical errors and sentence structure in numerous sections. Furthermore, we added a paragraph in the introduction about thoughts of self-harm, and moved the self-harm paragraph in the discussion to be after the paragraph about newborn death and former history of mental illness as risk factors for EPDS score 9 or more (risk of PPD). We also added a new reference to the reference list in line 441.

Additional Editor Comments

Comment 1: Though the title has been revised, the content of the paper seems to be at variance. Is the focus on "Risk of PPD", OR "Prevalence of PPD" or "Prevalence of risk of PPD"? That should come out clearly and consistently throughout the manuscript.

Response: Thank you for this comment. Our focus is the prevalence of mothers at risk of PPD (as measured by mothers with an EPDS score 9 or more). We have now edited the manuscript and title to be consistent in terminology.

ABSTRACT

Comment 2: Lines 40 & 41: Please revise the grammar and sentence structure to enhance clarity.

Response: Thank you for pointing this out, we have changed the sentence in line 22-24 to make it clearer.

Comment 3: Line 42: Should the sentence not be "The prevalence of PPD among mothers with EPDS score 9 and above was..."? To the best of my understanding, the prevalence is of PPD and not EPDS.

Response: Thank you for raising this issue. The EPDS is a screening tool to identify mothers at risk of PPD, but a final diagnosis is made through clinical interview with the mothers identified. Therefore, we believe that the sentence “mothers with an EPDS score 9 or more (risk of PPD)” is more precise than stating “mothers with PPD”; and we have adjusted the terminology throughout the manuscript to be consistent on this matter; in the risk factor sections we have changed the wording to “risk of EPDS score 9 or more”. As previous studies in Sri Lanka have used a higher cut-off for screening for PPD, the prevalence estimates cannot be compared correctly without stating which EPDS score is being used.

However, we can use the terminology “prevalence of PPD” consistently throughout the paper if Editor prefers, but we believe that this terminology is less precise.

Comment 4: Line 47: The "thoughts of self-harm" should be appropriately linked to PPD... Just mentioning it does not add any value to its presence...

Response: Thank you, this needed clarification. We have added a new paragraph in the introduction, to describe the relationship between serf-harm and PPD in line 86-92.

Comment 5: Line 48: can 9.4% be referred to as "a significant proportion"???

Response: We have adjusted the line 39.

Comment 6: Lines 50 & 51: Would it not have been simpler to just say as the data shows that "a higher cut-off produces reduced PPD prevalence estimates"? The current statement is quite ambiguous...

Response: We agree to this comment, and have adjusted the line 41-44.

INTRODUCTION

Comment 7: Line 64: "suggest" or "suggests"?

Response: We have corrected the line 56.

Comment 8: Lines 73 & 74: how is this related to PPD? that would be very helpful to contextualize the information on "self-harm".

Response: Thank you for pointing this out. We have added a paragraph in line 86-92 to make this more clear.

Comment 9: Line 94: "put focus" does not seem to be the right expression

Response: We agree to this point and have adjusted the sentence in line 80.

Comment 10: Line 98, 104 & 105: Revise the sentences grammatically and structurally to enhance its clarity

Response: The sentences has been adjusted in line 95 and line 102-103.

Comment 11: Lines 106: Is the focus of the study "Prevalence of mothers at risk of PPD" or "prevalence of mothers with PPD"????

Response: Thank you for addressing the inconsistent use of terminology. As stated in comment 3, the focus in our paper is mothers at risk of PPD defined as mothers with an EPDS score 9 or more and we have adjusted the manuscript and title to be consistent.

Comment 12: Line 113: Under point 1, which study is being "repeated"?

Response: We have changed the sentence in line 107-108.

Comment 13: Line 114: Postpartum women do not attend "antenatal clinics", so please revise

Response: We have changed the line 107-108.

PARTICIPANTS

Comment 14: Line 138: Would it not be best to add that this sample size (907) was close to the 975 that delivered over the period and as such they were all included in the study?

Response: Thank you for this comment. However, as the sample size estimation was completed before we had the actual number of births in 2019, we believe that our sentence describes this more clearly.

Comment 15: Line 140 & 141: Why not say "mothers with incomplete EPDS records were excluded from the study"?

Response: We have adjusted the sentence in line 133-134 accordingly.

STUDY TOOL

Comment 16: Line 152 & 153: The sentence seems to be contained in the next, so why not incorporate one into the other?

Response: We incorporated the sentences in line 141-144.

Comment 17: Line 156: Would be great if the influence of "thoughts of self-harm" on PPD could be clearly stated

Response: We added a paragraph in the introduction to describe the relationship between PPD and self-harm. We deleted the sentence in line 145.

DATA COLLECTION

Comment 18: Line 170: pregnant women in this study cannot have EPDS records as the information is collected after 4 weeks of delivery.

Response: Good point, we have changed the line 158 to be more precise.

DATA ANALYSIS

Comment 19: Lines 199 - 202: The non-use of the Pearson Chi-square test in data analysis was raised by Reviewer 1.... this was not reflected in the results section...

Response: As we have not used the results of the Pearson Chi-square in the tables, we have now deleted the sentence in line 186-188.

Comment 20: Lines 206 - 208: "Thoughts of self-harm" seem to be a very important feature of this study - could you clearly state the importance of this to PPD or its influence on the study outcomes clearly?

Response: We have inserted a paragraph in the introduction to describe the importance of self-harm thoughts in PPD and have adjusted the discussion in line 353-355 to be more precise.

RESULTS

Comment 21: Line 218: The study was not about "pregnant women", so please revise

Response: We have adjusted the line 205.

DISCUSSION

Comment 22: Line 276: which is the former publication??? any citations?

Response: Citation has been added in line 270.

Comment 23: Line 281: "Newborn death" is not a maternal characteristic

Response: We agree to this comment and have adjusted the line 272-273.

Comment 24: Line 286: What are you implying with the statement "our material"???

Response: We have changed the wording in line 276-278.

Comment 25: Line 293: The statement is contrary to that of the first sentence in the conclusion of the abstract...

Response: The conclusion has been changed in line 39-40.

Comment 26: Line 299: Check your grammar

Response: It has been corrected in line 292.

Comment 27: Line 360: the available "study population" was used... why not state that instead of "adequate sample size"?

Response: We have adjusted the sentence in line 372.

Comment 28: Line 366: is it "prevalence of PPD" or "risk of PPD" or "prevalence of risk of PPD"???? Please be consistent all through the manuscript...

Response: Thank you for pointing this out. We have adjusted the manuscript to make this clearer in numerous sections.

CONCLUSION

Comment 29: Line 381: Should the initial part of the sentence not be "the prevalence of PPD" or "risk of PPD at an EPDS score of 9 or more"???? Thought it was the "prevalence of PPD being measured with the EPDS" and "not the prevalence of EPDS".

Response: We have adjusted the sentence in line 394-396 to be more precise.

Comment 30: Line 384: In place of "had a substantial impact", why not state clearly that "there was a drop in prevalence rates with higher cut-offs for EPDS"??? That would be much more to the point

Response: We have rephrased the sentence in line 397-399 accordingly.

Submitted filename: Response to reviewers.docx

Click here for additional data file.

9 May 2022

Mothers at risk of postpartum depression in Sri Lanka: A population-based study using a validated screening tool

PONE-D-21-20712R2

Dear Dr. Solås,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Yeetey Akpe Kwesi Enuameh, MD, MSc, DrPH

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

The authors have addressed the issues raised. Just one final suggestion, could you replace "crosstabulation" on line 26 of the abstract with "univariate" and change "regression" to plural?

Congratulations and thanks for your patience.

Reviewers' comments: